Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The coordination chemistry of ruthenium porphyrin complexes Sishta, Chand 1990

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

Item Metadata

Download

Media
831-UBC_1990_A1 S58.pdf [ 11.69MB ]
Metadata
JSON: 831-1.0059668.json
JSON-LD: 831-1.0059668-ld.json
RDF/XML (Pretty): 831-1.0059668-rdf.xml
RDF/JSON: 831-1.0059668-rdf.json
Turtle: 831-1.0059668-turtle.txt
N-Triples: 831-1.0059668-rdf-ntriples.txt
Original Record: 831-1.0059668-source.json
Full Text
831-1.0059668-fulltext.txt
Citation
831-1.0059668.ris

Full Text

THE COORDINATION CHEMISTRY OF RUTHENIUM PORPHYRIN COMPLEXES by CHAND SISHTA B.Sc, The University of New Brunswick, 1984 M.Sc, The University of British Columbia, 1986 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES THE DEPARTMENT OF CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA February, 1990 © Chand Sishta, 1990 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Chemistry The University of British Columbia Vancouver, Canada Date March 14, 1990  DE-6 (2/88) Abstract This thesis work reports developments in the coordination chemistry of ruthenium porphyrin complexes, both in terms of the synthesis and chemistry of new compounds, as well as the study of the solution chemistry of some previously reported complexes. The synthesis, characterization and chemistry of ten new Ru(porp) coordination complexes in the oxidation states Ru111 and Rulv containing halide (Br, CI) and other axial ligands (pyridine, CH3CN, NH3 and SbF6) are described in this thesis. Some additional ten Ru(porp) complexes have been studied in situ. Measurement of the rate constants for forward and reverse reactions and the corresponding equilibrium constant by 'H NMR and UV/visible spectroscopy for the dissociation of PPh3 ligand from Ru(OEP)L(PPh3) (OEP is the octaethylporphyrinato dianion; L = CO, PPh3) in C7DS to generate the previously reported five-coordinate Ru(OEP)L complexes allowed for an estimation of the Ru-P bond strength (64 _+ 9 kJ mol"1) in these complexes. A study of PPh3 dissociation from Ru(OEP)CO(PPh3) in C-jDg and in CDC13 indicates that solvation effects play a major role, with CDC13 being more capable than C 7D 8 of solvating the Ru(OEP)CO complex. The presence of trace H20 in these systems was a major problem, and the coordination of H20 to Ru(OEP)L complexes to generate the in situ Ru(OEP)L(H20) complexes (L = CO, PPh3) is described. The formation of Ru(OEP)L(H20) and the observed difference in the solvation of Ru(OEP)CO by C 7H 8 and CHC13 indicate that truly Five-coordinate species may not exist in solution. The outer-sphere oxidation of Ru(OEP)PPh3 by 0 2 to give [Rurv(OEP)OH]20 was shown to occur only in the presence of H20. Mechanistic studies on the previously reported reaction of HCI with [Ru(OEP)]2 to generate Ru^OEPJCl, (C. Sishta, M.Sc.Thesis, University of British Columbia, 1986) show that solvent plays a major role in directing this oxidation reaction. A reaction stoichiometry of 4:1 between HCI and [Ru(OEP)]2 in C 6D 6 or C7Dg showed that HCI itself was the oxidant and not trace Cl 2 in HCI, as thought previously. A range of HX acids having plC, values in the range 38 to less ii than -10 (HX = H 2 , MeOH, H,0, H2S, CH3COOH, C 6H 5COOH, HF, CF3COOH, HN0 3, HBF4, HCI. HBr, and HSbFg) were tested for reactivity with [Ru(OEP)]2 in C 6 D 6 ; the data showed that a strong acid (pK^ < ca. 0) was necessary to initiate reactivity. The complex RuIV(OEP)(SbF6)2 was generated in situ by reacting HSbF6 with [Ru(OEP)]2. In CH2C12, a 1:1 stoichiometric reaction between HCI and [Ru(OEP)]2 was observed, instantly fanning a mixture of products, tentatively formulated as Rura(OEP)H and [Ru(OEP)]2CHCl2 based on spectroscopic data. The species proved impossible to separate. These same products were formed slowly by the reaction of [Ru(OEP)]2 with CH2C12 in the absence of HCI, and kinetic studies suggest that a direct reaction of [Ru(OEP)]2 with CH2C12 is likely, rather than reaction of [Ru(OEP)]2 with impurities in CH2C12. The product mixture generated Ru(OEP)Cl2 upon further reaction with HCI, both in the absence and in the presence of air. The complex RuIV(OEP)(BF4)-> was generated in situ by an analogous reaction of aqueous HBF4 with the product mixture. The required hydrogen-containing co-product from the reaction of HX (X = Br, CI) with [Ru(OEP)|2 in C 7 D g or CH2C12 was not detected, but was shown not to be H 2 . Oxidation of Ru(porp)(CH3CN)2 and Ru(OEP)py2 (py = pyridine; porp = OEP, TMP (the dianion of tetramesitylporphyrin)) by gaseous HX (X = Br, CI) in the absence of air yielded RuIV(porp)X2 complexes. The new compound Ru(TMP)Br2 was synthesized by this method using the bis(acetonitrile) precursor, and was characterized by spectroscopy; the chloride analogue Ru(TMP)Cl2 was generated in situ. The magnetic properties (susceptibility and moment) of Ru(OEP)Br2 from 6 to 300 K are unlike those reported for ruthenium(IV) non-porphyrin complexes, and reveal a significant contribution from temperature-independent paramagnetism. The reaction of Ru(OEP)X, (X = Br, CI) with NH 3 gave the complexes Ruin(OEP)X(NH3), which upon acidification under an inert atmosphere yielded the Rum(OEP)X compounds. These Ru111 complexes were characterized by spectroscopic techniques, and the solution chemistry of the five-coordinate species Ru(OEP)X was developed: the Rum(OEP)X(CH3CN) species were also iii characterized. Solvation of the five-coordinate species Ru(OEP)X (X = Br, CI) was observed in coordinating solvents to form the six-coordinate species Ru(OEP)X(solvent) (solvent = py, CH3CN and MeOH). Estimates of the equilibrium constants for the association of these ligands to Ru(OEP)X were obtained from UV/visible titration experiments in CH2C12. Similarly, the equilibrium constant for the association of Br to Ru(OEP)Br to generate in situ (rc-Bu)4N + [Rura(OEP)Br2]", was measured. Disappointingly, the complexes Ru(OEP)X were shown not to catalyze the oxidation of organic substrates such as cyclohexene. Electrochemical and spectroelectrochemical studies of the complexes Ru(OEP)X2 and Ru(OEP)X (X = Br, CI) showed that the Ru^/Ru111 couple occurred at 480-460 mV and 950-870 mV vs. NHE, respectively, and that the probable reductant for the reaction of Ru(OEP)X2 with NH 3 was NH3 itself. A facile reduction of Ru(OEP)(SbF6)2 gave the complex RuIU(OEP)SbF6, by a probable homolysis of the Ru-F bond. The outer-sphere oxidation of Ru(OEP)py2 by air in the presence of HX acids gave the isolated or in situ characterized complexes [Ruin(OEP)py2]+ X" (X = CI, Br, F , B F 4 ) . Similar oxidation of Ru(OEP)(CH3CN)2 formed [Ru(OEP)(CH3CN)2]+ Br-. Electrochenucal studies showed that 0 2 in acidic media was capable of oxidizing the Ru(OEP)(solvent)2 complexes (solvent = py, CH3CN) to the RuU I complexes, presumably generating H0 2 . iv Table of Contents Pace Abstract. ii Contents. v List of Tables. ix List of Figures. xi List of Schemes. xviii Abbreviations. xix Numerical key to Ru(porp) complexes discussed in this thesis. xxiv Acknowledgements. xxvi Chapter 1. Introduction. 1 1.1. Coordination chemistry at ruthenium porphyrin centers. 1 1.1.a. Ruthenium(II) carbonyl complexes. 1 1. l.b. Dioxygen and dinitrogen ruthenium compounds from Run(porp)(solvent)2 complexes. 3 1.1 .c. Ruthenium(II) phosphine complexes. 5 1.1 .d. Other ruthenium(II) complexes. 8 1.1 .e. Ruthenium(III) complexes. 8 1.1. f. Ruthenium(IV) complexes. 10 1.1. g. Ruthenium(VI) complexes. 11 1.2. Organoruthenium porphyrin complexes. 12 1.2. a. Compounds originating from [Ru(OEP)]2. 12 1.2.b. Compounds originating from 2 K + [Ru(porp)]~2. 12 1.2.c. Compounds originating from Ru(porp)X2. 13 1.3. Objectives of this dissertation. 13 Chapter 2. Experimental. 16 v 2.1. Reagents and solvents. 16 2.2. Instrumentation and techniques for synthesis. 17 2.3. Preparation of ruthenium complexes. 23 2.3.a. Spectroscopic parameters for compounds prepared from literature methods. 23 2.3.b. Synthesis and characterization of new Ru(porp) complexes. 27 2.4. General and specific reaction conditions used to study the solution chemistry of Ru(OEP) complexes 35 Chapter 3. The reactivity of the five-coordinate complexes carbonyl-(octaethylporphyrinato)ruthenium(II), (3a), and (octaethylporphyrinato)(triphenylphosphine)ruthenium(II), (5a). 38 3.1. Introduction. 38 3.2. Results and discussion. 39 3.2.a. Reaction of Ru(OEP)PPh3 (5a) with small molecules. 39 3.2.b. The Reaction of PPh3 with Ru(OEP)CO (3a) in C 7 H 8 . 45 3.2.c. The Reaction of PPh3 with Ru(OEP)CO (3a) in CHC13. 59 3.2.d. The Reaction of Ru(OEP)PPh3" (5a) and PPh3 in C 7 H 8 . 67 3.2.e. Discussion of die rate (k„ and k.n) and equilibrium (KJ constants for the reaction of PPh3 widi Ru(OEP)L (L = CO and PPh3). 75 3.2.f. Summary and discussion of activation and thermodynamic data for the reversible dissociation of PPh3 from Ru(OEP)L(PPh3) (L= CO, PPh3). 79 3.2.g. The decarbonylation of aldehydes using Ru(OEP)PPh3 (5a). 84 Chapter 4. Further studies on the oxidation of bis((octaethlyporphyriiiato)-vi ruthenium(II)) (7) by HX acids. 88 4.1. Introduction. 88 4.2. Results and discussion. 89 4.2.a. Oxidation of [Ru(OEP)]2 (7) by the mixture H20/02. 89 4.2.b. Measurement of the stoichiometry for the reaction between [Ru(OEP)]2 (7) and HCI. 92 4.2.c. The attempted isolation and purification of the intermediates obtained from the reaction of [Ru(OEP)]2 with CH2C12 and HC1/CH2C12. 96 4.2.d. Further characterization of the mixture 13/14. 100 4.2.e. Measurement of the kinetics of the reaction of CH2C12 with [Ru(OEP)]2. 110 4.2.f. Chemical behavior of the in situ mixture. 127 4.2.g. The reaction of [Ru(OEP)]2 (7) with chlorinated reagents. 133 4.2.h. The fate of H in HX upon oxidation of 7 by HX in C 7 H 8 andC6H6. 137 4.2.i. The effect of the acid HX used. 141 4.2.j. Generation of Ru(porp)X2 (8) from Ru(II) sources. 143 Chapter 5. Further characterization of Ru(OEP)X2 complexes and synthesis, characterization and chemistry of several new Rura(OEP) complexes. 149 5.1. Introduction. 149 5.2. Results and discussion. 150 5.2.a. Electrochemical studies of Ru(OEP)X2. 150 5.2.b. Further studies on the magnetic properties of Ru(OEP)Br2. 154 vii 5.2.c. Attempted generation of other Ru (OEP)X'2 complexes by metathesis reactions of Ru(OEP)X2 (X = Br, 8a; CI, 8b) with AgX' salts (X' = BF 4, 8c; SbF6, 8d, and PF6). 164 5.2.d. Synthesis, characterization and chemistry of Ru(OEP)X(L) (X = Br, L = vacant, 16a; X = Br, L = NH 3, 16c; X = Br, L = CH 3CN, 16e; X = CI, L = vacant, 16b; X = CI, L = NH 3, 16d; X = CI, L = CH 3CN, 16f). 165 5.2.d.l. Synthesis of 16a-b. 165 5.2. d. 2. Characterization of 16a-f. 168 5.2.d.3. Chemistry of Ru(OEP)X (X = Br,16a; CI, 16b). 183 5.2.e. Synthesis, characterization and chemistry of [Runi(OEP)(py)2]+ X- (X = CI, Ha; X = BF 4, 17b). 200 5.2.e.l. Synthesis of 17a-b. 200 5.2.e.2. Spectroscopic characterization of 17a-b. 201 5.2.e.3. Chemistry of [Ru(OEP)(py)2]+ X", (17a-b). 211 5.2.f. Summary of electrochemical data for Ru(OEP) complexes. 213 Chapter 6. General conclusions and recommendations for future studies. 219 Chapter 7. References. 222 Appendix 1. Information needed for the use of the NMR simulation program DNMR3. 231 Appendix 2. Data obtained from the titration of Ru(OEP)CO by PPh3 in C 7 D 8 as a function of temperature. 234 Appendix 3. Data obtained from the titration of Ru(OEP)CO by PPh3 in CDC13 as a function of temperature. 236 Appendix 4. Data obtained from the titration of Ru(OEP)PPh3 by PPh3 in C 7 D 8 as a function of temperature. 238 viii Appendix 5. Kinetic data for the reaction of [Ru(OEP)]2 with CH2C12 at 293 K. 240 ix List of Tables Table Page 3.1 Spectroscopic data for Ru(porp) complexes containing carbonyl and phosphine axial ligands. 41 3.2 Data for the simulation of *H NMR spectra for the chemical exchange system Ru(OEP)CO(PPh 3) ^ Ru(OEP)CO + PPh 3 > in C 7D 8 as a function of temperature. 52 3.3 Data for the simulation of 'H NMR spectra for the chemical exchange system Ru(OEP)CO(PPh 3)-Ru(OEP)CO + PPh 3 in CDC1 3 as a function of temperature. 63 3.4 Equilibrium constants ( K 2 ' ) measured for Ru(OEP)CO(PPh 3) ^ Ru(OEP)CO + PPh 3 as a function of temperature in CHC1 3. 68 3.5 Data for the simulation of 'H NMR spectra for the chemical exchange system Ru(OEP)(PPh 3) 2 ^  Ru(OEP)PPh 3 + PPh 3 in C 7D 8 as a function of temperature. 70 3.6 Equilibrium constants ( K 3 ' ) measured for Ru(OEP)(PPh 3) 2 H Ru(OEP)PPh 3 + PPh 3 as a function of temperature in C 7H g. 74 3.7 Rate constants (k o b s) for the dissociation of a ligand L' from Ru(porp)L(L') as in Ru(porp)L(L') -• Ru(porp)L + L' in C 7H 8. 76 3.8 Equilibrium (Kg ) data at 293 K for the dissociation of a ligand L' from Ru(porp)L(L') as in Ru(rx)ip)UL')«^Ru(porp)L + L' 78 3.9 Summary of the activation and thermodynamic data for the equilibria Ru(OEP)L(PPh 3) Ru(OEP)L + PPh 3 (L = CO, PPh 3) in C 7 H 8 and CHC1 3. 80 4.1 Summary of 'H NMR spectroscopic data for compounds relevant to the characterization of the products from the reaction of [Ru(OEP)] 2 with CH 2C1 2. 91 4.2 XPS binding energies for several Ru(OEP) complexes. 108 4.3 Raw data for the change in the position (6) of the resonances of [Ru(OEP)] 2 (2) with time upon reaction of 7 (4.9 mM) with CD 2C1 2 as measured from the *H NMR spectrum of the reaction mixture in CD 2C1 2. 111 4.4 Summary of data for the reaction of [Ru(OEP)] 2 with CH 2C1 2 at 293 K. 120 4.5 Summary of spectroscopic data for Ru I V(porp) complexes. 132 4.6 'H NMR spectroscopic data for [Ru(OEP)] 2 and products generated from the reaction of [Ru(OEP)] 2 with various solvents. 135 4.7 Data for the Curie plot for Ru(TMP)Br, (8e). 147 5.1 Corrected magnetic susceptibility and moment data for Ru(OEP)Br 2. 158 5.2 Microanalysis and conductivity data for several Ru m(OEP) complexes. 169 5.3 Summary of UV/visible spectroscopic and magnetic moment data for Ru m(OEP) complexes. 169 5.4 Summary of *H NMR spectroscopic properties of Ru m(OEP) complexes. 174 5.5 Data for the Curie plot of Ru(OEP)Br, (16a). 176 5.6 Data for the Curie plot for Ru(OEP)Br(NH 3), Q6c). 187 5.7 Summary of 'H NMR spectroscopic data for Ru 1 1 1 bis(solvent) complexes. 205 5.8 Data for the Curie plot for [Ru(OEP)(py) 2]Cl (17a). 207 5.9 Summary of electrochemical data for Ru(OEP) complexes. 214 xi List of Figures Figure 1.1. Structural features of (a) Ru(OEP)L2, and (b) RutTMP)!^ complexes, 2.1 UV/visible spectroscopy sample cells incorporating a Teflon-valve to seal the sample solution under an inert atmosphere or under vacuum. Cell A has no reservoir, while cells B and C have plain and calibrated reservoirs, respectively. 2.2 NMR sample tubes for use in the study of air-stable (A) or air-sensitive (B) samples. The side-arm in B allows the introduction of reagents to the sample without causing decomposition of the latter. Tube (C) incorporates a septum-port screw-cap fitting for use in experiments involving titration of samples in NMR tubes. 2.3 Exploded view of the apparatus used for the detection of H 2 by direct sampling of the reaction vapor phase via a gas sampler of a mass spectrometer. 3.1 (a) UV/visible spectral changes observed upon addition of MeOH (2.47 M) to Ru(OEP)CO (2.24 x 10"6 M) in C 7 H 8 at 293 K. (b) Plot of log [(A-A0)/(Aj-A)] vs. log [MeOH] rfor the above titration, measured at 393 nm. 3.2 (a) lU NMR spectra of a mixture of Ru(OEP)CO (5.1 mM) and PPh 3 (3.3 mM) in C 7 D 8 measured at 10 K intervals, and (b) spectra simulated using the parameters listed in Table 3.2 and Appendix 1. 3.3 Ey ring plot of In (k, IT) vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in C 7D 8. 3.4 Plot of In K[ vs. 1 IT (K"1) for the dissociation of PPh 3 from xii Ru(OEP)CO(PPh 3) in C 7D 8. 55 3.5 UV/visible spectrum of (a) Ru(OEP)CO, 3a and (b) Ru(OEP)(CO) 2 in C 7H 8. 58 3.6 (a) UV/visible spectral changes observed upon addition of PPh 3 (0.0831 M) to Ru(OEP)CO (3.87 x 10"6 M) in C 7 H 8 at 293 K. (b) Plot of log [(A-AD)/(Aj-A)] vs. log [PPh 3] f for the above titration, measured at 393 nm. 60 3.7 (a) 'H NMR spectra of a mixture of Ru(OEP)CO (8.77 mM) and PPh 3 (4.52 mM) in CDC1 3 measured at 10 K intervals, and (b) spectra simulated using the parameters listed in Table 3.3. 62 3.8 Eyring plot of ln (k2/T) vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in CDC1 3. 64 3.9 Plot of ln K 2 vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in CHC1 3. 65 3.10 (a) UV/visible spectral changes observed upon addition of PPh 3 (0.676 M) to Ru(OEP)CO (6.75 x 10"6 M) in CHC1 3 at 293 K. (b) Plot of log [(A-A 0)/(Aj-A)] vs. log [PPh 3] ffor the above titration, measured at 393 nm. 66 3.11 Plot of ln K 2' vs. 1/T (K _ 1) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3)inCHCl 3. 68 3.12 (a) 'H NMR spectra of a mixture of Ru(OEP)(PPh 3) 2 (3.88 mM) and PPh 3 (3.18 mM) in C 7 D 8 measured at 10 K intervals and, (b) spectra simulated using the parameters listed in Table 3.5. 69 3.13 Eyring plot of ln (k3/T) vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)(PPh 3) 2 in C 7D 8. 71 3.14 Plot of ln K 3 vs. 1/T (K _ 1) for the dissociation of PPh 3 from Ru(OEP)(PPh 3) 2 in C 7H 8. 72 xiii 15 (a) UV/visible spectral changes observed upon addition of PPh 3 (0.0422 M) to Ru(OEP)(PPh 3) 2 (2.78 x 10"6 M) in C 7 H 8 at 293 K. (b) Plot of log [(A-A 0)/(A ;-A)] vs. log [PPh 3] f for the above titration, measured at 395 nm. 16 Plot of In K 3' vs. 1/T (IC 1) for the dissociation of PPh 3 from Ru(OEP)(PPh 3) 2 in C 7H 8. 1 'H NMR spectra of [Ru(OEP)] 2 in C 6D 6 containing added (a) 0.1 H 20, and (b) 500 A»L 0 2. Spectra were measured at 293 K at 400 MHz (FT). 2 'H NMR spectrum of the products of the reaction of [Ru(OEP)] 2 with excess HCI in CD 2C1 2 vacuo, at 292K, 300 MHz. 3 'H NMR spectra of (a) [Ru(OEP)] 2 + BF 4" in CD 2C1 2 and, (b) Ru(OEP)Cl(NH 3) in C 6D 6. The spectra were measured at 292 K at 300 MHz (FT) under an N 2 atm. 4 !H NMR spectra of the product eluted from the attempted purification of 13/14 using chromatography. Spectrum was measured at 293 K, 300 MHz. 5 Fast atom bombardment (FAB) mass spectra of (a) 13/14 and, (b) the deuterated analogue of 13/14. 6 IR spectrum of the mixture 13/14 as a Nujol mull with KBr plates. 7 X-ray photoelectron spectra of (a) Ru(OEP)py 2, (b) Ru(OEP)CH 3, (c) Ru(OEP)Cl, (d) Ru(OEP)Br, (e) Ru(OEP)Cl 2 and (f) 13/14. 8 First order plot of ln(Pt-Pf) for the resonance C H 2 a vs. time (min) for the reaction of [Ru(OEP)] 2 with CD 2C1 2. 9 Plot of the observed chemical shift (ppm) of the individual porphyrin signals of 7 vs. equivalents of added HCI ([Ru] = 3.1 mM, [HCI] = 91 mM). 10 *H NMR spectra measured as a function of temperature for the product mixture obtained from the incomplete reaction of [Ru(OEP)] 2 with CD 2C1 2 xiv in CD 2C1 2 (16 mg/mL). The spectra were measured at 300 MHz in vacuo. 115 4.11 (a) Typical changes from the initial UV/visible spectrum of [Ru(OEP)] 2 (7) to that of the mixture 13/14 with time at 293 K. [Ru 2] = 0.39 mM, [CH 2C1 2] = 15.6 M. (b) Plot of ln(A t-A f) vs. time for the above reaction. 117 4.12 Dependence of k o b s on the concentration of [Ru(OEP)] 2 at high [CH 2C1 2]. 121 4.13 Dependence of k o b s on [CH 2C1 2] at 293 K, [Ru 2] = 0.6 +. 0.1 mM. 124 4.14 'H NMR spectrum of the in situ reaction mixture of 13/14 (3 mg) and pyridine (1 M L ) in CD 2C1 2 (0.6 mL) sealed under vacuum, measured at 293 K, 300 MHz. 129 4.15 The 'H NMR spectrum of in situ Ru(OEP)(BF 4) 2 (8c) in CD 2C1 2 at 293 K, 300 MHz. 131 4.16 1 9 F NMR spectra of (a) HBF 4 ( a q ) and, (b) Ru(OEP)(BF 4) 2. The spectra were measured at 293 K, at 282 MHz, and are referenced against («-Bu)4N + BF 4" (0.00 ppm). 134 4.17 Gas chromatograms of (a) the vapor phase sample from 7/HX reaction mixture, and (b) the corresponding H 2 control experiment. 138 4.18 MS Spectra obtained from the GC/MS of (a) the solvent from the reaction of 7 with HBr in C 7H 8, (b) the C 7 H 8 control and, (c) the hydrogenated/ brominated C 7 H g obtained upon reaction of C 7H 8/HBr with the silicon septum. 139 4.19 'H NMR Spectrum of Ru I V(OEP)(SbF 6) 2 formed in situ by die addition of HSbF 6to 7 in C 6D 6 under N 2. Obtained at 293 K, 300 MHz. 142 4.20 'H NMR Spectra of (a) Ru(TMP)(CH 3CN) 2 (6d) in C 7D 8 and, (b) Ru(TMP)Br, (8e) in C 6D 6. Obtained at 293 K, 300 MHz (FT) in vacuo. 145 4.21 Curie plot of the isotropic shift (ppm) vs. 1/T (K"1) for Ru(TMP)Br 2. 148 5.1 Cyclic voltammograms of (a) Ru(OEP)Br 2 and (b) Ru(OEP)Cl 2. Data measured xv in 0.1 M (n-Bu) 4N + C104" (TBAP) in CH 2C1 2 under an N 2 atmosphere, 293 K. 151 5.2 Bulk coulometry of Ru(OEP)Br 2 (8a). Trace A shows the optical spectrum of 8a before coulometry at -235 mV vs. Fc +/Fc in 0.1 M TBAP in CH 2C1 2. Trace B shows the generated optical spectrum, which corresponds to diat of Ru(OEP)Br given in Section 5.2.d. 153 5.3 (a) Changes observed in the optical spectrum of Ru(OEP)Br 2 (8a) upon bulk coulometric oxidation of 8a at 900 mV vs. Fc +/Fc in 0.1 M TBAP in CH 2C1 2. (b) ESR spectrum of the solution generated in part (a). 155 5.4 Plot of / i e f f (corrected) and xm (corrected) vs. T for Ru I V(OEP)Br 2. 160 5.5 Plot of xm (corrected) vs. T for several Ru(IV) complexes including K 2RuF 6, K 2RuCl 6, (NH 4) 2RuCl 6, Rt^RuClg, K 2RuBr 6 and 8a. 162 5.6 Plot of T 1 / 2 vs. Mefr(corrected) for Ru(OEP)Br 2. 163 5.7 '-H NMR spectra of (a) Ru(OEP)Br 2 (5.32'mM) in CDC1 3, (b) the reaction mixture 5 minutes after addition of NH 3 (five mole equivalents), (c) after 215 minutes, and (d) after 1170 minutes. The experiment was conducted aerobically, at 293 K and die arrows indicate the fate of each signal as the reaction progresses. 167 5.8 FAB-MS of (a) Ru(OEP)Br (16a) and, (b) Ru(OEP)Cl (16b). 170 5.9 UV/visible spectra of (a) Ru(OEP)Br (26.4 / i M ) , (b) Ru(OEP)Br(NH 3) (28.8 /xM), and (c) Ru(OEP)Br(CH 3CN) (29.2 / i M ) in C 7H 8 > cell pathlength = 1.00cm. 171 5.10 300 MHz, 'H NMR spectra of Ru(OEP)Br (16a) in C 6D 6 at 293 K. The spectrum of Ru(OEP)Cl (16b) is very similar to that of the bromide analogue. 173 5.11 Curie plot for Ru(OEP)Br (16a). Solid lines represent die least-square fit of the data, with extrapolation to 1/T = 0. 177 xvi 5.12 (a) Cyclic voltammogram of Ru(OEP)Br (16a) and, (b) 16a and added excess (n-Bu)4N + Br". X-axis scale is in Volts vs. the Fc +/Fc couple. 5.13 Changes observed in the optical spectrum of 16a upon bulk coulometric oxidation of the solution at (a) +400 mV vs. Fc +/Fc both in the absence and presence (25 equivalents) of added (n-Bu) 4N +Br", (b) coulometric oxidation at 900 mV vs. Fc +/Fc in the absence of added (n-Bu) 4N +Br". The experiments were conducted in CH 2C1 2 with 0.1 M TBAP electrolyte. 5.14 Infrared spectrum of Ru(OEP)Br(NH 3) in the 4000-2500 cm"1 region, as a KBr pellet. The IR spectrum of Ru(OEP)Cl(NH 3) is identical to that of the bromide analogue. 5.15 Curie plot for Ru(OEP)Br(NH 3), J6c. Solid lines represent the least-squares fit of the data, with extrapolation to 1/T = 0. 5.16 Infrared spectrum of Ru(OEP)Br(CH 3CN) as a Nujol mull. The IR spectrum of Ru(OEP)Cl(CH 3CN) is identical, except that the ^ C 3 N is observed at 2250 cm"1. 5.17 300 MHz, 'H NMR spectrum of [Ru n i(OEP)(py) 2] + Br", isolated from the reaction of Ru(OEP)Br with pyridine (see Figure 5.23b). 5.18 (a) UV/visible spectral changes observed upon addition of («-Bu)4N+ Br" (68 mM) to Ru(OEP)Br (2.95 x 10"6 M) in CHC1 3 at 293 K. (b) Plot of log [(A-A0)/(Aj-A)] vs. log [Br"] f for the above titration, measured at 397 nm. 5.19 JH NMR spectrum of N H 4 + [Ru m(OEP)Br 2]" generated in situ by the association of Br" to Ru(OEP)Br in CDC1 3 at 293 K. 5.20 UV/visible spectrum of the reaction mixture of Ru(OEP)Br (16a) with excess mCPBA in CH 3CN. 5.21 Gas chromatograms of (a) the reaction mixture from the reaction of Ru(OEP)Br with excess PhIO in the presence of excess cyclohexene and, (b) the corresponding cyclohexene blank. 5 . 2 2 (a) Optical spectra of [Ru(OEP)(py) 2] + CI" Q7a, [Ru] = 46.5 /iM) measured in C H C I 3 with a 1.0 cm pathlength cell, (b) FAB-MS of 17a. The corresponding UV/visible and FAB-MS spectra of [Ru(OEP)(py) 2] + BF 4~ (17b) were identical in pattern but not in intensity to those of the chloride analogue. 5 . 2 3 1H NMR spectra of (a) Ru n(OEP)(py) 2 and (b) [Ru m(OEP)(py) 2] + BF4". Spectra were recorded in CDC1 3 under aerobic conditions at 2 9 3 K, 3 0 0 MHz (FT). 5 . 2 4 Curie plot for [Ru(OEP)(py) 2] + CI", J7a. Solid lines represent the least-squares fit of the data with extrapolation to 1/T = 0. 5. 2 5 CV for (a) Ru"(OEP)(py) 2, (b) 17a and (c) 17b. Experiments were conducted in 0.1 M TBAP in CH 3CN, die X-axis scale is in mV vs. Fc +/Fc. 5 . 2 6 Bulk coulometry of [Ru(OEP)(py) 2] + BF 4" in 0.1 M TBAP in CH 3CN to generate (a) Ru n(OEP)(py) 2 and, (b) [ R u n , ( O E P + ) ( p y ) 2 ] + 2 . 5 . 2 7 Cyclic voltammogram of Ru(OEP)Br(PPh 3) in 0.1 M TBAP in CH 2C1 2. X-axis scale is in mV vs. the Fc +/Fc couple. xviii List of Schemes Scheme Page 2.1 Synthetic strategy used to prepare key Ru(OEP) complexes. 24 3.1 Reactivity of Ru(OEP)PPh 3. 40 3.2 Proposed mechanistic scheme for the reaction of Ru(OEP)PPh 3 with 0 2/H 20. 43 3.3 Alternate mechanistic scheme for the reaction of Ru(OEP)PPh 3 with 0 2/H 20. 43 4.1 Proposed mechanistic scheme for the reaction of HCI with [Ru(OEP)] 2. 97 5.1 Summary of the reactivity of Ru(OEP)Br (16a). 184 xix Abbreviations A absorbance AIBN 2,2'-azobisisobutyronitrile br broad, used in UV/vis or NMR spectroscopy CV cyclic voltammetry D or d deuterium d doublet, used in N M R spectroscopy 6 chemical shift relative to T M S A heat DMA N,N-dimethylacetamide DMF N,N-dimethylforrnamide e electron EI electron impact ionization technique in mass spectroscopy ESR electron spin resonance FAB fast atom bombardment ionization technique in mass spectroscopy Fc ferrocene, dicyclopentadieneiron(II) G Gauss GC gas chromatograph GC-MS gas chromatograph-mass spectrometer ° H° standard enthalpy of reaction A H ) enthalpy of activation H m meto-proton of an aryl or pyridyl ring H m e s o meso-proton of the O E P macrocycle H Q ORTHO-PROTON of an aryl or pyridyl ring H p para-proton of an aryl or pyridyl ring H ,rr pyrrole-proton of the T P P or T M P macrocycle xx h hour hf irradiation with UV/visible light IR infrared K degree Kelvin Kp equilibrium constant kinetic rate constant k o b s observed kinetic rate constant L ligand M metal, or molarity unit m multiplet, used in NMR spectroscopy mCPBA meto-chloroperbenzoic acid mJe mass-to-charge ratio NMR nuclear magnetic resonance OEP octaethylporphyrinato dianion (OEP + ) octaethylporphyrinato pi-cation radical anion PhCN benzonitrile PPh 3 triphenylphosphine P(n-Bu)3 tri-n-butylphosphine PhIO iodosylbenzene porp porphyrinato dianion ppm chemical shift in parts per million relative to a reference pyrr pyrrole q quartet, used in NMR spectroscopy R aryl or alkyl RT room temperature S ground spin state xxi s second, or singlet when used in NMR spectroscopy A S° standard entropy of reaction A S t entropy of activation sh shoulder T temperature t triplet, used in NMR spectroscopy TBAP tetra-n-butylarnmonium perchlorate TEMPO tetramethyl-l-piperidinyloxy THF tetrahydrofuran TMP tetramesitylporphyrinato dianion TPP tetraphenylporphyrinato dianion TTP tetra-p-tolylporphyrinato dianion UV/visible ultraviolet-visible v infrared frequency B.M. Bohr magneton M e f r effective magnetic moment e molar extinction coefficient X m magnetic susceptibility V Volts xxii Structure of porphyrinato dianions: R i R2 R i porp R[ R 2 OEP Ethyl ( C J H J ) H TPP H . Phenyl ( C 6 H 5 ) TMP H Mesityl ( 2,4 , 6 - { C H 3 ) 3 C 6 H 2 ) TTP H para-Tolyl ( 0 - C H 3 ) C 6 H 4 ) xxiii Numerical key to Ruthenium complexes discussed in this Thesis Number ComDlex 1 R u ^ l - j ^ O 2 (Ru°)3(CO)12 3a Ru u(OEP)CO 3b Ru n(OEP)CO(MeOH) 4a Ru n(OEP)(PPh 3) 2 4b Ru n(TPP)(PPh 3) 2 5a Ru"(OEP)PPh 3 5b Ru H(TPP)PPh 3 5c Ru n(OEP)PPh 3(H 20) 5d Run(TMP)P(«-Bu)3 6a Ru n(OEP)py 2 6b Ru n(OEP)(CH 3CN) 2 6c Ru"(TMP)py 2 6d Ru n(TMP)(CH 3CN) 2 7 [Ru n(OEP)] 2 8a Ru I V(OEP)Br 2 8b Ru I V(OEP)Cl 2 8c Ru l v(OEP)(BF 4) 2 8d Ru l v(OEP)(SbF 6) 2 8e Ru I V(TMP)Br 2 9 [Ru r v(OEP)OH] 20 10 Ru I I !(OEP)Br(PPh 3) i i Ru n(OEP)CO(PPh 3) 12 Ru n(TPP)PPh 3(P(n-Bu) 3) xxiv i l [Ru(OEP)] 2CHCl 2 i i Ru m(OEP)H 11 Ru m(OEP)SbF 6 16a Ru m(OEP)Br 16b Ru m(OEP ) C l 16c Ru f f l(OEP)Br(NH 3) 16d Ru m(OEP)Cl(NH 3) 16e Ru i n(OEP)Br(CH 3CN) 16f Ru m(OEP)Cl(CH 3CN) Ha [Ru n i(OEP)(py) 2] + C r 17b [Ru , n(OEP)(py) 2] + BF 4" iZc [Ru n i(OEP)(py-d 5) 2] + Cl-17d [Ru n i(OEP)(py),] + Br" I7e [Ru n i(OEP)(py) 2] + F" 121 [Ru n i(OEP)(CH 3CN) 2] + Br' XXV Acknowledgements. I thank Dr. Brian James and Dr. David Dolphin for their support and encouragement throughout my tenure here at the University of British Columbia as a graduate student. Further, I thank Mr. Carl Jacobson for measuring the ESR spectra discussed in this thesis. A teaching assistantship stipend from the Department of Chemistry is gratefully acknowledged. I thank my fellow graduate students for valuable discussions, and the support staff of the department, without whose help the work described in this thesis could not be completed. xxvi Chapter 1. Introduction. The study of ruthenium porphyrin complexes originated with models for natural protein and enzyme systems such as hemoglobin and cytochromes P-450. The active site in both natural systems was identified as an iron porphyrin center,1 and early studies of iron complexes were designed to mimic the observed reactivity of these systems. Studies of die globins, for example, examined the coordination chemistry primarily at low-valent porphyrin centers,2 while a second group of studies reflect the organometallic/high-oxidation state nature of die P-450 enzymes.3 The presumed similar reactivity of ruthenium to iron, coupled with the greater stability of the nithenium complexes, made die latter an obvious choice for modelling both the enzymatic and protein-free iron porphyrins. 1.1. Coordination chemistry at ruthenium porphyrin centers. Because several reviews ol the coordination chemistry of nithenium porphyrin complexes have recently been reported, 4 5 the scope of this Introduction will be to introduce to the reader only the areas of research that arc of immediate relevance to the work conducted for this thesis. Coordination chemistry, mediated by or centered at metalloporphyrin complexes, normally involves chemistry limited to the two axial positions of an octahedral geometry because the porphyrin macrocycle coordinates the four equatorial sites of an octahedron centered about a metal (see Figure 1.1). The porphyrin macrocycle is an essentially planar chelating ligand, offering rich chemistry because of the presence of delocalized jr-aroniaticity. The review is presented in sections indexed according to the oxidation state of ruthenium; the discussion of the divalent nithenium complexes is further divided into sections representing the four general classes of Ru(II) porphyrin coordination compounds. 1.1.a. Ruthenium(II) carbonyl complexes. An early aim was to model the 0^ and CO binding capabilities of hemoglobin, which has an imidazole group (part of a histidine residue) coordinated at one of the axial coordination sites available at the Fe. 6 Almost all reactions of the parent 1 Figure 1.1: Structural features of a six-coordinate (a) Ru(OEP)L2, and (b) Ru(TMP)L, complex showing that when a porphyrin free base coordinates to a ruthenium, the four equatorial sites of an octahedral arrangement about the metal are occupied, leaving the two axial sites for coordination of ligands, L. See the abbreviations section for crude representations of other porphyrin structures ( a ) L L ( b ) 2 porphyrin bases with Ru precursors yield Ru n(porp)CO as the major product.7 This proved to be a major hurdle to early workers in this field because of the strong Ru-CO bond and the consequent difficulty in decarbonylating these complexes. Hence, much of the early work involved the coordination chemistry trans to the carbonyl group:8 Ru(porp)CO + L - Ru(porp)(CO)L (1.1) While these works demonstrated the thermal stability of the Ru(porp)CO fragment, little information on the strength of the Ru-L bond trans to CO was reported. Further, the carbonyl ligand, being a strong jr-acid, inhibits binding of 0 2 in a trans six-coordinate geometry. Therefore, methods had to be developed for the removal of CO from these complexes so that the 0 2 binding question could be addressed. 1.1.b. Dioxygen and dinitrogen ruthenium compounds from Ru n(porp)(solvent) 2 complexes. One early method developed for decarbonylating Ru(porp)CO(L) complexes was the photo-assisted dissociation of CO from these complexes in coordinating solvents, resulting in bis(solvent) complexes:9 hi/, Ar Ru(porp)CO + solvent -» Ru(porp)(solvent)2 + CO (1.2) solvent: pyridine, acetonitrile, tetrahydrofuran Some of these complexes were found to bind 0 2 reversibly, presumably after dissociation of one axial ligand, but long reaction times resulted in oxidized, substitution-inert binuclear oxo-bridged complexes:10 0 2, solvent RT, slow Ru(OEP)(CH 3CN) 2 ^ Ru(OEP)(0 2)(solvent) - [Ru(OEP)(OH)] 20 (1.3) vacuum solvent = pyrrole, DMF, DMA A similar problem was encountered in the reaction of iron porphyrin complexes with 0 2 in which 3 the oxo- and peroxo-bridged dinuclear species were commonly formed. Strategies for obtaining complexes containing bound 0 2 focused on the synthesis of metalloporphyrin complexes having steric obstacles to "dimerization"; typically, this involved appending bulky hydrocarbon substituents at the periphery of the macrocycle. 1 1 3 Indeed, the iron "picket fence" porphyrin complex (in which several hydrocarbon chains are appended to the porphyrin ring, forming a kind of protective fence) remains the best synthetic model for the binding of 0 2 to myoglobin: 0 2 0 2 (0 2)Fe(PFP)(im) «- Fe(PFP)(im) *» (im)(PFP)Fe-0-Fe(PFP)(im) (1.4) PFP = "picket fence" porphyrin, im = an imidazole Similarly, ruthenium complexes of sterically encumbered porphyrins have recently been shown 10 bind 0 2 . n b Although substantial effort was expended on the development of these protcin-frcc 0 2 complexes, the analogy to the natural globins is limited by the non-protein nature of the model ruthenium and iron porphyrin compounds. Further, while the ruthenium complexes arc always of the low-spin state, iron complexes are capable of both low- and high-spin states (for example, the iron(II) high-spin in hemoglobin becomes iron(II) low-spin upon ligation of 0 ? or CO). 1 2 Some of these Ru 1 1 bis(solvent) complexes also bind N 2 reversibly: 1 3 , 1 4 Ru(porp)(L) 2 + N 2 - Ru(porp)(L)N 2 + L (1.5) porp = OEP, L = THF, DMF; porp = TMP, L = THF, E t 2 0 While the availability of these bis(solvent) complexes showed that rich chemistry mediated by ruthenium porphyrin complexes was possible, with two exceptions (see Sections 1.1.d and l.l.e), no further developments originating from these solvent-bound complexes were reported. Chapters 4 and 5 describe the chemistry derived from the oxidation of these latter complexes to yield 4 R u m and R u I V species. l . l . c . Rutheniumfn) phosphine complexes. A second method for displacing CO from the Ru(porp)CO species involved the addition of excess phosphine, yielding Ru(porp)(PR 3) 2 complexes: 1 5" 1 7 C 7 H 8 Ru(porp)CO + excess PR 3 Ru(porp)(PR 3) 2 + CO (1.6) Ar R = Ph, n-Bu; porp = OEP, TPP, TMP These bis(phosphine) complexes show chemical behavior similar to that of die precursor carbonyl complexes, suggesting that the strong Ru-CO bond has been replaced by a strong Ru-PR3 bond. The bis(phosphine) complexes are more substitution-inert and also inert to oxidation by 0 2 than the bis(solvent) complexes. For example, with the exception of Ru(porp)(py)2 complexes, all reported Ru(porp)(solvent)2 compounds are unstable in solution under aerobic conditions, while all of die reported Ru(porp)(PR 3) 2 species are air-stable in solution. 1 5' 1 8 Secondly, the addition of other reagents such as C O ^ or H B r ^ to the bis(phosphine) complexes usually results in a mixed coordination complex of the type Ru(porp)(L)PR 3 (where L = CO, B r ) . 1 6 a In contrast, addition of reagents to Ru(porp)(solvent)2 species usually yields products where both solvent ligands have been substituted (see, for example Chapters 4 and 5). In spite of the obvious disadvantages of using phosphines to displace CO from ruthenium porphyrins, some unexpected and potentially useful reactivity has been observed. Six-coordinate phosphine complexes yield 16e~, five-coordinate mono-phosphine species when subjected to such forcing conditions as pyrolysis under vacuum: 1 6 8 1 x 10~5 torr Ru(porp)(PR 3) 2 -> Ru(porp)(PR3) + PR 3 (1.7) 493 K, 2h porp = OEP, R = Ph; porp = TMP, R = n-Bu These five-coordinate complexes were the first Ru(porp)PR 3 species reported, although some 5 dinuclear ruthenium porphyrin mono-phosphine species have been reported recently. The monomeric five-coordinate complexes, because of their coordinatively unsaturated nature, were found to bind ligands trans to the phosphine, as do the Ru(porp)CO complexes. Measurement of die equilibrium constant for the dissociation of die phosphine from some of die generated six-coordinate complexes as a function of temperature allows for the assessment of the Ru-P bond strength (see Chapter 3). Another interesting type of reaction chemistry noted earlier involving diese Ru(porp)(PR 3) n complexes was the observation that Ru(TPP)(PPh 3) 2, in the presence of a slight excess of P(n-Bu) 3, could catalytically decarbonylate a wide range of aldehydes, for example, phenylacetaldehyde (> 10000 turnovers/h) and benzaldehyde (about 10 tumovers/h) in toluene. 1 9" 2 1 The procedure involved dissolving Ru(TPP)(PPh 3) 2 in CH 2C1 2/CH 3CN, adding 100-1000 equivalents of aldehyde followed by a brief purge of CO gas through the reaction mixture; this gave a solution whose optical spectrum has a Soret absorbance indicating the presence of Ru(TPP)(CO)(PPh 3). The reaction was then initiated by adding P(«-Bu)3 (about ten equivalents), which resulted in the formation of two other porphyrin species with Soret absorption bands at 437 nm (due to Ru(TPP)(P(7z-Bu)3)2) and 420 nm (of an unknown species). A variable initiation period for the onset of catalysis was encountered, and the formation of this unknown species with a 420 nm Soret band absorbance always resulted in the onset of catalysis. No decarbonylation was observed in the absence of this species, which was suggested to be Ru(TPP)PPh 3, thought to be formed by the equilibrium shown in Equation 1.8. The mono-phosphine complex Ru(TPP)PPh 3 was postulated to associate CO, generated by the decarbonylation of aldehyde, to yield Ru(TPP)(PPh 3)CO (Equation 1.9). Displacement of the carbonyl and/or phosphine ligand of die latter by P(«-Bu)3 would give the mixed phosphine complexes Ru(TPP)(PPh 3)P(n-Bu) 3 and Ru(TPP)(P(n-Bu) 3) 2 (Equations 1.10 and 1.11). The Ru(TPP)PPh 3 complex, which is known to have an absorption maximum at 420 nm,19 was proposed as the species most likely to be responsible for the observed catalysis. 6 Ru(TPP)(PPh 3) 2 £ Ru(TPP)PPh 3 + PPh 3 (1.8) Ru(TPP)PPh 3 + CO ^ Ru(TPP)CO(PPh 3) (1.9) Ru(TPP)CO(P(n-Bu)3) + PPh 3 Ru(TPP)CO(PPh 3) + P(/i-Bu)3 (1.10) W Ru(TPP)(PPh 3)(P(n-Bu) 3) + CO Ru(TPP)CO(P(«-Bu)3) + P(n-Bu)3 2 Ru(TPP)(P(/i-Bu) 3) 2 + CO (1.11) When phenylacetaldehyde was used as substrate, some biphenyl was produced, strongly suggesting that radical processes were occurring. The ESR spectrum of an in situ reaction mixture, and some cyclic voltammetry data, revealed the presence of Ru 1 1 1 species. Formation of the Ru 1 1 1 products was attributed tentatively to the abstraction of a hydrogen atom from RCHO: 1 9 Ru(TPP)PPh 3 + RCHO - H-Ru m(TPP) + C(0)R (1.12) The acyl radical dius formed could then dissociate to generate CO and R; reabstraction of H from the Ru 1 1 1 hydride by R would complete a catalytic cycle. 2 3 In a recent report of catalytic decarbonylation of aldehydes mediated by Fe(TPP)(P(n-Bu) 3) 2, 2 4 an alternative proposal for this catalysis was made, based on the formation of oxidized iron porphyrin products observed at the end of the catalysis. In this case, the presence of trace 0 2 was suggested to oxidize Fe 1 1 to F e m , giving 0 2 _, which then abstracts a hydrogen atom from the aldehyde, forming H0 2" and C(0)R. Unfortunately, in both the Ru and Fe systems, the reactions, especially the rates of product formation, were not reproducible, making mechanistic studies difficult. Consequently, these catalytic decarbonylation processes remain poorly understood. Further studies were undertaken, as described in Chapter 3 of this thesis, to assess the role of five-coordinate mono-phosphine complexes such as Ru(OEP)PPh 3 in catalytic aldehyde decarbonylation reactions, and to determine what conditions are conducive for the reproducible initiation of the decarbonylation reaction. 1.1.d. Other rutheniumfJJ) complexes. Pyrolysis of the Ru(porp)(py)2 complexes under vacuum results in loss of the axial pyridine ligands, forming coordinatively unsaturated metal-metal bonded dimers. 2 5 These dimers react widi a range of added ligands 'L' to form six-coordinate complexes; some examples include organic sulfides and nitroso derivatives as ligands, as shown in Equation 1.13:26'27 A, vacuum 4 L 2 Ru(OEP)(py) 2 - [Ru(OEP)] 2 - 2 Ru(OEP)(L) 2 (1.13) L = SR 2 (R = methyl, decyl), PhNO In contrast, pyrolysis of Ru(TMP)(CH 3CN) 2 under vacuum yields the monomer Ru(TMP), the steric bulk of the macrocycle preventing dimerization. The monomer also binds a wide range of ligands: 1 4 A, vacuum 2 L Ru(TMP)(CH 3CN) 2 - Ru(TMP) - Ru(TMP)(L) 2 (1.14) for example, L = OEt 2, DMF, THF, N 2 l . l . e . Ruthenium(III) complexes. Oxidation of Ru n(porp)(PR 3) n (n = 1, 2) compounds by single-electron chemical oxidants yields the mixed phosphine halide Ru 1 1 1 compounds:163'28 Ru"(porp)(PR 3) 2 + HX/air - Ru m(porp)X(PR 3) + 0 = P R 3 (1.15) porp = OEP, R = Ph, n-Bu, X = Br, CI; porp = TPP, R = Ph, X = Br porp = TMP, R = n-Bu, X = Br, CI Ru n(porp)(PR 3) 2 + 1/2X 2 - Ru m(porp)X(PR 3) + PR 3 (1.16) porp = OEP, R = Ph, n-Bu, X = Br, CI; porp = TPP, R = Ph, X = Br Ru n(porp)PR 3 + HX/air - Ru m(porp)X(PR 3) (1.17) porp = OEP, R = Ph, X = Br; porp = TMP, R = n-Bu, X = CI 8 Chemical or electrochemical oxidation of Ru(OEP)CO complexes yields products in which at first sight the ruthenium has a formal oxidation state of +3. 1 7 These compounds, however, have been shown to be of the formulation [ R u n ( O E P + ) C O ( L ) ] + (L = Br, py, vacant), the products of a ring-centered oxidation. 1 7 , 2 9 In contrast, oxidation of Ru(OEP)(P(n-Bu) 3) 2 by Br 2 yields [Ru r a(OEP)(P(/i-Bu) 3) 2] + Br": 1 7 Ru n(OEP)CO - [ R u n ( O E P + ) C O ( L ) ] + + e (1.18) Ru n(OEP)(P(7j-Bu) 3) 2 + 1/2 Br 2 - [Ru m(OEP)(P(n-Bu) 3) 2] + Br" (1.19) Ruthenium(III) bis-phosphine complexes can also be prepared by adding PR 3 to the n -cation radical species [Ru u(OEP +)CO(L)] + (L = py, Br, vacant; R = n-Bu, Ph). 1 7 Other Ru 1 1 1 porphyrin complexes not containing phosphines were generated by chemical oxidation of Ru(porp)C0 3 0 and Ru(TPP)(THF) 2, 3 1 and by the electrochemical oxidation of [Ru(porp)] 2: 3 2 Ru(porp)CO + A i r o r B r 2 + KCN - K + [Ru(porp)(CN)2]- + CO (1.20) porp = OEP, TPP Ru(TPP)(THF), + air + EtOH/CH 2Cl 2 - Ru(TPP)(OEt)(EtOH) (1.21) (1:1 v/v) [Ru n(porp)] 2 - [Ru(porp)] 2 + 2 2 X" + 2 e (1.22) porp = OEP, TPP; X = BF 4, SbCl 6, P F 6 The last-mentioned dimeric products are considered electrolytes in solution, although no conductivity data were presented.32 No other coordination compounds (i.e. excluding organometallic derivatives) of R u m porphyrins have been reported, prior to the synthesis of Ru(OEP)X (X= Br, CI, SbF 6) described in Chapters 4 and 5 of this diesis. 3 3 9 l . l . f . Ruthenium (TV) complexes. The first Ru porphyrin coordination complex reported, [Ru(OEP)OH] 20, was incorrectly characterized as the methoxide congener in solution, 3 1' 3 4 but was subsequently characterized by X-ray crystallography. 3 4 Several other /x-oxo dinuclear species have been reported since: 3 1 0 2/H 20 HX [Ru(porp)]2 - [Ru(porp)OH] 20 -> [Ru(porp)X] 20 (1.23) porp = OEP, TPP, TTP; X = OMe, Br, CI This net four-electron oxidation is thought to require the presence of trace 0 2 and H-,0,35 although experimental evidence is lacking. Attempts made with the complexes Ru(OEP)PPh 3 and [Ru(OEP)] 2 to ascertain the dependence of the oxidation reaction on 0 2 and on H,0 arc described in Chapters 3 and 4. In anodier net four-electron oxidation reaction, the dimcric species [Ru(porp)]2 was oxidized by HX acids in organic solvents in the absence of oxygen to give R u l v products: 3 6 [Ru(porp)]2 + 4 HX - 2 Ru(porp)X 2 + '4 H' (1.24) porp = OEP, X = Br, CI, F; porp = TPP, X = Br As dihydrogen could not be detected, trace X 2 in HX was consequently thought to be the oxidant. 3 6' 3 7 Further studies were initiated in the present work to identify the actual oxidant and the co-product in this reaction, and to ascertain the mechanism (see Chapter 4). The dibromide complex can also be prepared, although less effectively, via reaction 1.25:36'37 [Ru(OEP)] 2 + 2 Br 2 - 2 Ru(OEP)Br 2 (1.25) Synthesis of the TTP analogue of Ru r v(OEP)Br 2 by die thermal decomposition of Ru n(TTP +)(CO)Br in moderate yields was recently reported. 3 8 These monomeric Ru , v dihalidc 10 products have been used as precursors in the investigation of organoruthenium porphyrin complexes and their chemistry (vide infra). The halide complexes were the first reported monomelic R u I V complexes and, being of intermediate-spin (S = l, d 4), exhibit interesting magnetic behavior. Why these compounds are paramagnetic, while the aryl and alkyl R u I V analogues are diamagnetic, is not well understood; therefore, measurements of the magnetic moments of Ru(OEP)Br 2 as a function of temperature were made, as discussed in Chapter 5. Other monomelic Ru^Oporp) complexes reported previously include Ru(TMP)0 3 9 a and Ru(TMP)(OC 6H 4OH) 2, 3 9 b and an ill-characterized species generated by the reaction of Ru m(OEP)Br(PPh 3) with oxene reagents (e.g. meto-chloroperbenzoic acid (mCPBA) or iodosylbenzene (PhIO)): 4 0 Ru(OEP)Br(PPh 3) + [0] - [ 0 = R u ( O E P + ) ] + Br" + 0=PPh 3 (1.26) This oxo OEP complex (formally Ru v) was formulated as a Ru I V porphyrin w-cation radical, and was proposed as die catalyst responsible for oxidation of saturated and unsaturated hydrocarbon substrates to epoxides, alcohols and ketones. Catalytic oxidation of cyclohexene by Ru(OEP)Br(PPh 3) and excess PhIO in C 6 H 6 resulted in a distribution of oxidized products such as epoxides, alcohols and ketones;40 similar oxidized products were observed in systems involving c tochromes P-450 as a catalyst for the oxidation of organic substrates.3 Thus, the Ru(OEP)Br(PPh 3)/PhIO catalyst mimics to some extent the natural enzymes; however, a 'better' c tochromes P-450 mimic is Fe(TMP)Cl, which gives higher overall yields. 4 1 1.1.g. RutheniumCVT) complexes. The dioxo R u V I porphyrin complexes, Ru(porp)(0)? (porp = OEP, TMP, and OCP (octachloroporphyrinato dianion)) are the only R u v l porphyrins reported to d a t e.14b,39b,42,43 0 2 /-BuOOH Ru(TMP) -> Ru(TMP)(0) 2 <- Ru(TMP)CO (1.27) These Ru(porp)dioxo species are essentially models for the dioxygenases, and not for the 11 cytochromes / M 5 0 . 1.2. Organoruthenium porphyrin complexes. Again, because several comprehensive reviews have appeared recently in the literature, 4 5' 4 6 this section of the introduction will be brief and intends to show how ruthenium porphyrin coordination compounds have been utilized to generate organoruthenium porphyrin complexes. 1.2.a. Compounds originating from [Ru(OEP)]2. This complex has been instrumental is the development of both coordination chemistry and organometallic synthesis; for example:47 Ru(OEP)(CH2=CH2) C 2 H 4 '/ [Ru(OEP)] 2 (1.28) HRCC1 2 \ Ru(OEP)(=CHR) R = methyl, trimethylsilylmediyl The synthesis of the Ru(OEP)(CH 2=CH 2) complex by the addition of ethylene to Ru(OEP)(THF) 2 was reported previously; however, the proposed product was not well-characterized at that time. 1 3 Reduction of the dimer by Na/K amalgam results in the formation of a dianionic Ru° 'naked' monomer:47 [Ru(porp)]2 + Na/K amalgam -. 2 K + [Ru(porp)]"2 (1.29) 1.2.b. Compounds originating from 2 K + [Ru(porp)]"2. These Ru° monomers oxidatively add alkyl iodides, giving R u l v di-alkyl derivatives: 4 8 2 K +. [Ru°(porp)]-2 + 2 R-I - Ru I V(porp)(R) 2 + 2 KI (1.30) porp = OEP, TTP; R = methyl, ethyl While this method proved useful for the generation of alkyl products, formation of the di-aryl analogues has not been reported. Several R u n carbene complexes were also generated, by reaction of the Ru° dianions widi carbene precursors such as 1,1-dichloroethane and 12 (dichloromethyl)trimethylsilane. 1.2. c. Compounds originating from Ru(porp)X 2. The first reported route to Ru™ aryl derivatives was via metathesis of the Ru™ di-halide complexes, a process that is applicable also for the di-alkyl species: 3 6 , 5 8 Ru(porp)X 2 + RL i or RMgX -» Ru(porp)(R)2 + L i X o r M g X 2 (1.31) for example, porp = OEP; R = methyl, phenyl, tolyl, p-fluorophenyl These alkyl and aryl products homolytically cleave a Ru-C bond upon thermolysis, giving organoruthenium(III) products: 4 9 A . Ru(OEP)(R) 2 - Ru(OEP)R + organic products (1.32) R = methyl and phenyl Where R = ethyl, the Ru(IV) precursor decomposes at room temperature to give Ru n(OEP)(C(H)CH 3) and Ru m(OEP)Et via a series of steps involving both a- and p-H atom abstraction from an ethyl ligand of Ru(OEP)Et 2, by attack of a ethyl r a d i c a l . 4 8 , 5 0 1.3. Objectives of this dissertation. This thesis describes continuing efforts to develop the coordination chemistry of ruthenium porphyrin complexes explored partly in an earlier thesis. 3 7 While the initial aim of this work was to synthesize catalysts capable of oxidizing organic substrates (i.e. as mimics of cytochromes PASO), the observed inability of the ruthenium porphyrin halide complexes to oxidize organic substrates, with high turnovers in the presence of suitable oxygen donors, suggested that ruthenium porphyrin complexes, in general, are poor mimics of cytochromes P-450. Hence, the aim of this work became broader in scope, addressing such basic issues as (a) Are four- and five-coordinate ruthenium porphyrin complexes really coordinatively unsaturated in solution? (b) What factors influence substitution chemistry at these centers (e.g. axial ligand bond strengths, stability)? (c) Why are these compounds not good models 13 for the c tochromes P-450? It was hoped that while the coordination chemistry of these complexes was being developed, some answers to these issues and others mentioned earlier in this introduction would be found. Because this work is a continuation of the earlier Master's thesis work, a description of the new versus the old results is provided to ensure diat die new data arc properly highlighted. The experimental section (Chapter 2 ) describes the synthesis of the new complexes described in this thesis. The main body of the thesis work is presented in the three Results and Discussion chapters. The first of these chapters (Chapter 3 ) describes the reactivities of the monomelic, \€>e~ complexes Ru(OEP)(L) (L = CO, PPI13). Only the synthesis and characterization of these compounds were described in the Master's thesis. 3 7 The reaction of Ru(OEP)PPh 3 with the small molecules CO, 0 2 , H 2 0 , N 2, and H 2, and with PPh 3, is discussed here. These data include two measurements of the Ru-P solution bond strength. The possible role of the five-coordinate phosphine complex in the reported catalytic decarbonylation of aldehydes was tested. Chapter 4 describes efforts to determine the oxidant responsible for the "oxidation of [Ru n(OEP)] 2 by HX to yield Ru I V(OEP)X 2." In the earlier Master's work, the oxidant was suggested as trace X 2 in HX (X = Br, CI). 3 7 This suggestion is shown to be incorrect and the oxidant is identified as HX itself. The mechanism of the HX/[Ru(OEP)] 2 reaction, and the nature of the HX acid required for the reaction to be successful, were studied, allowing for the prediction of possible reactivity between [Ru(OEP)] 2 and other HX acids. Other routes to Ru r v(porp)X 2 species from Ru n(porp)(L) 2 are described (porp = OEP, TMP; L = CH 3CN, py). The last Results and Discussion chapter describes the further characterization of the Ru l v dihalide complexes described in the Master's thesis. 3 7 The initial findings of the inability of these R u I V complexes to act as catalysts for the oxidation of cyclohexene were further substantiated, and new studies were conducted to help ascertain the reasons. The poorly understood change in ground spin state with change in axial ligand from halide (S = l) to aryl and alkyl (S=0) within the R u I V oxidation state was studied further. Facile reduction of these monomeric Ru I V halide 14 complexes leads to novel R u m compounds of the formulation Ru(OEP)X (X = Br, CI), whose characterization and reactivity are documented. Oxidation of Ru n(OEP)(solvent) 2 compounds to generate [Ru m(OEP)(solvent) 2] + X" species (solvent = py; X = CI, BF 4) is also described. 15 Chapter 2: Experimental 2.1. Reagents and solvents. A l l solvents (CH 2C1 2, C 6H 6, C 7H 8, pentanes, hexanes, and THF) were distilled from the appropriate drying agents and stored under argon until needed.51 Aliquots of these solvents (excluding THF, but including all deuterated solvents required for NMR studies) were stored over a bed of activated molecular sieves (Davidson type 4A) in flasks sealed by Teflon valves under vacuum to avoid contamination by grease. THF was also stored under vacuo but over sodium benzophenone ketyl instead of molecular sieves. These anaerobic, anhydrous solvents were prepared for use by three successive freeze-pump-thaw cycles at lxlO" 3 torr, and manipulated using standard vacuum-transfer techniques when needed.52 The non-dcuteratcd solvents were obtained from Fisher Scientific, BDH, or Aldrich. Deuterated solvents were purchased from either MSD Isotopes (C 6D 6, C 7D 8, CD 2C1 2, CDC1 3, CD 3OD) or Stohler Isotope Chemicals (CD 3CN and C 5D 5N) and used as purchased. Al l gases (Ar, N 2, 0 2, CO, and anhydrous HX (X=F, CI, Br)) used for routine applications or synthesis were used as obtained from either Linde (Union Carbide Inc.) or Matheson Ltd. Prepurified N 2 for use in the inert atmosphere chamber (glove-box) was also supplied by Lindc and used directly. When argon or N 2 were required to be oxygen- and moisture-free, these cases were passed dirough a Ridox column (2.5 x 50 cm column, a Cu based catalyst used for removal of 0 2 from a gas, purchased from Fisher) and then through an activated molecular sieve column (2.5 x 50 cm, type 4A) to remove moisture. Dry 0 2 was obtained by passing the gas through an Aquasorb column (1 x 10 cm of P 2 0 5 on Vermiculite), purchased from Mallinckrodt. All gases required for use in gas chromatography (He, Air, H 2) were of the zero-gas grade from Lindc, indicating the removal of hydrocarbons to the < 0.5 ppm level, and were further purified by passing through activated charcoal columns (2.5 x 10 cm). Triphenylphosphine and tri-n-butylphosphine were obtained from Aldrich. The latter was stored 16 under argon after distillation under reduced pressure.51 Aqueous acids (HBr, HCI, HF, HSbF 6, HN0 3, H C O O C F 3 , HCOOCH3) were purchased from BDH or Fisher Scientific and used directly. Cyclohexene (passed dirough an activity I alumina column every six months to prevent peroxide formation) and benzaldehyde were supplied by Fisher. 2,2'-Azobisisobutyronitrile (AIBN, Aldrich), 5 3 a tetramethyl-l-piperidinyloxy (TEMPO, Aldri c h ) 5 3 b and tetra-n-butylammonium perchlorate (TBAP, Eastman Kodak) 1 7 b were purified following literature methods. RuCl 3.3H 20 (about 4 0 % Ru) was loaned from Johnson, Matthey Ltd., and OEP was kindly supplied by Dr. T. Wijesekera of this laboratory. 2.2. Instrumentation and techniques for synthesis. UV/visible spectra were recorded using either an HP 8452A, a Perkin-Elmer 552A or a Cary 17D spectrophotometer using either standard or specially developed quartz and pyrex cells of 0.01, 0.10 and 1.00 cm pathlength (see Figure 2.1). Infrared spectra (as Nujol mulls on Csl plates) were obtained using a Nicolet 5DX FT-IR spectrophotometer. Alternatively, a Pye Unicam SP1100 IR spectrometer was used when the sample to be studied was either a solution (0.1 mm pathlength cells) or a KBr pellet. NMR data were compiled using either a Varian (XL-300, 300 MHz FT) or a Bniker (WH-400, HXS-270 or WP-80; 400, 270 and 80 MHz FT respectively) spectrometer utilizing standard 5 mm (outer diameter) sample tubes, as well as specially developed sample tubes for work requiring an inert atmosphere or for NMR spectroscopic titration experiments (see Figure 2.2). A l l NMR spectra were referenced using an external standard: solvent: C 6D 6 Ph-CD 3 CDC1 3 CD 2C1 2 reference: 7.15 2.08 7.25 5.32 ppm (for !H NMR spectra) reference: 128.0 20.4 77.0 53.8 ppm (for 1 3C NMR spectra) A l l 3 1P{ ,H} NMR data were referenced to the external standard PPh 3 (6.05 ppm upfield relative to 8 5 % phosphoric acid). The data are presented using die diamagnetic convention where signals upfield of the reference are negative in sign. 17 Figure 2.1: UV/visible spectroscopy sample cells incorporating a Teflon-valve to seal the sample solution under an inert atmosphere or under vacuum. Cell A has no reservoir, while cells B and C have plain and calibrated reservoirs, respectively. 18 Figure 2.2: NMR sample tubes for use in the study of air-stable (A) or air-sensitive (B) samples. The side-arm in B allows for the introduction of reagents to the sample without causing decomposition of the latter. Tube (C) incorporates a septum-port screw-cap fitting for use in experiments involving titration of samples in NMR tubes. 19 Mass spectra were obtained using a Kratos-AEI MS902 mass spectrometer operating in the electron impact (EI, 70 eV) direct insertion mode at 473-573 K source temperatures. Fast atom bombardment (FAB) mass spectra were measured using a updated MS-9 mass spectrometer. GC/MS studies were conducted using a Nermag R-l0-10 spectrometer. Electrochemical studies were performed with the use of a EG&G model 175 universal programmer and a model 173 galvanostat/potentiostat from Princeton Applied Research. The standard three-electrode electrochemical 1 7 b and spectroelectrochemical54 cells required for use with die above equipment have been described previously. The electrodes used were a Ag/AgCl electrode for the reference, a Pt disk electrode for die working and a Pt wire for the counter electrode. The typical cyclic voltammetry (CV) experiment involved measuring the voltammogram of a sample solution (ca. 0.1 mM) in CH 2C1 2 or CH 3CN (depending on solubility of the sample in each solvent) containing 0.1 M tetra-n-butylammonium perchlorate (TBAP) under argon. After the CV of the sample was successfully obtained, ferrocene (Fc) was added to the cell and the CV of the solution measured again. The data were then reported with respect to the Fc + /Fc couple (632 mV vs. NHE). In all cases, no change in the reduction potential of die redox couples of the sample were observed upon (a) changing the scan rate, or (b) the initial scan direction, or (c) upon addition of Fc. The typical spectroelectrochemistry experiment involved pre-electrolyzing the solvent/ electrolyte mixture at the potential of interest, followed by addition of the sample. A purge of Ar through the solution was used to stir, as well as to remove oxygen, from the solution during the experiment. The concentration of the sample solution was then measured, and the sample itself was electrolyzed, the bulk coulometric experiment being followed with respect to time by observing the changes in the optical spectrum of the sample solution as the experiment progresses. Upon completion of die experiment, as judged by the lack of change in the optical spectrum of the electrolyzed solution, the potential was adjusted to test for reversibility of the 20 bulk coulometry. Equivalent conductivity measurements were made in either CH 2C1 2, CHC1 3 or CH 3CN using a cell (cell constant = 1.0 cm - 1) and a conductivity bridge (model RCM 15B1) from the A.H. Thomas Company. Measurements of the magnetic susceptibility of paramagnetic samples were affected using Evans' method with a Varian XL-300 MHz FT spectrometer and f-butanol as the reference.55 Diamagnetic corrections were made using data reported earlier for OEP, 3 7 and from totalling the contributions from each atom using Pascal's constants.56 The detection of H 2 by GC was done using a HP5890 GC with a 12' x 1/8" molecular sieve (13X) column and a thermal conductivity detector, 30 mL/minute column flow, 303 K oven and injector temperature, and a 333 K detector temperature. The retention times for H,, O^  and N 2 were 0.90, 2.25, and 5.20 minutes, respectively. In a typical experiment, a 10 mL flask with a side-arm septum port was loaded widi [Ru(OEP)] 2 (85 mg, 67 /xmol) in the glove box and the flask sealed with a Teflon valve; then, 5 mL of a HX/solvent solution was added by vacuum transfer to die flask, die mixture was allowed to react, and dien refrozen using a liquid N-, bath. With a gas-tight syringe, 500 fiL vapor samples were withdrawn and injected into the GC, the residual HX and solvent vapor being trapped in solution at 77 K. Because of the importance of the implications of diis study, die results obtained from GC studies were further corroborated by injecting vapor samples from the reaction flask into a gas probe of a MS and the vapor then analyzed for mle — 2. To address die possibility of leakage of H 2 from the syringe, an apparatus was devised which allowed the attachment of the reaction flask directly to the MS (see Figure 2.3). This allowed the reaction flask to be evacuated to approximately 10'7 torr and, by simply opening the Teflon valve, sample could be introduced to the gas probe of the MS. Samples of the thawed reaction liquid were also withdrawn from the flask and tested for the possible hydrogenation of the solvent by injecting 0.5 uL aliquots onto a OV101 capillary column (25 m x 0.5 mm). The temperature program used was 323 K for one minute, 10 K/minute for 10 minutes, 423 K for one minute, 25 K/minute for four minutes and lastly 523 K for one minute. 21 Figure 2.3: Exploded view of the apparatus used for the detection of by direct sampling of the reaction vapor phase via a gas sampler of a mass spectrometer. B-14 ball/socket joint with stopper metal o—rln to to r 1 I I § c rubber o-rlng 5 mm Teflon valve 6 mm nut gas probe of mass spectrometer reaction vessel (10 mL volume) 22 mm The injector and flame ionization detector (FID) temperatures were 523 and 623 K, respectively. These same samples were also tested for hydrogenation of solvent with a Nermag GC/MS. 2.3. Preparation of ruthenium complexes. Scheme 2.1 describes the syndietic strategy used to prepare the key complexes Ru(OEP)(PPh 3) (5a), Ru(OEP)X 2 (X = Br, 8a; CI, 8b), and Ru(TMP)(CH 3CN) 2 (6c) using reported methods. Thus, H 2TMP, 5 7 and the ruthenium complexes Ru 3C0 1 2, 5 8 Ru(OEP)CO(MeOH) (3b), 1 7 Ru(OEP)(PPh 3) 2 (4a), 1 5 Ru(TPP)(PPh 3) 2 (4b), 1 5 Ru(OEP)PPh 3, 1 6 a Ru(OEP)(py) 2 (6a),9 Ru(TMP)CO(MeOH), Ru(TMP)(CH 3CN) 2 (6c), 3 7' 1 4 a [Ru(OEP)] 2 (7), 2 5 Ru(OEP)Br 2 (8a) and Ru(OEP)Cl 2 (8b) 3 6 were characterized by (where applicable) UV/visible, IR and NMR spectroscopy and the data compared with the spectroscopic properties reported for each compound; die data obtained were in excellent agreement with those reported. Because die synthesis and characterization of the precursor OEP complexes have been presented in the Master's thesis, 3 7 only die characterization details for the precursor TPP and TMP compounds are provided in Section 2.3.a. The synthesis details for all new Ru(porp) complexes are provided in Section 2.3.b. 2.3.a. Spectroscopic parameters for compounds prepared from literature methods. Bis(triphenylphosphine)(tetraphenylporphyrinato)ruthenium(II) (4b). 1 5 90 % yield. Anal, calcd. for C 8 0H 5 8N 4P 2Ru: C,77.59; H.4.72; N.4.52. Found: C, 77.37; H, 4.84; N, 4.37. •H NMR spectrum (6, CDC1 3, aerobic): 8.06 (s, 8H, H p y r r); 7.55 (m, 4H, p-H p h e n v l); 7.53 (ni, 8H, m-H p h e n y l); 7.51 (m, 8H, o-H p h e n y l); 6.71 (t, 6H, p-H p h o s); 6.47 (t, 12H, m-Hphos) and 4.22 ppm (br, 12H, o-H p h o s). 3 1P{'H} NMR spectrum (CDC1 3, 13 mM): 6.87 (s). UV/visible spectrum (CH 2C1 2, A m a x nm): 433 (Soret), 515, 555. Tetramesitylporphyrin (H2TMP):57 6 % yield. ! H NMR spectrum (5, CDC1 3, aerobic): 8.60 (s, 8H, H p y r r); 7.25 (s, 8H, m-H); 2.61 (s, 12H, p-CH 3) and 1.84 (s, 24H, o-CH3). UV/visible spectnim (CH 2C1 2, A m a x nm): 409 (Soret), 511, 544, 589. 23 60 atm C 0 ( g ) » 4 2 3 K C7H5 r 6 f l u x » A r Run,CI3.H20 • Ru 3 C0 1 2 • Ru"(porp)CO 1 MeOH J2 H2(porp) 3^  porp = OEP, TPP, TMP to 4^ 10 PR3, CyHg reflux, Ar 1 x 10 ^torr, 473 K, 2h Ru"(porp)(PR3), Run(porp)PR3 + P R 3 CO t porp = OEP, R = Ph 4a porp = TPP, R = Ph 4b porp = OEP; R = Ph 5a Scheme 2.1: Synthetic strategy used to prepare key Ru(porp) complexes. Ru l l(porp)CO _3_ ex porp - OEP; S - py (6a), CH-jCN (6b) P o r P = 0 E F > : S " P V porp = TMP; S = py (6c), CH 3CN (6d) HX /_s, anaerobic conditions [Ru»(porp)] ™ • Ru l v(porp)X 2 + '2 H' porp = OEP, X = Br 8a porp = OEP. X « CI 8b hv, Ar > Ru'WpXS), - 5 1 x 10 torr, 493 K, 2h [Ru l !(porp)] porp = OEP 7 cess S COf 2 S Scheme 2.1 continued. Carbonyl(methanol)(tetramesitylpo This complex was synthesized via a modified procedure of an method reported for the OEP analogue.178'14" A solution of H2TMP (2.1g, 3.4 mmol) in mesitylene (300 mL) was stirred under a CO atmosphere for two hours to presaturate the solution with CO^y Ru 3 CO p (2.0g, 3.1 mmol) was added to the porphyrin/CO solution, and the reaction mixture refluxed under a CO atmosphere until the fluorescence due to the presence of H2TMP disappeared (several days). The reaction mixture was cooled to room temperature, and three-quarters of the solvent was removed under reduced pressure. This concentrated solution (ca. 80 mL) was then added to an alumina column (activity I, neutral grade, 50 x 4 cm), and the column was eluted with CH2C19. A purplish-red solution eluted first which was a mixture of H2TMP and Ru(TMP)CO. Upon further elution with CH,C17, an orange colored solution (containing just Ru(TMP)CO) was collected. Then, the first eluate was concentrated under reduced pressure and rechromatographed on the same type of column, further separating the H2TMP and Ru(TMP)CO. The combined eluates of Ru(TMP)CO were then concentrated, and chromatographed a diird time to remove any last traces of H,TMP. The Ru(TMP)CO solution was collected as it eluted from the column, the solvent removed under reduced pressure, and the residue collected by scraping the solid from the walls of the flask. This red solid was dien recrystallized from CH2Cl2/MeOH/pentanes and dried under vacuum at room temperature. Yield, 1.6g (62 %). ! H NMR spectrum (S, C 6 D 6 aerobic): 8.86 (s, 8H, H p v r r); 7.24 (s, 8H, m-H); 2.56 (s, 12H. p-CH3); and 2.25 ppm (s, 24H, o-CH3). UV/visible spectrum (CH2C12, A m a x nm): 414 (Soret), 532. Bis(acetonitrile)(tetramesitylporphyrinato)ruthenium (6c). This compound was prepared using the method described for the OEP analogue,37 but using a 1:1 CH 3CN:C 6H 6 mixture in place of neat CH3CN as is required for the synthesis of Ru(OEP)(CH3CN)2. The spectroscopic data were close 26 to those reported in Reference 14a. 'H NMR spectrum (6, C 6 D 6 vacuo): 8.67 (s, 8H, Hpyrr); 7.27 (s, 8H, m-H); 2.56 (s, 12H, p-CH3); 2.23 (s, 24H, o-CH3); -1.30 (s, 6H, CH^CN). UV/visible spectrum (CH3CN, A m a x mn): 402 (sh), 420 (Soret), and 510. 2.3.b. Synthesis and characterization of new Ru(porp) complexes. Dichloromethylbis((octaethylporphyrinato)ruthenium(II,III)) (13) and hydrido(octaethylDorphyrinato)ruthenium(III) (14): This mixture of compounds, which have not been unambiguously identified, can be prepared via two routes: Route A: 3/2 [Ru(OEP)], + CH2C12 - [Ru(OEP)]2CHCl2 + Ru(OEP)H (2.1) 2 13 14 [Ru(OEP)]2 (30 mg, 24 /xmol) was placed in a Teflon-valve equipped flask and CH2C12 (10 mL) was transferred onto the solid under vacuum. The resulting greenish solution was then heated to reflux under vacuum for 30 minutes, the solution color changing from green to red. The solvent was removed under vacuum, and the solid residue dried overnight under vacuum at ambient temperature. The reaction flask was then taken into the glove box, and the product was scraped off the walls of the flask, collected and stored in the glovebox. The deuterated analogues of J_3 and 14 were prepared using diis method with CD2C12 used in place of CH2C12. Route B: 3/2 [Ru(OEP)]2 + HC1/CH2C12 - [Ru(OEP)]2CHCl2 + Ru(OEP)H (2.2) 7 13 14 [Ru(OEP)]2 (30 mg, 24 /xmol) was placed in a Teflon-valve equipped flask and an HC1/CH2C12 solution (10 mL, made by bubbling HCl^ into CH2C12 for 30 seconds) was transferred onto the solid under vacuum. An immediate reaction took place upon dissolution of the green solid as 27 evidenced by the instantaneous change in color from green to red. The subsequent isolation procedure was the same as that described for route A. Anal, calcd for a 1:1 mixture of 13 and 14 formed from the reaction of 1_ with HCI: C, 65.86; H, 6.75; N, 8.46; CI, 3.52. Found: C, 64.39; H, 6.66; N, 8.30; CI, 4.46. Analysis found for the mixture generated from 7 and CH 2C1 2: C, 64.57; H, 6.56; N, 8.43. lU NMR spectrum (6, CD 2C1 2, 293 K, in vacuo): 13.06 (s, 4H, H m e s o ) ; 10.07 (m, 8H, CH 2); 5.78 (m, 8H, CH 2); 4.20 (m, 8H, CH 2'); 2.28 (t, 24H, CH 3); -3.45 (br, 12H, CH 3'); -8.20 (br, 2H, H m e s o'); see Section 4.2.b. Conductivity (10 mg sample in 5 mL CH 2C1 2, 293 K): 1 fimhos/cm. UV/visible spectrum (2 mg sample in 5 ml CH 2C1 2, A m a x nm): 400 (sh), 486 (Soret), 504, 542, 574. FAB-MS: 1282 mle ([Ru(OEP)] 2CH 2) +, 1268 ([Ru(OEP)] 2) +, 634 (Ru(OEP)) +. Solution molecular weight (3 mg sample of I3/J4, Ru(OEP)Cl-> used as the reference sample): 920 +_ 100 g/mole. Mefr (corrected, 2.2 mg sample in CD 2C1 2, at 293 K): 2.0 B.M. Hexafluoroantimonate(octaethylporphyrinato)ruthenium(III) (15): [Ru(OEP)] 2 + excess HSbF 6 - 'Ru I V(OEP)(SbF 6) 2' - Ru m(OEP)SbF 6 (2.3) 7 8d 15 [Ru(OEP)] 2 (7 mg, 5.5 /xmol) and HSbF 6 ( a q ) (1 /iL) were reacted in C 6 D 6 (0.5 mL) in an N M R sample tube under a N 2 atmosphere. A dark red solid precipitated immediately, which was dissolved by adding CDC1 3 to the sample tube under aerobic conditions. The product in solution at this stage was identified as 'Ru I V(OEP)(SbF 6) 2' (8d, see Section 4.2.i) by 'H N M R spectroscopy. This reddish colored solution was removed from die NMR tube, and n-pentanc (10 mL) was added to afford a brownish-red colored solid, which was collected by filtration and 28 reprecipitated from CH2Cl2/n-pentane (1 mL/10 mL). The resulting suspension was filtered to collect the product, which was then air-dried. Yield, 5 mg (70%). The microanalysis and 'H NMR spectroscopic data for 15 are consistent with those reported for Ru(OEP)SbF 6(THF) 3 7 with the exception that the THF ligand in the latter complex is absent in 15. Anal, calcd. for C 3 6H 4 4N 4RuSbF 6: C, 49.66; H, 5.06; N, 6.44. Found: C, 49.12; H, 5.18; N, 5.86. ! H NMR spectrum (5, CDC1 3, 293 K, aerobic): 18.22 (br, 8H, CH 2), 9.45 (br, 4H, H m e s o ) , 3.11 (br, 8H, CH 2), 1.71 (br, 24H, CH 3). Dibromo(tetramesitylporphyrinato)ruthenium(IV) (8e): excess HBr + Ru(TMP)(CH 3CN) - Ru(TMP)Br 2 -I- 2 CH 3CN + '2 H' (2.4) 6c 8e Ru(TMP)(CH 3CN) 2 (100 mg, 0.1 mmol) was loaded into a reaction flask equipped with a Teflon stopcock in die glove-box. In another flask, also equipped with a Teflon stopcock, C 6 H 6 (20 mL) was saturated with H B r ^ and this solution was degassed by diree freeze-pump-thaw cycles. This HBr solution was transferred onto the solid Ru(TMP)(CH 3CN) 2 under vacuum, the resulting solution being bright orange in color. The volume of this solution was reduced under vacuum to approximately five mL and then n-pentane (15 mL) was added under vacuum. The reaction mixture was cooled to 253 K to yield a brownish colored suspension, which was exposed to air and the product was collected by filtration. The brownish solid was then dried under vacuo at 353 K for 12h. Yield, 100 mg, 93 %. Anal, calcd. for C 5 6H 5 2N 4Br 2Ru: C, 64.68; H, 5.00; N, 5.39; Br, 15.21. Found: C, 64.58; H, 5.21; N, 5.57; Br, 15.08. lU NMR spectrum (6, C 6 D 6 vacuo, 292 K): 13.76 (br, 8H, m-H); 4.49 (br, 24H, o-CH3); 4.01 (br, 12H, p-CH3); -44.37 ppm (br, 8H, H p y r r). T t measurement (4 mM, 293 K, C 6D 6 vacuo): H D = 9 ms, H m = 121 ms, H p = 417 ms, that for 29 the H p y r r could not be measured accurately because the relaxation time was too short. UV/visible spectrum (CH 2C1 2, A m a x nm): 416 (Soret), 518. Mass spectrum (EI, 523 K source temperature): 960 vale (Ru(TMP)Br) +, 881 (Ru(TMP)) +, 80/82 (HBr) +. M e f f (corrected, 5 % *-butanol in C 7H 8, at 293 K): 2.73 B.M. (Ammine)bromo(octaethyIporphyrinato)ruthenium(III) (16c): Ru(OEP)Br 2 + 1 atmNH 3 ( g ) - Ru(OEP)Br(NH 3) + '1/2N 2H 4' + 'HBr' (2.5) 8a 16c A brownish-red suspension of Ru(OEP)Br 2 (50 mg, 0.06 mmol) in C 7 H g (50 mL) was saturated with NH 3 ( g ) and this reaction mixture was heated to 323 K and sonicated for five minutes, resulting in the formation of a clear deep red solution. This reaction mixture was cooled to room temperature, extracted with water (6 x 20 mL) and die aqueous phase was back-extracted with C 7 H g (3 x 20 mL). The combined organic phase was then dried over MgS0 4 for ten minutes, filtered and die filtrate added to a column (neutral alumina, activity III, 10 x 1 cm). Elution of the column widi CH 3CN yields a deep red solution, from which the solvent was removed under reduced pressure to yield a reddish residue. The reddish product was dissolved in C 7 H g (5 mL), precipitated out of solution by adding n-pentane (15 mL), collected by filtration, and reprecipitated from a C7Hg/n-pentane (5/15 mL) mix. The product was collected and dried under vacuum at 373 K for 6h. Yield, 36 mg 83%. Anal, calcd. for C 3 6H 4 7N 5BrRu.H 20: C, 57.75; H, 6.55; N, 9.35; Br, 10.56. Found: C, 57.43; H, 6.34; N, 8.89; Br, 10.13. The compound decomposes at 403 K without evolving H 20, as evidenced by the thermal gravimetric analysis of a 3 mg sample. JH NMR spectrum (5, CDC1 3, 293 K, aerobic): 11.43 (br, 8H, CH 2), 6.10 (br, 8H, CH 2), -2.28 (br, 24H, CH 3), -5.29 (br, 4H, H m e s o ) . UV/visible spectrum (C 7H g, 28.8 u-M, A m a x (log e) nm): 385 (4.41), 420 (4.52), 512 (3.79), 550 30 (sh). Mass spectrum (FAB-MS): 713 rale (Ru(OEP)Br) +, 634 (RuOEP)) + . Infrared spectrum (KBr pellet): v N . H = 3100 cm"1 (broad), J>oh = 3700-3300 cm"1. M e f f (corrected, 5 % f-butanol in C 6H 6, at 292 K): 1.92 B.M. (Ammine)Aloro(octoeUiylporphyrinato)ruthenium(ni) (16d): This complex was prepared using the method discussed for the bromide analogue, but using Ru(OEP)Cl 2 in place of Ru(OEP)Br 2. Yield, 36 mg (86%). Ru(OEP)Cl 2 + 1 atmNH 3 ( g ) -» Ru(OEP)Cl(NH 3) + '1/2 N 2H 4' + 'HCI* (2.6) 8b 16d Anal, calcd. for C 3 6H 4 7N 5ClRu.H 20: C, 61.10; H, 6.93; N, 9.90. Found: C, 61.51; H, 7.00; N, 9.62. 'H NMR spectrum (5, CDC1 3, 293 K, aerobic): 11.24 (br, 8H, CH 2), 6.19 (br, 8H, CH 2), -2.31 (br, 24H, CH 3), -5.07 (br, 4H, H m e s o ) . UV/visible spectrum (CHC13): identical to that of die bromide analogue within +_ 1 nm and _+ 0.01 units of log e. Mass spectrum (FAB-MS): 669 m/e (Ru(OEP)Cl) +, 634 (RuOEP)) +. Infrared spectrum (KBr pellet): i / N . H = 3105 cm"1 (broad), J/0_h = 3700-3300 cm"1. Bromo(octaethyIporphyrinato)ruthenium(III) (16a): Ru(OEP)Br(NH 3) + HF ( a q ) -» Ru(OEP)Br + 'NH 4 + F"' (2.7) 16c 16a HF ( a q ) (0.5 mL) was added to a reddish colored solution of Ru(OEP)Br(NH 3) (20 mg, 0.03 mmol) in CHC1 3 (50 mL), and die two-phase mixture was stirred vigorously for 10 minutes at ambient temperature under a N 2 atmosphere. This reaction mixture was concentrated to 5 mL 31 under reduced pressure, and then added to an alumina column (activity III, 3 x 2 cm). The column was eluted with CHC1 3 until a reddish eluate was obtained, which was collected and concentrated to 3 mL under reduced pressure. The product was precipitated out of solution by adding n-pentane (15 mL), the precipitate collected by fdtration, and reprecipitated from a C7H8/«-pcntanc (5/15 mL) mix. The product was collected and dried under vacuum at 373 K for 6h. Yield, 21 mg (95%). Anal, calcd. for C 3 6H 4 4N 4BrRu: C, 60.59; H, 6.17; N, 7.85; Br, 11.08. Found: C, 60.83; H, 6.27; N, 8.03; Br, 10.81. ] H NMR spectrum (6, CDC1 3, 293 K, aerobic): 14.15 (br, 8H, CH 2), -0.20 (br, 8H, CH 2), -0.90 (br, 24H, CH 3), -2.20 (br, 4H, H m e s o ) . UV/visible spectrum (C 7H g, 26.4 M M , A m a x ( l o g e)nm): 396 (4.61), 500 (3.77), 538 (3.75). Mass spectrum (FAB-MS): 713 rale (Ru(OEP)Br) +, 634 (RuOEP) +. Conductivity (CH 3CN, 4.12 mM, 293 K): 5 cm"1 ohm"1 M"1 /*eff (corrected, 5% r-butanol in C 7H 8, 293 K): 1.78 B.M. Chloro(octaethylporphyrinato)njthenium(IIT) (16b): This complex was prepared using the method discussed for the bromide analogue, but using Ru(OEP)Cl(NH 3) in place of Ru(OEP)Br(NH 3). Yield, 20 mg (95%). Ru(OEP)Cl(NH 3) + HF ( a q ) - Ru(OEP)Cl + N H 4 + F" (2.8) 16d 16b Anal, calcd. for C 3 6H 4 4N 4ClRu: C, 64.57; H, 6.58; N, 8.37; CI, 5.23. Found: C, 64.32; H, 6.51; N , 8.30; CI, 5.25. 'H NMR spectrum {8, CDC1 3, 293 K, aerobic): 14.13 (br, 8H, CH 2), -0.16 (br, 8H, CH 2), -0.93 (br, 24H, CH 3), -1.11 (br, 4H, H m e s o ) . 32 UV/visible spectrum (CHC13): identical to that of the bromide analogue. Mass spectrum (FAB-MS): 669 m/e (Ru(OEP)Cl) +, 634 (RuOEP) +. Conductivity (CH 3CN, 1.34 mM, 293 K): 7 cm ohm"1 M"1 M E F R (corrected, 5 % f-butanol in C 7H 8, 293 K): 2.03 B.M. (Acetonitrile)bromo(octaethylporphyrinato)ruthenium(III) (16e): This complex was syntlicsized following the method for the synthesis of 16a, except that the solvent used was CH 3CN. Also, after the chromatographic purification of the product, die CH 3CN solvent was removed under reduced pressure, and the residue precipitated out of solution using a C7Hg/n-pentanes mixture. Ru(OEP)Br(NH 3) + HF ( a q ) - Ru(OEP)Br(CH 3CN) + N H 4 + F" (2.9) 16c 16a Anal, calcd. for C 3 8H 4 7N 5BrRu: C, 57.51; H, 6.35; N, 9.07; Br, 10.23 Found: C, 57.37; H, 6.16; . N, 9.31; Br, 9.87. 'H NMR spectrum (<5, CD 3CN, 293 K, aerobic): 9.02 (br, 8H, CH 2), 5.80 (br, 8H, CH 2), -2.57 (br, 24H, CH 3), -5.77 (br, 4H, H m e s o ) . UV/visible spectrum (CH 3CN, 29.2 /xM, A m a x (log e) nm): 381 (sh), 396 (4.58), 519 (4.00), 538 (sh). Mass spectrum (FAB-MS): 713 mle (Ru(OEP)Br) +, 634 (RuOEP)) +-Infrared spectrum (Nujol mull, KBr windows): I^ C e N - 2257 cm-1. (Acetonitrile)chloro(octaethylporphyrinato)ruthenium(III) (16f): This analogue was synthesized from Ru(OEP)Cl(NH 3) using the method described for the bromide analogue 16e. Anal, calcd. for C 3 8H 4 7N 5ClRu.H 20: C, 62.64; H, 6.73; N, 9.62; CI, 4.81. Found: C, 62.86; H, 6.68; N, 9.90; CI, 5.21. JH NMR spectrum (6, CD 3CN, 293 K, aerobic): 9.20 (br, 8H, CH,), 5.78 (br, 8H, CH,), -2.70 33 (br, 24H, CH 3), -5.85 (br, 4H, H m e s o ) . UV/visible spectrum (CH 3CN): identical to that of the bromide analogue. Mass spectrum (FAB-MS): 669 mle (Ru(OEP)Cl) +, 634 (RuOEP)) +. Infrared spectrum (Nujol mull, KBr windows): i / C E N = 2250 cm - 1 (broad), v0_H = 3700-3300 cm"1. Bis(pyridine)(octaethylpoiphyrinato)nithenium(in) chloride (17a): Ru(OEP)(py) 2 + HCl ( a q )/air - [Ru(OEP)(py),] + CI" (2.10) 6a 17a A red-yellow solution of Ru(OEP)(py) 2 (50 mg, 0.06 mmol) and concentrated H C l ( a q ) (50 y.L, 0.6 mmol) in C 7 H 8 (10 mL) was sonicated for five minutes at room temperature or until a salmon colored suspension was generated. This suspension was filtered to collect the product, which was purified by two successive reprecipitation steps from CH2Cl2/n-pentanes (5/15 mL). The solid collected from die second reprecipitation step was dien dried under vacuum at 333 K for 12h. Yield 49 mg, 94%. The [Ru(OEP)(py-d 5) 2] + CI" analogue (17c) to J_7a was prepared by adding py-d5 (3 mL) to [Ru(OEP)] 2 (10 nig, 8 Mmol) under vacuum, tliis generating Ru(OEP)(py-d 5) 2 in situ. The py-d5 was then removed under vacuum, and the residual solid redissolved in CDC1 3. Some HCI(.,q) was added to this solution in air to generate die required Ru(III) product, which was isolated using the procedure described for 17a. Anal, calcd. for C 4 6H 5 4N 6ClRu: C, 66.75; H, 6.53; N, 10.16; CI, 4.23. Found: C, 66.43; H, 6.52; N, 10.05; CI, 4.15. >H NMR spectrum («, CDC1 3 aerobic, 293 K): 22.96 (br, 4H, m-Hpy); 17.21 (br, 16H, CH 2); 3.21 (br, 2H,p-H p v); -0.20 (br, 24H, CH 3); -1.26 (br, 4H, H m e s o ) ; -6.97 (br, 4H, o-H p v). /xeff (corrected, 5% C 6 H 6 in CHC1 3, 292 K): 1.99 B.M. 34 Conductivity (CH 3CN, 2.34 mM, 293 K): 125 M"1 ohm"1 cm"1. UV/visible spectrum (CHC1 3, 46.5 A » M , (log e) nm): 390 (5.09); 454 (sh); 503 (4.01); 529 (4.08); 628(2.97); 719(3.25). Mass spectrum for 17a (FAB-MS): 792 rale- (Ru(OEP)(py) 2) +; 712 (Ru(OEP)py) +; 634 (Ru(OEP)) +. FAB-MS of 17c: 802 rale- (Ru(OEP)(py-d 5) 2) +; 717 (Ru(OEP)(py-d 5)) +; 634 (Ru(OEP)) +. Bis(pyridine)(octaethylporphyrinato)ruthenlum(III) tetrafluoroborate (17b). This complex was prepared following the method discussed for the chloride analogue, but using HBF 4 ( a q ) in place of aqueous HCI. Yield, 50 mg (91 % ) . Ru(OEP)(py) 2 + HBF 4 ( a q )/air [Ru(OEP)(py),] + BF 4" (2.11) 6a 17b Anal, calcd. for C 4 6H 5 4N 6BF 4Ru.2 H 20: C, 60.33; H, 6.34; N, 9.18. Found: C, 60.37; H, 6.11; N, 9.13. ! H NMR spectrum (6, CDC1 3 aerobic, 293 K): 24.19 (br, 4H, m-Hpy); 18.05 (br, 16H, CH 2); 3.43 (br, 2H, /;-Hpy); -0.33 (br, 24H, CH 3); -1.88 (br, 4H, H m e s o ) ; -7.60 (br, 4H, o-H p v). /x e f f (corrected, 5 % C 6 H 6 in CHC1 3, 292 K): 2.44 B.M. Conductivity (CH 3CN, 1.11 mM, 293 K): 120 M"1 ohm"1 cm. UV/visible spectrum (CH 3CN, 13.9 fiM, A m a x (log £) nm): 390 (4.62); 454 (sh); 503 (3.59); 530 (3.65); 634 (2.52); 719(2.82). Mass spectrum (FAB-MS): 793 rale- (Ru(OEP)(py) 2) +; 713 (Ru(OEP)py) +; 634 (Ru(OEP) +. 2.4. General and specific reaction conditions used to study the solution chemistry of Ru(porp) complexes. The intense v -» rr* transition (labelled the Soret band) observed for metalloporphyrin complexes in the 380-420 nm region, and other bands typically observed in the 500-600 nm region, make UV/visible spectroscopy an extremely useful tool for the smdy of the solution chemistry of 35 diamagnetic Ru(porp) complexes. Solutions of Ru(porp) complexes of the concentration range 10"6-10"5 M were used to measure the equilibrium constants for die dissociation of PPh3 from Ru(OEP)L(PPh3) (L = CO, PPh 3, see Sections 3.2.b to 3.2.d), as well as for identifying species in solution whose UV/visible spectra are well known. UV/visible spectroscopy is generally less useful where paramagnetic Ru(porp) complexes are studied, because the Soret and odier bands observed in the UV/visible spectra of these species are generally of much lower intensity, and vary to a lesser extent in position (for example, the Soret band positions range over 300-400 nm) than the corresponding bands observed for diamagnetic complexes. Nevertheless, the use of UV/visible spectroscopy to study kinetic processes involving paramagnetic Ru(porp) complexes is possible, as exemplified by the study of die reaction of [Ru(OEP)] 2 (0.1 to 4 mM) with CH 2C1 2 (see Section 4.2.e). ]H NMR spectroscopy is an ideal technique for studying the solution chemistry of paramagnetic Ru(porp) complexes because die 'H resonances of these paramagnetic complexes are well separated from each other, appearing in the range -55 to 70 ppm relative to TMS. Typically, samples analyzed by 'H NMR spectroscopy were prepared by dissolving 2-3 mg of the Ru(porp) complex in 0.5-0.6 mL of an appropriate solvent, giving 1-5 mM concentration solutions. If the sample to be studied was air-sensitive, the solution was sealed under an inert atmosphere or under vacuum in NMR sample nibes such as those shown hi Figure 2.2. Where the conditions of NMR experiments differed from those described above, specific details are provided in the relevant accompanying text. In studies involving the reaction of H 2 0 widi Ru(porp) (see Sections 3.2.a, 4.2.a, 4.2.f, and 5.2.d.3), 1 tiL of H 2 0 was added to a solution of the appropriate Ru(porp) complex, and where O, was the reagent, a 0.5 mL sample was added to the NMR sample tube; in all cases, the subsequent reaction solutions were sealed under vacuum. Unless stated odierwise, gaseous reagents such as H 2 and CO (or other gases mentioned in the text) were introduced into the reaction solution by bubbling die gas through die solution under ambient pressure. 36 Studies of the possible stoichiometric decarbonylation of benzaldehyde or phenylacetaldchydc by Ru(OEP)PPh 3, Ru(TPP)(PPh 3) 2 or PPh 3 (Section 3.2.g) were conducted using 4-5 mM solutions of Ru(porp) complex and aldehyde (25-30 mM). The reagents 0 2 (0.5 mL), H 20 (1 //L) or P(n-Bu)3 (1 uL) were added separately to the reaction mixture of Ru(porp) complex and aldehyde using a syringe, and each reaction solution was then sealed under vacuum (total reaction solution volume was 0.5 +_ 0.1 mL in each case). In general, 1-2 uL samples of pyridine or HBF 4 were used in the reaction of these reagents with [Ru(OEP)] 2 (see Section 4.2.f). Samples of [Ru(OEP)] 2 (10-20 mg) were reacted with CH 2Br 2 (3 mL), CC1 4 (5 mL) or C 6H 5CHC1 2 (5 mL) under vacuum; the isolated solid from each reaction mixture obtained by the removal of die solvent under vacuum was analyzed by 'H NMR spectroscopy (Section 4.2.g). Solutions of [Ru(OEP)] 2 (1-5 mM) were also reacted with 1-5 fiL volumes of C 6H 5COOH and aqueous CH 3COOH, CF 3COOH, HF, HN0 3, HBF 4, HCI, HBr, and HSbF 6 or one atm of gaseous HF in NMR sample tubes under an N, atmosphere, and each resulting solution was subsequently analyzed by *H NMR spectroscopy (Section 4.2.i). The attempted oxidation of cyclohexene (70 /xmol) by PhIO or mCPBA (70 //mol) catalyzed by Ru(OEP)X (X = Br, CI; 1.4 //mol) was studied in C 6H 6, CH 2C1 2 or CH 3CN (2 mL, sec Section 5.2.d.3). 37 Chapter 3. The reactivity of the five-coordinate complexes carbonvKoctaethylporphyrinato)- rutheniumfJTJ). (3a). and (octaethylporphvrinato)(triphenylphosphine)rutheniiim(II), (5a). 3.1. Introduction. A number of six-coordinate ruthenium(II) porphyrin phosphine complexes are reported to dissociate a phosphine ligand in solution yielding five-coordinate intermediates which may be involved in catalytic systems (see Chapter i ) . 1 6 b . 1 9 - 2 1 The complexes Ru(OEP)(PPh 3)i (4a), Ru(TPP)(PPh 3) 2 (4b) and Ru(OEP)CO(PPh 3) (IT) have been found to dissociate PPh 3 to some extent in solution, yielding the five-coordinate complexes Ru(OEP)PPh 3 (5a), Ru(TPP)PPh 3 (5b) and Ru(OEP)CO (3a), respectively. 1 5 , 1 7 This chapter describes some solution coordination chemistry originating from 3a and 5a and the possible role of 5a in the catalytic decarbonylation of aldehydes.16" • The measurement of the dissociation rate of a ligand ' L ' from Ru and Fe porphyrin complexes containing phosphines and nitrogenous ligands such as imidazoles, isocyanates and nitriles, and the subsequent association rates of CO to the proposed five-coordinate intermediate species M(porp ) L have been documented: 1 3' 2 1' 5 9 - 6 2 -LJC.L +CO,k + c o M(porp ) ( L ) 2 «- M(porp ) L «- M(porp ) (L)CO (3.1) -fL,k^L ~CO,k_QQ The dissociation of a phosphine ligand from Ru(OEP)(PR 3) 2 complexes (i.e. k.L step in Equation 3.1) has been studied for only one system to date (R = n-Bu).62 The measurement of the dissociative rate constant (k n) and overall equilibrium (K,,) constant, as well as the corresponding activation and thermodynamic parameters for the dissociation of PPh 3 from the two species Ru(OEP )L(PPh 3) ( L = CO, PPh 3), are presented in Sections 3.2.b to 3.2.f: -PPh 3 >k n Ru(OEP)L(PPh 3) £ Ru(OEP )L + PPh 3 (3.2) + PPh 3,k n L = PPh 3, CO ('n' will be defined in the following Sections) 38 The complex Ru(TPP)(PPh 3) 2 (4b] has been reported to catalyze die decarbonylation of aldehydes in a variety of solvents, and it was suggested that the formation of a five-coordinate intermediate Ru(TPP)PPh 3 (5b) generated by the dissociation of a phosphine from 4b was cnicial for the catalysis to be observed.19 When attempts to synthesize the TPP species 5b failed, 1 6 a 5a was tested instead as a possible catalyst for the decarbonylation reaction. Due to the repotted ^reproducibility of the decarbonylation reaction catalyzed by 4b using the described experimental conditions, 1 9" 2 1 no attempts were made to repeat the previous work. Instead, the effects of trace impurities such as 0 2 and H20 on the decarbonylation reaction were studied and are documented in Section 3.2.g. 3.2. Results and Discussion. The syndiesis of Ru(OEP)(PPh 3) (5a) utilizing a vacuum pyrolysis technique has been described previously. 1 6 3' 3 7 This monomeric, 16e", five-coordinate complex is capable of reacting with several ligands to yield coordinatively saturated compounds as diagrammed in Scheme 3.1. 3.2.a. Reaction of Ru(OEP)PPh3 (5a) with small molecules. Solutions of 5a in C 6D 6 or CD 2C1 2 react with air at ambient temperatures to yield die M-OXO dinuclear species [Ru(OEP)OH] 20 (9) and 0=PPh 3 within minutes as observed by 'H NMR spectroscopy, and therefore 5a is stored as a solid in an inert atmosphere glove-box until required. An attempt was made to elucidate the mechanism of diis aerobic oxidation by adding either H 20 or dry 0 2 to C 7D g solutions of 5a. In the latter case, an aquo complex, Ru(OEP)(PPh 3)(H 20) (5c) is observed by ]H NMR spectroscopy (see Table 3.1), die species undergoing rapid exchange with free water at room temperature (water signal observed at -0.54 ppm, 50 Hz full-width at half-height, 400 MHz (FT)). This result agrees well with other data widiin ruthenium porphyrins where exchange of bound and free water is observed.27 The mixture containing 5a and dry 0 2 showed no reactivity at either 293 or 213 K and the lH NMR signals of the porphyrin and phosphine ligands remained sharp over this temperature range. A similar behavior was observed for Ru(TMP)(P(n-Bu)3) (5d) in C 7D 8 with dry 0 2 over die same temperature range, suggesting diat the more basic trialkylphosphine axial ligand 39 Scheme 3.1: Reactivity of Ru(OEP)PPh 3, 5a. 40 Table 3.1: Spectroscopic data for Ru(porp) complexes containing carbonyl and phosphine axial ligands. (a) ]H (300 and 400 MHz) and 3 1P{'H} (121 MHz) NMR spectroscopic data, measured at 293 K. complex solvent *H NMR macrocycle2 6 H meso 8 C H 2 6 C H 3 resonances phosphine phenyl-S p-H 6 m-H S o-U i _ 3 1 E Ru(OEP)CO, 3a C 7D 8 10.15 4.00 1.95 - -Ru(OEP)CO, 3a CDC1 3 9.92 4.03 1.92 - ~ Ru(OEP)(PPh 3) 2, 4a C VD 8 8.99 3.67 1.81 6.46 6.25 4.25 8.27 (s) Ru(TPP)(PPh 3) 2, 4b CDC1 3 c 6.71 6.47 4.22 6.87 (s) Ru(OEP)PPh 3, 5a C 7D 8 9.53 3.81 1.87 6.47 6.28 4.22 58.89 (s) Ru(OEP)PPh 3(H 20), 5c C 7D 8 9.36 3.75 (br) 1.86 6.64 6.28 -0.54 (br)4 4.21 Ru(TMP)(P(n-Bu)3), 5d C 7D 8 e f 53.09 (s) [Ru(OEP)OH] 20, 9 C 6D 6 9.43 4.40,4.04^ 1.89 -9.39 (OH) -Ru(TPP)(P(n-Bu) 3) 2 C 6D 6 h i -2.01 Ru(OEP)CO(PPh 3), 11 CDC1 3 9.50 3.96 (br) 1 .74 k 11.23 (br) - Unless stated otherwise, the multiplicity of each signal of the macrocycle is as follows: H m e s o = singlet (s), C H 2 = quartet (q), C H 3 = triplet (t), br = broad. - Unless stated otherwise, the multiplicity of each signal of the phosphine phenyl ring is as follows: p-H (doublet, d), m-H (t), o-H (t). £ TPP resonances: 8.06 (H p y r r, s), 7.55 (p-H, multiplet, m), 7.53 (m-H, m), 7.51 (o-H, m). - Trace free H 20 appears at 0.34 ppm in C 7D 8 as a sharp singlet. 5 TMP resonances: 8.31 (H p y r r, s), 7.21 and 7.05 (m-H, s), 2.57 (p-CH 3, s), 2.44 and 1.62 (o-CH 3, s). f Axial ligand resonances: 0.43 (-CH2CH3, br), -0.74 (P-CH2-CH2-, br), -1.59 (P-CH2-, br). 41 Table 3.1: Continued. 2 These signals appear as two sets of multiplets. * TPP resonances: 8.60 ( H p y r r , s), 8.36 (m-H, m), 7.57-7.48 (p- and o-H, m). i Axial ligand resonances: 0.55 (-CH^CH^, br), -0.92 (P-CH2-CH2-, br), -1.95 (P-CH2-, br). - Spectrum measured in the presence of a large excess of PPh 3. (b) UV/visible spectroscopic data for Ru(porp) complexes. complex solvent Amax> ( l o? Onm Ru(OEP)CO, 3a C 7 H g 393 (Soret, 5.40), 518 (4.41), 549 (4.65) Ru(OEP)CO, 3a CHC1 3 393 (Soret, 5.62), 515 (4.43), 547 (4.74) Ru(OEP)CO(MeOH), 3b C 7 H 8 " 392 (Soret), 517, 547 Ru(OEP)(PPh 3) 2, 4a C 7 H g 422 (Soret), 518, 532 Ru(OEP)PPh 3, 5a C 7 H g 395 (Soret), 520 Ru(OEP)(CO) 2 C 7 H 8 ~ 395 (Soret, 5.32), 512 (4.34), 546 (4.65) Ru(OEP)CO(PPh 3), J l C 7 H 8 £ 407 (Soret), 525, 555 2 Measured by adding excess MeOH to Ru(OEP)CO in C 7H g. - Measured under a CO atmosphere. - Measured in the presence of 100 equivalents of PPh 3 to Ru. 42 in 5d does not enhance the binding of 0 2; ^-acceptor/donor capacity of the axial ligand and solvent polarity are both known to affect 0 2 binding to Ru and Fe porphyrin complexes. 7 1 0 Although it was apparent that there is no association of 0 2 to 5a or 5d, a (porp)Ru(0 2) adduct must exist to yield eventually 9, which forms upon addition of 0 2 to the first mixture (5a and H 20) or the addition of H 20 to the second mixture (5a and 0 2). These results confirm that both oxygen and moisture are required for the aerobic oxidation of 5a to the M-OXO species 9, as observed for other Ru(OEP) complexes where there are either weakly bound (THF, CH 3CN) or no axial ligands. 9 , 1 0' 4 7 A similar reactivity was observed for the TMP case, where the addition of H 20 to a solution of Ru(TMP)(P(/z-Bu)3) (5d) and added 0 2 led to the oxidation of 5d; however, the oxidized Ru(TMP) product was not identified. Two mechanisms can be considered for the oxidation of 5a to 9 by O, and H 70. The first, based on a mechanism reported for the oxidation of PPh 3 by Ru(OEP)(PPh3)->,16b involves the oxidation of 5a by 0 2 to yield 0 = PPh 3 and 0=Ru(OEP), the latter porphyrin complex then reacting further widi anodier molecule of 5a to form [Ru(OEP)] 2 (7) and another equivalent of 0=PPh 3. The [Ru(OEP)] 2 then decomposes to the M-OXO dinuclear species upon reaction with H 20 and 0 2 (Scheme 3.2).35 The second mechanism is based on the H + catalyzed outer-sphere oxidation of 5a by 0 2, leading to [Ru(OEP)PPh 3] + and 02~. Ru(OEP)PPh 3 + 0 2 -> 0=Ru(OEP) + 0 = PPh 3 (3.3) 0=Ru(OEP) + Ru(OEP)PPh 3 - [Ru(OEP)] 2 + 0 = PPh 3 (3.4) [Ru(OEP)] 2 + 0 2 + H 20 -» [Ru(OEP)OH] 20 (3.5) Scheme 3.2 Ru(OEP)PPh 3 + 0 2<- [Ru(OEP)PPh 3] + + 02' (3.6) 0 2" + H + - H 0 2 (3.7) 2 H 0 2 - H 2 0 2 + 0 2 (3.8) 43 H 2 0 2 + 2 [Ru(OEP)PPh 3] + - 2 0=PPh 3 + [Ru(OEP)] 2 + 2 H + (3.9) [Ru(OEP)] 2 + 0 2 + H 20 - [Ru(OEP)OH] 20 (3.10) Scheme 3.3 In Scheme 3.2, the expected slow-step would be the formation of 0=Ru(OEP). Because species ol' die type 0=Ru(OEP) have been shown to be very reactive, 3 9*' 4 0 and because the dimer J generated in step 3.4 is known to be very reactive toward a i r , 2 5 ' 3 7 ' 4 7 steps 3.4 and 3.5 can be assumed to be fast. Because 7 does not react with dry 0 2 (see Chapter 4) to generate 9, the concentration of 7 in the reaction mixture should increase in the absence of added H-,0. According to Scheme 3.2, die loss of the intensity of the 'H NMR signals of 5a should be observable even in the reaction of 5a widi dry 0 2; this is contrary to the findings and, therefore, Scheme 3.2 is not the best model for the observed reactivity. In contrast, die outer-sphere oxidation of 5a by 0 2 would be the slow-step for Scheme 3.3, and because the 0 2" generated is diought to be stabilized by trace H +, 2 8 Scheme 3.3 reflects the need for the presence of H 20 in the oxidation of 5a to 9 by 0 2 and H-,0. The detection of [Ru(OEP)PPh 3] + would have enhanced the case for die latter mechanism, but no such species was observed, suggesting that the equilibrium depicted by Equation 3.6 favors reactants over products (as expected). Mixtures of 5a and H 2 or N 2 in C 6D 6 were unreactive as attested by *H NMR spectroscopy. As noted in the introduction, 5a reacts widi HBr^/air, resulting in Ru(OEP)Br(PPh 3) (10). 1 6 , 1 The reaction of 5a with CO immediately generates Ru(OEP)CO(PPh 3) (11), which under these conditions (see below) establishes an equilibrium involving dissociation of a phosphine: k - l Ru(OEP)CO + PPh 3 3a (3.11) 44 The species 3a instantly associates a further molecule of CO under diese conditions to give Ru(OEP)(CO) 2 (this latter equilibrium will be discussed in further detail later in this Chapter). 3.2.b. The reaction of PPh 3 with Ru(OEP)CO (3a) in C 7H g. Initial attempts in this laboratory to measure the rate constants (kj and k.]) detailed in Equation 3.11 following synthesis of JJ. by a literature method 1 7 failed due to the inability to prepare 1_1 analytically pure. 6 3 Kinetic studies initiated using impure 1_1 suggested that this system (1 l/3a/PPh3) was ideal for study by the technique of dynamic NMR spectroscopic analysis, 6 3 and so an alternate route involving reaction of PPh 3 with Ru(OEP)CO(MeOH) (3b) was devised, the system yielding the same overall K, (K, = k,/k.j). It is known that the dissociation of ethanol from Ru(OEP)CO(EtOH) is almost complete in CH 2C1 2 solution at /xmolar concentrations at room temperature, yielding Ru(OEP)CO and EtOH: 1 7 Ru(OEP)CO(EtOH) £ Ru(OEP)CO + EtOH, K e q = 1.8 x IO"3 M (3.12) Because die 'H NMR exchange mechanism is suggested to involve the association of PPh 3 to Ru(OEP)CO (3a) to give Ru(OEP)CO(PPh 3) (JD, the extent of dissociation of MeOH from 3b to give 3a was measured by titrating a 2.24 / i M solution of 3b with a MeOH:C 7H g solution (10:90 v/v) in C 7 H g by following the changes in die UV/visible spectrum of the solution upon addition of MeOH. The analysis of die data was done according to: Ru(OEP)CO(MeOH) ^ Ru(OEP)CO + MeOH (3.13) 3b 3a K ^ = [Ru(OEP)CO][MeOH]/[Ru(OEP)CO(MeOH)] (3.14) hence -log K e q = log {[Ru(OEP)CO(MeOH)]/[Ru(OEP)CO]} - log [MeOH] f (3.15) and log [(A-A 0)/(A rA)] = log {[Ru(OEP)CO(MeOH)]/[Ru(OEP)CO]} (3.16) where A 0 is the absorbance of the solution before the addition of any MeOH, A is the absorbance of die solution after die addition of an aliquot of MeOH, and A; is the absorbance of 45 the solution in the presence of a large excess of MeOH. [MeOH] f refers to the concentration oi" MeOH in solution after each addition, after correcting for the amount coordinated to ruthenium using: [MeOHlf = [MeOH], - [(A-A 0)/(A rA 0)][Ru] l o l a l (3.17) where [MeOH], is the total amount of MeOH added to the reaction cell, including that coordinated to the ruthenium. Thus, a plot of log [(A-A0)/(A;-A)] vs. log [MeOH] f should give a straight line with slope equal to one and the y-intercept will give K g q . Figure 3.1 shows the changes observed in the UV/visible spectrum of the solution upon addition of MeOH to a dilute solution of 3b, and the corresponding log plot to obtain K g q . The data used to determine K were measured at 393 nm and 293 K* a sample treatment of die data to give K e q is shown below: A(393 nm): 0.564 0.601 0.628 0.665 0.697 0.723 0.761 0.786 0.814 0.874 0.995 log[(A-A 0)/ (A;-A)]: -- -1.03 -0.76 -0.51 -0.35 -0.23 -0.07 0.03 0.14 0.41 - -Volume MeOH: 0.0 0.5 1.0 1.5 2.0 3.0 4.0 5.0 10.0 20.0 220 (in fiL) log[MeOH] f: - -3.62 -3.32 -3.14 -3.02 -2.84 -2.72 -2.62 -2.32 -2.02 The stock [MeOH] was 2.47 M, and the correction to [MeOH], (as in Equation 3.17) was minimal (< 1%) and, therefore, was ignored. From the log plot, K e q = 2.5 x 10"3 M and the slope was 0.9 +_ 0.1. This value for the dissociation of MeOH from Ru(OEP)CO(MeOH) (3b) shows that at 10"6 M concentrations, 3b is > 9 5 % dissociated to Ru(OEP)CO (3a) and MeOH. This result compares well with that for the extent of dissociation of EtOH from Ru(OEP)CO(EtOH) at similar concentrations (>97 % dissociated).1 7 However, at NMR sample concentrations (1-5 mM), calculations show that the MeOH should be ca. 5 0 % dissociated from 3b. The NMR spectrum of Ru(OEP)CO(MeOH) (3b) at 5.1 mM in C 7D 8 46 Figure 3.1: (a) UV/visible spectral changes observed upon addition of MeOH (2.47 M) to Ru(OEP)CO (2.24 x 1CT6 M) in C 7 H 8 at 293 K. The total added MeOH present in solution after each addition corresponds to 0, 107 , 214, 321, 428, 642 , 876, 1071, 2142, 4284 and ca. 45000 equivalents. The limiting spectrum is not shown because of significant dilution of the solution, (b) Plot of log [(A-A D)/(A rA)] vs. log [MeOH] f for the above titration, measured at 393 nm. Ca) u z < pp a PQ 0.5 0.0 <b) . o < 1 ^ao -1 < CD 2 -08-1 , 1 1 1 1 -a6 -Z8 -2.0 log D4eQH3f 350 400 WAVELENGTH 450 nm 47 containing added PPh 3 (3.3 mM, see Figure 3.2a), only shows H m e s o signals for two Ru(OEP) complexes at 10.145 and 9.776 ppm at 223 K corresponding to Ru(OEP)CO and RuOEP)CO(PPii 3), respectively, and no third signal corresponding to that for 3b. This result shows that the MeOH is fully dissociated at NMR sample concentrations, in apparent contradiction to the result based on the measured equilibrium constant for the dissociation of MeOH from 3b, and suggested that a more careful assessment of the reaction conditions in each type of experiment should be made. The UV/visible spectroscopic titration experiment involves / J M concentrations of 3b and, because at this concentration the trace [H 20] in dried C 7 H 8 could be higher {ca. 10-100 fiM) than the concentration of 3b, association of H 20 to Ru(OEP)CO is likely, forming Ru(OEP)CO(H 20). Thus, the analysis used to treat the titration data for die dissociation of MeOH from 3b probably needs to be modified to include the [H 20], an unknown but constant parameter for the system. Indeed, the addition of trace H 20 (14 mM) to a solution of Ru(OEP)CO(MeOH) (3b) (5 /xM in C 7H 8) prior to the addition of any excess MeOH caused no change in the UV/visible spectrum of 3b. This result shows that the generated species in solution upon dissolving 3b in C 7 H 8 is almost certainly Ru(OEP)CO(H 20), and not Ru(OEP)CO (3a). A similar result was observed in the Ru(OEP)CO(EtOH)/EtOH/CH 2Cl 2 titration system, 1 7 b suggesting that the EtOH system also needs to be reassessed. In die NMR spectroscopy experiment, the [H 20] is less than the concentration of 3b, as judged by the relative intensity of the *H NMR signals of 3b and H 20; therefore, the formation of Ru(OEP)CO(H 20) would be limited at these concentrations. The actual equilibrium constant for the dissociation of MeOH from 3b must be larger than the number measured by UV/visiblc spectroscopy. Indeed, the room temperature ]H NMR spectrum of neat 3b shows dissociated MeOH at die expected position for free MeOH in C 7 H 8 (5 3.30 (CH3OH), 4.78 (CH 3OH)), and the data show that, at NMR sample concentrations in dry C 7D 8, 3b fully dissociates to give Ru(OEP)CO and MeOH. The implications of these findings will be discussed later in this Chapter, after other evidence for this disparity in data between those obtained from UV/visible spectroscopic titration 48 Figure 3.2: (a) 1 H NMR spectra of a mixture of Ru(OEP)CO (5.1 mM) and PPh 3 (3.3 mM) In CyDg measured ot 10 K Intervals, and (b) spectra simulated usfng the parameters listed In Table 3.2 and Appendix 1. The temperature and the rate constant, k^ , are listed for each spectrum. ( a ) VO T (K) k 1 Cs"1) 333 3600 323 2000 313 950 303 450 293 200 283 60 273 40.0 263 7.0 253 2.0 243 0.7 233 0.1 223 0.04 (b) T 10 .10 9 . 9 0 ppm 10 .10 9 . 9 0 ppm and those from NMR spectroscopic experiments have been presented. Discussions on the theoretical aspects of the phenomenon of dynamic NMR processes are numerous,64 and the use of this technique to study the lineshape of an observed NMR signal can yield data such as rate constants and activation parameters. The phenomena described by this process indicate that two or more separate sharp NMR resonances would exist in the slow exchange rate region; the resonances would first broaden in the medium exchange rate region, and then finally collapse into a single sharp resonance witii increasing temperature when fast exchange is realized. The conditions under which this analysis apply to diamagnetic complexes arc that, in die absence of chemical exchange, (1) the lineshapes must be Lorentzian, (2) the linewidths at half-height arise from die homogeneous and inhomogeneous effects which result in the normal broadening of NMR signals; for the present study, the T 2 relaxation time measured from the solvent signal linewidth at half-height at low temperature (i.e. 223-233 K) was used as an approximation of the normal line-broadening that would be observed for the signals of interest, and (3) diat the separation between die two non-exchanging signals (in Hz) must exceed the natural linewidths of these same signals so diat the effects of chemical exchange can be easily observed. Additional factors diat simplify die measurements are that the ratio of the two exchanging species be as near as possible to 1:1 and that the observed exchange occurs between two resonances of singlet multiplicity. These additional factors are non-essential, but helpful. The 'H NMR spectra of a C 7D 8 solution of Ru(OEP)CO(MeOH) (3b) (5.1 mM) and PPh 3 (3.3 mM) sealed under vacuum were measured as a function of temperature, with an emphasis made to study the H m e s o resonances of each porphyrin species, 3a and Ru(OEP)CO(PPh 3) (11). From these spectra (Figure 3.2a), the chemical shifts of the non-exchanging H m e s o resonances at low temperature were measured. The measurement of the position of each of the H m e s o signals of 3a and J l (formed using an excess of added PPh 3) as individual samples in C 7D g at the two temperature extremes (223 and 333 K) showed that the absolute position of each resonance relative to the solvent standard increased with temperature by about 0.03 ppm overall, due to 50 some temperature dependence; however, the relative difference in position between the two samples against the internal solvent standard was found to be negligible. Thus, the positions measured at low temperature from the 'H NMR spectrum of the mixture were used as assignments for 3a and 11 over the entire temperature range of the study. The individual population of each complex (A = 3a, B = 11_, where A + B = 1.0) involved in the exchange mechanism at each temperature was measured several ways: (1) at temperatures below the coalescence temperature (for example, below 293 K in Figure 3.2), the population of each species was obtained from the ratio of integration for these two signals; (2) at temperatures above die coalescence temperature, the population of each species was calculated from the position of the coalesced signal, relative to the difference in position of the two signals in the absence of chemical exchange; (3) in the intermediate region (for example, between 283 and 303 K), the individual population of each signal was obtained by interpolation of a graph of population A (or B) vs. temperature, using die previously determined populations in the two other temperature regions. The estimated error in the population of each species is +_ 5 %. The populations and assigned positions of the H m e s o resonances of 3a and N. at each temperature (Table 3.2) were utilized in the program DNMR3 6 5 (details of this program arc provided in Appendix 1) to simulate spectra as a function of a single rate constant, k,: k, Ru(OEP)CO(PPh 3) - Ru(OEP)CO + PPh 3 (3.18) By varying the value for k j , spectra which matched those measured at each temperature were obtained (Figure 3.2b). The error limit for each value of kj is estimated to be +_ 5%, as suggested by the range of values that yield a simulated spectrum which had a reasonable fit to the measured spectrum. This method gives reliable values for the first order rate constant k,, and the data were treated by the Eyring plot analysis to obtain die activation parameters AH i and AS t (as in Equation 3.19).66 51 Table 3.2: Data for the simulation of 'H NMR spectra for the chemical exchange system Ru(OEP)CO(PPh 3) -• Ru(OEP)CO + PPh 3, in C 7D 8 as a function of temperature.£ Temp.k- Population of 3a- K , , ( x l 0 ; 223 0.360 0.04 -233 0.360 0.10 -243 0.369 0.7/ 0.51 -253 0.378 2.0/ 1.23 7.8 263 0.386 7.0/ 2.87 10.6 273 0.395 40.0/ 6.35 14.0 283 0.404 60.0/ 12.71 17.7 293 0.413 200 / 29.90 21.6 303 0.422 450 / 50.24 25.7 313 0.431 950 / 90.50 30.2 323 0.440 2000 /156.50 34.9 333 0.449 3600 /265.00 39.9 - The T 2 value used (0.1 s) was obtained from die linewidth of the solvent signal at 223 K, die low temperature extreme in diis solvent. - +_ 1 K uncertainty in temperature. £ 5 Hmeso = 1 0 - 1 4 5 P P m f o r 3a and 9.776 ppm for 11. - +. 5 % uncertainty in kj. Data provided after the '/' mark were taken from Reference 63 and represent the kj values obtained for this same system using an impure sample of Ru(OEP)CO(PPh 3) (11). - Calculated from the ratio of populations of 3a and JJ., and from the initial concentrations of 3a and PPh 3, see Text; _+ 2 0 % uncertainty in K,. 52 In (k„/T) = In (kb/h) - AE a/RT + AS } /R (3.19) where * 5 is the Boltzman constant (1.38 x 10"23 J K" 1), h is Planck's constant (6.63 x 10"34 J s) and R is the gas constant (8.31 J K"1 mol"1). A plot of In (kj/T) vs. 1/T will yield A E a from the slope and AS} from the intercept by calculation (Figure 3.3). For a solution reaction, A E a and AH} are related by the expression given in Equation 3.20.66 AH} = A E a - RT (3.20) The AH} (62 _+ 6 kJ mol"1) and AS} (18 + 4 J K"1 mol"1) values for this reaction and their implications will be discussed in detail later in this chapter. The calculation of at any one temperature can be accomplished with a knowledge of [PPh 3] 0 and [3a] c (the initial concentrations of phosphine and 3a, respectively), and of the ratio of the populations of 3a and 11 (denoted as 'n' here): k, Ru(OEP)CO(PPh 3) ^ Ru(OEP)CO + PPh 3 (3.21) 11 k j 3a K, = k,/k , = [Ru(OEP)CO][PPh 3]/[Ru(OEP)CO(PPh 3)] (3.22) Kj = ([3a] 0 - X)([PPh 3] G - X) / X (3.23) where X is the f i l l at equilibrium. Therefore, at equilibrium: n = [ 3 j ] e q / [ l l ] e q = ([3a] 0-X)/X (3.24) X = [3a] 0 / (1 + n) (3.25) This gives K j with an estimated +_ 2 0 % error at one temperature (see Table 3.2). Applying this analysis to the other temperatures where 'n' is known allows Kj to be determined as a function of temperature. These values of Kj as a function of temperature were applied to the van't Hoff equation to obtain die AH° (16 ± 6 kJ mol"1) and AS° (-16 ± 8 J K"1 mol"1) values (Figure 3.4), 53 Figure 3.3: Eyring plot of ln (kj/T) vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in C 7D 8. 4.0 —5 0.0 H " c •4.0 •8.0 H 0.0030 0.0038 1 / T, K -1 0.0046 54 Figure 3.4: Plot of ln Kj vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in C 7D 8. which will be discussed later in Section 3.2.f of this Chapter: ln K,, = -AH°/RT + AS°/R (3.26) As an independent check of the K, values obtained from the NMR experiment, K,' (used to differentiate these values from those determined by NMR spectroscopy) was measured by a titration of 3a by PPh 3 in C 7 H g at several temperatures followed by UV/visible spectroscopy.15 Equations 3.27-3.30 describe the analysis used to determine Kj': Ru(OEP)CO(PPh 3) ^ Ru(OEP)CO + PPh 3 (3.27) K j ' = [Ru(OEP)CO][PPh 3]/[Ru(OEP)CO(PPh 3)] (3.28) -log = log {[Ru(OEP)CO(PPh 3)] / [Ru(OEP)CO]} - log [PPh 3] f (3.29) and log (A-A0/Aj-A) = log [PPh 3] f - log K,' (3.30) where A is the measured absorbance of the solution at any given [PPh 3] f, A Q is the initial absorbance of the solution in die absence of added PPh 3, and A-t is the absorbance of the solution at the end of the titration. A plot of log (A-A 0/A rA) vs. log [PPh 3] f should give a straight line with a slope of one and the intercept of log Kj'. Although 3a does not decompose appreciably as a concentrated solution in air, dilute solutions were found to oxidize fully to [Ru(OEP)OH] 20 (9) within three hours under aerobic conditions in C 7H 8; thus, the titrations were conducted under an inert atmosphere (N 2). Several complexities were encountered during this study: The addition of too much PPh 3 to a solution of 3a in C 7 H g results in the formation of Ru(OEP)(PPh 3) 2 (4a) as judged by the increase in the intensity of the Soret absorbance of 4a at 422 nm via the further equilibria: Ru(OEP)CO(PPh 3) <- Ru(OEP)PPh 3 + CO (3.31) i i 5a 56 Ru(OEP)PPh 3 + PPh 3 <- Ru(OEP)(PPh 3) 2 (3.32) 5a 4a The contribution from the competing equilibria (Equations 3.31 and 3.32) was minimized by using two methods: (i) the titration was attempted under an atmosphere of CO because this would help prevent the formation of 5a by loss of CO from 1_1, and (ii) m the absence of CO, by analyzing only at the wavelength for the loss of 3a (393 nm) instead of at both 393 nm and at 407 nm (the latter wavelength representing the formation of i i ) . The UV/visible bands of 3a in C 7 H 8 (14.6 fiM) at 518 and 549 nm shifted to 512 and 546 nm, respectively, when the solution was saturated with CO(g ) ( and die Soret band shifted from 393 to 395 nm (Figure 3.5). Under these conditions, an equilibrium between 3a and Ru(OEP)(CO) 2 is present as shown in Equation 3.33:67 - CO Ru(OEP)(CO) 2 ^ Ru(OEP)CO (3.33) Quite surprisingly, die former method (method i) results in the increase in the intensity of the Soret band at 393 nm upon addition of two equivalents of PPh 3. Further addition of PPh 3 then leads to the loss of intensity at 393 nm, and an increase in the intensity at 407 nm, the Soret band position for Ru(OEP)CO(PPh 3). Analysis of die overall data as described earlier gives K, = 5 x 10*5 M widi the slope of the log plot line equal to 0.9 +.0.1 at 293 K. The equilibrium depicted in Equation 3.33 cannot account for the increase in die amount of 3a upon addition of PPh 3, and as such, we are unable to explain these initial spectral changes under CO. Consequently, method (ii) was emphasized, die data being analyzed at only one wavelength (393 nm). Another problem encountered was the reactivity of the C 7 H 8 solvent with the silicon septum used to seal the cuvette under an inert atmosphere. This problem was averted by using Teflon-faced septa so diat presumably no leaching of plasticizer occurred. Yet another problem experienced was that if the cell was left in die cell-holder of the instrument when the wavelength of light passing through die anaerobic solution, either in the presence or absence of PPh 3, was below that of the Soret band of 3a (< 393 nm), loss of 0.1 absorbance units in five 57 Figure 3.5: UV/visible spectrum of (A) Ru(OEP)CO, 3a and (B) Ru(OEP)(CO) 2 in C 7H 8. The Ru(OEP)(CO) 2 was made in situ by bubbling C O ^ into a C 7 H 8 solution of 3a under a N 2 atmosphere. 1.0 WAVELENGTH minutes for the Soret band was observed,68 and is attributed to die photolytic decarbonylation of 3a/l 1. This last complication was avoided by keeping the wavelength of light well above that of the Soret band (> 600 nm) when (a) not recording the spectrum, (b) while adding titrant and (c) while waiting for the solution to equilibrate. Figure 3.6 shows the changes in the optical spectrum of the system with the addition of PPh3 to 3a, and the corresponding plot of log [(A-A0)/(A;-A)] vs. log [PPh 3] f. The data give straight-line plots widi slope near one (0.8 +_ 0.1, see Appendix 2). The K,' values obtained, which arc presented below, show that the equilibrium constants are essentially temperature independent, and because of this, no van'T Hoff plot was attempted: Temperatures K t x 104. M- K t' x 104. M 293 2.16 1.42 j f 0.10 303 2.57 1.25 _+ 0.15 313 3.02 1.39 ±0.01 323 3.49 1.54 + 0.05 - _+ 0.2 K, data tabulated in Appendix 2. - These values of were taken from Table 3.2. Discussion of these equilibrium constants (K,') will be presented later in this Chapter. 3.2.c. The reaction of PPh 3 with Ru(OEP)CO (3a) in CHC1 3. The H m e s o NMR signal of 3a is observed to vary in position depending on the solvent used (10.145 ppm in C 7D 8 and 9.918 ppm in CDC1 3). In contrast, the 'H NMR signals of six-coordinate complexes such as Ru(OEP)(py),, which are known not to dissociate a axial ligand in solution, are essentially invariant in chemical shift in various solvents (6 H m e s o = 9.3 _+ 0.1 ppm in C 6D 6, CDC1 3 or C 7D 8). The solution IR v c o band of 3a also varies with solvent ( v c o = 1926 cm"1 in CHC1 3, 1927 cm"1 in CH 2C1 2, 1929 cm"1 in CH 3CN, 1940 cm"1 in Et 20, and 1945 cm"1 in C 7H 8). These data suggest that significant solvation 59 Figure 3.6: (a) UV/visible spectral changes observed upon addition of PPh 3 (0.0831 M) to Ru(OEP)CO (3.87 x 10"6 M) in C 7 H 8 at 293 K. The total added PPh 3 present in solution after each addition corresponds to 0, 5.2, 16, 53, 160, 291 and 420 equivalents, (b) Plot of log [(A-A 0)/(A rA)] vs. log [PPh 3] f for the above titration, measured at 393 nm. Ld O Z < CD QL O 00 CD < ° ) 0.6 0.0 350 400 450 nm WAVELENGTH 60 effects are present in reaction systems where the possibility of generating "five-coordinate" species in situ exists. Therefore, the reaction of PPh 3 with 3a was studied in CDC1 3 to detect if a solvent dependence was present, both by *H NMR spectroscopy to measure the first order rate constant k2, and the equilibrium constant K 2, and by UV/visible spectroscopy to obtain K7'. The analysis of the data obtained from these studies was done analogously to that for k j , K 1 ; and K,' in the previous section: The measured and simulated JH NMR spectra for a mixture of Ru(OEP)CO (8.77 mM) and PPh3 (4.52 mM) are presented in Figure 3.7 and the data are tabulated in Table 3.3. From the experimental data, it is apparent that the k 2 value varies to a lesser extent in CDC1 3 than the k, value in C 7D g, although die coalescence temperatures are similar. From the Eyring plot for the k 2 data in CDC1 3 (see Figure 3.8), the AS* value obtained (-63 _+ 13 J K"1 mol"1) is more negative dian diat in C 7D 8 (18 J K"1 mol"1). The A H i value (37 _+ 4 kj mol"1) in CDC1 3 is less than that in C 7D 8 (62 kj mol"1). The calculated K 2 from the NMR data are presented in Table 3.3, and the AH° (21 ± 8 kj mol"1) and AS° (21 ± 11 J K"1 mol"1) values were determined from a plot of hi K 2 vs. 1/T (Figure 3.9). The measurement of K2', the equilibrium constant in CHC1 3, was done by UV/visible spectroscopy as outlined in the previous section. The CHC1 3 solutions of 3a were not air-sensitive, and so die titration of 3a with PPh 3 was done aerobically, by following the loss of the Soret band at 393 nm (corresponding to the loss of 3a), and die same data were obtained within experimental error when the titration was done under a N 2 atmosphere. Figure 3.10 shows a sample titration experiment and the corresponding log plot of relative absorbance vs. added PPh3. Ru(OEP)CO(PPh 3) 11 Ru(OEP)CO + PPh 3 3a (3.34) K 2 = [Ru(OEP)CO][PPh 3] / [Ru(OEP)CO(PPh 3)] (3.35) 61 CN Figure 3.7« <a> *H NMR spectra of Q mixture of RuCDWCO <8.77 mM) and P P h 3 (4.52 mM) In CDCI3 measured at 10 K Intervals, and <b) spectra simulated using the parameters listed In Table 3.3. The temperature and the rate constant, kg, are listed for each spectrum. (°) ^ . , - u (b) T ( K ) k (s-1) - 2 323 980 313 670 303 480 293 330 283 170 273 60 263 25 253 12 243 7 233 4 -A - A _ 2 5 3 1 2 _ J \ JL_ 2 4 3 ?  - A - X .  9.90 9.50 ppm 9.90 ' 9.50 ppm Table 3.3: Data for the simulation of 'H NMR spectra for the chemical exchange system Ru(OEP)CO(PPh 3) Ru(OEP)CO + PPh 3 in CDC1 3 as a function of temperature.3 Temp> Population of 3a- K 2, (x 10' 233 0.484 4 -243 0.484 7 -253 0.514 12 2.7 263 0.544 25 6.2 273 0.574 60 10.6 283 0.604 170 16.0 293 0.639 330 24.0 303 0.664 480 31.1 313 0.694 670 41.6 323 0.724 980 55.1 - The T 2 value used (0.45 s) was obtained from the linewidth of the solvent signal at 233 K, the low temperature extreme in diis solvent. - + I K uncertainty in temperature. - 6 Hmeso = 9 - 9 1 8 P P m f o r 3a and 9.501 ppm for 11. - +_ 5 % uncertainty in k2. - Calculated from the ratio of populations of 3a and i i , and from die initial concentrations of 3a and PPh 3; _+ 2 0 % uncertainty in K2. 63 Figure 3.8: Eyring plot of ln (k2/T) vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in CDC1 3. 1 j — r— 0.0032 0.0040 1 / T, K 64 Figure 3 . 9 : Plot of C H C I 3 . ln K 2 vs. 1/T ( K 1 ) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in Figure 3.10: (a) UV/visible spectral changes observed upon addition of PPh 3 (0.676 M) to Ru(OEP)CO (6.75 x IO"6 M) in CHC1 3 at 293 K. The total added PPh 3 present in solution after each addition corresponds to 0, 39, 160, 597, 1090, 1578 and 2917 equivalents, (b) Plot of log [(A-A 0)/(A rA)] vs. log [PPh 3]f for the above titration, measured at 393 nm. a ° - t ! 1 350 400 450 nm WAVELENGTH 66 The raw data are tabulated in Appendix 3 and the calculated K 2' data are presented in Table 3.4. A plot of In K 2' vs. 1/T allowed the determination of the thermodynamic parameters AH° (10 _+ 1 kJ mol"1) and AS° (-18 +_ 4 J K"1 mol"1, see Figure 3.11). 3.2.d. The reaction of Ru(OEP)PPh 3 (5a) and PPh 3 in C 7H g. The addition of excess PPh 3 to 5a at fiM concentrations yields Ru(OEP)(PPh 3) 2 (4a) while the addition of less than a stoichiometric amount of PPh 3 yields a solution whose *H NMR spectrum has a very broad (> 150 Hz linewidth at 293 K) H m e s o signal (Figure 3.12), signifying another system undergoing chemical exchange (as described by Equation 3.36). Using the method described earlier for the Ru(OEP)CO/PPh 3 reaction in C 7D g and CDC1 3, the individual rate constants (k 3) and equilibrium constants (K 3 and K-.') were calculated. Ru(OEP)(PPh 3) 2 ^ Ru(OEP)PPh 3 + PPh 3 (3.36) 4a k.3 5a The mixture of 5a (3.88 mM) and PPh 3 (3.18 mM) in C 7D 8 exhibited 'H NMR spectra quite similar in behavior to those of the Ru(OEP)CO/PPh 3 system in C 7D 8 (Figure 5.12). The AH f (66 ± 1 kJ mol"1) and AS} (49 _+ 10 J K"1 mol"1) values were obtained from the Eyring plot of (k3/T) vs. 1/T (Table 3.5 and Figure 3.13). The AH° (18 +_ 5 kJ mol"1) and A S 0 (-10 +_ 5 J K"1 mol"1) values were obtained from the Van't Hoff plot of In K 3 vs. 1/T (Table 3.5 and Figure 3.14). The measurement of the K 3' by UV/visible spectroscopy as described earlier, by titrating dilute (10"6 M) solutions of Ru(OEP)(PPh 3) 2 with PPh 3, was done under inert conditions, by following die loss in intensity of the Soret band of Ru(OEP)(PPh 3) (5a) at 395 nm and the increase in the intensity of the 422 nm Soret band of Ru(OEP)(PPh3)-, (4a). There were no observed competing equilibria, and the analyses of the data were straightforward. Figure 3.15 shows a sample titration experiment and the corresponding plot of relative absorbance vs. added PPh 3. The raw data are tabulated in Appendix 4 and die calculated K 3' data are presented in Table 3.6. The thermodynamic parameters A H 0 (27 +_ 3 kJ mol"1) and AS° (-4 +_ 1 J K"1 mol"1) were subsequently obtained from a plot of In K 3' vs. 1/T (Figure 3.16). 67 Table 3.4: Equilibrium constants CK,'1) measured for Ru(OEP)COfPPh 3) ^ Ru(OEP)CO + PPh 3  as a function of temperature in CHC1 3. Temperature^ K 2 x 103. M- K 2' x 103. M 293 2.40 2.00 ±0.10 303 3.11 2.27 ±0.10 313 4.16 2.52±0.01 323 5.51 2.91 + 0.02 - ± 0.2 K, data tabulated in Appendix 3. - These data were taken from Table 3.3. Figure 3.11: Plot of ln K 2' vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in CHC1 3. - 5 . 8 -0.0032 0.00340 1 / T . K"1 68 Ffgure 3.12: (a) 1 H NMR spectra of a mixture of Ru(OEP)PPh3 (3.88 mM) and P P h 3 (3.18 mM) in CyDg measured at 10 K intervals, and (b) spectra simulated using the parameters listed In Table 3.5. The temperature and the rate constant, k3, are listed for each spectrum. (a) T_(K) k, (s~ 1 ) (b) 7V <3\ vO  k 3 ( s z H 333 18000 323 12000 313 7000 303 3000 293 1400 283 300 273 90 263 22 253 6.0 243 2.0 233 0.7 - A . A A Table 3.5: Data for the simulation of 'H NMR spectra for the chemical exchange system k3 Ru(OEP)(PPh 3) 2 - Ru(OEP)PPh 3 + PPh 3 in C 7D 8 as a function of temperature.2 Temp.- Population of 5a- K,. (x 10: 233 0.180 0.7 -243 0.198 2 -253 0.216 6 3.8 263 0.234 22 6.4 273 0.252 90 9.4 283 0.270 300 12.9 293 0.288 1400 15.5 303 0.306 3000 21.5 313 0.324 7000 26.8 323 0.342 12000 32.6 333 0.360 18000 39.2 a The T 2 value used (0.1 s) was obtained from die linewidth of die solvent signal at 233 K, the low temperature extreme in diis solvent. - + 1 K uncertainty in temperature. £ 6 Hmeso = 9 - 5 4 1 P P m f o r & ^ 8 - 9 9 3 P P m f o r ^ - +_ 5 % uncertainty in k3. - Calculated from the ratio of populations of 5a and 4a, and from die initial concentrations of 5a and PPh 3; ± 2 0 % uncertainty in K3. 70 Figure 3.13: Eyring plot of In (k3/T) vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)(PPh 3) 2 in C 7D 8. 6.0 -I 0.0030 0.0034 0.0038 0.0042 1 / T, K 71 Figure 3.14: Plot of ln K 3 vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)(PPh 3) 2 C 7H g. Figure 3.15: (a) UV/visible spectral changes observed upon addition of PPh 3 (0.0422 M ) to Ru(OEP)(PPh 3) 2 (2.78 x 10"6 M) in C 7 H 8 at 293 K. The total added PPh 3 present in solution after each addition corresponds to 0, 3.8, 7.5, 11, 15, and 150 equivalents, (b) Plot of log [(A-A c)/(A rA)] vs. log [PPh 3] f for the above titration, measured at 395 nm. ( a ) 0.75 -UJ o < CD DC O (/) GO < 0.25 1 350 400 WAVELENGTH 450 nm 73 Table 3.6: Equilibrium constants (K 3') measured for Ru(OEP)a >Pb3)o ^ Ru(0EP1PPh 3 + PPh 3 as a function of temperature in C 7H g. Temperatures K 3 x 104 K j ' x 105, M 293 1.55 0 . 8 6 ± 0 . 1 0 303 2.15 1.22+0.07 313 2.68 1.69 ± 0 . 0 7 323 3.26 2.43 + 0.10 - +_ 0.2 K, data tabulated in Appendix 4. - These data were taken from Table 3.5. Figure 3.16: Plot of In K 3' vs. 1/T (K"1) for the dissociation of PPh 3 from Ru(OEP)(PPh 3) 2 in C 7H g. -10.8 -* -11 .2 -11 .6 -0.0030 1 / T , K 0.0034 -1 74 3.2.e. Discussion of the rate (k^ and k.n) and equilibrium ( K n and Kn') constants for the reaction of PPh 3 with Ru(OEP)L (L = CO and PPh 3). Table 3.7 summarizes the rate constants (k o b s) measured in this present work and provides a listing of comparable rates for other ruthenium and iron porphyrin complexes containing pyridine, nitrile and phosphine axial ligands. Comparison of k o b s obtained in this study for die dissociation of PPh 3 from Ru(OEP)(L)PPh 3 shows that PPh 3 dissociates seven times faster from Ru(OEP)(PPh 3) 2 (4a) than from Ru(OEP)CO(PPh 3) QI) in C 7D 8 at 293 K. Also, PPh 3 dissociates 1.7 times faster from J i in CHC1 3 than in C 7D g. Comparison of the data obtained in this study with those reported earlier shows that PPh 3 dissociates > 10 5 times faster from Ru(OEP)(PPh 3) 2 than other L ligands do from Ru(OEP)(L) 2 in C 7 H g at 303 K (L = CH 3CN, P(«-Bu)3)62 and 298 K (L = PhCN). 6 0> 6 9 The k o b s value for the dissociation of PPh 3 from J i measured in this study (200 s"1) compares reasonably well widi that reported earlier for die same system (107 s"1) at 293 K.63 However, in the previous work, 6 3 one error apparent in the method used for calculation of the rate constants from the data involved the estimation of k o b s from the linewidth (w) of the observed H n u ; s o signals of i i and 3a. These data for use in a NMR spectral simulation program based on measured linewidths of a signal should be expressed in units of 2ir radians instead of Hz, reflecting the angular momentum nature of w.Mc Thus, in the previous work, k o b s more correctly calculated should be 672 s"1. The data from this previous study are therefore justifiably suspect. Direct comparison of the dissociation rate constant (k n) for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) with other M(porp)CO(PPh 3) systems is possible for one case only, that being the previously mentioned work involving Ru(OEP)CO/Ru(OEP)CO(PPh 3)/C 7H g. 6 3 The other entries in Table 3.7 for the dissociation of a ligand L' from M(porp)L(L') and two other non-porphyrin systems are provided to show die range of values that have been reported. The rate of association of a PR 3 ligand to a ruthenium porphyrin complex has, to our knowledge, been reported for only one case to date; that of the association of PPh 3 to Ru(OEP)CO in C 7H g, studied using impure samples of Ru(OEP)CO(PPh 3). 6 3 The association rate 75 Table 3.7: Rate constants (ko5s) for the dissociation of a ligand L' from Ru(porp)L(L ')S as in •obs Ru(porp)L(L') Ru(porp)L + L' Ru(norD)L L' TemD. K reference Ru(OEP)CO PPh 3 293 200 tw Ru(OEP)CC£ PPh 3 293 107 63 Ru(OEP)CO^ PPh 3 293 330 tw Ru(TPP)CO^ 4-(f-Bu)py 298 6.2 x IO"2 61 Fe(PPIX-DME)CC£ Melm 298 1500 59 Fe(PPIX-DME)C02 P(«-Bu)3 298 20 59 Ru(OEP)PPh 3 PPh 3 293 1400 tw Ru(OEP)P(n-Bu)3 P(n-Bu)3 304 2.11 x 10"3 62 Ru(OEP)CH 3CN CH 3CN 303 2.41 x 10"3 62 Ru(OEP)(PhCN) PhCN 298 2.56 x 10'3 60 Ru(OEP)(PhCN)-f PhCN 298 5.87 x 10"3 69 RuCl(C(0)C 7H 9)(CO) 2PPh 3 PPh 3 323 2.60 x IO"5 70 RuCl(C(0)C7H9)(CO)2PPh3£ PPh 3 323 1.40 x IO"5 70 - Measured in C 7H g, unless noted otherwise; the last two entries are non-porphyrin systems; tw = diis work. - Considered less reliable than data from the present work (see Text). £ Measured in CHC1 3. - Measured in 1,1,2,2-tetrachloroediane. - PPIX-DME refers to the dianion of protoporphyrin(IX) dimethyl ester, Melm refers to methylimidazole. - Measured in PhCN. £ Measured in N,N-dimethylacetamide. 76 constant k.j (obtained from kj/Kj, see Table 3.2) for the association of PPh 3 to Ru(OEP)CO in C 7 H 8 (9.3 x 10 5 M"1 s"1) compares well with that reported earlier for the same system at 293 K (4.3 x 10 5 M"1 s" 1); 6 3 and is ca. 7 times larger than the corresponding value (k_2) in CDC1 3 (1.4 x 10 5 M"1 s"1, see Table 3.3). Also, the rate constant k.j for die association of PPh 3 to Ru(OEP)CO is ca. 10 times smaller than that for association to Ru(OEP)PPh 3 (k.3 = 9.0 x 10 6 M' 1 s 1 , Table 3.5) in C 7 H 8 at 293 K. The K^ values calculated at 293 K from the NMR data (see Table 3.8) for the dissociation of PPh 3 from Ru(OEP)L(PPh 3) (L = CO, PPh 3, see Sections 3.2.b to 3.2.e) agree quite closely to those (K,,') obtained from UV/visible spectroscopic titration experiments at 293 K where L = CO. but not for L = PPh 3. The agreement is fortuitous because the thermodynamic parameters for the K n and K„' systems are quite different: thus, as seen, the equilibrium constants measured by NMR spectroscopy varied with temperature to a greater extent than those from UV/visible spectroscopy (Table 3.8), giving rise to different A H 0 and A S 0 values for the same 'apparent' equilibrium. These differences almost certainly arise from the effects of trace H 20 in the system. In light of the previously mentioned problem involving die probable association of H-,0 to five-coordinate Ru(OEP)CO in the Ru(OEP)CO/MeOH/C 7H 8 titration experiment, the equilibrium being studied by NMR spectroscopy differs from that measured by UV/visible spectroscopy in the following manner: UV/visible spectroscopic studies: Ru(OEP)L(PPh 3) + H 2 0 1 Ru(OEP)L(H,0) + PPh 3 (3.37) NMR spectroscopic studies: Ru(OEP)L(PPh 3) ^  Ru(OEP)L + PPh 3 (3.38) (L = CO, PPh 3) Thus, the equilibrium constants measured by UV/visible spectroscopy (K„') cannot be compared directly to those measured from NMR spectroscopy (K^). The near temperature-independence of K j ' measured by UV/visible spectroscopy for die dissociation of PPh 3 from JJ. in C 7 H g (see 77 Table 3.8: Equilibrium (K^) data at 293 K for the dissociation of a ligand L' from Ru(porp)L(L') as in Ru(porp)L Ru(porp)L(L') V Keq ^ Ru(porp)L + L'. solvent K „ , M —eq' reference Ru(OEP)CO PPh 3 C 7 H 8 2.16x IO"4 tw2 Ru(OEP)CO PPh 3 C 7 H 8 1.42 x IO"4 twt Ru(OEP)CO PPh 3 C 7 H 8 2.44 x IO"4 63 Ru(OEP)CO PPh 3 CHC1 3 2.40 x IO"3 tw2 Ru(OEP)CO PPh 3 CHC1 3 2.00 x IO"3 tw^ Ru(OEP)CO PPh 3 CH 2C1 2 3 x IO"5 71 Ru(OEP)CO PPh 3 CH 2C1 2 3 x IO"6 17t£ Ru(OEP)CO PPh 3 CH 3CN 7.1 x IO"4 71 Ru(OEP)CO PPh 3 E t 2 0 2.6 x IO"4 71 Ru(TPP)CO PPh 3 C 7 H 8 2 x IO"6 15 Ru(OEP)CO MeOH C 7 H 8 2.5 x 10"3 t\v!2 Ru(OEP)CO EtOH C 7 H 8 1.8 x IO"3 17b^ Ru(OEP)PPh 3 PPh 3 C 7 H 8 1.55 x IO"4 tw2 Ru(OEP)PPh 3 PPh 3 C 7 H 8 8.6 x IO"6 twi! Ru(OEP)PPh 3 PPh 3 C 7 H 8 1.2 x IO"5 15 Ru(OEP)P(n-Bu) 3 P(«-Bu)3 C 7 H 8 < IO"8 15 Ru(TPP)PPh 3 PPh 3 C 7 H 8 1.35 x IO"4 15 - tw = this work. Calculated from data obtained from NMR spectroscopic studies. - Obtained from UV/visible spectroscopic titration studies. 78 Section 3.2.b) suggests that PPh 3 and H 20 bind to the Ru(II), trans to a CO, with similar bond strength. A similar temperature-invariance in K,,q was observed for the dissociation of PPh 3 from 11 in CH 2C1 2 measured using an analogous method to that described in this Chapter,71 showing diat the dissociation product in both systems exhibits similar chemistry. Trace H 20 from the solvent as suggested by the present study, or MeOH from the precursor (Ru(OEP)CO(MeOH) was also used in the work reported in Reference 71) probably compete with die solvent or PPh3 for the remaining axial coordination site of 3a. That H 20 or MeOH competes effectively with PPh3 for die last coordination site of 3a could result from the strong trans influence of the CO ligand; 7 2 the CO and PPh 3 will compete for available TT-electron density, while the weak TT -donor properties of H 20 and MeOH could aid dieir binding trans to CO. The reasons for the large differences between die K,,' for the dissociation of PPh 3 from Ru(OEP)CO(PPh3) measured by UV/visible spectroscopy in CH 2C1 2 (3 x 10"5 M,71 3 x 10"6 M ) 1 7 b and CHC1 3 (2 x 10"3 M) arc presently unknown. The K,,' values measured for the same equilibrium in different solvents by UV/visible spectroscopy show die expected trend in decreased binding of PPh 3 to Ru(OEP)CO in coordinating solvents such as CH 3CN or Et 20. The equilibrium constant measured by NMR spectroscopic studies for the dissociation of PPh3 from Ru(OEP)(PPh 3) 2 (4a) is die same as that for the dissociation of PPh 3 from Ru(OEP)CO(PPh3) (JJ_) in C 7 H 8 at 293 K, suggesting that in these systems, CO and PPh 3 have similar strong trans influences. Comparisons, noted 'in previous studies, 1 7 b also observed here include the stronger binding of PPh3 to Ru(TPP)L than to Ru(OEP)L complexes (L = CO , P P l i 3 ) , 1 5 , 1 7 b and the stronger binding of P(/z-Bu)3 to Ru(OEP)P(n-Bu)3 than PPh 3 to Ru(OEP)PPh 3 1 5 because of the increased basicity and generally smaller steric bulk of the alkylphosphines compared to the arylphosphines. 3.2.f. Summary and discussion of activation and thermodynamic data for the reversible dissociation of PPh 3 from Ru(OEP)L(PPh 3) (L= CO, PPh 3). While the data were introduced and cursorily discussed in the preceding three sections, a summary (see Table 3.9) is required to emphasize the findings and dieir implications. The activation entropy for a reaction that generates two products 79 Table 3.9: Summary of the activation and thermodynamic data for the equilibria K Ru(OEP)UPPh ?) ^  Ru(OEP)L + PPh, (L = CO. PPh ?) in C 7 H 8 and CHCU. (a) Activation parameters obtained from k n by NMR spectroscopic studies. L. solvent A H t. kj mol'1 A S t , J K'1 mol''2 A G i . k J m o r ' s CO, C 7 H 8 62 ± 6 18 ± 4 57 ± 1 4 CO, CHC1 3 37 ± 4 -63 ± 13 55 ± 1 3 PPI13, C 7 H 8 66 ± 7 49 ± 10 52 ± 12 (b) Thermodynamic parameters obtained from K,, by NMR spectroscopic studies. L. solvent AH°. kJ mol'1^ AS 0. J K'1 mol''a AG 0, kj mol' 1 2 CO, C 7 H 8 16 ± 6 -16 ± 8 21 ± 1 3 CO, C H C I 3 21 ± 8 21 ± 1 1 15 ± 1 0 PPh 3, C 7 H 8 18±5 -10± 5 2 1 ± 7 (c) Activation parameters obtained for k.n from K„ and kn. L. solvent AHt. kJ mol'1 A S i , J K'1 mo!''2 A G t , kJ mol' 1 2 CO, C 7 H 8 46 ± 18 34 ± 18 36 ± 2 4 CO, C H C I 3 16 ± 8 -84 ± 4 7 40 ± 2 8 PPh 3,C 7H 8 48 ± 15 59 ± 3 2 31 ± 13 80 Table 3.9 continued. (d) Thermodynamic parameters obtained from K„' by UV/visible spectroscopic studies. L, solvent AH°. kJ mor 1 A j ^ l K ^ j n o J - 1 2 AG°, kJ mol''a CO, C 7 H 8 ca. 0 ca. -9(£ ca. -27 CO, CHC1 3 10 ± 1 -18 ± 4 15 ± 4 PPh 3, C 7 H 8 27 _+ 3 -4 ± 1 28 ± 8 a Calculated at 293 K. ^ Estimated error in A H 0 from NMR studies is +. 30%, that for AS° is +. 50%. - Calculated assuming A H 0 is ca. zero. 81 from one reactant (i.e. a dissociative process), ignoring solvation effects, qualitatively should be positive: The variation in ASi. values obtained by NMR spectroscopy in C 7 H 8 and CHC1 3 suggest that, for the above reactions, significant solvation of the activated complex relative to the reactants may be present. For example, the AS J value for the dissociation of PPh 3 from JJ. in CHC1 3 is quite negative (-63 J K"1 mol"1) in comparison to that for dissociation of PPh 3 from U in C 7 H g (18 J K"1 mol"1), suggesting diat CHC1 3 solvates die activated complex to a greater extent than docs C 7H 8. The activation enthalpies for the same reaction are also different in C 7 H 8 (62 kJ mol"1) and in CHC1 3 (37 kJ mol"1), indicating diat solvation energies are contributing more to the observed activation enthalpy values in the latter case. For Reaction 3.39, the activated complex for die dissociation of PPh 3 from Ru(OEP)L(PPli 3) complexes is pictured as involving somewhere between a slightly stretched Ru-P bond, generating a 'reactant-like' activated complex, and an almost dissociated 'five-coordinate like' activated complex. Based on the activation parameters obtained for the dissociation of PPh3 from Ru(OEP)CO(PPh 3) in C 7 H 8 and CHC1 3, die latter type of activated complex seems likely, solvation of this 'five-coordinate' like activated complex accounting for the observed differences in activation parameters in different solvents for the same reaction: k, n Ru(OEP)L(PPh 3) L = CO, 11; PPh3,4a Ru(OEP)L + PPh 3 (3.39) Y e a c t a n - t - U k e ' ' f i v e - c o o r d i n a t e l i ke ' 82 Based on the assumption that C 7 H 8 does not solvate significantly die 'five-coordinate' like activated complex, and that the activated complex resembles closely die five-coordinate product, the Ru-P bond dissociation energy may be approximated as 64 ± 9 kJ mol"1; this is for a bond trans to a CO or PPh3, ligands of strong trans effect.72 A similar solvent dependence has been observed for die dissociation of PPh3 from RuCl(C(0)C7H9)(CO)2(PPh3)2 to form a five-coordinate kinetic intermediate, diis reaction being a rare case where activation parameters have been measured in more than one solvent:70 RuCl(C(0)C7H9)(CO)2(PPh3)2 - RuCl(C(0)C7H9)(CO)2(PPh3) + PPh3 (3.40) Solvent = C 7 H 8 , N,N-dimethylacetamide (DMA) The activation parameters for Reaction 3.40 are presented below:70 solvent AH}, kJ mol"1 AS}. J K"1 mol"'2 C 7 H 8 128 ± 12 62 ± 6 DMA 69 + 7 -126 + 13 *• Calculated at 293 K. The negative entropy of activation (-126 J K"1 mol"1) obtained when Reaction 3.40 was conducted in DMA was ascribed to the increased solvation by DMA compared to C 7 H 8 of the anticipated "five-coordinate like" activated complex formed from the partial dissociation of PPh3 ligand trans to a PPh3 in RuCl(C(0)C7H9)(CO)2(PPh3)2, because the activation entropy was found to be positive (62 J K"1 mol"1) when the same reaction was done in C 7 H 8 . The activation enthalpy (AH} ) obtained in C 7 H 8 was, therefore, suggested to represent the upper limit of the Ru-P bond strength. One other estimation of die M-P bond strength may be obtained from the dissociation of PPh3 from rrans-RuCl2(CO)(PPh3)3 in chlorobenzene (AH} = 129 ± 2 kJ mol"1 and AS} = 33 ± 8 J K"1 mol"1).738 83 Axial ligands in 'octahedral' metalloporphyrin complexes show quite generally increased lability compared to related non-porphyrin macrocycle systems and other six-coordinate complexes.7311 The lower AHt values noted in the Ru(porp) systems (64 kJ) compared to die carbonyl phosphine non-porphyrin systems (128 kJ) are consistent with this generalization. The positive AH° values obtained for the dissociation of PPh 3 from U in CHC1 3 (10 _+ 1 kJ mol"1) and 4a in C 7H 8 (27 +_ 3 kJ mol"1) by UV/visible spectroscopy are consistent with a process involving breaking a Ru-P bond (bond dissociation energy of about 64 kJ mol"1) and formation of an aquated product. Indeed, the enthalpies of reaction determined for the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) in C 7 H 8 and CHC1 3 by UV/visible spectroscopic titration are ca. 10-15 kJ mol"1 more exothermic than diose measured by NMR spectroscopic studies. These differences may be due to die association of H 20 to Ru(QEP)CO generated by the dissociation of PPh 3 from Ru(OEP)CO(PPh 3) at UV/visible spectroscopy concentrations (/iM). The close to zero AS° values for the 3a/PPh 3/CHCl 3 system (-18 ± 4 J K"1 mol"1) and for the Ru(OEP)PPh 3 5a/PPh 3/C 7H 8 system (-4 _+ 1 J K"1 mol"1), are considered consistent with the replacement of one coordinated ligand by another, as in Equation 3.37. The enthalpy of activation for the association of PPh 3 to 'Ru(OEP)CO' in C 7 H 8 (46 +_ 18 kJ mol"1), calculated from K,, and k n (data obtained from the NMR experiment), is quite different from that in CHC1 3 (16 _+ 8 kj mol"1), showing again that solvation of reactants and/or the activated complex probably plays a major role in these systems. The enthalpy and entropy of activation for the association of PPh 3 to 'Ru(OEP)CO' (3a) in C 7 H g are similar to those for the association of PPh 3 to 5a (Table 3.9). These data indicate diat the association of PPh 3 to 5a and to 3a in C 7 H 8 are equally favorable, suggesting that in these Ru(OEP) complexes, CO and PPIi 3 have similar trans effects. The entropies of activation for the association of PPh 3 to 3a in CHC1 3 (-84 + 47 J K"1 mol"1) and in C 7 H 8 (34 _+ 18 J K"1 mol"1) also reflect the difference in solvation by CHC1 3 and C 7 H 8 of Ru(OEP)CO compared to the activated complex. 3.2.g. The decarbonylation of aldehydes using Ru(OEP)PPh3 (5a). Five-coordinate intermediates 84 such as 5a have been suggested to play perhaps a role in the catalytic decarbonylation of aldehydes by binding some free CO generated upon decarbonylation of the aldehyde.19"21 When Ru(TPP)(P(n-Bu)3)2 (ca. mM concentration) was reacted with a stoichiometric amount of phenylacetaldehyde (PhCH2CHO) in CH2C12> the formation of a stoichiometric amount of C 7 H 8 was detected by GC analysis.21 At lower [Ru], the formation of some Ru(TPP)CO(P(n-Bu)3) was noted from UV/visible spectroscopic studies:21 *-obs Ru(TPP)(P(/i-Bu)3)2 + PhCH2CHO - Ru(TPP)CO(P(n-Bu)3) + PhCH3 + P(n-Bu)3 (3.41) This reaction was found to be only 70% complete because die starting bis(phosphine) complex was slowly oxidized in a side reaction by impurities from die aldehyde to Ru111 products, as judged by UV/visible spectroscopy and cyclic voltammetric measurements.21 While the decarbonylation of PhCH2CHO (3.4 x 10"2 M) by Ru(TPP)(P(n-Bu)3)2 ((1.7-5.1) x 10"5 M) gave straight-line first-order plots for die loss of die bis(phosphine) species, the k o b s values obtained were not reproducible. The addition of trace 0 2 to die reaction mixture did not affect the k o b s value obtained nor its range of irreproducibility; however, the rigorous removal of 0 2 by argon purge led to die complete inhibition of the stoichiometric decarbonylation of PhCH2CHO. The role of the 0 2 in this stoichiometric decarbonylation was tentatively suggested to be the abstraction of a hydrogen atom from the aldehyde, giving C(0)R. The addition of a radical inhibitor (2,6-di-/-butyl-p-cresol) also led to die inhibition of the decarbonylation reaction.21 A similar stoichiometric decarbonylation of PhCH2CHO by Ru(OEP)(CH3CN)2 was noted; however, this system was found to be unapplicable for the catalytic decarbonylation of PhCH2CH0.2 1 These results were used to devise a catalytic decarbonylation system using Ru(TPP)(PPh3)2 (4b), CH 3CN, CH2C12, P(«-Bu)3 and aldehyde. The use of 4b was dictated by the lack of significant dissociation of a phosphine from Ru(TPP)(P(/i-Bu)3)2 in solution, and because of the noted side-reaction involving this complex. Complex 4b is known to dissociate PPh3 in solution, giving Ru(TPP)PPh3 (see Table 3.8). The catalytic system devised involved bubbling 'a small 85 amount' of C O ^ into a CH 2C1 2:CH 3CN (1:60 v/v) solution of 4b (2 /imol) and PhCH 2CHO (2 mmol) to generate Ru(TPP)CO(PPh 3), as judged by UV/visible spectroscopy. The addition of P(«-Bu) 3 (20-40 nmol in total) to the reaction mixture then initiated the decarbonylation reaction. The different complexes presumed to be generated in solution at this point include Ru(TPP)CO(P(n-Bu) 3), Ru(TPP)(PPh 3)(P(n-Bu) 3) (12) and Ru(TPP)(P(n-Bu) 3) 2 via the equilibria discussed in Section 1.2.c. The role of CH 3CN as the solvent for this catalytic system was to enhance the stability of any five-coordinate species generated in situ by occupying the sixth coordination site (as in Ru(TPP)L(CH 3CN)), thereby preventing decomposition of these five-coordinate species. This orange-colored 'soup' was stirred at ambient temperatures, yielding decarbonylated aldehyde and eventually a brown-green colored inactive solution. The irreproducibility observed in the stoichiometric decarbonylation reaction was observed in this catalytic system. 1 9' 2 1 Thus, because the presence of five-coordinate species was deemed necessary for the catalytic system, and because die vacuum pyrolysis of 4b at lxlO" 5 torr and 473 K resulted only in its sublimation- and not in the formation of Ru(TPP)PPh 3, 1 6 a the stoichiometric decarbonylation of an aldehyde by 5a was attempted, widi the aim of studying die catalytic systems once the stoichiometric decarbonylation reaction was well understood. Further, because a stoichiometric amount of toluene was detected from the reaction of Ru(TPP)(P(n-Bu) 3) 2 with PhCH 2CHO at m M concentrations, the stoichiometric reaction of Ru(OEP)PPh 3 (5a) with PhCH 2CHO and PhCHO was studied by 'H NMR spectroscopy, to ascertain which Ru(OEP) complexes were present in this reaction, and to determine if any spectroscopically observable intermediates existed. Solutions of 5a and benzaldehyde in C 6D 6 (see Section 2.4 for further experimental details) under vacuum were found to remain unchanged even after heating at 373 K for 24h. Because the previous stoichiometric system was not always reproducible, 1 9' 2 1 the presence of trace impurities such as air, moisture or free phosphine was thought perhaps to be responsible for the initiation of the stoichiometric reaction. To address these possibilities, 5a and benzaldehyde were mixed with added H 20 or dry 0 2 in sealed rube NMR experiments. Again, no decarbonylation of 86 aldehyde was observed, even when the solutions were heated at 373 K for 22h. When the contents of each sealed tube were exposed to air, the it-oxo dinuclear species [Ru(OEP)OH] 20 (9) was formed, without any decarbonylation of the added aldehyde. In a further experiment, a mixture of PPh 3 and benzaldehyde in the absence of 5a was also tested for decarbonylation of the added aldehyde in the same deuterated solvent, with similar negative results. The same reactivity was observed for the reaction of 5a with PhCH 2CHO in C 6D 6, even when trace 0 2 or H 20 were added. Several important conclusions can be drawn from these experiments: (1) the role of five-coordinate species such as 5a in die stoichiometric reaction may not be critical, (2) the presence of trace moisture or trace oxygen alone is not enough to initiate decarbonylation, and (3) in the absence of trace 0 2, H 20, or other added reagents, no stoichiometric decarbonylation of aldehyde is expected. A system involving Ru(TPP)(PPh 3) 2 (4b) (which has the same K e q as that for the dissociation of PPh 3 from Ru(OEP)(PPh 3) 2 to yield Ru(OEP)PPh 3 (5b) and PPh 3 in solution (see Table 3.8)), was studied using the reaction conditions devised for the attempted stoichiometric decarbonylation of benzaldehyde by 5a, to assess whether die conditions described earlier involving 5a would lead to decarbonylation of benzaldehyde by a mixture of 4b/5b. A solution of 4b and benzaldehyde or phenylacetaldehyde in C 6 D 6 was found, to be unchanged even after being heated at 373 K for 20h in vacuo. The addition of dry 0 2 or H 20 or P(n-Bu)3 to 4b and benzaldehyde (conducted in sealed tube experiments) induced no reactivity. Indeed, in the reaction mixture containing P(n-Bu) 3, a bis-phosphine complex Ru(TPP)(P(n-Bu) 3) 2 can be observed in a 70:30 ratio with Ru(TPP)(PPh 3) 2 (see Table 3.1). This suggests that in a 10:1 ratio of P(n-Bu)3 to 4b, as in the original paper, 1 9- 2 0 Ru(TPP)(P(«-Bu)3)2 (4c) is certainly present as well as Ru(TPP)(PPh 3) 2. 87 Chapter 4. Further studies on the oxidation of bis(octaethvlporphyrinato)ruthenium(II)) (7) by H X acids. 4.1. Introduction. The oxidative addition of HX to coordinatively unsaturated non-porphyrin metal complexes (e.g. of M°) to generate products of the type M n(H)(X)(L) n and M n ( X ) 2 ( L ) n (where L is some ancillary ligand) has been reported previously with both cis- and rrans-isomer products being observed.74 Within metalloporphyrin complexes, examples of the oxidative addition reaction include die reaction of K2[Ru°(OEP)] with CH 3I to give Ru I V(OEP)Me2 (double oxidative-addition),48 and the protonation of [M^OEP)]" generating the hydride complexes M m ( O E P ) H (M = Rli, I r ) . 7 5 ' 7 6 The reaction of excess HX with [Ru(OEP)] 2 (7) in aromatic solvents generates Ru(OEP)X 2 (X= Br, 8a; X = CI, 8b), possibly via a similar oxidative addition route, but the oxidant responsible for the oxidation of Ru 1 1 to Ru I V had not been identified definitively. 3 6' 3 7 The presence of trace X 2 in HX was tentatively postulated as the probable oxidant based on two results: (1) no H 2 was observed to be formed during the reaction, and (2) Br 2 instead of HBr could be used to oxidize 7, again yielding 8a. 3 6' 3 7 Subsequently, more detailed studies were carried out in this present work to identify both the oxidant and the mechanism for the generation of 8 from 7. The stoichiometry between HCI and 7 was measured by a titration of HCI vs. 7 which proved unambiguously that H + was the oxidant in aromatic solvents (Section 4.2.b). Consequently, tests for the formation of H 2 were repeated and others devised to detect H 2; however, all of diese tests indicated that H 2 was not generated during this reaction. These results are discussed in Sections 4.2.h. Varying the HX acids used (HX = H 2, MeOH, H 20, H 2S, CH 3COOH, C 6H 5COOH, HF, CF 3COOH, HN0 3, HBF 4, HCI, HBr, and HSbF 6) allows for an estimation of the strength of acid (i.e. p K J required for die reaction of 7 with HX to proceed. Knowing the pK a required for this reaction in aromatic solvents then allows for die prediction of the possible products utilizing this reaction with other acids. In CH 2C1 2, two intermediates Q3 and 14) are generated rapidly when excess HCI is added to 88 7, both of which yield 8b upon further oxidation with HCI. A much slower formation of these intermediates was found, in fact, to not require HCI, and needed only the solvent; the reactivity of 7 with other chlorinated solvents confirms this solvent dependence. The isolation and characterization (both spectroscopic and chemical) of these intermediates, and their further reaction with HX to yield 8b are presented. The generation of Ru™ products from Ru 1 1 sources other than 7 is also presented in this chapter. Anaerobic oxidation of Ru(OEP)(S) 2 (S = py, 6a; CH 3CN, 6b) by HX (X = Br, CI) was found to give 8a or 8b in quantitative yields and the conditions under which these reactions proceed are documented. Similarly, the HX/7 (X = Br, CI) reaction can be generalized in application to other porphyrin (TMP, TPP) complexes,58 and details for the synthesis and spectroscopic characterization of Ru(TMP)Br 2 are provided. 4.2. Results and Discussion. 4.2.a. Oxidation of 7 by the mixture H 20/0 2. The aerobic oxidation of 7 was postulated to need the presence of both 0 2 and moisture to yield [Ru(OEP)OH] 20 (9), 3 1' 3 5 but this hypothesis has never been proven. A solution of 7 in C 6D 6 (see Section 2.4 for specific details of the reaction conditions) did not react with added N2-flushed H 20 as studied by 'H NMR spectroscopy (Figure 4.1). The 'H NMR spectrum of a solution of similar concentration of 7 in C 6D 6 containing added dry 0 2 indicated that ca. 5 0 % of /x-oxo dinuclear species (9) had formed, but the remaining porphyrin residue was essentially 7. Some change in the position of the C H 2 resonance of 7 at S 25-27 ppm was detected in each spectrum (7 and H 20, or 7 and 0 2) relative to that observed for the same set of protons in 7 in C 6D 6 in the absence of H 20 and 0 2 (ca. 26 ppm, Table 4.1), and these shifts in position are considered to be due to the effects of the added reagents on the bulk solution. When either reaction mixture (7 with H 20, or 7 with 0 2) was exposed to the other reagent ( 0 2 or H 20, respectively), 9 was fully generated instantly, confirming the proposed postulation. 3 1' 3 5 Thus, 7 was stored in a glove-box as a solid and, when needed in solution, the subsequent reaction conditions are rigorously anaerobic but not necessarily anhydrous. The source 89 Figure 4.1: »H NMR spectra of [Ru(OEP)] 2 in C 6D 6 wntaining added (a) 0.1 ftL H 20, and (b) 500 M L 0 2. Spectra were measured at 293 K at 400 MHz (FT), s = C 6H 6. 90 Table 4.1: Summary of 'H NMR spectroscopic data for compounds relevant to the characterization of the products (13/14) from the reaction of rRuOEP)l 2 with CH2CJ2.5 complex solvent i-£H 2 a A C H 2 b i-Hmeso I C H 3 [RuOEP)] 2 > 7 C 6D 6 26.25 11.20 10.18 3.49 7 + H 20 C 6D 6 27.45 11.09 10.20 3.53 2 + 0 2 C 6D 6 25.37 10.98 10.13 3.59 [Ru(OEP)OH)] 20, 9 C 6D 6 4.45 (m) 4.08 (m) 9.45 1.98 13b CD 2C1 2 9.99 5.79 13.00 2.27 CD 2C1 2 10.07 5.78 13.06 2.28 [Ru(OEP)] 2BF 4^ CD 2C1 2 10.05 5.73 13.01 2.22 [Ru(OEP)] 2BF 45 CD 2C1 2 9.92 5.65 12.91 2.15 Ru(OEP)H, 1 4 ^ CD 2C1 2 4.20 (br) -8.20 (br) -3.45 (br) Ru(OEP)Cl-f C 6D 6 14.13 -0.20 -2.20 -0.90 Ru(OEP)Cl(NH 3)- f C D C I 3 11.24 6.19 -5.07 -2.31 The diird component, X£ CD 2C1 2 3.90 9.73 1.82 - Measured at 300 or 400 MHz at 293 K. Unless stated otherwise, the multiplicity each signal of die macrocycle is as follows: C H 2 (quartet, q), H m e s o (singlet, s), C H 3 (triplet, t), br = broad, m = multiplet. ^ Generated from the reaction of [Ru(OEP)] 2 with CH 2C1 2. £ Generated from the reaction of [Ru(OEP)] 2 with HCI in CH 2C1 2. - Generated from the reaction of [Ru(OEP)] 2 with HBF 4 in CH 2C1 2, see Section 4.2.f. - From Reference 32. - From Chapter 5 of this thesis; all signals appear broad. £ Refers to a minor component of the mixture formed by the reaction of CH,Clo HC1/CH 2C1 2 with [Ru(OEP)] 2; see Text, Section 4.2.b. 91 of the water in the dry 0 2 reaction is clearly the C 6 D 6 solvent, despite rigorous attempts to dry die solvent (see Chapter 2). 4.2.b. Measurement of the stolchiometry for the reaction between [Ru(OEP)]2 (7) and HCI. Gaseous HCI dissolved at one atm in C 7D 8 (83 mM corrected for the solvent vapor pressure at 293 K ) 7 7 was added in single equivalents to the dimer 7 in C 7D 8, and the instantaneous loss in intensity of the 'H NMR signals of 7 due to the formation of 8b was observed. Complete precipitation occurred upon addition of four equivalents of HCI to 7, (8a-b are known to, be sparingly soluble in aromatic solvents), 3 7 suggesting a 4:1 stoichiometry between HCI and 7. The product 8b was solubilized by the addition of CDC1 3, and the 'H NMR data confirmed the quantitative yield reported for this reaction. 3 7 The same stoichiometry was measured in the same manner at 243 K in C 7H 8, and in C 6 H 6 at room temperature, indicating that (1) the reaction is instantaneous even at 243 K, (2) no spectroscopically observable intermediates exist and, (3) the same reactivity occurs in toluene and benzene. The titration of a C 7 H 8 solution of H C l ^ vs. 7 in CD 2C1 2 was done in partial equivalents analogously to diat for the same titration in aromatic solvents, but complete rapid loss of the ' H NMR signals of 7 now corresponded to a 1:1 stoichiometry between HCI and 7 at either 243 or 293 K; further, the addition of excess H C l ^ to 7 in CD 2C1 2 yielded the same products. The generated product was a mixture of three porphyrin species, two of which are present in approximately equal proportions and comprise ca. 9 0 % of die mass balance based on the relative integration ratios of the two sets of *H NMR signals (see Figure 4.2). These intermediates were stable in solution for weeks under inert conditions (under N 2 or vacuo in CH 2C1 2), and the two major intermediates are tentatively assigned the formulations [Ru(OEP)],CHCl 9 (13) and Ru(OEP)H ( i i ) . The *H NMR spectrum of the mixture in CD 2C1 2 (see Figure 4.2) suggests that two of the components of the mixture are paramagnetic because the resonances for OEP are not within the usual range of positions for diamagnetic Ru(OEP) complexes. 9' 1 6' 1 7 One component (J_3) has a 92 Figure 4.2: 'H NMR spectrum of the products of the reaction of [Ru(OEP)] 2 with excess HCI in CD 2CI 2 vacuo, at 292 K, 300 MHz. (13) has a *H NMR spectrum very similar to those reported by Collman et al. for [Ru(OEP)] 2 + X" (X = SbCl 6, PF 6, B F 4 ) , 3 2 with the same coupling constant ( J H . H = 7.4 Hz) for die C H 3 (5 2.28 ppm) protons of the ethyl group of OEP (compare the signals for 13 in Figure 4.2 with Figure 4.3a), and two resonances at 10.07 and 5.78 ppm for the anisochronous methylene protons of OEP (see Table 4.1).78 The observation of the J H . H coupling in 13 is significant because the only paramagnetic ruthenium porphyrin complexes, where the coupling is always observed, have a dinuclear structure. No paramagnetic ruthenium porphyrin monomer, including those to be discussed in Chapter 5, has observable J H . H coupling in its JH NMR spectrum. These results indicate that 13 has the probable formulation of a dinuclear species; for example, [Ru(OEP)] 2X. The other component (14) is suggested to have a formulation similar to Ru U I(OEP)X(L) (16a-f. X = Br, CI; L = vacant, NH 3, CH 3CN, see Chapter 5) because the positions of die H m e s o (S -8.20), C H 3 (6 -3.45) and C H 2 (6 4.20) resonances of the OEP macrocycle in 14 are similar in pattern to those for the corresponding protons in 16a-f. The C H 2 resonance of 14 at S 4.20 integrates to the same number of protons (16H) as that for the single methylene resonance of J_3 at 5.78 ppm, consistent with the proposed generation of JJ3 and 14 in equal proportions during the reaction. However, \A has only one broad methylene OEP resonance, instead of the two widely separated (approx. 5 ppm) resonances observed for Ru(OEP)Cl(NH 3), the Ru 1 1 1 halide complex having the smallest difference between die positions of die two methylene resonances of the macrocycle (compare the signals for 14 in Figure 4.2 with Figure 4.3b). Of interest, the 'H NMR spectnim of K + [Ru(OEP)H]~ also has only one methylene resonance, indicating that the hydride ligand does not significantly alter the mirror plane of symmetry offered by the macrocycle. 4 5 b The formulation Ru m(OEP)H for J4 is tentatively suggested based on the Ru 1 1 1 'paramagnetic' type of NMR spectrum observed, and because the methylene 'H NMR shift appears as a broad signal at 293 or 193 K. Because of the paramagnetic nature of J3 and 14, the normally observable proton NMR resonances for the CHC1 2 and H fragments were not detected in L3 and 14, respectively, (nor 94 Figure 4.3: *H NMR spectra of (a) rRu(OEP)]2+ BF4" in CD2C12; the spectrum reported by Collman et al. for the same complex does not have the impurities, labelled V, present, and (b) Ru(OEP)Cl(NH3) in C6D6. The spectra were measured at 292 K at 300 MHz (FT) under an N 2 atm. r n il ' ' H [ t i i i|iiti [ i i H | i H i | i i i i|tivTpTTM " ' n | i M i|"H i;rtn|TiH [ t i t T - [ i"i | i i i i|mTj-n 12 8 A 0 - 4 P P M 111111111111111 ii11111nTTrm11111; i;11111111111!111111111111111111111111; 11111;11111111 n 1111! 11111' 12 10 4 £ 4 i 6 -2 -4 -fe PPM 95 were those for Ru-CDC12 or for Ru-D by 2 H NMR spectroscopy). In diamagnetic complexes, for example, the resonances for the CH 2C1 fragment in Rh m(TPP)(CH 2Cl) can be observed by 'H NMR spectroscopy at -2.81 and -2.82 ppm (JR^H = 2.9 Hz), 7 9 and the hydride resonance for K + [Ru(TPP)H]" was observed at -59 ppm.48 However, that for the C H 3 resonance in Ru n ,(OEP)CH 3 was not observed because of the paramagnetic nature of the Ru 1 1 1 complex. 4 9 The 75 MHz, 1 3C NMR spectrum of 13/14 in CD 2C1 2 (16 mg/mL) showed only broad resonances between -100 and 300 ppm relative to TMS, all of which were too broad to assign; consequently, the resonance for the CHC1 2 could not be assigned. Further chemical characterization data for \3 and 14 will be presented later in tliis Chapter. The diird component (labelled 'x' in Figure 4.2) of the mixture is diamagnetic and, because it has only one mediylene resonance (6 3.90 ppm, quartet), is consistent with a Ru(OEP)L2 formulation, although it is hard to envision what 'L' would be. Subsequent characterization of the mixture will concentrate on the identification of the two major paramagnetic species L3 and 14. The intermediates YS and j 4 were found to react slowly with further HCI in the absence of air to yield 8b only (requiring about two hours for completion), but instantly in the presence of both air and HCI. The analogous reaction of excess HBr with 7 under anaerobic conditions in CH 2C1 2 gave die dibromide analogue 8a directly. During die titration experiments, it was observed that while 7 was indefinitely stable in C 6D 6 or C 7D g under inert conditions at 293 or 373 K, a slow reaction of 7 with CH 2C1 2 was found to occur (vide infra). This result was startling in that the products generated by this slow reaction were spectroscopically identical to those of the in situ species generated by the reaction of excess HCI with 7 in CH 2C1 2. The data unambiguously suggest a dual pathway to die generation of 8b from 7 and HCI, dependent on the type of solvent used, as shown in Scheme 4.1. 4.2.c. The attempted isolation and purification of the intermediates obtained from the reaction of [Ru(OEP)] 2 with CH 2C1 2 and HC1/CH 2C1 2. The intermediates were isolated by transferring a solution of CH 2C1 2 saturated widi HCl^) or neat CH 2C1 2 onto 7 (see Section 2.3.b) under 96 CI 13 Scheme 4.1: Proposed mechanistic scheme for the reaction of HCI with [Ru(OEP)] . vacuum; this instantly formed a reddish solution when HC1/CH 2C1 2 was used, and a subsequent freeze-drying step yielded a brownish-red powder. This solid was then collected and stored in an inert-atmosphere glove-box. This mixture of isolated intermediates dissolved in CD 2C1 2 had the same ! H NMR spectroscopic characteristics as the mixture, formed in situ from the reaction of 7 with HC1/CH2C12. Many attempts were made to separate the intermediates formed, all of which failed. Chromatography was attempted under rigorously anaerobic conditions, but a large portion (ca. 60%) of the reaction mixture adhered irreversibly to the chromatographic support (activity IV alumina, 1 x 10 cm, CH 2C1 2 eluent). The only porphyrin residue collected from the column was found to be a mixture (Figure 4.4), a component of which had JH NMR spectroscopic characteristics similar to those of Ru(OEP)X(L); for example, the C H 3 and H m e s o resonances arc found at -2.2 ppm and -4.6 ppm, respectively (see Table 4.1 and Section 5.2.d.2), and the spectrum was different from diat of the one component J4 of the in situ mixture. Although attempts were made to separate the components of the mixture using other chromatographic supports (silica gel or Florisil) and other solvent systems, all gave similar or worse results. Precipitation techniques were attempted also. The slow addition of n-pentane to a CH 2CI 2 solution of the mixture led to complete precipitation of the porphyrin mixture present, instead of the preferential precipitation of one species. Similar results were obtained using other common solvent mixtures (CH2Cl2/hexanes, THF/n-pentane or OEt2/n-pentane) used for precipitation of porphyrin complexes. The attempted dissolution of a solid sample of die mixture in C 6D 6 led to the formation of a reddish solution with a suspended precipitate. The JH NMR spectrum of the reddish filtrate showed it to contain a Ru m(OEP)X-lilce species (see Section 5.2.d.2) as the major ruthenium porphyrin species of a mixture, and this major species was the same as the product eluted in low yields during the attempted separation of the mixture using chromatography (Figure 4.4). The precipitate from die C 6D 6 solution was collected by filtration and dissolved in CD 2C1 2. The lU NMR spectrum of the CD 2C1 2 solution suggested a mixture also, with Ru(OEP)Cl 2 98 Figure 4.4: 'H NMR spectrum of the product eluted from the attempted purification of 13/14 using chromatography. Spectrum was measured at 293 K, 300 MHz. as the major fraction. Neither the C 6D 6 nor the CD 2C1 2 solutions gave 'H NMR spectra similar to those obtained for the in situ species shown in Figure 4.2, showing that chromatography or selective precipitation techniques do not work in this case. 4.2.d. Further characterization of the mixture 13/14. Because of the difficulty in separating the components of the mixture, subsequent chemical and spectroscopic characterization was done on die mixture in attempts to identify each of the species, which tentatively have been assigned the formulations [Ru(OEP)] 2CHCl 2 Q3) and Ru(OEP)H (14). Microanalysis data obtained for the proposed 13/14 mixture (see Section 2.3.b) formed by the reaction of 7 with HCI do not fit the analysis calculated for a 1:1 mixture of j_3:J4, ignoring the contribution from the diird component of die mixture. However, die microanalysis obtained for the mixture synthesized from die reaction of 7 with HCI in CH 2C1 2 was similar to that obtained for the products from the reaction of 7 with CH 2C1 2 or CD 2C1 2, showing that the same products are formed in all three cases, and diat diese products contain chloride. The highest-mass peak observed in the FAB-MS spectra of the mixture and its deuterated analogue (made by reacting 7 with CD 2C1 2) was at 1282 vale' (Figure 4.5), suggesting a complex of dimeric structure [Ru(OEP)] 2 (1268 vale') having a single carbon hydrocarbon ligand (CH n, n = 1,2) is present (([Ru(OEP)] 2CH 2) +, 1282 mle~). The slight differences in intensity of the signals observed in the spectrum of each sample may be due to differences in sample sizes rather than differences in the samples themselves. Due to the large uncertainty of die peak position (+_ 3 m/e~) at this extreme high-end region of the FAB-MS spectral range, no deuterium incorporation could be discerned. Also, the same mass spectrum showed a signal for [Ru(OEP)] + (634 m/e"), bin no parent peak for [Ru(OEP)Cl] + (669 mle~). An authentic sample of Ru(OEP)Cl, when studied separately by FAB-MS (see Section 5.2.d), always showed the parent mass peak in the FAB-MS. Thus, the mass spectral data also rule out the formulation Ru(OEP)Cl for 14, and are consistent with the presence of [Ru(OEP)] 2CHCl 2 (13) in the mixture. The low resolution EI-MS of the 13/J4 mixture shows peaks for [Ru(OEP)] 2 + (1268 rale'), [Ru(OEP)] + (634 vale'), and HCI (36/38 m/e). 100 Figure 4.5: Fast atom bombardment (FAB) mass spectra of (A) the mixture including 13/14 and (B) the proposed deuterated analogue of the mixture. 1 2 6 8 m/e~ —+ 12 6 8 B 101 No signal due to (CHC1 2) + (83/85 mle) was observed. Of note, the EI-MS of CH,C1 2 also show signals for HCI (36/38 mle'). Thus, the observed signal for HCI in the EI-MS of the mixture could be due to the presence or generation of CH 2C1 2. The measured solution, molecular weight in CH 2C1 2 using the Signer method 8 0 gave a value of 920 _+ 100 g/mole for the mixture, in agreement with the expected value of 993 g/mole for a 1:1 mixture of Ru(OEP)H (635 g/mole) and [Ru(OEP)] 2CHCl 2 (1351 g/mole). The infrared spectra of the mixture as a Nujol mull (Figure 4.6) or a CH 2C1 2 solution had no bands assignable to either a terminal hydride (vRU-H) o r a terminal chloride ( u R l , _ C ] ) . 3 7 : 8 1 Further, the ring oxidation marker band near 1500 cm"1, commonly observed for the n -cation radical in metalloporphyrin complexes,82 was missing indicating that a metal-centered rather than a porphyrin-centered ( O E P + ) oxidation had occurred in forming 13 and 14. The matrix bands for Nujol or C H 2 C i 2 , and die porphyrin ring bands prevented the direct observation of u c . H for Ru-CHC1 2 by IR spectroscopy. The IR spectrum of the mixture of products obtained from 1_ and CD 2C1 2 should have a v c . D band between 2300 and 2000 cm"1 ( u c _ H is normally observed between 3050 and 2800 cm"1) for die proposed CDC1 2 analogue of 13; 8 1 b however, none was detected. Two possible reasons why die t> c_ D band was not detected are that (1) only one C-D bond is present in a sample having ca. 60 carbon atoms, and (2) die region of interest usually has significant interference from the presence of moisture in the sample and the sample cavity. The UV/visible spectrum of die mixture of products in CH 2C1 2 is also missing the ring oxidation marker typified by a broad absorption centered at 580 nm.17 The possible mixed-valence nature of J_3 (Ru n i/Ru n) suggests diat a metal-to-metal charge transfer (MMCT) band may exist as observed in odier mixed-valence dinuclear complexes;83 however, no bands were observed for 13/14 in the 700-1600 nm region. The near-zero specific conductivity measured for the 13/14 mixture in CH 2C1 2 (see Section 2.3.b) shows that 13/14 are non-electrolytes, consistent with the suggested formulations. Attempts to measure the redox potentials for 13/14 by cyclic voltammetry in CH 2C1 2 with 102 Figure 4.6: IR spectrum of tlic mixture 13/14 as a Nujol mull with KBr plates. The IR spectrum of the mixture 13/14 measured using Csl plates showed no assignable band for the u R u - C i m m c 500-250 cm"1 region. -f- -f- -4- -f-1 3BOO. O 3200. O 2BDO. D »DOO- O 17DD. O 1400. O 1 lOO. O WAVENUMBERS <CM-J> BOO. OO SOO. OO 0.1 M tetra-n-butylammonium perchlorate (TBAP) as electrolyte in order to compare the data with those obtained, for example, for [Ru(OEP)] 2 + X" (X = BF 4, PF 6, S b C l 6 ) 3 2 failed due to the reaction of 13/14 with the solvent/electrolyte matrix, resulting in irreversible voltanimetric waves (peak-to-peak separations of > 200-400 mV). 8 4 The reaction of the mixture of products with the solvent system ( 5 % f-butanol in CH 2C1 2) precluded the measurement of the solution magnetic susceptibility using this system; however, that measured using a sample of 13/14 in neat CD 2C1 2 and a formula weight of 920 +_ 100 g/mol gave ne{T (corrected) of 2.0 B.M., consistent with the observation that the product mixture has unpaired spin present (a spin-only value of 1.83 B.M. is expected for a R u m , d 5, low-spin configuration, S = 1/2). The mixture of 13/14 was studied by X-ray photoelectron spectroscopy85 because it was hoped that direct, unambiguous identification for the axial ligand in 13 (CHC1 2) could be obtained. The XPS spectra of die well-characterized complexes Ru(OEP)Cl 2 (8_b), Ru(OEP)Br (16a. see Chapter 5), Ru(OEP)Cl (16b). Ru(OEP)CH 3 and Ru(OEP)py 2 (6a) were used as a database because, to our knowledge, no other studies involving ruthenium porphyrin complexes have been reported. From the XPS spectrum of each sample (see Figure 4.7), the signals due to the ionization of an electron from the Ru 4p, 3d 5 / 2, 3p 3 / 2 and 3p 1 / 2, die carbon Is, and the N Is orbitals were identified. Where possible, die signal from the chlorine 2p orbital and that from the bromine 3p and 3d orbitals were also identified. These assignments were made by comparison of the observed data to those of a general database of non-porphyrin complexes, and are presented in Table 4.2.86 Because of the high ratio of carbon to ruthenium, nitrogen and halide in these complexes, the signal corresponding to the carbon Is orbital was always the highest intensity signal, and always obscured the signal due to the ionization of an electron from the Ru 3d 3 / 2 orbital. Because the signal for die C Is orbital was so intense, it was not possible to detect a specific signal due to the carbon atom of the axial ligand attached to Ru(OEP)CH 3 or that for the proposed CHC1 2 ligand in 13. The CI 2p signal was studied as an alternative because the kinetic energy of the electrons removed from this orbital varies when a chlorine is attached to carbon 104 Figure 4.7: X-ray photoelectron spectra of (a) Ru(OEP)py2, (b) Ru(OEP)CH3, (c) Ru(OEP)Cl, (d) Ru(OEP)Br, (e) Ru(OEP)Cl2 and (f) 13/14. The spectra were externally referenced against the carbon Is signal of polystyrene (284.6 eV), x = trace 0 2 contaminant. ? 1600 • z 80 0-(a)Ru(OEPXpy)2,6a. •Ru3p3/2 \ .013 •Bu3p1/2 5 0 0 4 0 0 B I N D I N G E N E R G Y , e V • C l l 3 0 0 ••160 0 a o to Z ui 800 • (b) Ru(OEPX:H3 •Ru3p3/2 T T 5 00 4 00 B I N D I N G E N E R G Y , e V 105 • C LS 300 (c) Ru(OEP)Cl, lft. I + C l s 120 j *nu3p3/2 M w > - i 1 1 1 — 4 0 0 2 0 0 B I N D I N G E N E R G Y , eV (d) Ru(OEP)Br, J6a. a. u 2 0 _ 1 2 to 4 4 • C l B *3u3p3/2 • 018 \ • N I B • flu 3p 1/2 4 0 0 •Br3p • Br3d —I r 2 0 0 B I N D I N G E N E R G Y , e V 106 (e) Ru(0EP)Cl2, Sb. (f) the mixture of products ircluding 13/14. BINDING ENERGY,eV 107 Table 4.2: XPS binding energies for several Ru(OEP) complexes. binding energies, eV-complex- C Is Ru 3d 5 n N Is Ru 3p 3 / 2 Ru 3p ] / 2 halogen-Ru-py2, 6a 286.2 275.8 400.3 462.7 487.0 -R11-CH3 285.0 275.8 399.7 464.1 486.8 -Ru-Cl, 16b 283.9 275.7 404.4 472.4 489.7 197.3 (CI 2p) Ru-Br, 16a 285.3 275.4 398.6 463.5 485.2 181.2 (Br 3p) 70.4 (Br 3d) Ru-Cl 2, 8b 284.1 271.1 399.7 464.7 486.8 197.1 (C12p) 13/14 284.8 272.3 401.5 467.9 489.7 197.0 (CI 2p) - Excitation energy 1486.6 eV, 3-6 scans per sample, energy steps of 400 meV, errors in binding energies are _+ 0.1 eV. ^ Ru denotes Ru(OEP). - Represents the signal from the appropriate halogen atom and the origin of the electron. 108 (observed at approximately 200-201 eV) or to a metal (197-199 eV). 8 6 The CI 2p signal was found at 197.1 eV for 8b, 197.3 eV for 16b, and 197.0 eV for 13/14. These data suggest that the chlorine is attached to ruthenium rather than carbon, in contrast with the conclusions reached from the NMR and MS data presented previously. No other significant trends in the data can be observed; for example, the Ru 3d 5 / 2 signal from Ru n(OEP)py 2 (275.8 eV) is the same as that from Ru m(OEP)CH 3 (275.8 eV), and is similar to that from Ru m(OEP)Cl (275.7 eV) or Ru m(OEP)Br (275.4 eV). The Ru 3d 5 / 2 signal for Ru I V(OEP)Cl 2 (271.1 eV) would have been expected to be detected at higher binding energy than that for Ru H(OEP)py 2, 8 6 but the reverse occurs. Unfortunately, because of the lack of a larger database for nithenium porphyrin complexes needed for comparison with the data obtained in this present study, the XPS results offer no insight into the structures of the components of die mixture. In summary, die spectroscopic data show diat diree products are generated by the reaction of 7 with HCI in CH 2C1 2 or with CH 2C1 2 itself; two of which (13,14) are formed in equal proportions and comprise 9 0 % of the product mixture (observed by 'H NMR spectroscopy). Further, these two products Q3 and 14) are paramagnetic, with 13 likely having a dinuclear structure (from 'H NMR and FAB-MS data) and a Ru n/Ru m mixed-valance state. The FAB-MS data suggest that a 'CHn' fragment is attached perhaps to 13. The other paramagnetic species (14) is likely a monomelic R u m species not containing chloride (as assessed by FAB-MS and 'H NMR spectroscopy). The solution magnetic susceptibility confirms die presence of paramagnetic species. The component J_3 contains chloride, judged by die microanalysis and 'XPS data to be either coordinated to the metal or within an axial ligand, because the product mixture dissolved in CH 2C1 2 does not generate ionic chloride. These data, taken together, are consistent widi die tentative formulations of [Ru(OEP)] 2CHCl 2 for 13 and Ru(OEP)H for _14. The third component remains unidentified. Unfortunately, no evidence for the coordinated CHC1 2 in 13 or the hydride ligand in 14 was obtained from IR and !H NMR spectroscopy or from XPS studies, preventing the unambiguous identification of the 109 components of the product mixture. Therefore, the chemistry of the product mixture was studied to characterize further the components of the mixture (see Sections 4.2.e and 4.2.f). 4.2.e. Measurement of the kinetics of the reaction of C H 2 C I 2 with [Ru(OEP)] 2. Although kinetic data are best obtained for systems in which the reactants and products are fully characterized,66 it was thought that, in the present [Ru(OEP)] 2/HCl/CH 2Cl 2 system, kinetic studies might establish better the nature of the product mixture, and to clarify the roles of CH 2C1 2 and/or HCI as reactants. The reaction of 7 with CD 2C1 2 was first observed at *H NMR sample concentrations (1-5 mM) and was noted to be slow (approximately one day for complete reaction): Of interest, the H NMR resonances due to 7 did not diminish in intensity as the reaction progressed, but instead broadened and moved with time. For example, the chemical shift for one of the C H 2 signals of 7 was found to change from 26.05 ppm to 9.99 ppm (the position of one of the C H 2 resonances of 13). Concomitantly, the intensity of die signals of J4 increases as the reaction proceeds, and the ratio of the formation of 13:14 from 7 is always 1:1. First order analysis of die change in position of one of the signals of 7 (In Pt-Pf, where Pt is the position of the signal at any time 't' and P f is die final position of this same signal) versus time gave k o b s of 4.3 x 10"5 s"1 at 293 K (see Table 4.3 and Figure 4.8). A similar shift in the resonances of 7 was observed for the titration of 7 with HCI in CD 2C1 2, followed by iH NMR spectroscopy (see Figure 4.9 for details of die experiment); the spectral changes toward formation of 13/14 were now instantaneous. Figure 4.9 shows the change in the equilibrium position of each signal of 7 as partial equivalents of HCI were added to 7. These results indicate that as 7 is converted to the mixture by reacting with CD 2C1 2 or HC1/CD2C12, the remaining amount of 7 is involved in a chemical exchange mechanism widi 13 as in Equation 4.2; surprisingly, complex J4 does not appear to be involved in this exchange. The exhange of Ru(OEP)H (14) with [Ru(OEP)] 2 (7) would k. obs 3/2 [Ru(OEP)] 2 + CH 2C1, 7 [Ru(OEP)] 2CHCl 2 + Ru(OEP)H 13 14 (4.1) 110 Table 4.3: Raw data for the change in the position (5) of the resonances of [Ru(OEP)] 2 (7) with time upon reaction of 7 (4.9 mM) with CD 2C1 2 as measured from the *H NMR spectrum of the reaction mixture in CD 2C1 2 at 293 K. time, min i _ C H 2 a £ C H 2 b £ C H 3 £ H m e s o 0 26.05 11.06 3.39 10.77 75 22.01 9.78 3.10 11.32 90 21.30 9.47 3.30 11.45 105 20.74 9.32 3.00 11.51 120 20.50 9.18 2.97 11.57 150 19.66 8.89 2.93 11.66 180 19.07 8.72 2.88 . 11.78 210 18.49 8.55 2.84 11.85 240 17.98 8.36 2.80 11.93 270 17.45 8.20 2.77 12.01 300 17.00 8.04 2.74 12.07 330 16.47 7.89 2.71 12.14 360 16.03 7.74 2.68 12.20 390 15.59 7.61 2.65 12.26 420 15.32 7.49 2.62 12.32 450 14.63 7.35 2.60 12.38 480 14.54 7.24 2.57 12.43 510 14.01 7.12 2.55 12.48 540 13.46 6.83 2.49 12.61 570 13.10 6.83 2.49 12.61 750 11.41 6.28 2.37 12.78 2160 9.99 5.79 2.27 13.00 111 Figure 4.8: First order plot of ln(Pt-Pf) for the resonance C H 2 a vs. time (min) for the reaction of [Ru(OEP)] 2 with CD 2C1 2. Figure 4.9: Plot of the observed *H NMR chemical shift (ppm) of the individual porphyrin signals of 7 vs. equivalents of added HCI ([7] = 3.1 mM, [HCI] = 91 mM). The titration was conducted under an N 2 atmosphere at 292 K, 300 MHz (FT). £ o. a X o Ld X u u > u CO m • 25 15 -5 -0.2 0.6 LEGEND O CH2 a o C H 2 b ' C H 3 * Hmeso 1.0 -e e — o EQUIVALENTS DF HCI ADDED v s . CRuCDEP)] * Represents chemical sh i f t s of CRuC0EP)]g. * * Represents chemical s h i f t s of CRu<0EP)]2CHClg. 113 [Ru(OEP)] 2CHCl 2 + [*Ru(OEP)]2 ^ [Ru(OEP)] 2 + [*Ru(OEP)] 2CHCl 2 (4.2) 11 1 1 11 require the cleavage of the metal-metal double bond in 7 and the Ru-H bond in 14, perhaps making the process unfavorable. Taking advantage of the noted long reaction time between 7 and CD,C12, one can cool a half-reacted solution of 7 in CD 2C1 2 (16 mg/mL, effectively quenching the reaction) and observe the changes in the *H NMR spectra as a function of temperature (Figure 4.10). Because the two species involved in die chemical exchange mechanism arc paramagnetic, a full analysis of the data by dynamic NMR analysis techniques (as was done for several systems described in Chapter 3) becomes very difficult; however, there is little doubt that a chemical exchange mechanism between 7 and 13 is present. For example, the resonance due to a methylene proton of 7 is observed at ca. 40 ppm at 193 K (see Figure 4.10), but this signal disappears as the temperature is raised instead of shifting upfield, as expected for a neat sample of [Ru(OEP)] 2. 2 5 Similarly, the resonance due to one of die methylene protons of j_3 is observed at approximately 12 ppm at 193 K, and diis signal also disappears as the temperature is raised. In place of bodi of these signals, one observes a broad resonance at about 15 ppm at 303 K, which is partially overlapped with the H m e s o resonance of 13. The C H 3 resonance of 14 at -3 ppm at 303 K essentially varied linearly with temperature over the same range, showing that j_4 is not involved in the chemical exchange mechanism. In principle, the loss of the CHC1 2 fragment from 13 will give 7, and a reaction of CHC1 2 with either 7 or the solvent (i.e. exchange with the solvent as in Equation 4.5) followed by a reaction with 7, will yield 13: 2 [Ru(OEP)] 2CHCl 2 2 [Ru(OEP)] 2 + 2 CHC1 2 114o (4.3) Figure 4.10: 'H NMR spectra measured as a function of temperature for die product mixture obtained from the incomplete reaction of [Ru(OEP)] 2 with CD 2C1 2 in CD 2C1 2 (16 mg/mL). The spectra were measured at 300 MHz in vacuo; arrows define the fate of each signal specified. TE M P.,K 3 0 3 2 8 3 2 6 8 2 5 3 2 4 3 2 3 3 2 2 3 2 1 3 2 0 3 1 9 3 i i ; i i i i i i i | I | I I ! -10 PPM I I I I I I | I I I I 1 I 40 30 i I I I I I I I I I | I I I I I I I I 20 10 115 CHC1 2 + [*Ru(OEP)]2 2 [*Ru(OEP)] 2CHCl 2 (4.4) CHC1 2 + *CH 2C1 2 2 CH 2C1 2 + -*CHC12 (4.5) *CHC12 + [*Ru(OEP)]2 ^ [*Ru(OEP)] 2*CHCl 2 (4.6) 2 [Ru(OEP)] 2CHCl 2 + *CH 2C1 2 + 2 [*Ru(OEP)]2 ^ Net Reaction: 2 [*Ru(OEP)] 2CHCl 2 + CH 2C1 2 + 2 [Ru(OEP)] 2 (4.7) Because of the large excess of CD 2C1 2 present relative to 7 (approximately 10 6 equivalents), the exchange with solvent as described in Equation 4.5 cannot be observed directly. If an exchange of CHC1 2 with the solvent is present, a solution of [Ru(OEP)] 2CDCl 2 in CH 2C1 2 (using the deuterated analogue of 13/14) should give an observable deuterium signal for CD 2C1 2 when studied by 45 MHz 2 H NMR spectroscopy; however, only natural abundance CD 2C1 2 in CH 2C1 2 was detected and not one equivalent of CD 2C1 2 based on a blank of CH 2C1 2/CD 2C1 2. Thus, no exchange with solvent seems to be occurring and, on neglecting a contribution from reaction 4.5, Equation 4.7 is simplified to Equation 4.2. The possibility of exchange of the axial ligand by a radical cage mechanism where J_3 generates [Ru(OEP)] 2 and CHC1 2 as a loose radical pair in CH 2C1 2 seems plausible. 2 3 The added complication of observing a chemical exchange mechanism while studying the rate of reaction of 7 with CD7C1? by !H NMR spectroscopy was avoided by studying the same reaction by UV/visible spectroscopy, because die exchange of 7 with 13 would be too slow to be observed by diis technique.64*1 A typical experiment involved making up a solution of 7 in CH 2C1 2 (0.17 - 4.4 mM) in a UV/visible cell in die glovebox, sealing the cell under an atmosphere of N 2, and then transferring the cell from the glovebox to the spectrophotometer. Because the amount of 7 being weighed out in the glove-box was 0.5 .+ 0.1 mg, the uncertainty in the concentration of 7 is > 20%. The temperature differential between the glovebox and the cell holder of the spectrophotometer was _+ 2 K, and the experiments were conducted at 293 K. Figure 4.11a shows the changes in the 116 Ld CJ <Z • pq Figure 4.11: (a) Typical changes from the initial UV/visible spectrum of (Ru(OEP)I2 (7) to that of the mixture 13/14 with time at 293 K. [Ru 2] = 0.39 mM, [CH 2C1 2] = 15.6 M (neat solvent), time between scans was 84 seconds, (b) Plot of ln(A,-Af) vs. time for the above reaction; the data were measured at 540 nm and 293 K and are presented in Appendix 5. 1.0 -i 0,5 -0.0 480 540 600 nm W A V E L E N G T H TIME, s 118 UV/visible spectrum of 7 in CH 2C1 2 with time and the associated first-order plot of relative absorbance vs. time; the changes in absorbance at 516 and 540 nm were used for the kinetic analysis of the data. Because of the extreme air-sensitivity of 7 in solution, some of the kinetic runs were aborted when the color of the reaction mixture turned in minutes from green to green-purple instead of red; the spectral changes differed from the type shown in Figure 4.11a, and the final spectrum ( A m a x = 565 nm) corresponded to that of the /*-oxo dinuclear species 9.31 Interestingly, if CH 2C1 2 solutions of 7 were handled in the dark, the qualitative rate of conversion of 7 or 13/14 to 9 was considerably slower (hours instead of minutes), while that of formation of 13/14 from 7 appeared to be unaffected by room light. These results show that the oxidation of 7 to 9 by trace H 20/0 2 is light-dependent (Section 4.2.a). While die reaction of 7 with CH 2C1 2 was conducted under die best possible anaerobic atmosphere, the presence of trace air could not be avoided. Thus, some conversion of 7 or 13/14 to 9 was always observed in the UV/visible spectrum of the reaction mixture. Where there was a significant increase in the absorbance of the reaction solution at 565 nm (where 9 has a absorbance maximum)31 compared to that at 540 nm, these data were also discarded; data passing these criteria are presented in Appendix 5. A first-order analysis of die data gave straight line plots of ln(A,-Af) vs. time (see Figure 4.11b), where A, is the absorbance of the solution at eitiier 516 or 540 nm at any time Y and A r is the absorbance of die solution at die end of die reaction. Surprisingly, the k o b s for die reaction decreased as [ 7 ] increased from 0.17 to 4.4 mM while the [CH 2C1 2] was kept constant (Table 4.4, see Figure 4.12), showing that the reaction cannot be a simple overall first-order process because k o b s should be independent of [ 7 ] for a first-order reaction. 6 6 Kinetic studies where a first order rate constant (k o b s) decreased with increasing dimcr concentration have been reported before. 8 7 In one case, 8 7 a the data analyzed well for first-order kinetics within each experiment, but k o b s decreased with increasing dimer concentration, showing that the data only approximate an overall first-order behavior. A dimer/monomer equilibrium was 119 4.4. Summary of data8- for the reaction of [Ru(OEP)] 2 with CH 2C1 2 at 293 +_ 1 K. entry fRu 2l x 104, M rCH2Cl21. M ko b s * 104, s-1 1 1.7 15.6 40 2 1.7 15.6 47b 3 1.9 15.6 46 4 1.9 15.6 4lk 5 2.8 15.6 21 6 3.7 15.6 9.1 7 3.9 15.6 12 8 4.6 15.6 14 9 4.6 15.6 13k 10 5.7 15.6 7.0 11 9.1 15.6 8.0 12 9.2 15.6 7.0 13 9.2 15.6 7.4*-14 13 15.6 3.5 15 44 15.6 1.2 16 6.1 8.9 2.7 17 7.3 2.0 2.0 18 7.3 2.0 1.9*-19 12 0.9 0.5/1. C£ 20 12 0.9 0.3/0.6^ - Data measured at 540 nm used for analysis, +_ 30 % uncertainty in k o b s. - Data measured at 516 nm used for analysis, _+ 3 0 % uncertainty in k o b s. - Data after the '/' mark are adjusted using the data from the above Table to reflect a k o b s when [Ru 2] = 0.6 mM, see Text. 120 ure4.12: Dependence of k o b s on the concentration of [Ru(OEP)] 2 (7) at high [CH 2Ci 2] (15.6 M) at 293 K. 121 suggested for this reaction. In the other system, 7 b was observed also to decrease with increasing metal concentration, and an analysis of the data was proposed for the reaction of a monomer generated from the disproportionation of a dimeric species with added reagent. While both the reactants and the products of the reactions were identified for these two analogous systems, lending credibility to the quantitative kinetic analysis of the data suggested for each system, the products of the reaction being examined in the present study have not been identified. Thus, only a tentative analysis of the data can be provided here. At high [ 7 ], k o b s appears to be independent of [ 7 ], indicating a first-order dependence of the rate on 7. However, at low [ 7 ], a less than first-order dependence of the rate on 7 is apparent (Figure 12). These results were quite similar to those derived from kinetic studies of the reactions described in Equations 4.8 and 4.9:87a k, [HRuCl(PPh 3),] 2 £ 2 HRuCl(PPh 3), (4.8) HRuCl(PPh 3) 2 + PPh 3 -> HRuCl(PPh 3) 3 . (4.9) At low [PPh 3], the k 2 was found to be the rate-determining step; therefore die rate expression was shown to be: rate = k 2[PPh 3][HRuCl(PPh 3) 2] (4.10) k, / {k.,k2[PPh3][Ru],} rate = (4.11) 2[HRuCl(PPh 3) 2] + kj/k.j where {2[HRuCl(PPh 3) 2] + k,/k ,} [Ru]t = [Ru] l o t a l = [HRuCl(PPh 3) 2] x (4.12) (k,/k.i) 122 Thus, for increasing monomer concentration (increasing [HRuCl(PPh 3) 2]), the k o 5 s for this system will decrease because the former term in the denominator of rate expression (Equation 4.11) will be dominant. If a similar equilibrium involving the monomer [Ru(OEP)] and 7 (Equation 4.13) is present for the system being studied, a similar rate expression will apply here, and a decrease in k0t>s a s [ 2 ] increases would be expected: [Ru(OEP)] 2 ^ 2[Ru(OEP)] (4.13) No discernable change (in either A m a x or extinction coefficient, O in the UV/visible spectmm of 7 was observed over the concentration range 10"3-10"5 M in C 6H 5, indicating diat Equilibrium 4.13 lies well to the left, favoring dimerization. The dependence of the rate of the reaction on [CH 2C1 2] could not be measured quantitatively because stock solutions of 7 in C 6 H 6 were unstable even when stored in the glovebox, thus leading to variation in [ 7 ]. Thus, because both [ 7 ] and [CH 2C1 2] are varying in each experiment, a quantitative dependence of the reaction rate on [CH 2C1 2] alone cannot be determined. However, comparison of the observed k o b s obtained for reactions where [ 2 1 w a s similar but [CH 2C1 2] varied shows that k o b s decreased with decreasing [CH 2C1 2]. For example, for [ 7 ] ca. 0.6 mM (entries 10 and 16 in Table 4.4), k o b s decreased from 7 x 10"4 s"1 to 2.7 x 10"4 s"1 as [CH 2C1 2] decreased from 15.6 to 8.9 M, respectively. Also, for [ 7 ] ca. 1.2 mM (entries 14, 19, and 20), k o b s decreased from 3.5 x 10"4 s"1 to ca. 4 x 10 - 5 s"1 as [CH2CI 21 decreased from 15.6 to 0.9 M, respectively. If the measured k o b s values are 'scaled' to a constant [ 2 ] (0-6 +_ 0.1 mM) using the previously determined dependence of k o b s on [ 2 L a linear dependence of k o b s on the [CH 2C1 2] is obtained as a crude estimate (see Figure 4.13). The increasing k o b s with increasing [CH 2C1 2] could in principle be due to either a reaction of 7 with CH 2C1 2 or an impurity in CH 2C1 2. For CH 2C1 2, and for all the halogenated solvents used for synthesis and discussed later in this Chapter except CH 2Br 2 (which was used as obtained), the 123 Figure 4.13: Dependence of k o b s on [CH 2C1 2] at 293 K, [Ru 2] = 0.6 ± 0.1 mM. 8 A [CH 2 CI 2 ] , M 124 procedure for purification involved first elution from a column (alumina, activity I, 3 x 50 cm) to remove acidic impurities, followed by distillation from CaH 2 under N 2 into an flask equipped with a Teflon valve and containing CaH 2. The solution was then degassed by three successive freeze-pump-thaw cycles and stored under vacuo. The presence of impurities was tested for by G C and none was found (HP-1 column (15 m), TCD detector). In light of the quantitative nature of the reaction of 7 with CH 2C1 2, the presence of a stoichiometric amount of impurity being responsible for the observed reactivity at the 10"3 M level after this solvent purification treatment is discounted. Indeed, when the reaction of 7 with undried CH 2C1 2 was studied, the slow formation of the same ratio of products was observed, showing that trace H 20 does not affect the formation or stability of 13/14. Furthermore, the addition of more 7 to the solution at the end of a kinetic run re-initiates the same reaction to generate 13/14 in the same 1:1 ratio as before. These findings suggest that die reaction of 7 with CH 2C1 2 is not caused by the presence of a stoichiometric amount of an impurity in the solvent; catalysis by an unknown trace impurity is not ruled out by this experiment. One possible route toward the synthesis of 13/14 from 7 is that H C I and the C H C I T radical are formed from the homolytic cleavage of the C-Cl bond in C H 2 C 1 2 (Equations 4.14 and 4.15). This route takes into consideration the C-H (412 kJ mol"1) vs. C-Cl (338 kJ mol"1) bond strengths:88 CH 2C1 2 ^ CH 2C1 + CI (4.14) CI + CH 2C1 2 - CHC1 2 + HCI (4.15) the subsequent reaction of HCI with the 14e" species [Ru n(OEP)] yields R u m(OEP ) H and C I , and die reaction of the CHC1 2 radical with 7 gives [Ru(OEP)] 2CHCl 2: HCI + [Ru(OEP)] - Ru(OEP)H 4- CI CHC1 2 + [Ru(OEP)] 2 - [Ru(OEP)] 2CHCl 2 125 (4.16) (4.17) The effect of added HCI may be to promote the formation of Ru(OEP)H (and therefore, CHC1 2). While the addition of HCI to 7 in CH 2C1 2 does give the mixture 13/14 rapidly, the HCI is not a catalyst in the traditional sense because addition of less than a stoichiometric amount of HCI to 7 does not generate the fully formed mixture 13/14. Instead, only partial conversion of 7 to the mixture 13/14 is observed. This result shows that the above mechanism is not correct because HCI is formed as part of the proposed mechanism (step 4.15). Another possible route involves the direct reaction of [Ru(OEP)] with CH 2C1 2 to give Ru(OEP)H and CHC1 2: CH 2C1 2 + [Ru(OEP)] - Ru(OEP)H + CHC1 2 (4.18) Aldiough this mechanism requires the cleavage of the stronger C-H bond instead of the C-Cl bond in CH 2C1 2, the driving force for die reaction could be die formation of a Ru-H bond; M-H bonds are known to be stronger by 60-100 kJ/mol than M-alkyl bonds.89 Further, the radical combination reaction 4.17 (2 is a diradical species, S = l ) 2 5 would be expected to be diffusion-controlled. 5 3 The cleavage of a C-H bond in C H 4 by [Rh(TMP)] has been recently reported, 9 0 3 forming the previously known complexes Rh m(TMP)H and Rh U I(TMP)CH 3. The diamagnetic nature of these R h m complexes allows for the identification of the axial ligands present in each complex. The steric bulk of the TMP macrocycle in [Rh(TMP)] prevents dimerization; dierefore, the monomer can be studied in solution readily. Although the [Rh(TMP)] system is quite different from the [Ru(OEP)] 2 system, the activation of C H 4 by the previous system does provide some precedent for die latter system. Similarly, [Rh(OEP)] 2 has been reported to abstract an H from ((CH 3)C 6H 4R) at elevated temperatures to give Rh(OEP)(CH 2(C 6H 4R)) and Rh(OEP)H (R = H, p-CH 3, m-CH 3). 9 0 b Other examples of metalloporphyrin reactivity with CH 2C1 2 include the reaction of [Rh(TPP)]-, with CH 2C1 2 to yield Rh(TPP)(CH 2Cl), which is thought to proceed via a dissociation of the dimer 126 to the monomer [Rh(TPP)], 7 9 and the reaction of 2 K + [Ru(OEP)]"2 with CH 2C1 2 in THF to give Ru(OEP)(THF) 2 and Ru(OEP)(H 2C=CH 2), via a presumed alkylidene (OEP)Ru=CH 2 intermediate.48 Several other non-porphyrin systems have been reported to abstract a CI atom from chlorinated solvents to give complexes containing products generated from the cleavage of a C-Cl bond of die chlorinated solvent. 9 1' 9 2 This present kinetic study suggests that CH 2C1 2 itself is involved in the reaction with 7 and not some impurity that was initially present or generated in situ; indirectly, these results support the proposed formulations of 13 and 14. 4.2.f. Chemical behavior of the in situ mixture. Unless stated otherwise, all of the following reactions involving 13/14 were conducted under a N 2 atm or under vacuum, and the solvent used was CH 2CI 2 (see Section 2.4 for details of the reaction conditions). Under these conditions, 13/14 are indefinitely stable in solution. The mixture of products 13/14 dissolved in, and reacted in minutes with, CDC1 3 to give initially a mixture with Ru(OEP)Cl 2 (8b) as the major fraction; this was followed by die complete destruction of 8b, as well as die porphyrin ring, as monitored by 'H NMR spectroscopy over two days; 8b itself is indefinitely stable in CDC1 3, showing that radicals are generated in situ upon dissolution of 13/14 in CDC1 3 which lead to the decomposition of 8b. It was hoped that the decomposition of 13/14 in CDC1 3 would generate CH 2Cl 2-like products which could be detected by 'H NMR spectroscopy, but because the spectrum of the decomposed products had ca. fifty signals in the 0-10 ppm region, no readily discernable product signals were observed. The addition of trace H 20 purged with N 2 to 13/14 in CD 2C1 2 caused no reaction; however, when 0 2 and H 70 were added together to the mixture, formation of the n-oxo dinuclear species 9 was noted in minutes by *H NMR spectroscopy. Solutions of 13/14 in CD 2C1 2 were found to be unreactive to added gaseous H 2, dry 0 2, C 2H 4, and N 2, but rapidly reacted with CO to form Ru n(OEP)CO as the major product, along with some 8b and unidentified minor paramagnetic products as studied by ]H NMR spectroscopy. The ability of CO to reduce Ru I V(OEP)Cl 2 rapidly in chlorinated solvents to give Ru(OEP)CO has been noted before, although the mechanism 127 involved is unknown. The lH NMR spectrum of the reaction mixture formed upon adding pyridine to a solution of 13/14 in CD 2C1 2 under N 2 (Figure 4.14), shows that some Ru"(OEP)(py) 2 (6a), Ru m(OEP)X and [Ru m(OEP)(py) 2] + X" species are formed: [(OEP)Ru n-Ru m(OEP)]CHCl 2 + Ru m(OEP)H + py -• Ru n(OEP)(py) 2 + [Ru m(OEP)(py) 2]Cl + Ru r a(OEP)Cl(py) + 'others' (4.19) These three products were identified in the in situ spectrum by comparison of the observed chemical shifts with those reported for Ru(OEP)(py) 2 (6a). 3 7 [Ru(OEP)(py) 2] + CI" (see Section 5.2.e), and Ru(OEP)Cl(L) (X = CI; L = vacant, NH 3, CH 3CN, py, MeOH, see Section 5.2.d). No Ru I V paramagnetic products are formed as judged by the absence of a signal at ca. 60 ppm. If one assumes that pyridine does not oxidize or reduce the precursors it reacts with, but instead " only binds to the metal either via substitution of coordinated ligands or coordination to a vacant axial coordination site, then die products obtained from die reaction of 13/14 with pyridine show that the oxidation states of the ruthenium porphyrin species of the mixture including 13/14 must include Ru 1 1 and Ru 1 1 1. Unfortunately, the signals of die 'liberated' axial ligands of .13 and J_4 cannot be discerned among the myriad of peaks in the 0-12 ppm range. The mixture 13/14 reacted with HCI to give Ru(OEP)Cl 2 within two hours, as judged by UV/visible spectroscopy; however, the reaction was instantaneous when HCl/air was used to oxidize the mixture to 8b. The addition of HBF 4/CH 2C1 2 to 7 under argon gives [Ru(OEP)] 2 + BF 4~ (observed both by 'H NMR (see Figure 4.3a) and UV/visible spectroscopy, with the data in complete agreement with those reported by Collman et al.). 3 2 This route to synthesize [Ru(OEP)] 2 + BF 4" is superior to that devised by Collman et al. because there is no need to control the stoichiometry of addition of HBF 4 to 7. Collman's method requires the stoichiometric addition of AgBF 4 to 7; further addition 128 Figure 4.14: In situ 'H NMR spectrum of the reaction mixture of 13/14 (3 mg) and pyridine ( l /*L) in CD 2C1 2 (0.6 mL) scaled under vacuum, measured at 293 K, 300 MHz. 6 A 2 0 - 2 - 4 PPM of AgBF 4 results in the formation of [Ru(OEP)] 2 2 BF4". Exposure to air of a solution of [Ru(OEP)] 2 + BF 4" in CD 2C1 2 containing excess HBF 4 generates a new in situ species (8c) with a 'H NMR spectrum similar to that of Ru(OEP)Cl 2 (8b) (Figure 4.15, Table 4.5). The in situ 'H NMR spectrum of this species (8c) shows the OEP methyl (6.95 ppm), methylene (59.10) and the H m e s o (10.42 ppm) resonances are similar to those for Ru(OEP)Cl 2 (8b 6 C H 3 = 6.34, C H 2 = 59.75, Hmeso = 8.20).36'37 This new species is tentatively assigned the formulation Ru I V(OEP)(BF 4), (8c): excess HBF 4 air, HBF 4 [Ru(OEP)], -» ' [Ru(OEP)] 2 + BF 4" - 2 Ru(OEP)(BF 4) 2 (4.20) 7 " CH 2C1 2 8c The reaction of HBF 4 widi in situ 13/14 under anaerobic conditions (complete within 2h) or in air (instantaneous) yields the same single product; this product has an lH NMR spectrum identical to that of 8c: excess HBF 4 CH,C1, + [Ru(OEP)], - 13/14 - Ru(OEP)(BF 4) 9 (4.21) 8c Just as HCI reacts with 13/14, die reaction of aqueous HBF 4 widi 13/14 appears to give some 8c. Of note, the only Ru(OEP) complexes (in any oxidation state) which have a single methylene resonance at 60 ppm in their !H NMR spectrum are complexes of the type 8a-b. Further, the in situ spectrum of 8c is very similar to that for Ru(OEP)(SbF 6) 2 (see Section 4.2.i and Table 4.5). Complex 8c was isolated as a solid from the HBF 4/CH 2Cl 2/7 aerobic reaction mixture by the addition of «-pentane; the precipitated reddish solid was collected by filtration, which was then reprecipitated again from CH2Cl2/n-pentane, collected by filtration and air-dried at room temperature. The *H NMR spectrum of the isolated solid indicated that it was impure. The optical spectrum of 8c is dissimilar to diat for 8b and the molar conductance measured for 8c shows it to be a non-electrolyte (see Table 4.5). Unfortunately, die i/ B_ F bands normally observed at 900-130 Figure 4.15: The 'H NMR spectrum of in situ Ru(OEP)(BF 4) 2 (8c) in CD 2C1 2 at 293 K, 300 MHz. The other peaks in the Figure are suggested to arise from other Ru(OEP) complexes, as well as impurities from HBF 4 ( a q ). ' W I H 2 Q i I i i i i | i i i i | i i i i | 2 P P M 0 i i i I i i i i | i i i i 10 I i i i i | i i 8 i i I i i i i | i i i i I i i i i | i i 6 4 Table 4.5: Summary of spectroscopic data for Ru^tporp) complexes. lH NMR spectrum, CDC1 3 complex 6 C H 2 6 C H 3 i-Hmeso UV/visible spectrum, nm-Ru(OEP)Br 2, 8a 60.12 7.10 3.50 365 (sh), 398 (Soret), 505, 535 Ru(OEP)Cl 2, 8b 59.75 6.34 8.20 364 (sh), 398 (Soret), 505, 535 Ru(OEP)(BF 4) 2, 8c> 59.10 6.95 10.42 346 (sh), 384 (Soret), 408, 510 Ru(OEP)(SbF 6) 2, 8d 57.25 6.45 10.03 -Ru(TMP)Br 2, 8e c ~ - 416 (Soret), 518 Ru(TMP)Cl 2, 8f d — — -2 In CHC1 3 or CH 2C1 2. £ 6 1 9 F (8c) = 4.54 ppm vs. («-Bu)4N+ BF 4" (0.00 ppm); 1 1B{ 1H} = 4.28 vs. («-Bu)4N (0.00 ppm). Molar conductance = 4.6 cm"1 M"1 ohm"1, 5 mM solution in CH 2C1 2 at 293 K. - SUm= 13.76, S o -CH 3 = 4.4, c5p-CH3 = 4.01, 5 H p y r r = -44.37 (see Section 4.2.j). ^ <5Hm = 11.21, So -CH 3 = 3.83 5p-CH 3 = 3.52, SH = -52.22 (see Section 4.2.j). 132 1000 cm" , l a are eclipsed by the bands from the porphyrin macrocycle (see Figure 4.6 for a typical example of an IR spectrum of a Ru(OEP) compound). The 282 MHz 1 9 F NMR spectrum of the in situ species Ru(OEP)(BF 4) 2 (8c) (Figure 4.16) is dissimilar to that for (/i-Bu) 4N + BF 4" or HBF 4 ( a q^ in CD 2C1 2, 9 3 showing that the BF 4" ligands in 8c could be coordinated to the Ru. The 96 MHz "B^H} spectra of 8c, HBF 4 and (n-Bu) 4N + BF 4" also confirm this result; 9 4 while the spectra are not shown, the data are provided in Table 4.5. If the B F 4 ligands in 8c (and presumably the SbF 6 ligands in Ru(OEP)(SbF 6) 2 (8d)) are not coordinated to the Ru™ centre, the ensuing compound would tend to dimerize, as for the dimeric species [ R u m ( O E P ) ] 2 + 2 (X") 2. 3 2 Further, based on the MO diagrams used to explain the observed magnetic properties of [Ru n(OEP)] 2 (7), [Ru(OEP)] 2 + and [ R u U I ( O E P ) ] 2 + 2 , 2 5 ' 3 2 both 8c and 8d would be diamagnetic if they are dimeric in structure, in spite of die R u I V oxidation state. The data favor the paramagnetic Ru I V(OEP)X 2 instead of the diamagnetic {[RuIV(OEP)]-> (X"),} + 2 formulation or die 1:2 electrolyte [Ru^OEP)]" 1" 2 (X") 2 formulation for the complexes 8c and 8d. Aerobic oxidation of 13/14 in CD 2C1 2 containing five equivalents of («-Et)4N+ CI" does not yield 8b but a species having a Ru m-like 'H NMR spectrum (-3.6 (br), C H 3 (24H); -7.8 ppm (br), Hmeso (4H); the C H 2 resonance could not be assigned easily because of the presence of many peaks in the 0-10 ppm range, see Chapter 5), showing the need for the presence of HCI to form 8b from 13/14. The role of the H + in the aerobic oxidation of 13/14 in the presence of HCI could be attributed to the stabilization of the 0 2" (as H0 2) formed during the outer-sphere oxidation of 13/14 by 0 2. 2 8 4.2.g. The reaction of [Ru(OEP)] 2 (7) with chlorinated reagents. The inability to determine unambiguously the nature of the axial ligation of the compounds comprising the mixture led to attempts to prepare other complexes of similar character by reacting 7 with a variety of chlorinated solvents. Table 4.6 lists die macrocycle 'H NMR resonances of 7 in various solvents. The difference between the two methylene resonances (CH 2 a-CH 2 b) calculated from Table 4.6 can be used to perhaps judge die 'degree of association' of solvent ligands to 7 at the vacant axial 133 Figure 4.16: , 9 F NMR spectra of (a) HBF 4 ( a q ) and, (b) Ru(OEP)(BF 4) 2. The spectra were measured at 293 K, at 282 MHz, and are referenced against (n-Bu) 4N + BF 4" (0.00 ppm) H B F . H B F 4 8c i I I I ' 1 i I I 1 | I I I 1 | i I I 111 I I I|IiII | I I I 11 1 11 11 I 1 I i|I I I I|I 1 I 11 I I I I | I I I I | I I 1 I | I I I 11I 11 I| I i ( I | I I I I | 1 I I 11 I i i I 7 6 5 4 3 2 1 0 - 1 P P M 134 Table 4.6: *H NMR spectroscopic data for rRu(OEP)l 2 and products generated from the reaction of [Ru(OEP)12 with various solvents.5 solvent * C H 2 a 6 C H 2 b Uimeso 8 C H 3 CD 2C1 2 26.05 11.06 10.77 3.39 C 6 D 6 26.25 11.20 10.18 3.49 C 7D 8 26.25 11.21 9.80 3.59 THF-d 8- 25.72 11.20 9.80 3.36 MeOH-d 4 22.13 9.59 10.76 4.90 11- 9.99 5.79 13.00 2.27 14£ 4.20 (br) -8.20 (br) -3.45 (br) 10.07 5.78 13.06 2.28 CDC13£ 10.23 5.96 13.15 2.40 7 + C C l / 4.2 (br) 10.1 (br) 2.1 (br) - Measured at 300 MHz at 293 K. Unless stated otherwise, the multiplicity for each signal of die macrocycle is as follows: C H 2 (quartet, q), H m e s o (singlet, s), C H 3 (triplet, t), br = broad. - From Reference 45b. £ From die reaction of [Ru(OEP)] 2 with CH 2C1 2. - From the reaction of [Ru(OEP)] 2 widi HCI in CH 2C1 2. J4 was also syndiesized from this reaction, but the data are not shown. £ From the reaction of [Ru(OEP)] 2 widi CHC1 3. - From the reaction of [Ru(OEP)] 2 widi CC1 4. Product is diamagnetic. 135 coordination sites because upon such ligand coordination, the metal is known to move more into the porphyrin plane.5" Although the differences in position (A <5) of die two mediylene resonances (CH 2 a-CH 2 b) in CD 2C1 2, C 6D 6, or C 7D 8 are similar (14.99, 15.05, and 15.04 ppm respectively), those for THF-d 8 (14.52 ppm) and MeOH-d 4 (12.54 ppm) suggest stronger association of these two solvents. In contrast, the resonances for the protons of the OEP macrocycle for Ru(OEP)(py) 2, a six-coordinate complex which is known not to dissociate a pyridine ligand in solution, vary only 0.1 ppm in the solvents C 6D 6, C 7D 8, CD 2C1 2 and CDC1 3 (see Section 3.2.c). While these results suggest that THF-d 8 or MeOH-d 4 coordinate to 7, the possibility that CD 2C1 2, C 7D 8 or C 6D 6 may also be coordinated, albeit weakly, cannot be discounted. For example, dichloromethane as a coordinated ligand in a silver/palladium heterobinuclear complex has recently been reported. 9 5 No reaction was observed between 7 and CU3Cl(^ in C 6D 6, but when 7 was dissolved in CDC1 3, an instantaneous reaction occurred (as judged by die difference in color between the reddish solution formed in CDC1 3 and the greenish color of 7 in aromatic solvents) to give a mixture of species, the major fraction of which had a ]H NMR spectrum similar to that of J_3 (see Table 4.6); of interest, no 14 was detected. The reaction of 7 with CH 2Br 2 also was observed to be instantaneous; however, the 'H NMR spectrum of an isolated crude product from this reaction in C 6D 6 had signals from five different compounds as judged by the number of H m c s o and C H 3 resonances present in the 9-10 and 2 ppm regions, respectively. When 7 was dissolved in CC1 4, an instantaneous reaction ensued, but the product solution did not give a 'H NMR spectrum similar to that of 13/14 (Table 4.6). A l l these in situ products (except those from the 7/CH 2Cl 2 system) were not stable widi time, die 'H NMR spectrum for each system showing ca. twenty signals between 0-10 ppm after a few days. A possible explanation for the complex reactions observed between 7 and CHC1 3, CH 2Br 2 or CC1 4 is that radicals are generated in situ which attack the macrocycle, thus reducing the overall symmetry of die OEP ring. An instantaneous reaction occurred when 7 was dissolved in C 6H 5CHC1 2 at room temperature, and the isolated product, when dissolved in C 6D 6, had an 1H NMR spectnim which showed that 136 the product was a mixture of four diamagnetic products. The same products were detected when two equivalents of C 6H 5CHC1 2 were added to 7 in C 6D 6, the reaction being much slower in this latter case. The four porphyrin species were dissimilar to 13/14. and because of the complicated nature of this mixture, the isolation of these products was not pursued further. 4.2.h. The fate of the H in HX upon oxidation of 7 by HX in toluene and benzene. It was initially thought that H 2 would be generated during the oxidation of 7 by HX in aromatic solvents, but diis possibility was ruled out when initial attempts to detect H 2 by GC failed. 3 7 However, because of the later established 4:1 stoichiometry between HCI and 7 (Section 4.2.b), and because better GC equipment was available, 9 6 these initial experiments to detect H 2 were repeated (see the experimental details presented in Chapter 2); however, no H 2 was detected by GC (see Figure 4.17). Because of the key nature of this negative result, vapor samples taken from the reaction flask containing 7 and HBr in C 7 H 8 were injected into a mass spectrometer to analyze for H2. Again, no dihydrogen was detected in the reaction mixture aldiough blanks indicated that H 2 could be observed easily at the expected levels. To address die possible loss of H 2 from the syringe used to inject the vapor samples from the reaction flask into the gas probe of the MS, a direct sampling apparatus was designed (see Figure 2.1). The use of this apparatus showed again that dihydrogen was not present in the reaction vapor phase. Samples of the filtrate of the reaction mixture from the above study involving the possible detection of H 2, and the HBr/C 7H 8 solution without 7, were injected into both a GC and a GC/MS in attempts to detect the loss of H 2 due to possible solvent hydrogenation. When 7 was reacted with HBr in C 7 H 8 (two attempts), C 6D 6 (once) or C 6 H 6 (one attempt) under the conditions previously described in die experimental Chapter, all die resulting solutions except one of the C 7 H 8 reaction mixtures tested negative for hydrogenated toluene or benzene (see Figure 4.18a/b). In the one exception, the solvent was both hydrogenated and brominated (Figure 4.18c); however, this appears to have occurred as the solution (with and without 7 present) reacted with die silicone septum used to seal the flask. A similar reaction of Ru(OEP)CO in C 7 H 8 with the 137 Figure 4.17: Gas chromatograms of (a) the vapor phase sample from the 7 (85 mg, 67 //mol) and HBr/C 7H g (5 mL) reaction mixture, and (b) the corresponding H 2 control from a H 2/HBr/C 7H 8 mixture (500 /iL sample, 12 /imol). An additional control experiment conducted was that involving addition of H 2 to the 7/HBr/C 7H 8 solution after completion of the synthesis reaction. The GC trace from this last control experiment was identical to that shown in trace (b) below. The reaction of 7 with HBr should generate 134 /imol of H 2. to CO in CM CM A CM B 138 Figure 4.18: Mass spectra obtained from the GC/MS of (A) a 0.5 uL sample of the solvent from the reaction of 7 with HBr in C 7H 8, (B) the C 7 H 8 control, and (C) the hydrogenated/brominated C 7 H 8 obtained upon reaction of C 7H 8/HBr with the silicon septum. i » •M-1 M r C 7 H 8 . i | ' » ' i i ' i 1 i B >» _ B« _ 76 _ IB _ SB _ ie _ 38 _ IB _ II _ f •M-1 LwJl 'I ' " j ' I'" I" i ' i — i — i — p r Iff us i n 171 C »f • f 71 tl it IB 3B 2t I f f 13 M+1 - r - " * i — | — r Iff M+1+Br i n 139 septum was documented in Section 3.2.b. The three other attempts were done in a Teflon valve-sealed flask, and no hydrogenation or bromination of these solutions was observed. A second method of testing for the possible hydrogenation of solvent by the HX/7 mixture was to react a concentrated sample of 7 in C 6D 6 (10 mM) with HBr in a sealed tube, and to study the in situ system (which contained a suspension) by 'H NMR spectroscopy. Even after 60,000 scans (FT), no signals other than those for the solvent and excess HBr were detected. This experiment was repeated using DC1 and C 6H 6, and the resulting "suspension" was analyzed by 2 H NMR spectroscopy, yielding the same conclusion in that only the signal for DC1 was seen (only 5,000 scans were accumulated, however). The reaction of 7 with excess HBr in the presence of two mole equivalents of cyclohexene or 1 atm of C 2 H 4 in C 7D 8 gave Ru(OEP)Br 2 (8a) without any hydrogenation of the added olefin as shown by the ]H NMR spectra of these solutions. A l l these results suggest that, with one exception, no hydrogenation of the solvent or an added olefin occurred during the reaction of 7 with HBr. When HCI was added to a mixture of 7 and 100 equivalents of TEMPO (TEMPO is tetramethyl-l-piperidinyloxy, a free radical trap) 5 3 b in C 6D 6 at NMR sample concentrations, the product was not Ru(OEP)Cl 2 (8b), but instead the reaction gave a ° greenish solution whose UV/visible spectrum ( A m a x = 680 nm) suggested that reduction of a double bond of an OEP pyrrole ring had occurred. These ring-reduced compounds are termed 'chlorins' and have a characteristic sharp, strong absorbance at 680 nm in their visible spectrum.97 This result does imply, however, that the HX/7 reaction in aromatic solvents occurs via a radical pathway, as does the reaction of 7 and HCI or CH 2C1 2. In light of the complicated radical-based reactivity of the 7/HX or 7/HX/CH 2Cl 2 systems, the hydrogen atoms presumed to be formed during the reaction of HX with 7 could be trapped possibly by the solvent and other species formed in situ, diluting die amount of any one product to an undetectable level. Although this explanation is weak, with the experiments performed to date, the fate of the required 'hydrogen' co-product remains unsolved. 140 4.2.i. The effect of the acid HX used. Although B r 2 reacts with 7 (in a 2:1 stoichiometry) to give 8a, 3 7 this synthesis route is inferior to the HX method because addition of more than a stoichiometric amount of Br 2 to 7 results in the bromination of OEP macrocycle. 3 7 As expected, aqueous acids, HX( a q^, also react with 7 under anaerobic conditions to yield monomeric Rn ! V complexes analogous to 8. When X = Br or CI, 8a or 8b respectively were generated, and when X = SbF 6, the new compound Ru I V(OEP)(SbF 6) 2 (8d) was obtained. The 'H NMR spectrum of 8d (sec Figure 4.19) is similar to those of 8a-c. This new product rapidly decomposes in air, yielding Ru m(OEP)(SbF 6) (15) upon attempts to purify 8d by precipitation techniques (see Section 2.3.b); the JH NMR spectrum of 15 is consistent with that reported for Ru(OEP)(SbF 6)THF. 3 7 This latter complex was prepared by reacting AgSbF 6 with 8b in THF under anaerobic, dark conditions; 3 7 presumably 8d is generated first, which then undergoes Ru-F bond homolysis to generate L5. The acids attempted to date and their pK^'s in aqueous phase (unless stated otherwise) arc presented below: 9 8' 9 9 HX: H 2 MeOH H 20 H 2S CH 3COOH C 6H 5COOH pK,: 38s- 15 15 7 4 4 HX: HF C F 3 C O O H HNO3 HBF 4 HCI HBr HSbF 6 pK,: 3 0.2 -2 -3 -7 -9 <-10 - Measured in the gas phase. It was found that excess H F ^ or HF ( a q^ does not react with 7 in C 6D 6 while excess CF 3COOH does and, presumably in the pKj range 3 to 0.2, reactivity of 7 toward HX is first observed. Ii should be emphasized that the pKj scale used here represents estimated values in aqueous solution, 9 8' 9 9 while the studies presented here were conducted in C 6D 6. However, in general, the scale appears to work well for predicting which HX reagents are likely to react with 7. For example, HI (pK,, < -10) 9 8 would be expected to react with 7. An exception to this rule is the reaction of H 2S with 7, which resulted in a green residue that was sparingly soluble in CH 2C1 2 and whose 'H NMR spectrum (6 C H 3 = 1.92 (t); C H 2 = 4.05, 141 Figure 4.19: 'H NMR Spectra of Ru I V(OEP)(SbF 6) 2 formed in situ by the addition of HSbF 6 to 7 in C 6D 6 under N 2. Obtained at 293 K, 300 MHz. The other signals observed in the spectrum arc suggested to arise from impurities from the HSbF 6( a q^ or from other Ru(OEP) complexes formed in low yields. i i i •[ l i i i I i i M I i i T r p i i i I i i i i | i i T T - r r r i i | i i i i i i i n | i i i r - r r r i i | i i i i i i i i i | i i i i i i i i i | i i i i i i i i i | 1 4 1 2 1 0 8 6 4 i 6 - 2 3.97 (m); H m e s o = 10.37 (s); UV/visible spectrum in CH 2C1 2 = 370 (sh), 392 (Soret), 516, 548 nm; compare these data with those provided in Table 4.1) was very similar in pattern (but not in position) to that of the bridging U-OXO dinuclear species [Ru(OEP)OH] 20 (9). A possible product which would generate the observed spectrum is (HS-)(OEP)Ru-S-Ru(OEP)(-SH). The products from the reaction of aqueous CF 3COOH, HBF 4 or H N 0 3 with 7 in C 6 D 6 were in each case found to be a mixture of compounds and, as such, were not identified. The synthesis of 'Ru(OEP)F2' from 7 and H F ^ (the route analogous to that used for 8a-b) was reported during the Master's thesis work, 3 6 but could not be reproduced. From the results of the reaction of 7 with various aqueous acids, it is apparent that Ru(OEP)F 2 cannot be generated by die reaction of 7 with HF (either gaseous or aqueous) in C 6D 6; thus, some other unknown oxidant must be responsible for the observed formation of Ru(OEP)F 2 from 7. Thus the data show diat a HX acid with a pK a less than 0.2 will react with 7, and those that have a pIC, of -7 or lower will react with 7 to form complexes of the type Ru I V(OEP)X 2. The data show diat 7 is a weak base, requiring strong HX acids to initiate reactivity. 4.2.j. Generation of Ru(porp)X2 (8) from Ru(II) sources. While 8a and 8b can be easily prepared from the dimer 7, the HX (X = Br, CI) oxidation of monomeric Ru(OEP)(L) 2 complexes can also yield 8a-b. Solutions of Ru(OEP)(py) 2 (6a) in C 6D 6 react rapidly with gaseous HCI under inert conditions to give an intermediate whose 'H NMR spectrum (6 C H 3 = -1.75; C H 2 = 12.10, 10.22; Hmeso = -3-98, all resonances appear as broad signals; refer to Table 5.4, Section 5.2.d.2) looks very similar to that of Ru(OEP)Cl(py). Further reaction to give 8b occurs rapidly when this mixture is exposed to air. The two step formation of 8b from 6a via the generation of the intermediate can be avoided by removing the coordinated pyridine of Ru(OEP)(py) 2 by heating solid samples of this complex at high temperatures under high vacuum to yield 7.25 In contrast, Ru(OEP)(CH 3CN) 2 (6b) directly yields 8b within six hours upon anaerobic oxidation by HCI in C 6D 6; however, the synthesis of 6b from Ru(OEP)CO(MeOH) always results in some trace, unreacted Ru(OEP)CO(MeOH) contaminant,37 and the use of 6b as a precursor is not the best 143 route to synthesize species such as 8a, 8b. An interesting discovery made during these studies of the oxidation of 6a by HCI was that aerobic oxidation of 6a in the presence of HX yields novel species formulated as [Ru m(OEP)(py) 2] + X" (X = CI, 17a; BF 4, 17b, see Section 5.2.e) instead of 8b or 8c, respectively. Other Ru I V(porp) complexes, as exemplified by the anaerobic oxidation of Ru(TMP)(CH 3CN)-, (6d) by HBr to generate Ru(TMP)Br 2 (8e), can also be synthesized. Figure 4.20 shows the changes observed in the 'H NMR spectrum of Ru(TMP)(CH 3CN) 2 upon bubbling H B r ^ into the solution under anaerobic conditions; Ru(TMP)Cl 2 (8f) can be generated analogously using HCI via the same method (but only the bromide analogue was isolated and fully characterized, see Table 4.5). Surprisingly, the reaction of Ru(TMP)(py) 2 (6cJ with HCI does not yield 8f, in contrast to the findings for OEP system. The TMP complexes 8e and 8f can be also generated by the reaction of HX f e ) (X = Br, CI) with Ru(TMP) or Ru(TMP)(N 2) 2 in aromatic solvents. However, because both Ru(TMP) and Ru(TMP)(N 2) 2 are more unstable than Ru(TMP)(CH 3CN) 2 towards oxidation by trace H 20/0 2, die easiest route experimentally into Ru(IV)TMP compounds is via the Ru(TMP)(CH 3CN), route. The formulation for 8e was based on similarity of the synthesis procedure for 8e to that used for the Ru(OEP)X 2 (X = Br, CI) complexes, and from the microanalysis data. The infrared spectrum of 8e as a Nujol mull using Csl windows offers no details of the bonding of ligands in the axial positions (no w Ru-Br band was observed), while die EI-MS of 8e shows the peaks tor Ru(TMP) (881 vale') and Ru(TMP)Br (960 vale'). The assignments of the 'H NMR resonances of 8e were made by correlating the observed linewidth of each resonance with the distance of the proton from the metal (see Section 2.3.b),64 and by integration. The trend in the change in position of the proton signals of TMP from their position in diamagnetic complexes to those in 8e and 8f is similar to those reported for other Ru I V(TMP) complexes,39 as well as that observed for Ru(TPP)Br 2, 5 a with the exception that additional resonances for die ortho- and para-mediyl protons are present for 8e instead of the 144 Figure 4.20: !H NMR Spectra of (a) Ru(TMP)(CH3CN)2 <6d) in 0,Dt and, (b) Ru(TMP)Br2 (8e) in C6D6. Obtained at 293 K, 300 MHz (FT) in vacuo. CHgCN u —t Ru p y r r CH 3 CN p y r r > C7 H8 o-CH. CH 3CN l l | I I l I | I l I I | I I I I [ I I I I | I I I I | I I I I | M I I | I I I I | I I l l | l I I I | I I I 8 " 6 4 2 0 P P M H, p y r r o-CH, p y r r I I I | I I I I | I II I | I I I I | I I M | -44 -45 -46 m C 6 H 6 P-CH. .•J j M I I | M M I I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I I I I | I -,4 12 10 8 6 4 P P M 145 ortho- and para-protons of Ru(TPP)Br 2. The presence of only one 'H NMR signal for the o-methyl groups of the mesityl group of TMP indicates that the complex is symmetrically substituted in the axial positions. From Figure 4.20b, it is apparent that 8e and Ru(TMP)Cl 2 (8f) (see Table 4.5) are paramagnetic, and the corresponding plot of isotropic shift vs. 1/T for the data measured from 185 to 373 K for 8e in C 7D 8 indicates that only one spin state is occupied over this temperature range (see Table 4.7 and Figure 4.21). The solution Evans' method 5 5 magnetic susceptibility for 8e (2.73 B.M. at 293 K) is consistent with a spin-only value of 2.83 B.M. for a S = 1 system (Ru l v, d 4, intermediate spin-state).56 Thus, these data are consistent with the suggested formulation of Ru I V(TMP)Br 2 for 8e. This compound differs from the OEP analogues (8a and 8b) in that it is unstable toward aerobic oxidation, yielding Ru(TMP)(0) 2 1 4 b' 4 2 within a day in C 7H g, as judged by UV/visible spectroscopy. The OEP analogues (8a and 8b) were air-stable for weeks in solution. Another difference is the thermal stability of 8e in solution. Complex 8e can be heated in C D C 1 3 (or C 7D 8) under vacuo at 373 K for days widiout decomposition, while 8a-b decompose to an insoluble green product when heated to reflux in CDC1 3 after only a few hours. Lastly, 8e-f are quite soluble in aromatic solvents, unlike the OEP analogues 8a and 8b. 146 Table 4.7: Data for the plot of isotropic shift (ppm) vs. 1/T (K'1) for Ru(TMP)Br 2 (Se)A T(K) 1/T(K) «Hm(obs/iso^ «o-CH3(obs/iso)^ 5p-CH3(obs/iso)^ c5Hr„rr(obs/iso)^ 185 5.41xl0-3 17.14/9.88 5.72/3.51 5.28/2.74 -68.18/-76.83 204 4.90 16.20/8.94 5.37/3.16 4.93/2.39 -63.29/-71.94 224 4.46 15.38/8.12 5.09/2.89 4.83/2.30 -58.66/-67.31 243 4.12 14.69/7.43 4.84/2.64 4.40/1.87 -54.22/-62.87 264 3.79 14.09/6.84 4.62/2.42 4.13/1.59 -50.18/-58.83 284 3.52 13.56/6.22 4.43/2.22 4.00/1.46 -46.41/-55.06 303 3.30 13.15/5.88 4.27/2.07 3.88/1.35 -43.24/-52.89 323 3.10 12.77/5.51 4.15/1,94 3.75/1.21 -40.22/-48.87 353 2.83 12.48/5.21 4.04/1.84 3.66/1.12 -37.69A46.34 373 2.67 12.21/4.94 3.94/1.74 3.57/1.03 -35.52/-44.16 °- Obtained for a 5 mM solution of 8e in C 7D 8 in vacuo, _+ 1 K at 300 MHz (FT). - (Observed/isotropic) chemical shift in ppm, diamagnetic correction made from data for Ru(TMP)(CH 3CN) 2: d5Hm = 7.26 ppm, 6 o-CH3 = 2.21 ppm, 6 p-CH 3 = 2.54 ppm, 5 H p v r r = 9.26 ppm. 147 Figure 4.21: Plot of the isotropic shift (ppm) vs. 1/T (K - 1) for Ru(TMP)Br 2. Solid lines represent the least-square fit of the data with extrapolation to 1/T = 0. 148 Chapter 5. Further characterization of Ru(OEP)X2 complexes and synthesis, characterization and and chemistry of several new Rum(OEP) complexes. 5.1. Introduction. In the earlier Masters work,37 the inability to oxidize chemically RuOEP)X2 (X = Br, 8a; CI, 8b) either the metal to the Ru v oxidation state or the macrocycle to a »r -cation radical in 8a-b to generate higher oxidation compounds of ruthenium porphyrins was noted, as was die facile reduction of 8a and 8b to generate Ru m complexes. These results suggested at the time that the reduction potentials for the Ru^/Ru™ and Rurv(OEP+)/RuIV(OEP) couples were both relatively high. Consequently, studies were initiated to measure these reduction potentials, and these results are discussed in Section 5.2.a. Also of interest is the observation diat while the dihalogeno derivatives of RuIV(OEP) are paramagnetic with S = 1 (die first examples widiin monomeric Rurv(porp) complexes), the diaryl and dialkyl derivatives are diamagnetic (S = 0). The measurement of the magnetic properties of 8a over the temperature range 6-300 K was thus undertaken in order to understand more fully the differences in the electronic ground states for these Ru I V complexes. These data, which are presented in Section 5.2.b, suggest that 8a has magnetic properties unlike those measured for other Ru I V noii-porphyrin complexes, in having contributions from effects other than temperature-independent paramagnetism. As discussed in the introduction (Chapter 1) of tiiis thesis, die number of Ru111 coordination complexes of porphyrins are f e w . 1 6 a > 1 7 ' 2 8 > 3 0 ~ 3 2 Sections 5.2.d and 5.2.e outline attempts made to synthesize a variety of Ru m complexes of OEP from 8a-d, utilizing the knowledge of the reduction potential for die Ru^/Ru111 couple. Attempts to generalize the synthetic route first encountered in the reaction of 8a-b with AgX' salts (X' = SbF6, PF 6 and BF 4 ) 3 7 showed that this route was inferior to that developed using amines as reducing agents. Section 5.2.d documents the synthesis, spectroscopic characterization and chemistry of the new complexes RuIll(OEP)X (X = Br, 16a; CI, 16b). which were synthesized by reducing 8a-b with amines. Because these five-149 coordinate, I5e' complexes have been sought as possible catalysts for the oxidation of organic substrates, analogous to the reported behavior of Fem(TMP)Cl,41 particular emphasis was made to study the oxidation chemistry of 16a-b. Section 5.2.e describes the aerobic oxidation of Ru(OEP)(py)2 in the presence of HX acids to generate products of the formulation [Ru(OEP)(py)2]+ X - (X = CI, 17a; X = BF 4, 17b). These complexes are six-coordinate, 1:1 electrolytes. The last section (5.2.0 summarizes the electrochemical data measured for Ru™ and Ru m complexes discussed in this chapter, with an emphasis toward explaining their lack of activity as catalysts for the oxidation of cyclohexene by oxygen atom sources. 5.2. Results and Discussion. 5.2.a. Electrochemical studies of Ru(OEP)X, (X = Br, 8a; CI, 8b). The reaction of 8a± with oxidants such as J W C P B A or PhIO resulted in die complete destruction of the macrocycle, as evidenced by the bleaching of the colored solutions of 8a-b.37 The decomposition of 8a-b upon chemical oxidation by PhIO or mCPBA was surprising and, therefore, cyclic voltammetric studies of these complexes in CH2C12 (see Figure 5.1) were pursued to perhaps understand why. Each voltammogram shows two reversible waves assigned (see below) to the Ru , v/Ru i n and RuIV(OEP+)/RuIV(OEP) couples at roughly -160 and 740 mV, respectively, vs. the internal ferrocene standard: reduction potentials* redox couple X = Br X = CI RuIV(OEP)X2/[Rum(OEP)X2]- -154 -170 [Ru IV(OEP+)X2] + /RuIV(OEP)X2 782 694 * _+ 10 mV vs. Fc+/Fc in CH2C12 in 0.1 M («-Bu) 4N + C104" at 293 K; the Fc +/Fc couple in CH2C12 is at 632 mV vs. NHE. The peak-to-peak separation for each wave is approximately 70 mV, indicating that each redox 150 Figure 5.1: Cyclic voltammograms of (a) Ru(OEP)Br 2 and (b) Ru(OEP)Cl 2. Data measured in 0.1 M («-Bu)4N+ C10 4" (TBAP) in CH 2C1 2 under an N 2 atmosphere, 293 K. X-axis scale is in Volts relative to the Fc +/Fc couple. 151 couple is a reversible, one-electron process. 4 The only other redox datum available for a Ru™ porphyrin complex is diat for the /x-oxo dinuclear species [Ru(OEP)OMe] 20 (9'), which was found to have only one unassigned reduction wave at approximately -1 V vs. Ag/AgCl. 3 1 However, there are many measurements on non-porphyrin ruthenium complexes, with reduction potentials ranging from 555 to -135 mV vs. Fc +/Fc for the Ru^/Ru 1 1 1 couple in complexes with tetraazacycloalkyl- 1 0 0" 1 0 2 and p y r i d y l - 1 0 3 " 1 0 5 based ligand systems; and much of these data fall within the -85 to -135 mV range vs. Fc + /Fc, comparable to the values found for 8a-b. The absence of a delocalized aromatic ring system such as a porphyrin within the tetraazacycloalkyl systems suggests that the porphyrin perhaps plays a minor role in determining the reduction potential of the Ru I V/Ru I U couple. The Fe , v/Fe'" couple for Fe I V(TMP)0/Fe m(TMP)OH + H + was reported as 665 mV vs. Fc +/Fc at 2 3 3 K, but this reduction potential also involves protonation of the Fe 1 1 1 species, and therefore may not be a true estimate of the F e I V / F e m couple. 1 0 6- 1 0 7 The Os^/Os 1 1 1 couples for [OsIV(porp)Br(P(/?-Bu) 3)] +/Os m(porp)Br(P(w-Bu) 3) were reported as -135 mV vs. Fc +/Fc for the OEP system and 3 6 5 mV for die TPP analogue. 1 0 8 The implication of the measured potentials for the Ru^/Ru 1 1 1 couples will be discussed after electrochemical data from several Ru 1 1 1 complexes have been presented. A further confirmation of the assignment of the Ru I V/Ru m couple is that controlled coulometric reduction of 8a by one-electron equivalent at -235 mV vs. Fc +/Fc yields Ru(OEP)Br (see Chapter 2 for the details needed to perform the coulometry experiment and later in this Chapter for details of the spectroscopic properties of Ru(OEP)Br) as monitored by spectroelectro-chemistry (see Figure 5.2). This process was found to be irreversible on attempting coulometric oxidation at 265 mV vs. Fc +/Fc, while the CV data show diat a reversible one-electron process is present. This seeming contradiction can be explained by the difference in timescale of the experiment involved between die CV scan (seconds) and the bulk coulometry ( 3 0 minutes) experiments. If there is appreciable dissociation of a halide from die generated [Ru m(OEP)Br 2]" 152 Figure 5.2: Bulk coulometry of Ru(OEP)Br 2 (8a). Trace A shows the optical spectrum of 8a before coulometry at -235 mV vs. Fc +/Fc in 0.1 M TBAP in CH 2CI 2. Trace B shows the generated optical spectrum, which corresponds to that of Ru(OEP)Br given in Section 5.2.d. wavelength to give Ru(OEP)Br (as in the coulometric experiment), then the reduction potential for this product will be different from that for [Ru(OEP)Br 2]\ The reduction potentials at 800-700 mV vs. Fc +/Fc for 8a-b are assigned to a ring oxidation process. The bulk electrolysis of Ru(OEP)Br 2 (8a) at 900 mV vs. Fc +/Fc in CH 2C1 2 with 0.1 M TBAP electrolyte was followed using a spectroelectrochemical cell. Bulk electrochemical oxidation of the solution with time results in the loss of the intensity of the absorbance of the solution in the wavelength range 300-700 nm (see Figure 5.3a), but the solution retains a greenish brown color after this oxidation, showing that the porphyrin ring remains intact. The ESR spectrum of this in situ product measured at 77 K (Figure 5.3b) showed a signal at g = 2.01 (3200 G, 20 G linewidth), assigned from the position of the signal to a TT-cation r a d i c a l . 2 9 ' 1 0 9 " 1 1 1 Also, a study of the difference in potential for the first ring oxidation ((OEP +)/OEP) and first ring-reduction (OEP/(OEP')) couple for a range of Ru(OEP) complexes shows important trends: 1 1 2 (1) the first ring-oxidation occurs between 965 and 565 mV vs. Fc +/Fc, (2) the first ring-reduction couple occurs near -1435 mV vs. Fc + /Fc, and (3) this difference in potentials is invariably 2000-2400 raV. A reversible potential near the solvent discharge range (ca. -1335 mV vs. Fc +/Fc, not shown in Figure 5.1) is tentatively assigned to the first macrocycle reduction potential for 8a-b, because the difference between this first-reduction potential to that for the first-oxidation potential of 8a-b is approximately 2100 mV. 5.2.b. Further studies on the magnetic properties of Ru(OEP)Br 2. The Ru I V(OEP)X 2 complexes (X = Br, 8a; CI, 8b; BF 4, 8c and SbF 6, 8d) and their organoruthenium(IV) porphyrin derivatives Ru(OEP)R 2 are d 4 systems, but the latter are diamagnetic while the former (8a-d) are paramagnetic (S = 1). Three ground-state d-orbital occupancy types for these d 4 complexes are possible: (a) _ ( L 2 (b) _ d x 2 . y 2 (c) _ d ^ — W _ cL2 _ d z 2 The x/y plane for the above diagrams is defined by that which includes the four Ru-N bonds. The 154 Figure 5.3: (a) Changes observed in the optical spectrum of Ru(OEP)Br 2 (8a) upon bulk coulometric oxidation of 8a at 900 mV vs. Fc +/Fc in 0.1 M TBAP in CH 2CI 2. Time between scans was 6 minutes, (b) ESR spectrum of the solution generated in part (a). 4 0 0 6 0 0 W A V E L E N G T H .nm Figure 5.3b: 9.03 GHz, 10 mW microwave power, 100 G sweepwidth. The spectrum was measured at 77 K as a frozen CH 2C1 2 glass sample. Ru organometallic complexes are diamagnetic and so type (c) occupancy is present. Because X-ray crystallographic data for ruthenium porphyrin complexes containing a variety of axial ligands generally show that tetragonal elongation distortion is present,53 occupancy type (b) is favored for the paramagnetic complexes 8a-d; however, the other type of occupancy (a) cannot be discounted. The magnetic moment of Ru(OEP)Br 2 (8a) measured as a solid sample at 300 K (2.65 B.M., see Table 5.1) agreed well with that from the solution Evans' method measurement (2.55 B.M.) made earlier for the same complex at 293 K. 3 7 However, from the plot of x m a n d Mcrr (corrected) vs. T, (see Figure 5.4) it is apparent that 'non-ideal' behavior for a paramagnet is present because the corrected magnetic moment ( M e f f c o n ) decreases as the temperature drops, and l / x m (which can be inferred from Figure 5.4) appears to deviate from a linear relationship with temperature as predicted by the Curie law: X m = C/T (5.1) In the earlier Masters work, 3 7 the reported Curie plot of isotropic shift vs. 1/T for 8a-b in CDC1 3 for the temperature range 223-323 K showed that significant deviation of the data from ideal Curie behavior was present. Normally, this deviation from die expected theoretical behavior is explained by invoking anti-ferromagnetic coupling between two or more paramagnetic centers,56 leading to a decrease in the magnetic moment for die complex. Antiferromagnetic coupling is common in polymers where there is significant two- and three-dimensional structure present,56 but in porphyrins this effect is usually observed in coordinatively unsaturated complexes, the molecules of which are closely separated in both the solid state and in solution. 1 1 3 These conditions do not apply to 8a because it is coordinatively sadirated, and the steric bulk of the macrocycle prevents two or more molecules from getting close together. An alternative to die anti-ferromagnetic coupling explanation is that the t 2 g orbitals (d x z, d x z, and d x y) of the ruthenium in 8a are closely spaced in energy and that a thermal population 157 Table 5.1: Corrected magnetic susceptibility and moment data for RufOEP^BrT.-T_1K£. y m . x 103 c.g.s. units£ i t c f f . BMA 6.00 16.85 0.90 7.00 15.36 0.93 8.00 14.19 0.95 10.00 12.49 1.00 12.00 11.37 1.04 15.00 10.21 1.11 20.00 9.00 1.20 25.00 8.32 1.29 30.00 7.85 1.37 35.00 7.51 1.45 40.00 7.26 1.52 45.00 7.03 1.59 50.01 6.91 1.66 55.27 6.77 1.73 60.35 6.65 1.79 65.00 6.54 1.84 69.98 6.42 1.90 75.30 6.15 1.92 80.40 6.11 1.98 90.00 6.05 2.09 99.90 5.84 2.16 110.70 5.62 2.23 120.68 5.41 2.28 158 Table 5.1 continued. xm. x 103 C .E .S . units^  iicff. B.M.4 130.30 5.23 2.33 140.68 5.04 2.38 161.48 4.65 2.45 180.00 4.64 2.58 200.00 4.02 2.53 219.90 3.84 2.60 239.93 3.51 2.60 259.90 3.31 2.62 279.87 3.10 2.63 299.82 2.92 2.65 - Data measured by Dr. C.A. Reed, University of Southern California, Los Angeles. ^ +_ 0.01 K uncertainty in temperature. - +_ 0.01 x 10"3 c.g.s. units uncertainty in die magnetic susceptibility. ^ _+ 0.01 B.M. uncertainty in the magnetic moment. 159 Figure 5.4: Plot of M efr (corrected) and x m (corrected) vs. T for Ru I V(OEP)Br 2 (8a). 0.02 o o o . o eff CD 1 • o o • o o • o 9 o • CO W 0.01 CD o E m T r 100 200 T, K 0.00 300 160 of these orbitals exists such that the population represented by types a and b decreases with temperature and the population represented by type c occupancy increases. This would lead to a decrease of the magnetic moment as the temperature is lowered. There are many examples of this type of behavior in porphyrin and other macrocyclic systems of i r o n . 1 1 4 ' 1 1 5 The presence of spin-orbit coupling in 8a will alter these d-orbital levels such diat die newly-generated 'J' levels are separated in energy by the amount indicated by the spin-orbit coupling constant. 1 1 6 Because this value is large (700 cm"1 for Ru™), 1 1 6 the possibility of a spin-state thermal equilibrium in 8a can be ruled out. The reported magnetic susceptibility of the Ru™ complexes K 2RuCl 6, (NH 4) 2RuCl 6, Rb 2RuCl 6, K 2RuBr 6 and K 2RuF 6 show that they are of the intermediate spin state (d 4), S = 1 at 300 K."7 Variable temperature magnetic susceptibility measurements indicated that there was little variance of x m with temperature (300-80 K) for the non-porphyrin compounds,117 whereas xm for 8a varies greatly with temperature (see Figure 5.5), showing that the R u I V non-porphyrin complexes have a significant contribution from temperature-independent paramagnetism (TIP). In 8a, the contribution from TIP must be smaller than that for die non-porphyrin R u l v complexes. Indeed, a plot of / i e f f vs. T 1 / 2 (as in /x e f f = 2.83(x m x T ) 1 / 2 , see Figure 5.6) shows a linear dependence except for die high-temperature region, where there is evidence that a contribution from TIP is present, the extent of which cannot be easily estimated because of a lack of data generally within ruthenium(IV) systems. Thus, while the magnetic behavior for 8a remains unexplained, the contributions from thermal spin equilibria or temperature-independent paramagnetism have been largely discounted. Further discussion of the data will have to await the availability of a larger database for Ru systems. An additional factor that could effect the relative energy levels of die ruthenium d-orbitals is the extent of the 7r-charge transfer from the axial ligands to the metal, because this would increase the energy of the d ^ and d ^ orbitals of the t 2 g set without affecting the energy level of the non-bonding d x v orbital. The diamagnetic nature of the organoruthenium(IV) analogues 161 Figure 5.5: Plot of x m (corrected) vs. T for several Ru(IV) complexes including K2RuF6, K 2RuCl 6 > (NH4)2RuCl6, Rt^ RuClg, K2RuBr6 and Ru(OEP)Br2 (8a). OT *C OT o ro O 6 5 4 H 3 4 LEGEND o KgRuF^ D KgRuCt6 Z 2 <NH4) RuCl6 A KgRuBr6 x Ru<DEP>Br0 x o A A O • A A X • 150 250 350 TEMPERATURE, K 162 Figure 5.6: Plot of Tm vs. *»eff (corrected) for Ru(OEP)Br 2. 163 suggests that less »r-charge density is transferred to the metal from the aryl and alky] ligands (an expected result, considering the er-only (alkyl) and weak rr -acceptor (aryl) nature of these ligands when they are bound to Ru) compared with that transferred from the halide axial ligands to ruthenium. Normally, from the 'H NMR spectrum of each type of rudienium porphyrin complex, one can qualitatively assess the extent of porphyrin-to-ruthenium charge transfer; 1 1 8 but in this case, the fact that the organometallic complexes are diamagnetic whereas die coordination complexes are paramagnetic makes even a qualitative estimate impossible. 5.2.c. Attempted generation of other Ru^OEPJX'j complexes (X' = BF 4, 8c; SbF 6, 8d, and PF 6) by metathesis reactions of Ru(OEP)X 2 (X = Br, 8a; CI, 8b) with AgX' salts. Because of the changes in the magnetic properties of these ruthenium porphyrin complexes with changes in the axial ligand coordination, the generation of Ru™ complexes other than 8a-b was attempted to ascertain if there was an explainable trend in magnetic behavior. One route previously attempted was to exchange die axial ligand of 8a-b by metadiesis with AgX' (X' = SbF 6, P F 6 and BF 4); 3 7 the other was to generate Ru(OEP)X' 2 species from the dimer [Ru(OEP)] 2 (7). In Chapter 4, the synthesis of some of these Ru I V(OEP) complexes from 7 was accounted. Here, the details of the reactions of 8a-b with AgX' salts are presented, hi die Masters study, 3 7 the attempted metathesis of 8a;b with AgSbF 6 in THF was shown to give Ru i n(OEP)(SbF 6)THF, and it was suggested that this reduction of Ru I V to Ru 1 1 1 occurred either via the oxidation of Ag 1 to Ag 1 1 or via the homolysis of die Ru-F bond within Ru-SbF6 by some undetermined manner. The auto-reduction of Ru(OEP)(SbF 6) 2 to give Ru(OEP)SbF 6 (15) was documented in Chapter 4; therefore the route for the reduction of Ru™ to Ru 1 1 1 is considered to be via bond homolysis and not via the oxidation of Ag 1. In testing the generality of this AgX' metathesis reaction, it is observed that facile reduction occurs in two of three cases (with X = P F 6 and SbF 6). From analysis of the 'H NMR spectrum of the reaction mixture of two equivalents of AgPF 6 with 8a in CDC1 3, it was apparent that two major products were formed in a 2:1 ratio. The ]H NMR spectrum of this reaction mixture (24.5 ppm, H m e s o (4H); 18.4, C H 2 (16H); 0.0, C H 3 (24H); and 32.7, H m e s o (4H); 10.1, C H 2 164 (8H); 5.6, C H 3 (24H); -7.0, C H 2 (8H)) suggests the formation of R u m species (see data presented later in this Chapter). The components of this mixture had similar solubilities (soluble in chlorinated solvents but not in aromatic solvents), and could not be separated by precipitation techniques; nor could they be separated by chromatography due to decomposition of the mixture on die column (alumina, activity I, neutral, 1 x 10 cm, CH 2C1 2 eluent). In the reaction of 8a with AgBF 4, Ru r v(OEP)(BF 4) 2 (8c) was generated as evidenced by the 'H NMR spectrum of the reaction mixture; this spectrum was identical to that generated by adding HBF 4 ( a q ) to the 13/14 mixture formed by the reaction of [Ru(OEP)] 2 widi CH 2C1 2 (Chapter 4). The latter route is preferable for die synthesis of 8c because the difficult task of separating the Ag products from the reaction mixture of 8a-b with AgBF 4 can be avoided. Thus, the best route to syndiesize Ru I V(OEP)X' 2 (X' = SbF 6, B F 4 and presumably PF 6) is via oxidation of 7 with HX' acids in aromatic solvents, and not by the reaction of 8a-b with AgX' salts. 5.2.d. Synthesis, characterization and chemistry of Ru(OEP)X(L) (X = Br, L = vacant, 16a; X = Br, L = NH 3, 16c; X = Br, L = CH 3CN, 16e; X = CI, L = vacant, 16b; X = CI, L = NH 3, 16d; X = CI, L = CH 3CN, 16f). 5.2.d.l. Synthesis of 16a-b. Attempts to react 8a-b with NH 4 +OH" in order to prepare the dihydroxy analogue Ru(OEP)(OH) 2 gave instead new products having spectroscopic characteristics like diose of Ru(OEP)R (R = Me, Ph). 5 a' 4 9 Subsequently, these new products were identified as Ru(OEP)X(NH 3) (X = Br, 16c; CI, 16d). Comparison of the known potential of the N 2H 4/NH 4 + couple in basic pH solutions (-0.10 V vs. S C E ) 1 1 9 with those of die Ru I V/Ru H I couple in 8a-b (ca. 0.35 V vs. SCE) indicates that 8a-b could be reduced to R u m products by NH 3 or NH 4OH: Ru(OEP)X 2 + 2 NH 3 - Ru(OEP)X(NH 3) + 1/2 'N2H4' + 'HX' (5.2) 165 Figure 5.7 shows the changes in the *H NMR spectrum of Ru(OEP)Br 2 (8a) upon addition of about a five-fold excess of a solution of N H 3 in CDC1 3 (19.3 mM N H 3 in CHC1 3 at 293 K ) . 1 2 0 Although the larger-scale syntheses were conducted in C 6 H 6 or C 7H 8, CDC1 3 was used for the NMR study because both reactants and products are soluble in CDC1 3 while die reactant 8a is sparingly soluble in C 6D 6. The chemical shifts of 8a are observed at approximately 60 (CH 2), 7 (CH 3) and 4 ( H m e s o ) ppm in CDC1 3 (Figure 5.7a). As the reaction progresses at 293 K, the loss of the intensity of these signals occurs with a concomitant increase in the intensity of the product signals at S 11.4 (CH 2 a), 6.1 (CH 2 b), -2.2 (CH 3) and -5.4 ( H m e s o ) ppm (Figure 5.7d). The presence of roughly an equal mixture of reactant and product is observed after approximately 215 minutes at 293 K (Figure 5.7c), suggesting that the reaction would be complete within 1100 minutes at 293 K, as is clearly observed (Figure 5.7d). If the ratio used for a reaction of NH 3 with 8a was less than 2:1, incomplete conversion of 8a to Ru(OEP)Br(NH 3) (16c) occurred; for example, when a 1:1 ratio of NH 3 to 8a was used, 5 0 % conversion of 8a to 16c occurred. The products of the reaction of 8a widi NEt ( 3. n)H n (n = 0, 1, 2) had NMR spectra similar to those from the reaction of 8a with NH 4OH or NH 3; however, these alkylamine reactions were not pursued in favor of using NH- , as the reductant. The Ru(OEP)Cl(NH 3) complex (16d) was synthesized in the same manner as Ru(OEP)Br(NH 3) by reacting N H 3 widi Ru(OEP)Cl 2 (8b). The data suggest that one equivalent of NH 3 is required to reduce Ru(OEP)Br 2 to give Ru(OEP)Br (presumably the slow step for this reaction), which immediately coordinates the N H 3 present in solution, to give Ru(OEP)Br(NH 3): slow fast Ru(OEP)Br 2 - Ru(OEP)Br - Ru(OEP)Br(NH 3) (5.3) 8a N H 3 16a NH 3 16c The co-product from this reaction was not identified, but may be N 2 H 4 or N 2 H 5 + B r , both of which could be the one-electron oxidation products of NH 3. The five-coordinate species Ru(OEP)X (X = Br, 16a) was generated from 16c by protonation of the N H 3 ligand of J6c with HF ( a q^ under anaerobic conditions in CHC1 3, with the generation of 166 Figure 5.7: 'H NMR spectra of (a) Ru(OEP)Br 2 (5.32 mM) in CDC1 3 > (b) die reaction mixture 5 (a) minutes after addition of N H 3 (five mole equivalents), (c) after 215 minutes, and (d) after 1170 minutes. The experiment was conducted aerobically, at 293 K and the arrows indicate the fate of each signal as the reaction progresses. i a i CH. i m H20 0 m in (b) 5 min 1 JUL T JL (O (d) 1 i i imni | i ) i i | i i i i | i i i i | i i i i | i i i i | i i i i | i i i i | i i i i | 64 58 I 1 1 i w 2 1 5 min 1170 min | i i i i [ i i i i | i i i i [ i i i i | i i i i | i i i i | i i i i | i i i i | i i i i j i i i i | i i i i | i i i i | i i i i | i i i i | ' ' ! ' j i i i : | i i i i j n i i | i i i n i i n 10 5 6 ~ 5 167 NH 4F. The corresponding reaction of J6c with HF in air also gave 16a, but some Ru 1 (OEP) complex was also observed in the 'H NMR spectrum of the reaction mixture in CDC1 3. This R u I V complex was not Ru(OEP)Br 2, and may indeed be Ru(OEP)F 2 (6 C H 2 = 53.3; C H 3 = 6.7; H m e s o = 10.3 ppm, see Section 4.2.h). The chloride analogue Ru(OEP)Cl (16b) was syndiesized in the same manner from Ru(OEP)Cl(NH 3) (16d) by reacting 16d with aqueous HF under anaerobic conditions. 5.2.d.2. Characterization of 16a-f. The microanalyses (C, H, N, and halogen) for the compounds 16a-f (halogen analysis was not done for 16d) are presented in Table 5.2, and are consistent with the proposed formulations for each complex. In the case of Ru(OEP)X(NH 3) and Ru(OEP)Cl (CH 3CN), these products were found to have a broad IR band between 3700-3400 cm" attributed to non-coordinated H 20. The compounds 16a-b were found to be non-electrolytes in CH 3CN solution, giving near-zero molar conductance at mM concentrations at 293 K (see Table 5.2). The EI-MS of Ru(OEP)Br(L) (L = vacant, 16a; NH 3, 16c; CH 3CN, 16e) show peaks for Ru(OEP)Br (713 m/e"), Ru(OEP) (634 m/e") and the decomposition peaks for Ru(OEP); the parent peak for either Ru(OEP)Br(NH 3) (730 mle) or Ru(OEP)Br(CH 3CN) (754 mle) was not observed in the EI-MS or FAB-MS. The same pattern in the EI- and FAB-MS spectra of Ru(OEP)Cl(L) (L = vacant, NH 3, CH 3CN) was observed, with die parent peak for Ru(OEP)Cl (669 mle) being the highest-mass signal. Again, no parent peak for Ru(OEP)Cl(NH 3) or Ru(OEP)Cl(CH 3CN) was detected. The FAB-MS of Ru(OEP)Br (16a) and Ru(OEP)Cl (16b) are shown in Figure 5.8. The UV/visible spectra of 16a-f show broad Soret band absorbencies centered at 390-400 nm and weak visible bands near 500 and 530 nm (see Figure 5.9 and Table 5.3). Within one type of complex (i.e. _16a and 16b or 16c and 16d) the optical spectrum is identical in shape and intensity for each halide analogue. The UV/visible spectrum of Ru n(OEP +)CO(Br) (d 6) has a fairly intense visible band at 630 nm (log e = 4.04), 1 7 b and C o m ( O E P + ) B r 2 (d 6) has a similar band at 680 nm (log £ = 3.18), 1 0 9 , 1 1 0 suggesting that if 16a-f were complexes of the type R u n ( O E P + ) X , a characteristic n -cation radical band between 630-680 nm would be observed, along with a low-intensity broad Soret band. Clearly, these features are not observed in die optical spectra of 168 Table 5.2: Microanalyses and conductivity data for several Ru m(OEP) complexes. microanalyses data (found/calculated) complex- %C %H % N % halogen c o n d u c t i v i t y -RuBr, 16a 60.83/60.59 6.27/6.17 8.03/7.85 10.81/11.08 5 RuCl, 16b 64.32/64.57 6.51/6.58 8.30/8.37 5.25/5.23 7 RuBr(NH 3).H 20, 16c 57.43/57.75 6.34/6.55 8.89/9.35 10.13/10.55 7 RuCl(NH 3).H 20, 16d 61.51/61.10 7.00/6.93 9.62/9.90 — ~ RuBr(CH 3CN), 16e 57.37/57.71 6.16/6.35 9.31/9.07 9.87/10.23 -RuCl(CH 3CN).H 20, 16f 62.86/62.64 6.68/6.73 9.90/9.62 5.21/4.81 — 5 Ru denotes Ru(OEP). ^ In units of M"1 cm ohm"1 at 293 K. Table 5.3: Summary of UV/visible spectroscopic and magnetic moment data for Ru m(OEP)  complexes. complex- A m a x (log e) nm*- corrected nef£ RuBr, 16a 396 (4.61), 500 (3.77), 538 (3.75) 1.78 RuCl, 16b 395,500,538 2.03 RuBrfNHQ. 16c 385 (4.41), 402 (4.52), 512 (3.79), 550 (sh) 1.92 RuCl(NH ?). 16d 383,402,513,550 RuBr(CH 3CN), 16e 381 (sh), 396 (4.60), 519 (3.74), 538 (sh) RuCl(CH 3CN), 16f 382,396,519,538 S-Ru denotes Ru(OEP). - Measured in C 7H 8. - Measured using 5 % f-butanol in C 6D 6 or C 7 H 8 at 293 K, in units of Bohr magnetons (B.M.). 169 Figure 5.8: FAB-MS of (a) Ru(OEP)Br (16a), and (b) Ru(OEP)Cl (16b). 6 3 4 I 71 3 T 1 1 1 1 1— •80 6 3 0 670 H I / * " 170 Figure 5.9: UV/visible spectra of (a) Ru(OEP)Br (26.4 /tM), (b) Ru(OEP)Br(NH 3) (28.8 /JM) , and (c) Ru(OEP)Br(CH 3CN) (29.2 uM) in C 7H 8, cell padilength = 1.00 cm. WAVELENGTH of 16a-f. Further, 16a-b are EPR silent at either 293 or 77 K as 2-4 mM C 7 H 8 solutions or C 7 H 8 glass samples, respectively. 1 1 1 The EPR spectra of known ir-cation radical metalloporphyrin complexes have a sharp signal at g = 2.00 at either 293 or 77 K,17,109'110 assigned to the presence of an organic free radical. Lastiy, Evans' method 5 5 determination of the solution magnetic susceptibility of 16a (1.78 B.M.), 16b (2.03 B.M.) and 16c (1.92 B.M., see Table 5.3) using 5 % f-butanol in C 6 H 6 or C 7 H 8 agree well with the spin-only value of 1.83 B.M. expected for a R u m (d 5, S = 1/2) oxidation state. 5 6 The infrared spectra of Ru(OEP)X(NH 3) (16c and 16d) show the v N _ H stretch as a broad band (+_ 60 cm - 1) centered about 3100 cm'1, as expected for M-NH 3 complexes, 8 1 8 and a broad band between 3700-3400 cm"1 for the i/^ of free H 20. 8 1 The IR spectra of Ru(OEP)X(CH 3CN) (16c and 16f) have bands at ca. 2260 cm"1 for vc — N of the coordinated CH 3CN, 8 1 a while the IR spectra of Ru(OEP)X (16a and 16b) are missing the "N.H. "OH v c = N bands, as expected. The IR bands for f M . x could not be assigned. The IR spectra of 16a-f had no ring oxidation marker bands at 1550 cm - 1, 8 2 consistent with the absence of a porphyrin cation-radical and the -presence of a Ru i n(OEP)X(L) formulation. The 'H NMR spectra of JjSa and 16b in C 7D 8 (see Figure 5.10, and Table 5.4) have four signals assigned by integration to two types of methylene protons (6 14 and 0 ppm), and one signal each for the methyl (-1 ppm) and H m e s o (-2 ppm) protons of the OEP macrocycle. The presence of the paramagnetic Ru 1 1 1 center causes the normal relaxation times for the protons of the OEP macrocycle to shorten, 1 2 1 the effects of which can be seen in the *H NMR spectra of 16a-f. resulting in broad (100-200 Hz fullwidth at half-height) resonances. Therefore, the J H . H coupling (ca. 6-10 Hz) for the ethyl protons of OEP normally observed in diamagnetic Ru(OEP) complexes are completely eclipsed in 16a-f. The spectra show that the mirror plane of symmetry of the porphyrin is not present in 16a-b. the molecules having two dissimilar axial ligands, and therefore, two methylene resonances.78 The 'H NMR spectra of 16a-f and of many other Ru i n(OEP) compounds are similar in pattern, but not necessarily in position (see Table 5.4). 172 Figure 5.10: 300 MHz, >H NMR spectrum of Rtf(OEP)Br (16a) in C 6D 6 at 293 K. The spectrum of Ru(OEP)Cl (16b) is very similar to that of the bromide analogue. Br Table 5.4: Summary of *H NMR spectroscopic properties of Ru m(OEP) complexes.-complex- 5 Hmeso * C H 2 a 6 CH, b 6 C H 3 RuBr, 16a -2.61 14.15 -0.20 -0.90 RuCl. 16b -2.11 14.13 -0.16 -0.93 RuBr(NH 3), 16c -5.29 11.40 6.10 -2.28 RuCl(NH 3), 16d -5.07 11.24 6.19 -2.31 RuBr(CH 3CN), 16e -5.67 10.92 5.80 -2.75 RuCl(CH 3CN), 16f -5.85 11.15 5.78 -2.70 RuMe£ -0.24 11.62 5.70 -1.78 RuPh- 1.14 14.13 5.81 -1.02 Ru 2, 7 3.49 26.25 11.20 10.18 RuBr(PPh 3), ip> 9.90 18.52 8.80 0.53 RuBr(py)^ -5.98 8.50 4.95 -2.75 [Ru(py)2] + Br"-f -1.82 18.18 -0.22 RuCl(MeOH)2 -5.31 11.15 7.05 -2.40 ( N H 4 ) + [RuBr2]-, 1 6 ^ 4.03 59.68 6.69 - Unless stated otherwise, the spectra were measured at 293 K in C 6D 6, 300 MHz. ^ Ru denotes Ru(OEP). - From References 5a and 49. - From Reference 28. - Spectrum obtained in C 5D 5N, see Section 5.2.d.3. - Spectrum obtained in CDC1 3, see Sections 5.2.d.3 and 5.2.e.2. s Spectrum obtained in CD 3OD, see Section 5.2.d.3. - Spectrum obtained in CDC1 3 > see Section 5.2.d.3. 174 The separation of the two methylene resonances (about 14 ppm) in the H NMR spectrum of Ru(OEP)X (16) is similar, for example, to the separation of the two methylene resonances of the dimer [Ru(OEP)] 2 (7) (about 16 ppm),25 or Ru(OEP)Br(PPh 3) (about 10 ppm). 2 8 , 3 7 The presence and position of the two methylene resonances of J_6a and 16b (ca. 14 and 0 ppm) are quite different from that of the single methylene resonance observed at 60 ppm for Ru(OEP)X 2 (X = Br, 8a; CI, 8b), reflecting both the differences in magnetic properties between 8a-b (S = 1) and 16a-b (S = 1/2) as well as the centrosymmetric nature of 8. These differences in observed chemical shifts between, for example, 8a and 16a can perhaps be explained in terms of the type of TT -charge transfer involved in 8a (ligand-to-metal charge transfer, L-»M CT) and J_6a (M-»L CT). As suggested for 8a, 3 7 the weak 7r-donor capacity of die halide ligands cause the metal's d x z, d y 7 orbitals to increase slightly in energy, such diat the transfer of spin density from the porphyrin 3e g (TT) orbitals to the metal is possible. 1 1 8 The effect of this spin-density transfer is observed in the 'H NMR spectrum of 8a, with the methylene resonances of OEP shifting to 60 ppm from the normal diamagnetic position of 4 ppm. In 16a, the further occupancy of die d x z, d v z orbitals by one electron (d 5) causes the energy of these orbitals to increase relative to those in 8a (6A), consequently bringing these orbitals closer in energy to that of the empty 4e„ (TT) orbital of the porphyrin. The effect of the M-»L CT can be observed in the •H NMR spectrum of J_6a, with the dramatic upfield shift of the H m e s o resonance from 10 ppm in diamagnetic complexes to -5 ppm in J_6a. The shift of the resonances of the other protons of OEP in 8 and in J6 relative to their diamagnetic positions are ascribed to correlation effects. 1 1 8 The chemical shifts of the protons of the OEP macrocycle of 16a-b are temperature-dependent, as expected for paramagnetic species. Table 5.5 shows the variation of the'H NMR resonances of 16a as a function of temperature (the chloride analogue exhibited identical behavior, and therefore, is not shown), and Figure 5.11 shows the plot of the isotropic shift vs. inverse temperature. While each data set yields a straight-line plot of isotropic shift vs. inverse temperature, significant deviation from the zero-intercept at 1/T = 0 is present. This deviation 175 Table 5.5: Data for the plot of isotropic shift (ppm) vs. 1/T (K"1) for Ru(OEP)Br (16a).8-T(K) 1/T(K) 5CHo n(obs/iso^ «CB,b(obs/iso^ 5CH,(obs/iso)^ * Hmeso( o b s / i s°) ! 223 4.48 16.91/12.71 9.19/4.99 -3.95A5.89 -12.12/-22.43 243 4.12 16.58/12.38 10.15/5.95 -2.56/-4.50 -8.87/-19.18 263 3.80 16.26/12.06 10.21/6.01 -1.98/-3.92 -6.74/-17.05 284 3.52 15.81/11.61 10.21/6.01 -1.44/-3.38 -4.19/-14.50 303 3.30 15.69/11.49 11.04/6.84 -0.80/-2.74 -3.10/-13.41 323 3.10 15.42/11.22 11.06/6.86 -0.58/-2.52 -2.12/-12.43 S Obtained for a 3.52 mM solution of 16a in CDC1 3 in vacuo, ± 1 K at 300 MHz (FT). The positions of the resonances of 16a are different in CDC1 3 at 293 K than those reported in Table 5.4 for 16a in C 6D 6, in keeping with the observed solvent-dependent nature of the !H NMR chemical shifts of J_6a. See Section 5.2.d.3 and die other entries in Table 5.4 for further details. - (Observed/isotropic) chemical shift in ppm, diamagnetic correction made from Rh(OEP)Cl: SCH 2 = 4.20ppm, 6CH 3 = 1.94ppm, 6 H e s o = 10.31 ppm. 1 2 2 176 Figure 5.11: Plot of isotropic shift (ppm) vs. 1/T (K"1) for Ru(OEP)Br (16a). Solid lines represent the least-square fit of the data, with extrapolation to 1/T = 0. 177 may originate from such effects as spin-orbit coupling or temperature-independent paramagnetism. The cyclic voltammograms of 16a-b were measured in 0.1 M TBAP in CH 2C1 2 and CH 3CN. Two reversible (peak-to-peak separations of 65 _+ 5 mV), one-electron oxidation processes are observed in the CV of each complex 16a-b. assigned to the Ru^/Ru 1 1 1 (242-312 mV vs. Fc +/Fc) and the Ru I V(OEP +)/Ru I V(OEP) (779-825 mV vs. Fc +/Fc) redox processes. These data are presented below: reduction potentials2 redox couple X = Br X = CI [Ru I V(OEP)X] +/Ru m(OEP)X 312 242 305^ 255^ [R u I V ( O E P + ) X ] + 2 / [ R u I V ( O E P ) X ] + 817 779 825^ 795^ 2 +_ 10 mV vs. Fc +/Fc in CH 2C1 2 in 0.1 M (n-Bu) 4N + C104- at 293 K. ^ ± 10 mV vs. Fc + /Fc in CH 3CN in 0.1 M (n-Bu) 4N + C104" at 293 K. The metal-centered couple was assigned based on the observation that the potential of the reversible wave at 312 mV for 16a varied as (n-Bu) 4N + B r was added to the cell, following the expected behavior for a Nernstian process (Figure 5.12a):84 Ru I V(OEP)Br 2 + e - Ru m(OEP)Br + Br" (5.4) E = E°' + (RT/nF)ln([Ru(OEP)Br2] / [Ru(OEP)Br][Br"]) (5.5) Figure 5.12b shows the changes in the CV of 16a upon addition of excess (n-Bu) 4N + B r to the cell, die corresponding potential of the Ru^/Ru 1 1 1 couple varying from 312 mV to -165 mV vs. the Fc +/Fc couple, the final potential being similar to that for the Ru I V/Ru n i couple in Ru(OEP)Br 2 (8b, -154 mV vs. Fc +/Fc). The irreversible couple observed at 250 mV vs. Fc + /Fc is 178 Figure 5.12: Cyclic voltammograms of (a) Ru(OEP)Br (16a) and, (b) 16a and added excess (n-Bu) 4N + Br". X-axis scale is in Volts vs. the Fc +/Fc couple. due to the Br 2/2 Br" couple because the intensity of the CV waves at this potential increased as more (n-Bu) 4N + Br" was added to the c e l l . 8 4 However, data for a Nernst plot of E vs. -In [Br] could not be obtained because the Br 2/Br" couple at 250 mV, being irreversible, prevented the quantitative determination of the change of E from 312 mV to -165 mV for the R u I V/Ru m couple. The bulk coulometric oxidation of l_6a in the absence of (n-Bu) 4N + Br" at 400 mV vs. the Fc +/Fc couple results in the overall loss of intensity of the UV/visible spectrum of 16a as die coulometry experiment progresses (Figure 5.13a). However, bulk coulometric oxidation of .16a at 400 mV vs. Fc +/Fc in CH 2C1 2 and 0.1 M TBAP in the presence of 25 equivalents of added (n-B u ) 4 N + Br" yields the optical spectrum of 8a. These results show that the couples at 242-312 mV for 16a-b are due to the Ru I V/Ru m couple. While the values for the measured potentials of the Ru I V/Ru U I couple in 16a-b are within the reported values for such compounds, 1 0 0" 1 0 8 they are ca. 400-500 mV more positive than that for the corresponding couple in 8a4). This difference in the value of the Ru^/Ru 1" couple may be due in part to the different types of charge transfer apparent in 8a-b and 16a-b. In 8a-b. the L-»M CT will increase the spin density at the metal, making die reduction potential of the metal less favorable. In j_6, the M-»L CT removes spin density from the metal, making any further oxidation of die metal unfavorable. Also, the effect of one halide ligand in 16a-b vs. two in 8a-b may be significant; the addition of an extra halide is expected to stabilize R u I V vs. Ru 1 1 1, consistent with the lower reduction potentials observed for the Ru(OEP)X 2 systems. The implication of this large shift in the Ru I V/Ru m couple in going to 16a-b from 8a-b will be discussed later in this Chapter. The second reversible oxidation process observed at approximately 779-825 mV vs. Fc + /Fc in the CV of each of 16a-b is assigned to the ring oxidation step because bulk coulometric oxidation of 16a at 900 mV vs. Fc +/Fc led to changes in the optical spectrum of 16a which paralleled those observed during the coulometric oxidation of 8a at 900. mV vs. Fc +/Fc (see Figure 5.13b). The 180 Figure 5.13: Changes observed in the optical spectrum of 16a upon bulk coulometric oxidation of the solution at (a) +400 mV vs. Fc +/Fc both in the absence and presence of 25 equivalents of (n-Bu) 4N +Br~, (b) coulometric oxidation at 900 mV vs. Fc +/Fc in the absence of added (n-Bu) 4N +Br~. The experiments were conducted in CH 2Cl2 with 0.1 M TBAP electrolyte. 3 0 0 4 5 0 6 0 0 WAVELENGTH, nm 182 couple at ca. -1 V vs. Fc +/Fc may be due to either the Ru r a/Ru n or the Ru m(OEP)/Ru n i(OEP") couple, but is not definitively assigned. 5.2.d.3. Chemistry of Ru(OEP)X (X = Br, 16a; CI, 16b). The five-coordinate 15e" complexes 16a-b are very stable in solution. For example, a solution of 16a in C 7 D 8 sealed under vacuum was heated at 373 K for 24h and then photolyzed for 2h (450 W Hg vapor lamp); the lU NMR spectrum of the solution remained unchanged. The compounds 16a-b are very soluble in aromatic and halogenated solvents, and are sparingly soluble in hexanes. In contrast, 8a-b are insoluble in hexanes, sparingly soluble in aromatic solvents and quite soluble in halogenated solvents. Solutions of 16a and 16b could be stored in closed containers for months without any observed decomposition. The compounds 16a and 16b. being coordinatively unsaturated, bind various ligands to form six-coordinate species. Because the reactivity of 16a and 16b with the same reagents was identical, only die reactivity of Ru(OEP)Br (16a) is summarized in Scheme 5.1, and discussed below (see Section 2.4 for details of die reaction conditions which apply to the reactions described below). Aerobic oxidation of 16a in C 6D 6 in the presence of HBr yields Ru(OEP)Br 2 (8a) in a quantitative yield within two hours as judged by *H NMR spectroscopy; the reaction is inhibited if 16a is reacted with HBr alone, suggesting 0 2/HBr is the oxidant. Indeed, !_6a does not react with dry 0 2 in C 6D 6. Addition of trace H 20 to this reaction mixture induces no further reactivity, while acidification of the reaction mixture by addition of HBr yields 8a. The optical spectrum of 16a at A * M concentrations in CHC1 3 or C 7 H g immediately changes upon bubbling NH 3 through the solution, yielding the optical spectrum of Ru(OEP)Br(NH 3) (16c). This latter compound and its chloride analogue (16d) were isolated and fully characterized. The KBr pellet IR spectrum of 16c and 16d have broad (+_ 60 cm"1) bands, assigned to the i ^ N . H stretch, centered at 3150 cm"1 (see Figure 5.14). These data, and the microanalyses, are the only direct proof of coordination of NH 3 to Ru in 16c-d. because the parent peaks of 16c-d were not 183 Scheme 5.1: Summary of the reactivity of Ru(OEP)Br (16a). Figure 5.14: Infrared spectrum of Ru(OEP)Br(NH 3) in the 4000-2500 cm'1 region, as a KBr pellet. The IR spectrum of Ru(OEP)Cl(NH 3) is identical to that of the bromide analogue. detected by FAB-MS nor were the resonance due to the NH3 protons detected by 'H NMR spectroscopy. The lU NMR spectra of 16c and 16d in C 6D 6 are similar in appearance to those of 16a-b. having two methylene resonances at 9.2 and 3.8 ppm, the methyl resonances at -2.3 ppm and the Hmeso resonances at -5.9 ppm. The positions of the resonances of die OEP macrocycle in 16c-d were found to be similar (_+_ 0.2 ppm) to each other, and both complexes exhibited temperature-and solvent-dependent spectra. The inability to detect the NH 3 signal in 16c-d may be due to the extreme broadening of the linewiddi due to the proximity of the N H 3 ligand to the paramagnetic centre. 1 2 1 Figure 5.15 shows the relationship between the isotropic shift and 1/T for the OEP resonances of 16a in the temperature range 203-363 K (see Table 5.6). The isotropic shifts essentially vary linearly with inverse temperature, but have significant non-zero intercepts at 1/T -+ 0, showing that non-Curie behavior was present. The origin of this non-Curie behavior may be related to aggregation effects, spin-orbit coupling or temperature-independent paramagnetism. The 'H NMR spectra of 16c-d indirectly support the formulations proposed for 16c-d in that the separation between the two methylene resonances of 16a-b (ca. 14 ppm) is reduced to ca. 5 ppm in 16c-d upon bubbling NH 3 dirough a solution of 16a-b. consistent with the coordination of a sixth ligand. 5 3 A similar change in the separation of the two methylene resonances of Ru(OEP)Ph (8.32 ppm) was observed upon coordination of pyridine (3.70 pprii). 5 3 The compounds 16c-d are air-stable in both the solid state and in solution, and their solubility in various solvents is similar to that of 16a-b. Solutions of 16c-d at A * M concentration in CHC1 3 or C 7 H 8 are unreactive toward added 0 2 , H 2, and N 2 as judged by UV/visible spectroscopy. No substitution of either the coordinated halide or the amine ligand was observed upon addition of approximately ten equivalents of PPh 3 to 16c in C 7H 8, as judged by UV/visible spectroscopy. The protonation of the amine ligand of 16c or 16d in CHC1 3 with an acid such as HF yields a reddish solution and an off-white precipitate; filtration of die reaction mixture followed by 186 Table 5.6: Data for the plot of isotropic shift (ppm) vs. 1/T (K'1) for Ru(OEP)Br(NH 3) (16c).?-T(K) 1/T(K) £CH2n(obs/iso)^ 5CH,h(obs/iso)^ <5CH3(obs/iso)^ 5 Hmeso(° b s / i s°) ! 203 4.93 9.48/5.28 -1.14/-5.34 -4.75/-6.69 -17.18/-27.49 223 4.48 9.03/4.83 0.39/-3.81 -4.41/-6.35 -14.02/-24.33 243 4.12 8.83/4.63 1.64/-2.56 -3.86/-5.80 -11.00/-21.31 263 3.80 8.81/4.61 2.54/-1.66 -3.33/-S.27 -8.66/-18.97 284 3.52 8.86/4.66 3.32/-0.88 -2.80/-4.74 -6.53/-16.84 303 3.30 8.89/4.69 3.86/-0.34 -2.40/-4.34 -4.90/-15.21 323 3.10 8.93/4.73 4.31/0.11 -2.06/-4.00 -3.61/-13.92 343 2.92 8.96/4.76 4.70/ 0.50 -1.74/-3.68 -2.37/-12.68 363 2.75 8.97/4.77 4.89/ 0.69 -1.60/-3.54 -1.80/-11.91 5- Obtained for a 2 mM solution of 16c in C 7D g in vacuo, +_ 1 K at 300 MHz (FT). - (Observed/isotropic) chemical shift in ppm, diamagnetic correction made from data for Rh(OEP)Cl: c?CH 2 = 4.20 ppm, 5CH 3 = 1.94 ppm, « H m e s o =10.31 ppm. 1 2 2 187 Figure 5.15: Plot of isotropic shift (ppm) vs. 1/T (K"1) for Ru(OEP)Br(NH 3), 16c. Solid lines represent the least-squares fit of the data, with extrapolation to 1/T = 0. 188 workup yields Ru(OEP)X (T6a-b). and the precipitate is presumed to be N H 4 + F \ The preparative reaction for the synthesis of Ru(OEP)X(NH 3) is, therefore, reversible: N H 3 Ru(OEP)X ^ Ru(OEP)X(NH 3) (5.6) H + Recrystallization of 16a-b from CH3CN/CH2Cl2//i-pentanes or acidification of a CH 3CN solution of 16c by HF yields Ru(OEP)X(CH 3CN) (X = Br, 16e; CI, 16$; the products, being isolated and fully characterized, serve as examples of the association of coordinating solvents to Ru(OEP)X. Again, the only direct evidence for the association of CH 3CN to Ru(OEP)X are the microanalytical data and the " C= N bands observed in the IR spectra of !6e and 16f as Nujol mulls (see Figure 5.16). The *H NMR spectra of 16e-f in CD 3CN are quite similar to those for Ru(OEP)X(NH 3) (see Table 5.4), each having two methylene resonances, and one methyl and H m e s o signal. Again, the position of the latter two signals appear to be relatively unaffected by the coordination of a sixdi ligand, while the separation of the two methylene signals decreases to ca. 5.4 ppm from ca. 14 ppm in 16a-b. The CH 3CN resonance is unobserved due to the presence of die Ru 1 1 1 centre. 1 2 1 The UV/visible spectrum of 16e is similar to that of 16a, differing only slightly in position and extinction coefficient (see Figure 5.9 and Table 5.3). 1U NMR and UV/visible spectra very similar to those of 16e-f were observed when 16a was dissolved in C 5D 5N or CD 3OD (see Table 5.4 for lU NMR data), suggesting that similar six-coordinate species are generated in these solvents. The relatively small changes in A (nm) and log e for Ru(OEP)X(L) species (X = Br, CI; L = CH 3CN, py and MeOH), in comparison to data for l_6a, are similar to the changes observed in the optical spectrum of Ru(OEP)CO(MeOH) (3b) upon coordination of diese same and other similar ligands (see Table 3.1 and Reference 71). The reasons for these unexpected small changes are not readily apparent; for example, the optical spectrum of Ru(OEP)Ph undergoes immense changes upon coordination of pyridine. 5 3 In any case, the observed small changes precluded die accurate 189 Figure 5.16: Infrared spectrum of Ru(OEP)Br(CH 3CN) as a Nujol mull. The IR spectrum of Ru(OEP)Cl(CH 3CN) is identical, except that the i>c=N is observed at 2250 cm'1. measurement of the for the association equilibria shown in Equation 5.7 by UV/visible spectroscopy, as was done for several systems in Chapter 3: Ru(OEP)Br + L ^  Ru(OEP)Br(L) (5.7) 16a L = CH 3CN, py, MeOH However, addition of 1-5 fiL of the reagents CH 3CN, py, or MeOH to 20-24 / i M solutions of l_6a in CH 2C1 2 (total volume = 4.0 mL) generates the optical spectra of the fully formed corresponding six-coordinate species, leading to qualitative K g q values of the order of 102-103 M" 1. Complex 16a dissolved in neat pyridine is suggested also to yield Ru(OEP)Br(py), based on the UV/visible spectrum generated, which was identical to that obtained for a solution of 16a in CH 2C1 2 containing added pyridine. Of interest, when 16a was dissolved in pyridine and the solution filtered through a short alumina column or refluxed for ten minutes, the end-product was [Ru(OEP)(py) 2] + Br, and not Ru(OEP)Br(py) as expected (Figure 5.17). This six-coordinate 1:1 electrolyte complex can also be prepared by the aerobic oxidation of Ru(OEP)(py) 2 in the presence of HBr acid (see Section 5.2.e). These findings show that substitution of the coordinated halide ligand in 16a is possible only at high temperatures (boiling point of pyridine = 388 K) or by ligand exchange mediated by chromatographic supports. The Ru system is therefore more substitution inert than die analogue Fe(OEP)Cl, which instantaneously forms bis(ligand) 1:1 electrolyte complexes upon addition of pyridine or imidazole to Fe(OEP)Cl at room temperature.123 The addition of (n-Bu) 4NBr to a solution of 16a in CHC1 3 leads to changes in the UV/visible spectrum of 16a which are large enough to measure accurately an equilibrium constant. Figure 5.18 shows the changes in the UV/visible spectrum of 16a as B r is added, and the corresponding log plot of relative absorbance changes vs. added Br. The measured at 397 nm and 293 K 191 Figure 5.17: 300 MHz, !H NMR spectrum of [Ru n ,(0EP)(py) 2] + Br", isolated from the reaction of Ru(OEP)Br with pyridine (see Figure 5.23b). I I | I I I I | I I I I | I I I I | I I I I | I I I I | 25 20 15 10 5 0 I I I I | I I I I | I -5 PPM -10 Figure 5.18: (a) UV/visible spectral changes observed upon addition of (n-Bu)4NBr (68 mM) to Ru(OEP)Br (2.95 x 10s M) in CHCI3 at 293 K. The total added (n-Bu)4NBr present in solution after each addition corresponds to 0, 0.3, 3.4, 10, 24, 52, and 500 equivalents, (b) Plot of log [(A-A0)/(ArA)] vs. log [Br]f for the above titration, measured at 397 nm. 400 450 500 W A V E L E N G T H , nm Figure 5.18b: A 3 9 7 n m : 1.202 1.193 1.162 1.148 1.130 1.117 1.093 Volume of Br7CDCl 3: 0 1.0 10 30 70 150 * (in A»L) Solid N(n-Bu) 4Br added. 194 from the line drawn in Figure 5.18b, was 4.8 x 1CT4 M 1 although the slope of this line from the log-log plot was 0.7 ±_ 0.1. Ru(OEP)Br + (n-Bu) 4NBr ^ (n-Bu) 4N + [Ru(OEP)Br 2]" (5.8) 16a 16g The 'H NMR spectrum of 16a in CDC1 3 containing added excess NH 4Br (this being used in place of (n-Bu) 4NBr because the presence of the latter would prevent the observation of the resonances of 16g) shows only one signal each for the methylene (59.68 ppm), medvyl (6.69 ppm) and H m e s o (4.03 ppm) protons of the OEP macrocycle (Figure 5.19 and Table 5.4). The presence of only one mediylene resonance in the *H NMR spectrum of 16g confirms die generation of a centrosymmetric six-coordinate complex upon association of Br - to 16a. These data corroborate the changes observed in die CV of 16a upon addition of Br" (Figure 5.12). Similar reactivity has been reported for die association of F" to Fe m(TPP)F upon addition of HF to [Fe(TPP)] 20 in the presence of excess imidazole, the reaction forming [Fe(TPP)F 2]". 1 2 4 The reaction of J_6a with C O ^ in CH 2C1 2 generates Ru(OEP)(CO) 2 immediately, as judged by UV/visible spectroscopy (see Section 3.2.b for the optical spectrum of Ru(OEP)(CO) 2). A similar reactivity was observed for the reactions of Ru(OEP)X 2 3 7 and Ru(OEP)(Ph) 2 5 a with CO in CHC1 3. The reductant in diese reactions has not yet been identified, but could be the axial anion(s) or CO, die latter requiring trace OH" (CO + OH" C 0 2 + H + + 2e"). No reaction was observed upon addition of a stoichiometric amount of mCPBA to 16a in either CH 2C1 2, C 6 H 6 or CH 3CN but when 50 equivalents of inCPBA or PhIO was added, a greenish colored solution was obtained instantly. The optical spectrum of this solution (Figure 5.20) has a single broad band, which remained unchanged when stored for five days under vacuo. This optical 195 Figure 5.19: 'H NMR spectrum of N H 4 + [Ru m(OEP)Br 2]" generated in situ by the association of Br" to Ru(OEP)Br in CDC1 3 at 293 K. C H 2 I I 111 IJ j I I I j 1111 j M I I ] JI 1 I | I 61.0 60.0 59.0 NH 4 Br(? ) C H C I 3 m H 2 Q -A. I I I 1 ; 1 I I 1 I I I I I j I ! I I j I i I I I I I I 1 I I I I I I I I I I J I I I I I I I I I j I I I 1 I I 10 2 PPM Figure 5.20: UV/visible spectrum of the reaction mixture of Ru(OEP)Br (16a) with excess mCPBA in CH 3CN. 2 .19 ui O Z < CD oe O to CD < 1.0 0 0 .1 5 3 0 0 5 5 0 W A V E L E K G T H . n m 8 0 0 197 spectrum was similar to that reported by Leung et al. for the proposed hydrocarbon oxidation catalyst [0=Ru r v(OEP +)Br]. Exposure of this solution to air led to the bleaching of the solution with the formation of a translucent white precipitate, indicating that the porphyrin ring was destroyed. Attempts to isolate the green Ru porphyrin species formed in situ using die methods described for the isolation of the intermediates 13/14 in Chapter 4, gave only the /i-oxo dinuclear species [Ru I V(OEP)OH] 20 (9). Of interest, this greenish solution, at the mM concentration, did not have any observable 'H NMR resonances in the range +_ 150 ppm at 293 K. Morishima et a l . 1 2 5 have shown that the presence of unpaired spin at die macrocycle (i.e. a IT -cation radical) generates marked changes in the *H NMR spectrum of a precursor such as Ru(OEP)CO (3a). For example, the resonances of the OEP macrocycle of 3a (5 H m e s o = 10.15; C H 2 = 4.00; C H 3 = 1.5) shift to 6 = -35 ppm for the H m e s o, to 24 and 30 ppm for the CH 2, and to 3 ppm for the C H 3 upon formation of the TT-cation radical species [Ru n(OEP +)CO(Br)] + . 1 2 5 If additional unpaired spin is present in the same complex (i.e. there is both metal- and ring-centered oxidation), die OEP macrocycle resonances may be further shifted and broadened, perhaps resulting in the observation of a featureless 'H NMR spectrum. The addition of 50 equivalents of cyclohexene to a solution of 16a containing 50 equivalents of PhlO did not alter die optical spectrum of the solution shown in Figure 5.20, except for a small dilution effect. Further, even after the solution containing Ru(OEP)Br, PhlO and cyclohexene in C 6 H 6 (or CH 2C1 2) was stirred for five days, the gas chromatogram of the reaction solution (Figure 5.21) showed that no oxidation of cyclohexene had occurred. A similar result was obtained when the solvent used was CH 3CN. Alternative methods used to study die 'oxidation' reaction include stirring cyclohexene with Kja in C 6 H 6 prior to addition of PhlO, heating the solution at 323 K, and using Ru(OEP)Cl (16b) in place of 16a. In all cases, no oxidation of cyclohexene was observed. Cyclohexene was used as the substrate for these studies because of the large database reported for the oxidation of tliis substrate with various metalloporphyrin catalysts. 4 0" 4 2' 4 4 198 Figure 5.21: Gas chromatograms of (A) the initial reaction mixture from the reaction of Ru(OEP)Br (1 mg, 1.4 ^mol) with excess PhIO (15.6 mg, 70 /jmol) in the presence of excess cyclohexene (7.3 uL, 70 Mmol) in CH ;C1 : (2 mL), and (B) the final reaction nuxture. A HP-17 capillary column (15 m x 0.5 nun) was used with a FID detector (573 K), the inlet temperature was 473 K, and 0.5 t*L sample volun>es were tested. The temperature program used for each run is shown below. 'Gci Zrr , C 6 H 1 0 C H 2 C I 2 1 minute 423 K \ 3 2 3 K 30 K /minute 199 These reactions involving 16a show that 16a-b are capable of chemistry related to both die vacant coordination site trans to the halide, as well as substitution of the halide ligand itself. The implications of the studies involving 16a-b as possible catalysts for the oxidation of cyclohexene by oxygen atom sources will be discussed in the last section of this Chapter. 5.2.e. Synthesis, characterization and chemistry of [Ru m(OEP)(py) 2] X (X = CI, 17a; X = BF 4, 17b). 5.2.e.l. Synthesis of 17a-b. The oxidation of Ru n(OEP)(py) 2 (6a, red-yellow solution) under aerobic conditions in the presence of an HX acid yields a deep reddish colored solution when the reaction is conducted in halogenated solvents (CH 2C1 2 or CHC1 3), and a salmon colored suspension when attempted in aromatic solvents ( C 6 H 6 or C 7H g), the solvents of choice. An essential component of this reaction is air or 0 2: Ru u(OEP)(py) 2 + HX/0 2 - [Ru m(OEP)(py) 2] X + "H0 2" (5.9) X = CI, 17a; BF 4, VJb; CI (py-d5), 17c; Br, 17d; F, 17e. [Ru(OEP)(CH 3CN) 2] + B r = 17f. When gaseous HX (X = CI, Br) was bubbled through a solution of 6a, its effect was to act as a purge gas for die removal of air from the solution, and some Ru(OEP)X 2 species (8a-b) was formed along with Ha and 17d, respectively. If a solution of 6a in C 6 H 6 was saturated with HCl^-, and left for several hours in the absence of 0 2, 8b was formed as die exclusive product of this reaction. Thus, for simplicity, the method used for the syndiesis of 17a-e was to add aqueous HX (X = CI, BF 4, Br, F) to a C 7 H g solution of Ru(OEP)(py) 2 in air. This mixture was then sonicated to facilitate complete mixing of the two-phase media, and die resulting suspension was filtered to collect the product. The residue was dien purified as described in Chapter 2. The use of dilute solutions of Ru(OEP)(py) 2 and HX( a q^ in C 7H g, or heating the reaction mixture to reflux in air for five minutes also gave 17a-e. the effect of dilution or heating being the increased solubility of die trace added aqueous acid in C 7H 8. 200 Only the products from the reaction of HCI or HBF 4 with Ru(OEP)(py) 2 (6a) were isolated, and the spectroscopic characterizations of these two species serve as an example for the others (17c-e). which were generated in situ at 2-5 mM concentrations. Further, similar oxidation of Ru(OEP)(CH 3CN) 2 (6b) by HBr/air is proposed to give the analogue [Ru(OEP)(CH 3CN) 2] + Br" (_17f), based on the similarity of the in situ *H NMR spectrum of this species to those of 17a-e. 5.2.e.2. Spectroscopic characterization of 17a-b. The microanalyses data for each complex (C, H, N analysis for both species and CI analysis for 17a, see Section 2.3.b) show that the products from the aerobic oxidation of Ru n(OEP)(py) 2 in the presence of HX ( a q ) acids (X = CI and BF 4) have the formulation [Ru m(OEP)(py) 2] + X" (X = CI, 17a; BF 4, 17b). In the case of 17b, microanalyses data obtained on samples from two separate syndieses indicate diat two waters of hydration are present in this product (see also the IR data below). In CH 3CN, die molar conductivity data measured for 17a-b and 17f at 293 K (for 17a, 125 M"1 ohm"1 cm; 17b, 120 M"1 ohm"1 cm; J7f, 125 M"1 ohm"1 cm, measured in situ) indicate that diese complexes are 1:1 electrolytes, because the values were similar to that measured for the 1:1 electrolyte standard n-Bu 4N +C10 4" in CH 3CN (120 M'1 ohm"1 cm). The infrared spectrum of 17b as a KBr pellet shows a broad band between 3700-3300 cm"1 assigned to the presence of free water, as suggested by the microanalysis data. This broad band is missing in the IR spectrum of l_7a, which from the microanalysis data is known not to have any water of hydration. Neither compound has die typical strong band at 1550 cm"1 for the OEP 7r-cation radical, 8 2 indicating that in the synthesis the oxidation is metal-centered (Ru 1 1 1) instead of macrocycle-centered (Ru u(OEP "*"•)). The v B _ F IR bands could not be assigned in 17b because these bands are typically found between 1130-1040 cm"1,81 a region where several porphyrin bands are concentrated. The optical spectrum of 17a-b (see Figure 5.22a) confirms that the metal-centered oxidation product is obtained because the appearance of the spectrum is unlike the broad band type spectrum normally observed for a JT-cation radical complex (see, for example. Figure 5.3a).17 The general appearances of the spectra for the two analogues 17a and JJb are 201 Figure 5.22: (a) Optical spectra of rRu(0EP)(py)2]+ CI" (17a, [Ru] = 46.5 fiM) measured in CHC13 with a 1.0 cm pathlength cell, (b) FAB-MS of 17a. The corresponding UV/visible and FAB-MS spectra of [Ru(OEP)(py)2]+ BF4" (17b) are identical in pattern, but not necessarily in intensity, to those of the chloride analogue. (a) LU O < CD o CO CD < 0.50-0.25-0.00 400 600 n m (b) WAVELENGTH 71 2 I 79 2 1 6 2 0 67 0 — I 7 2 0 l 7 70 m / c 202 very similar, further confirming the Ru oxidation state. Ruthenium(II) n -cation radical complexes having bromide or chloride anions give 2 A l u radical species while those having BF4", C10 4" or PF 6' anions give 2 A 2 u radical species. 1 7 , 1 0 9' 1 1 0 The terms 2 A l u and 2 A 2 u denote the porphyrin orbital from which the electron was removed, the resultant effect being to generate two characteristic but very different optical spectra for each type of radical s p e c i e s . 1 7 , 1 0 9 " 0 Further, 17a-b are EPR silent at either 293 or 77 K as mM CH 2C1 2 solutions or frozen glasses, respectively, 1 1 1 confirming the absence of a ff-cation radical. The Evans' method 5 5 for magnetic measurements gave M E F F of 1.99 B.M. for JL7a and 2.44 B.M. for 17b, reasonably consistent with the spin-only value of 1.83 B.M. for S = 1/2, low-spin, R u m oxidation state. 5 6 The low-resolution, electron-impact mass spectra (EI-MS) of 17a-b show porphyrin peaks due only to [Ru(OEP)] n (n = 1 (634 m/e'), 2 (1268 mJe')) and the normally observed 'decomposition peaks' of Ru(OEP). A peak at 79 m/e' for the pyridine ligands was observed for each sample. The fast atom bombardment (FAB) mass spectra of 17a-b (Figure 5.22b and Section 2.3.b) were identical in the pattern of observed signals, but with small differences (+_ 1 m/e') in the peak positions observed. No parent peak for l_7a or 17b was observed. The FAB-MS of [Ru(OEP)(py-d 5) 2]Cl (17c) has peaks at the expected position for the sample having deuterium incorporated (see Section 2.3.b), confirming the assignments made. The 'H NMR spectra of Ru(OEP)(py) 2 (6a) and [Ru(OEP)(py) 2]BF 4 (17b) are given in Figure 5.23 (see also Figure 5.17); the corresponding data for complexes 17a-f are presented in Table 5.7. The presence of one methylene resonance in the spectrum of each complex suggests a structure having D 4 h symmetry; the relative intensities and integration of die coordinated pyridine ligands to those of the macrocycle resonances also confirm this result. The resonances of the porphyrin ligand are quite dissimilar to those reported by Morishima et al. for the n-cation radical moiety within R u n(OEP + )CO complexes and, 1 2 5 therefore, are consistent with the proposed Ru 1 1 1 formulation for 17a-b. The resonances for the coordinated pyridine in Ru n(OEP)(py) 2 (Figure 5.23a) are shifted 203 Figure 5.23: JH NMR spectra of (a) Ru°(OEP)(py)2 and (b) [Ru I I I(OEP)(py) 2] + BF4". Spectra were recorded in CDC1 3 under aerobic conditions at 293 K, 300 MHz (FT). (a) I py / [ C H, C HCI . P y H 2 Q i i i i i i I i i i i I i i i i I i i i i j i i i i I 1 1 1 1 | 1 1 1 1 1 1 1 ' 1 | 1 1 ' 1 1 1 ' 1 1 I 10 B 6 4 2 PPM 0 (b) py I H „ C H , m \ H. H . Ru w "H T n„ + BF 4 C HCI , m ' p J. C H. J jr. i j ^ i i i i ^ i i i i y i i l i ^ i l i i ^ l i ' i j ' ' ' r-r-i—I—I—i—|—I—I—i—I—f PPM -10 204 Table 5.7: Summary of *H NMR spectroscopic data for rRum(0EP)(pv)21+ and rRum(OEP)(CH?CN)-,l+ complexes.8-macrocycle resonances axial pyridine resonances complex- lUmeso SCU2 «CH3 SUpara 6Wmeta 6 Wortho Ru n, 6a 9.74 3.98 2.08 4.69 4.18 2.30 rRumlCl. 17a -1.26 17.21 -0.20 3.21 22.96 -6.99 [Ru m]BF 4, 17b -1.88 18.05 -0.33 2.96 24.19 -7.60 [Ru-d 1 0)]Cl, 17c -1.25 17.25 0.23 - - -fRulBr, 17d -1.82 18.18 -0.23 3.26 24.32 -7.65 [Ru]F, 17e -1.79 18.20 -0.23 3.17 24.32 -7.65 [Ru]Br, 17£ 23.36 8.48 -2.60 — *- Measured at 293 K in CDC1 3, 300 MHz (FT). ^ Ru denotes Ru(OEP)(py) 2. c- Ru denotes Ru(OEP)(CH 3CN) 2. 205 upfield of those observed for free pyridine ( H p 8.71, H m 7.55 and H 0 7.19 ppm) because of the ring current effect of the macrocycle, as well as the usual upfield shift of the resonances of the ligand upon coordination to a metal. 1 6 , 1 7' 2 7 The H 0 and H p resonances for the coordinated pyridines in 17a-e (Figure 5.23b) are shifted upfield of the positions observed for Ru n(OEP)(py) 2 in CDC1 3 at 293 K, presumably due to the transfer of unpaired spin density from the metal to the pyridine (M-»L CT); the H m resonance is shifted some 18-20 ppm downfield. The general assignments of the three resonances for the coordinated pyridine ligands in 17a-e were made using data for the per-deuterated analogue of 17a. Assignments of the individual resonances to the H 0 and H p of coordinated pyridine were made utilizing die known effects of the ring current. 1 1 8 The protons closest to the macrocycle arc affected the most by the ring current effect, being shifted upfield the furthest. Thus, the H c resonance of coordinated pyridine is observed to be the furthest upfield resonance and the H p is least shifted from its 'free' position. Loss of spin density at the mcto-position of the pyridine ring causes die corresponding resonance for the attached proton to appear well downfield of its expected position. 1 1 8 The lH NMR resonances, both for the macrocycle and the coordinated pyridines, arc temperature-dependent, due to die presence of an unpaired spin at the ruthenium center. The chemical shifts of 17a were measured as a function of temperature and these data are presented in Table 5.8. Only die data for _17a are presented because those for 17b were quite similar to 17a. Isotropic shifts were calculated from die observed shifts by subtracting the known diamagnetic shift contribution based on data for Ru u(OEP)(py) 2. The corresponding plot of the isotropic shift vs. 1/T is displayed in Figure 5.24. A l l the data sets obtained for 17a shown in Table 5.8 gave straight line plots but deviated, in some cases significantly, from the expected zero-intercept at 1/T = 0 for Curie-type behavior. Similar effects were observed in the case of 17b. The influence of factors such as temperature-independent paramagnetism may lead to the observed deviation from Curie law behavior for 17a-b. 206 Table 5.8: Data for the plot of isotropic shift (ppm) vs. 1/T (K'1) for rRu(OEP)(py)21Cl (a) macrocyclic resonances T(K) 1/T(K) c5Hmeso(obs/iso£ 5CH-,(obs/iso)^ c5CH,(obs/iso)^ 222 4.50xl0-3 -7.79/-17.53 18.94/14.96 -1.64/-3.71 244 4.11 -5.57/-15.31 18.43/14.45 -1.00/-3.07 267 3.74 -3.49/-13.23 17.99/14.01 -0.68/-2.75 294 3.41 -1.26/-11.00 17.21/13.23 -0.20/-2.27 313 3.19 -0.44/-10.18 17.33/13.35 -0.03/-2.10 331 3.02 0.09/ -9.65 17.33/13.35 0.09/-1.98 (b) axial ligand resonances T(K) 1/T(K) 5Hp(obs/iso)^ SH m(obs/iso)^ 6HJobs/iso)t 222 4.50xl0-3 4.69/ 0.00 31.23/27.054 -7.96/-10.26 244 4.11 4.12/-0.59 28.21/24.03 -7.60/ -9.90 267 3.74 3.49/-1.20 25.62/21.44 -7.60/ -9.90 294 3.41 3.21/-1.48 22.96/18.78 -6.99/ -9.29 313 3.19 2.96/-1.73 22.16/17.90 -7.24/ -9.54 331 3.02 2.90/-1.79 21.39/17.21 -7.24/ -9.54 S Obtained for a 3.8 mM solution of 17a in CDC1 3, ± 1 K at 300 MHz (FT). - (Observed/isotropic) chemical shift in _+ 0.01 ppm, diamagnetic correction made from Ru(OEP)(py) 2: c S H m e s o = 9.74 ppm, SCU2 = 3.98 ppm, 5CH 3 = 2.08 ppm, 5 H p = 4.69 ppm, <5Hm = 4.18 ppm, 6H0 = 2.30 ppm. 207 Figure 5.24: Plot of isotropic shift (ppm) vs 1/T (K"1) for [Ru(OEP)(py)2]+ CI", 17a. Solid lines represent the least-squares fit of the data with extrapolation to 1/T = 0. 208 The 1 9 F and n B NMR spectra of 17b were measured to confirm the presence of ionic BF 4 . The NMR data obtained in CDC1 3 by observing either nuclei indicate diat the BF4" in JJb is essentially the same as those in the series of compounds C + B F 4 " (see below). A broad singlet at + 7.07 ppm relative to (n-Bu) 4NBF 4 in the 1 9 F spectrum of 17b is assigned to die BF4" counterion. These experiments were done using (n-Bu) 4NBF 4 as the external standard (set to -146.9 ppm (diamagnetic convention) vs. CFC1 3). 9 3 C +BF 4": H + NH 4+ C s + K + N a + R b + 1 9F5,ppm: -145.6 -146.9 -141.9 -152.9 -158.9 -149.9 The 1 IB{ 1H} NMR spectrum of 17b in CDC1 3 run against the standard («-Bu) 4 NBF 4 showed a sharp resonance at 3.51 ppm superposed over a broad resonance which has the appearance of a broad triplet (+48, 0 and -43 ppm, linewidths of approximately 10 ppm at 96 MHz). These data are analogous to those reported for the BF 4" anion, 9 4 and concur with the conductivity data in supporting die postulated ionic formulation. The cyclic voltammograms of 17a-b (see Figure 5.25 and data presented below) show two reversible one-e' couples (peak-to-peak separation of approximately 65 + 5 mV) which arc assigned to the Ru m/Ru u and the Ru i n(OEP +)/Ru m(OEP) redox processes. These assignments redox process8-[Ru 1 1 1]^!!!! 1 1] [ R u n i ( + ) ] + 2 / H i u I n ] + reduction potentials  for 17a^ for 17t£ -319 -410^ 710 690£ -345 -460£ 610 600H for 6a^ -360 -485^ 690 637£ - Ru denotes Ru(OEP)(py) 2. ^ Values _+ 10 mV vs. the Fc +/Fc in 0.1 M TBAP hi CH 2C1 2. £ Values ± 10 mV vs. the Fc +/Fc in 0.1 M TBAP in CH 3CN. 209 210 were made based on the null potentials and by monitoring the changes generated in the optical spectrum of 17a-b upon bulk coulometry as described for the Ru(OEP)Xn (n = 1, 2) cases (see Sections 5.2.a and 5.2.d.2). The first reduction potential (-345 mV for 17a, -319 mV for Hb and -360 mV for Ru(OEP)(py)2 (6a) vs. Fc+/Fc) for each complex is assigned to die Ru i n/Rii n couple because upon one-e" reduction of 17b at a potential of -500 mV vs. Fc +/Fc, die optical spectrum of Ru(OEP)(py)2 was generated (see Figure 5.26a). The optical spectrum of 17b could be regenerated by bulk oxidation of this in situ solution at 300 mV vs. Fc+/Fc. The changes in the optical spectrum show several isosbestic points, indicating clean reversible conversion of ]7b to Ru(OEP)(py)2. These reduction potentials for l_7a and 17b are similar to that reported for the Rum/Ru n couple in Ru(OEP)py(CH3CN) (-300 mV vs. Fc + /Fc in 0.1 M (n-Bu)4N P F 6 in CH 2C1 2). 1 2 6 Similar reversible redox behavior was observed for the bulk coulometric oxidation of J7b at 1200 mV vs. Fc+/Fc (Figure 5.26b). The generated product has an optical spectnim that typifies the strong Soret absorption bands normally observed when the w-cation radical (OEP+ ) is formed with a BF4" counterion.17'109,110 The optical spectrum of a solution of 17a electrolyzed at the same potential as 17b was quite similar to those shown earlier for the Ru l v(OEP+ ) and for the Ru n i(OEP +) halide species (Figures 5.3a and 5.13b) Therefore, diis redox process is assigned to the Ru i n (OEP + )] + 2 / [Ruin(OEP)]+ couple. The second reversible reduction potential reported for Ru(OEP)py(CH3CN) (743 mV vs. Fc+/Fc) was not assigned because die product formed upon bulk coulometric oxidation of Ru(OEP)py(CH3CN) decomposed upon generation.126 A similar result was observed for the Ru(TPP)(py)2 analogue.126 Another voltammetric couple could be observed in the CV of 17a-b at +1050-1070 mV vs. Fc+/Fc, but this could not be identified by bulk coulometry because the redox couple was too close to the solvent discharge range. 5.2.e.3. Chemistry of [Ru(OEP)(py)2]+ X", (17a-b). The substitution chemistry of these complexes is limited because the two pyridine ligands of 17a-b do not appear even to partially dissociate. 211 Figure 5.26: Bulk coulometry of [Ru(OEP)(py)2]+ BF4" in 0.1 M TBAP in CH3CN to generate (a) Ru^OEPXpyfe, and (b) rRura(OEP+Xpy^l+Z. 0.8 U z < CD OC O CD < I' '. ,' I. \ I l' !. 1 •. t w I 0.4 r ••"/ --~~.v. w . 0.2 1 — 40 0 1 5 5 0 W.A A.' V 7 0 0 1 .2 S o r e I O z < CD Ot O to CO < 0 3 I I 4-f i -t- -t- -t- -t- -t-4 Q 0 ' 5 5 0 7<J0 WAVELENGTH.nm 212 These compounds are soluble in chlorinated and coordinating solvents such as THF and C H 3 C N , stable in solution even at 10"5 M for weeks under aerobic conditions as judged by UV/visiblc spectroscopy, and display no susceptibility to light in solution. The complexes decompose when reacted with oxidants such as PhlO or mCPBA in CHC1 3 (about 3 mM) in air, as observed by the bleaching of the reaction mixture within minutes. This decomposition was observed also when the reaction was conducted in CH 3CN or under argon in either of the two solvents or in the presence of the organic substrate cyclohexene (100 equivalents). One source of interest is that these complexes can be reduced to the precursor Ru(OEP)(py) 2 by vacuum pyrolysis at 413 K and 1 x 10"5 torr for one hour. The mechanism of this reduction is not known, but outer-sphere reduction of the cation in 17a-b by the anion seems likely. This reactivity is dissimilar to that reported for the reduction of Ru I V(OEP)Br 2 (8a) upon pyrolysis under the same conditions to give [Ru n(OEP)] 2 because the halide is uncoordinated to the metal in 17.36>37 The partial auto-reduction of [Ru n i(OEP)(P(n-Bu) 3) 2] + B r to Ru"(OEP)(P(«-Bu)3)2 in the solid state over several weeks has been noted earlier, 1 7 the process believed to involve initial dissociation of P(«-Bu)3, followed by coordination of Br" and a reduction step involving phosphine as the reductant:17 slow [Ru m(OEP)(P(n-Bu) 3) 2] + Br" - Ru I U(OEP)Br(P(n-Bu) 3) + P(n-Bu) 3 (5.12) P(n-Bu)3 + 2 [Ru m(OEP)(P(7z-Bu) 3) 2] + Br" -> 2 Ru n(OEP)(P(n-Bu) 3) 2 + P(n-Bu) 3Br 2 (5.13) The reduction process was accelerated by recrystallization of [Rum(OEP)(P(«-Bu)3)2]+ B r from CH 7Cl 9/hexane. 1 7 A similar process cannot occur in the case of j j a and 17b because even partial dissociation of die coordinated pyridine is not observed in solution, and die pyrolysis reaction quantitatively gives Ru(OEP)(py) 2. 5.2.f. Summary of electrochemical data for Ru(OEP) complexes. Table 5.9 summarizes the electrochemical data discussed earlier in this Chapter: Further additions to this Table arc the data for Ru m(OEP)Br(PPh 3) (10). The CV for 10 in CH 2C1 2 is shown in Figure 5.27, and the two 213 Table 5.9: Summary of electrochemical data for Ru(OEP) complexes. reduction potentials5-complex^ fRu I V(OEP +')l / r R u I VOEP)l r R u m(OEP +')l / r R u m(OEP)l Ru^/Ru" 1 Ru m/Ru" Ru ^ B r ^ S a 782 - -154 Ru^Clj.Sb 694 - -170 Ru mBr. 16a . 8 1 7 - 312 R u ^ l . 16b 779 - 242 EMD(py)2,6a - 690 - -360 [ E i^CpyyCl, 17a - 600 - -345 [R»m(py)2]BF4> 17b - 694 - -319 RumBr(PPh3), 10 602 - -222 ?- In mV vs. Fc +/Fc (632 mV vs. NHE) in CH 2C1 2, 0.1 M TBAP electrolyte at 293 K. ^ Ru denotes Ru(OEP). 214 Figure 5.27: Cyclic voltammogram of Ru(OEP)BrfPPh3) in 0.1 M TBAP in CH2C12. X-axis scale is in mV vs. the Fc+/Fc couple. 215 reversible couples observed are assigned to the ring- (602 mV vs. Fc +/Fc) and to die metal-centered (-222 mV vs. Fc +/Fc) oxidation. These assignments are based on the observation diat the redox processes observed at 600-800 mV vs. Fc +/Fc for Ru(OEP) complexes arise from the ring-centered oxidation couple. The metal-centered couple was assigned because this couple is observed on the oxidation side of the null potential, and because the potential (-222 mV vs. Fc + /Fc) is similar to that for the Ru I V/Ru r a couple in Ru(OEP)X 2 (X = Br, CI). From Table 5.9, it is apparent that the chloride ligand is more capable than die bromide ligand of stabilizing die R u I V oxidation state as judged by the cathodic shift of the Ru I V / R u i n couple as the ligand varies from chloride to bromide; such an effect is as expected. Also, within Ru 1 1 1 precursor complexes, those having axial halide ligands coordinated to the metal (16a-b and 10) give Ru I V/Ru n i couples, while diose having axial pyridine ligands (17a-b) coordinated to the metal favor die- reduction process Ru m/Ru n. This trend is in keeping with the known ability of halide ligands to stabilize higher oxidation states at the metal while pyridine ligands stabilize a lower oxidation state. 1 2 7 Indeed, the first metal-centered (Ru m/Ru n) oxidation potentials for Ru(porp)(PR3)2 complexes (592-462 mV vs. Fc +/Fc) show that phosphines also stabilize the lower oxidation states of Ru. 1 7 Lastly, die first oxidation process of RuI1(porp)CO complexes yields [ R u n ( O E P + ) C O ] + and not the metal-centered oxidation product. 1 7 These data show that the ability of various ligands to stabilize a higher oxidation state of Ru increases in the order CO < PR 3 < py -< Br" < CI". The reduction potentials for the ring-oxidation process (700 _+ 100 mV vs. Fc +/Fc) vary to a lesser extent than that for the metal-centered oxidation process (-220 to 312 mV vs Fc + /Fc for the Ru I V/Ru m couple and -319 to -360 mV for the Ru m/Ru n couple) showing that axial ligand substitution has relatively little effect on the potential for die ring-oxidation process. A similar large variance for the potential for the metal-centered oxidation process ( F e n i / F e n ) 1 2 8 and relative invariance of the potential for the ring-centered process (porp +7porp) has been noted from electrochemical studies of Fe(porp)X complexes where X is varied. 1 2 9 Thus, the trends 216 observed here parallel those observed for the iron porphyrin systems. Of interest, the reduction potential for the 0 2 7 0 2 couple has been shown to vary from -165 mV vs. Fc +/Fc to +555 mV as the pH is adjusted from 7 to 1 in DMF. 1 3 0 While this potential will vary with different solvents and pH, the proposal that HX/air is the oxidant for the oxidation of Ru 1 1 precursors to yield R u I U complexes in CHC1 2 or C 7 H 8 appears to be correct, because the reduction potential of the Ru I U/Ru n couple (-320 to -360 mV vs. Fc +/Fc) is in die correct range for oxidation of Ru 1 1 to R u m by H +/0 2. The complexes 8a-b and 17a-b decompose when reacted with wCPBA or PhIO, while complexes Ru(OEP)X (16a-b) and 10 form green-colored solutions which are stable under inert conditions. Thus, no predictable trend in die reactivity of these Ru(OEP) complexes with /rcCPBA or PhIO can be derived from the electrochemical data. Comparison of the reduction potentials of the Ru(OEP) complexes shown in Table 5.9 indicates that Ru(OEP)Br(PPh 3) QO) has the most favorable potentials for the formation o f a species such as [Ru I V(OEP +)X(L)] + . Further, K) is the only Ru(OEP) complex reported to catalyze the oxidation of cyclohexene by PhIO; 4 0 clearly, it is possible that the presence of the phosphine ligand in K> plays a major role in its ability to act as a catalyst for the oxidation of cyclohexene. In die reaction of JO widi mCPBA or PhIO, die oxidation of PPh 3 to 0=PPh 3 was suggested to be die first step, followed by the oxidation of the remaining Ru(OEP)Br fragment to 0 = Ru I V(OEP +)Br. 4° Thus, species having 0=PPh 3 ligands may be present in situ; nithenium non-porphyrin complexes having coordinated 0=PPh 3 ligands are precedented.131 The relative ability of the 0=PPh 3 ligand to stabilize a high-oxidation state at Ru is unknown; however, considering that 0=PPh 3 would be a weak 7r-acceptor when bound to Ru, lower metal oxidation states should be favored. Secondly, die 0=PPh 3 ligand would have to compete with Br" for the vacant sixth-coordination site in [ 0 = R u r v ( O E P + ) ] + . Lastly, the trans effect of the oxo ligand may play an important role in determining which ligand is coordinated to Ru in tins system. Of interest, several five-coordinate tetra(perfluorophenyl)porphyrin (an analogue of 217 tetraphenylporphyrin, TPP) complexes of Fe, Mn and Cr having axial halide or azide ligands show remarkable activity as catalysts for the selective oxidation of, for example, isobutane to /-butyl alcohol at thousands of turnovers at room temperature utilizing just dioxygen. 1 3 2 These systems avoid the use of expensive oxygen atom sources such as m-CPBA or PhlO. The Ru(OEP)Br(PPh 3) system 4 0 for the oxidation of cyclohexene compares poorly with those reported using dioxygen as die oxygen atom source because few turnovers are observed, and the oxidation mechanism is non-selective. 4 0 Thus, further studies aimed at developing catalysts for the oxidation of alkanes and alkenes should perhaps be concentrated on the tetra(perfluorophenyl)porphyrin complexes of ruthenium. 218 Chapter 6: General conclusions and recommendations for future work. The measurement of the forward and reverse rate constants and equilibrium constants for the systems involving the dissociation of a PPh 3 ligand from Ru(OEP)L(PPh3) (L = CO, 11; PPh 3, 4a) has allowed for a crude estimation of the Ru-P bond strength in these complexes; die maximum value of 64 kJ mol"1 indicates a weak bond, but its strength must be influenced by the strong trans effect of the CO or PPh 3 ligand. The thermodynamic and activation parameters measured for these systems, as well as the solution chemistry of these complexes, show that the dissociation of a PPh 3 ligand from U and 4a to yield the formally five-coordinate species Ru(OEP)CO (3a) and Ru(OEP)PPh 3 (5a) is greatly influenced by solvent effects. The effect of the presence of trace H 20 in solution is shown to be negligible at *H NMR spectroscopy sample concentrations (mM), but at UV/visible spectroscopy sample concentrations (^L), the formation of aquated six-coordinate species in situ occurs. These results show diat while the structure of 3a and 5a may be five-coordinate, square-pyramidal in die solid state, in solution, six-coordinate octahedral species are formed even if weakly coordinating reagents are present. The extension of the these studies to other ligand systems should be pursued so that other Ru-L bond strengths (particularly, die Ru-CO bond strength) may be obtained. The mechanism of the oxidation of [Ru(OEP)] 2 (7) by HX (X = Br, CI) was found to be more complicated than diought initially. In aromatic solvents, die oxidant is identified as HX itself and not trace X 2 in HX, directly yielding Ru I V(OEP)X 2 quantitatively when the pK a of the HX acid used is equal to or below -7 (X = Br, 8a; CI, 8b; SbF 6, 8d). In CH 2C1 2, the formation of two major intermediates is observed in the reaction of 7 with HCI, the mixture quantitatively yields 8b upon aerobic oxidation in the presence of excess HCI. These same intermediates are generated from die reaction of 7 with CH 2C1 2. Because of the highly reactive nature of the mixture of intermediates, the individual species could not be separated or isolated; however, the species are believed to be [Ru(OEP)] 2CHCl 2 Q3) and Ru(OEP)H Q4), as determined by comparison of their 219 spectroscopic properties with those of known compounds. Preliminary kinetic studies on the reaction of 7 with CH 2C1 2 suggest a direct bimolecular reaction. Radical-based reactivity ot 7 with HX in aromatic solvents and in CH 2C1 2, or of 7 with CH 2C1 2 itself, are indicated by the solution chemistry. The required hydrogen containing co-product in diese reactions giving Ru I V(OEP)X 2 has not been identified; however, it is not H 2. The use of 3 H X may help elucidate the mechanism of this reaction in either type of solvent, as well as identify the co-products generated. The in situ, non-ionic, species Ru(OEP)(X') 2 have been generated by the reaction of HX' with 7 (X' = SbF 6, 8d) or with the mixture of intermediates from the 7/C H ? C1T reaction ( X ' = BF 4, 8c). One species, 8d, undergoes facile auto-reduction to give Ru I I I(OEP)SbF 6 Q5). The characterization and chemistry of these (X') species should be studied further. Also, Ru(TMP)Br, (8e) was isolated and characterized, and its chloride analogue generated in situ. The organometallic chemistry of Ru(TMP) complexes should be developed from 8c, because few organoruthenium complexes of diis porphyrin have been reported. The additional steric bulk that the mesityl ligand imparts to the macrocycle, relative to the substituents of the OEP and TPP porphyrins, may yield novel chemistry. An electrochemical study of Ru(OEP)X 2 (X = Br, CI) reveals that the Ru^/Ru" 1 couple in these complexes occurs at a low reduction potential, and some chemistry of Ru(OEP ) X 2 ( X = Br, CI, SbF 6) confirms this result. The new complexes Ru(OEP)X (X = Br, J_6a; C I , JjSb) have been synthesized by reduction of the parent Ru(OEP)X 2 species using gaseous N H 3 . Initial ' H N M R spectroscopy studies indicate that die reduction of Ru(OEP)X 2 (X = Br, CI) by N H 3 may follow clean pseudo-first order kinetics, and as such, the complete study should be undertaken to obtain the solution Ru-X bond strength. The coordination chemistry at the vacant axial coordination site in 16a-b is extensive, yielding the isolated six-coordinate species Ru(OEP)X(L) ( X = Br, C I ; L = CH 3CN, NH 3) and the in situ species (n-Bu) 4N + [Ru m(OEP)Br 2]". Substitution of the coordinated halide ligand in 16a-b occurs only at high temperatures or by the assistance of an 'exchange media' such as a chromatographic support. The in situ species [Ru m(OEP)(py) 2] + B r was 220 generated by either method. The inability of 16a-b to perform as catalysts for the oxidation of cyclohexene by oxene reagents was particularly disappointing. The use of complexes of highly fluorinated porphyrins such as the tetra(perfluorophenyl)porphyrins132 should be pursued i f the development of suitable catalysts for the oxidation of cyclohexene and other organic substrates is desired. These halogen-containing metalloporphyrin complexes have already shown great promise in utilizing dioxygen as the reagent for the oxidation of organic substrates, instead of the expensive oxene reagents. The metadiesis chemistry of Ru™ and Ru r a(OEP)X complexes (8a-b and of ]6, respectively) to generate complexes of alkoxides, azides or hydrides should be studied. The outer-sphere oxidation of Ru(OEP) complexes by 0 2/H + has also been extended, and the new complexes [Ru(OEP)(py) 2] + X" (X = CI, Ha; BF 4, 17b) isolated and characterized. Similarly, the outer-sphere oxidation of die coordinatively unsaturated complexes [Ru n(OEP)] 2 and Ru n(OEP)PPh 3 by trace air to give [Ru I V(OEP)OH] 20 has been shown to require both O, and H 20. Electrochemical studies of Ru 1 1 1 and Ru^COEP) complexes have shown that, whereas pyridine ligands favor reduction of die metal to the lower oxidation states (Ru"), coordinated halide ligands favor higher oxidation states (Ru I V). The ring-centered oxidation process is less favored, requiring a much higher oxidation potential than the metal-centered oxidation process. Further electrochemical studies involving other Ru I V and Ru m(porp) complexes should be pursued. 221 Chapter 7; References. 1. Iron Porphyrins. I^ver,A.B.P.,Gray,H.B.,Eas.,Addison-Wesley,Massachusetts,Parts 1 and 2,1981. 2. (a) Collman.J.P. Acc.Chem.Res., 1977,10,256. (b) Perutz.M.F. ScientincAm. .1978.239.92. (c) Collnm,J.P.;Braurrian,J.I.;Fitzgerald,J.P.;Sparapany,J.VV.;Ibers,J.I. J.Am.Chem.Soc, 1988.110.3486. 3. (a) Watanabe, Y.;Ishimura,Y. J.Am.Chem.Soc.,1989.111,410. (b) Mansuy.D. Pure and Appl. Chem., 1987,59,759. (c) Collman,J.P.;Kodadak,T.;Brauman,J.I. J.Am.Chem.Soc. 1986,108,2588. 4. Mashiko.T.,Dolphin,D. in Comprehensive Coordination Chemistry.Wilkinson.G..Ed..Pergamon Press,Oxford,Vol. 2,1987,p.813. 5. (a) Ke.M. Ph.D. Dissertation, University of British Columbia, 1988. (b) Pacheco.A.A. M.Sc. Dissertation, University of British Columbia, 1986. 6. (a) Jones,R.D.;Summerville,D.A.;Basolo,F. Chem.Rev., 1979,79,139. (b) Lever.A.B.P. /.Mo/.Sfmcr«re,1980,59,123. • 7. (a) Chow,B.C.;Cohen,I.A. Bioinorg.Chem., 1971,1,57. (b) Bonnet.J.J.;Eaton,S.S.;Eaton,G.R.;Holm,R.H.;Ibers,J.A. J.Am.Chem.Soc, 1973,95,2141. 8. (a) Eaton,S.S.;Eaton,G.R.;Holm,R.H. J.Organomet.Chem.,1971,32,C52 and references therein, (b) Kadish,K.M.;Leggett,D.J.;Chang,D. Inorg.Chem., 1982,21,3618 and references therein. 9. Antipas,A.;Buchler,J.W.;Gouterman,M.;Smith,P.D. /.Am.Cfte/n.5oc..l978.100.3015. 10. Farrell,N.;Dolphin,D.H.;James,B.R. y.Am.CAem.Soc, 1978,100,324. 11. (a) Collman.J.P.;Brauman,J.I.';D.oxsee,K.M.;Halbert,T.R.;Bunnenberg,E.;Linden,R.E.; La Mar,G.N.;Del Gaudio,J.;Lang,G.;Spartalain,K. J.Am.Chem.Soc.. 1980.102,4182. (b) Collman.J.P. ;Brauman,J.I. ;Fitzgerald,J.P.;Sparapany,J.W. ;Ibers,J.A. J.Am.Chem.Soc, 1988,110,3486. 12. (a) Perutz.M.F. Nature. 1972.237.495 (b) Perutz,M.F. Ann.Rev.Biochem., 1979,48,327. 13. James,B.R.;Addison,A.W.;Caims,M.;DolpWn,D.;Farrell,N.P.;Paulson,D.R.;W in Fundamental Research in Homogeneous Catalysis.Tsutsui,M.,Ed.,Vol. 3,Plenum, New York, 1979,p.751. See also reference 10. 14. (a) Camenzind,M.J.;James,B.R.;Dolphin,D.;Sparapany,J.W.;Ibers,J.A. Inorg.Chem., 1988,27,3054. (b) Camenzind,M.J.;James,B.R.;Dolphin,D. J.Chem.Soc.,Chem.Commun.,\9S6,\\37. 15. Ariel,S.;Dolphin,D.;Domazetis,G.;James,B.R.;Leung,T.W.;Rettig,S.J.;Trotter,J.; Williams,G.M. Can.J. Chem., 1984,62,755. 16. (a) Sishta,C.;Camenzind,M.J.;James,B.R.;Dolphin,D. Inorg.Chem., 1987,26,1181. (b) James,B.R.;Mikkelsen,S.R.;Leimg,T.W.;Williams,G.W.;Wong,R. Inorg.Chim.Acta, 1984,85,209. 17. (a) Barley,M.;Becker,J.Y.;Domazetis,G.;Dolphin,D.;James,B.R. Ca«.y.C/iem.,1983,6i,2389. (b) Barley,M.H. Ph.D. Dissertation, University of British Columbia, 1984. 18. Collman,J.P;Kim,K.;Leidner,C.R. Inorg.Chem., 1987,26,1152. 19. Domazetis,G.;James,B.R.;Tarpey,B.;Dolphin,D. ACS Symp.Ser..1981.152.243. 20. Domazetis,G.;Tarpey,B.;Dolphin,D.;James,B.R. J.Chem.Soc,Chem.Commun., 1980,939. 21. Tarpey.B. M.Sc. Dissertation,University of British Columbia, 1982. 22. Barringer.L.F., Jr. ;Rillema,D.P.;Ham,J.H., IV J Anorg.Biochem., 1984,21,195. 23. (a) Sheldon,R. A.;Kochi,J.K. Metal Catalyzed Oxidation of Organic Compounds. Academic Press, New York, 1981,Chapter 12. (b) Koenig,T.;Fischer,H. in Free Radicals,Vol.l.Kochi.J.K..Ed.,Wiley-Interscience,New York, 1973,Chapter4. 24. (a) Belani,R.M.;James,B.R.;Dolphin,D.;Rettig,S,J. Can.J.Chem., 1988,66,2072. (b) Belani.R.M. M.Sc. Dissertation,University of British Columbia, 1985. 25. Collman,J.P.;Barnes,C.E.;Swepston,.P.N.Ibers,J.A. J.Am.Chem.Soc, 1984,106,3500. 26. (a) James,B.R.;Pacheco,A.;Rettig,S.J.;Ibers,J.A. Inorg.Chem., 1988,27,2414. (b) James,B.R.;Pacheco,A.;Rettig,S.J.;Thorburn,I.S.;Ball,R.G.;Ibers,J.A. J.Mol.Catal., 1987,41,147. 223 27. Crotti,C.;Sishta,C.;Pacheco,A.;James,B.R. Jnorg.Chim.Acta,198S,Ul, 13. 28. James.B.R.jDolpWn.D.jUungJ.W.jEinstem.F.W.B.iWillis.A.C. Can.J.Chem., 1984,62,1238. 29. Bajley,M.;Becker,J.Y.Doinazetis,G.;Dolphin,D.;James,B.R. J.Chem.Soc.,Chem.Commun., 1981,982. 30. Smim,P.D.;Dolphm,D.;James.B.R.y.Organo/ncf.CAefft.. 1981.208.239 and references therein. 31. Collman,J.P.;Barnes,C.E.;Brothers,P.J.;Collins,T.J.;Ozawa,T.;Gallucci,J.C.;Ibers,J.A. J.Am. Chem.Soc., 1984,106,5151. 32. Collman,J.P.;Prodolliet.J.W.;Leidner,C.R. J.Am.Chem.Soc, 1986.108.2916. 33. Sishta,C.;James,B.R.;Dolphin,D.,Third Chemical Congress of North America,Toronto, Ontario.June 5-10,1988,P-INOR-635. 34. Masuda,H.;Taga,T.;Osaki,K.;Sugimoto,H.;Mori,M.;Ogoshi,H. J.Am.Chem.Soc.. 1981,103.2199. 35. Collman,J.P.;Barnes,C.E.;Collins,T.J.;Brothers,P.J.;Gallucci,J.;Ibers,J.A. J.Am.Chem.Soc, 1981.103.7030. 36. Sishta,C.;Ke,M.;James,B.R.;Dolphin,D.H. J.Chem.Soc.,Chem.Commun., 1986,787. 37. Sishta,C. M.Sc. Thesis,University of British Columbia, 1986. 38. Rachlewicz,K.;Latos-Grazynski,L. Jnore.Chim.Acta. 1988.144.213. 3 9 . (a) Groves,J.T.;Ahn,K.-H. Inorg.Chem., 1987,26,3831. (b) Rajapakse,N.;James,B.R.;Dolphin,D. in New Developments is Selective Oxidations,Ccnti.G.; Triffiro,F.,Eds.,Elsivier Science,Amsterdam, 1990,in press. 4 0 . Leung,T.W.;James,B.R.;Dolphin,D. Inorg.Chim.Acta,l983,79,lS0. 41. Groves,J.T.;Haushalter,R.C.;Nakamura,M.;Nemo,T.E.;Evans,B.J. J.Am.Chem.Soc,1981.103,2884. 42. Groves,J.T.;Quinn,R. Inorg.Chem., 1984,23,3844. 43. Che,C.-M..Sixth International Symposium on Homogeneous Catalysis,Vancouver,British Columbia, August 21-26,1988.C-1. 44. (a) Marchon,J.-C.;Ramasseul,R. J.Chem.Soc.,Chem.Commun., 1988,298. (b) Groves,J.T.;Quinn,R. J.Am.Chem.Soc. 1985.107.5790. 45. (a) Brothers,P.J.;Collman,J.P. Acc.Chem.Res., 1986,19,209. 224 (b) Brothers.P.J. Dissertation Abstracts g,1985,2657-B,order # DA8522115. 46. Setsune,J.-I.;Dolphin,D. Can.J.Chem.,1987,65,459. 47. Collman,J.P.;Bromers,P.J.;McElwee-White,L.;Rose,E.;^^ 48. Collinan.J.P.jBrothers.P.J.jMcElwee-VVhite.L.jRose.E. J.Am.Chem.Soc. 1985.107.6110. 49. Ke,M.;Rettig,S.J.;James,B.R.;Dolphin,D. J.Chem.Soc.,Chem.Commun., 1987,1110. 50. Collman,J.P;McElwee-VVhite,L.;Brothers,P.J.;Rose,E. J.Am.Chem.Soc. 1986.108.1332. 51. Perrin,D.D.;Armarego,W.L.F.;Perrin,D.R. Purification of Laboratory Chemicals,2— Ed.. Pergamon Press.Oxford, 1980. 52. Shriver,D.F.;Drezdzon,M.A. The Manipulation of Air-Sensitive Compounds.J.Wiley.New York, 1986. 53. (a) The Merck Index.Stecher,P.G..Ed.,8^ Ed..Merck and Co.,Railway,New Jersey, 1968,p. 115. (b) Nigam,S.;Asmus,K.-D.;Willson,R.L. J.Chem.Soc,Faraday Trans., 1976,72,2324. 54. Lexa,D.;Saveant,J.M.;Zickler,J. J.Am.Chem.Soc, 1977,99,2786. 55. (a) Evans,D.F. J.Chem.Soc, 1959,2003. (b) Live,D.H.;Chan,S.I. Anal.Chem., 1970,42,791. 56. Earnshaw.A. Introduction to Magnetocliemistry.Academic Press.New York, 1968. 57. Groves,J.T.;Nemo,T.E. J.Am.Chem.Soc. 1983.105.6243. 58. Bruce,M.I.;Matisons,J.G.;Wallis,R.C.;Patrick,J.M.;Skelton,B.W.;White,A.H. J.Chem.Soc,Dalton Trans.,1983,2365. 59. Stynes,D.V.;Fletcher,D.;Chen,X. Inorg.Chem., 1986,25,3483. 60. Pomposo,F.;Carruthers,D.;Stynes,D.V. Inorg.Chem., 1982,21,4245. 61. Eaton,S.S.;Eaton,G.R. Inorg.Chem., 1977,16,72. 62. Walker,S.G.,M.Sc. Dissertation.University of British Columbia, 1980. 63. Chesko.J.D.M. B.Sc. Honours Thesis.University of British Columbia, 1986. 64. (a) Sandstrom,J. Dynamic NMR Spectroscopy.Academic Press,London, 1982. (b) Martin,M.L.;Martin,G.J. Practical NMR Spectroscopy.Heydeu.London.1980,Chapter 8. 225 (c) Abragam, A. The Principles of Nuclear Magnetism.Clarendon.Oxford. 1961.Chapter X. (d) Sanders,J.K.M.;Hunter,B.K. Modern NMR Spectroscopy.Oxford University Press,Oxford, 1987 .Chapter 7. 65. Binsch,G.;Kleier,D.A. DNMR3 Program (General NMR Line-shapes Program with Symmetry and Magnetic Equivalence Factoring),University of Notre Dame,Notre Dame,Indiana,46556. 66. Laidler.K.J. Chemical Kinetics.3^ Ed.,Harper and Row,New York, 1987. 67. Eaton.G.R.;Eaton,S.S. J.Am.Chem.Soc, 1975,97,235. 68. Barley,M.;Dolphin,D.;James,B.R.;Kirmaier,C.;Holten,D. J.Am.Chem.Soc, 1984,106,3937. 69. Paulson.D.R.jBhakta.S.B.jHyun.R.Y.jYeim.M.jBeaird.C.E.jIxe.S.C.-.Kim.I.jYbarra.J.Jr., Inorg. Chim.Acta, 1988,151,149. 70. Joshi,A.M.;James,B.R. Organometallics, 1990,9,199. 71. Lakowski,B.C. B.Sc.Honours Thesis,University of British Columbia, 1989. 72. Huheey.J.E. Inorganic Chemistrv.2^ Ed..Harper and Row.New York,1978,pp. 489-498. 73. (a) Krassowski,D.W.;Nelson,J.H.;Brower,K.R.;Hauenstein,D.;Jacobson,R.A./nor^.C/'e/'7., 1988,27,4294. (b) Buchler, J.W. in The Porphyrins.Vol. 1.Dolphin.P..Ed..Academic Press,New York, 1978, Chapter 10. 74. Collman.J.P.;Hegedus.L.S.;Norton.J.R.:Finke.R.G.Principles and Applications of Organotransition  Metal Chemistry.University Science Books,Mill Valley,California, 1987,Chapter 5. 75. Setsune,J.-I.;Yoshida,Z.-I.;Ogoshi,H. J.Chem.Soc.Perkin 7,1982,983. 76. Ogoshi,H.;Setsune,J.-I.;Yoshida,Z.-I. /.Organomet.C/tem. 1978.159,317. 77. Seidell,A. Solubilities of Inorganic and Metal Organic Compounds.Vol. 1,3— Ed, Van Nostrand, New York,1940,pp.567-580. 78. Busby,C.A.;Dolphin,D. J.Magn.Reson.1916,23,2U. 79. Anderson,J.E.;Yao,C.-L.;Kadish,K.M. Inorg.Chem.\9%6,25, 80. Clark.E.P. Indust.Eng.Chem.,Anal.Ed.,1941,\3,%20. 226 81. (a) Nakamoto.K. Infrared Spectra of Inorganic and Coordination Compounds.2— Ed,Wiley-Interscience.New York, 1970. (b) Maslowsky,E.,Jr. Vibrational Spectra of Organometallic Compounds.Wilev-Interscience.New York, 1977. 82. Shimomura,E.T.;Phillippi,M.A.;Goff,H.M.;Scholz,W.F.;Reed,C.A. /.Am.CAem.Soc. 1981,J03,6778. 83. Mixed Valence Compounds.D.B.Brown.Ed.:D.Reidel.Dordrecht.Holland.l979. 84. Bard.A.J.;Faulkner,L.R. Electrochemical Methods.Fundamentals and Applications.John Wiley and Sons.New York, 1980. 85. The X-ray photoelectron spectra were run at the University of British Columbia by Dr. P. Wong, whose help is gratefully acknowledged. 86. Wagner,C-D.;Riggs,W.M.;Davis,L.E.;Moulder,J.F. Handbook of X-ray Photoelectron  Spectroscopy.Muilenberg.G.E..Ed..Perkin-Elmer Corporation,Eden Prairie,Minnesota,1979. 87. (a) Thorburn.I.S. Ph.D. Dissertation, University of British Columbia, 1985. (b) Mimoun,H.;Saursine,L.;Menage,S.;Girerd,J.-J. in New Developments is Selective  Oxidations.Centi.G.:Triffiro.F..Eds..Elsivier Science,Amsterdam,1990,in press. 88. Stark,J.G.;Wallace,H.G. Chemistry Data Book.John Murray.London. 1970. 89. Labinger,J.A.;Bercaw,J.E. Organometallics,1988,1,926. 90. (a) Wayland,B.B.;Sherry,A.E. The 1989 International Chemical Congress of Pacific Basin Societies.Honolulu,Hawaii,December 17-22. Abstract ORGN-0703. (b) Del Rossi,K.J.;Wayland,B.B. J.Am.Chem.Soc, 1985,107,7941. 91. (a) Ghilardi,C.A.;Midollini,S.;Moneti,S.;Orlandini,A.Ramirez,J.A. J.Chem.Soc,Chem.Commun., 1989,304. (b) Jackson,N.R.C.;Harrison,W.D.;Goodall,D.C. J.Chem.Soc.,Chem.Commun., 1988,729. (c) Marder,T.B.;Fultz,W.C.;Calabrese,J.C.;Harlow,R.L.;Milstein,D. J.Chem.Soc,Chem.Commun., 1987,1543. 92. (a) Sellman,D.;Waeber,M.;Binder,H.;Boese,R. Z.A^far/brscn..l986,41B,1541. (b) Labinger,J.A.;Osborn,J.A.;Colville,N.J. Inorg.Chem., 1980,19,3236. 93. Wray.V in Annual Reports on NMR Spectroscopy.Webb.G.A..Ed..Vol 14,Academic Press,London, 1983,p.331. 94. Eaton.G.R. ;Lipscomb.W.N. NMR Smdies of Boron Hydrides and Related Compounds. W.A.Benjamin.New York,1969,pp.436-445. 95. Newbound.T.D. ;Colsman,M.R.; Mi l l e r , M . M . ;Wulfsberg,G.P.; Anderson,0. P.; Strauss, S.H. J.Am.Chem.Soc..\9$9 A 11.3762. 96. A Carle GC was used with a 6' Porapak Q column, 298 K oven temperature, 20 mL/minute gas flow, TCD detector. Retention times were 2.3 minutes (H 2) and 3.3 minutes ((VN,), sec Chapter 2. 97. Scheer.H. in The Porphyrins.Vol.2.Dolphin.P..Ed..Academic Press,New York, 1978,Chapter 1. 98. Gordon,A.J.;Ford,R.A. The Chemist's Companion.Wiley-InterScience.New York, 1972,pp.54-80. 99. Cotton,F.A.;Wilkinson,G. Advanced Inorganic Chemistry.4^ Ed.,Wiley-Interscience,New York, 1980,Chapter 9. 100. Che,C.-M.;Tang,T.-W.;Poon,C.-K. J.Chem.Soc.,Chem.Commun., 1984,641. 101. Che,C.-M.;Wong,K.-Y.;Poon,C.-K. Inorg.Chem., 1986,25,1809. 102. Che,C.-M.;Lai,T.-F.;Wong,K.-Y. Inorg.Chem., 1987,26,2299. 103. Llobet,A.;Doppelt,P.;Meyer,T.J. Inorg.Chem., 1988,27,514. 104. Moyer,B.A.;Meyer,T.J. Inorg.Chem., 1981,20,436. 105. Binstead,R.A.;Meyer,T.J. J.Am.Chem.Soc, 1987,109,3287. 106. Groves,J.T.;Gilbert,J.A. Inorg.Chem., 1986,25,123. 107. Calderwood,T.S.;Bruice,T.C. Inorg.Chem.,1986,25,3722. 108. Che,C.-M.;Chung,W.-C. J.Chem.Soc.,Chem.Commun.,l9S6,3S6. 109. Dolphin,D.;Felton,R.H. Acc.Chem.Res., 1974,7,26. 110. Dolphin,D.;Forman,A.;Borg,D.C.;Fajer,J.;Felton,R.H. Proc.Natl.Acad.Sci., 1971,68,614. 111. EPR spectra were run at the University of British Columbia by Mr. Carl Jacobsori, whose 228 help is gratefully acknowledged. 112. Fiuirhop,J.-H.;Kadish,K.M.;Davis,D.G./.Am.CAem.Soc., 1973,95,5140. 113. (a) Scheidt,W.R.;Geiger,D.K.;Lee,Y.L.;Reed,C.A.;Lang,G. J.Am.Chem.Soc, 1985,107,5693. (b) Schol2,W.F.;Reed,C.A.;Lee,Y.L.;Scheidt,W.R.;Lang,G. J.Am.Chem.Soc. 1982.104.6791. (c) l^dmm,J.T.;Grimmett,D.;Haller,K.J.;Scheidt,W.R.;Reed,C.A. J.Am.Chem.Soc.,1981, 103.2640. 114. Kennedy,B.J.;McGram,A.C.;Murray,K.S.;Skelton.B.W.;White,A.H. Inorg.Chem.,1981,26,483. 115. (a) Scheidt,W.R.;Geiger,D.K.;Hayes,R.G.;Lang,G. J.Am.Chem.Soc. 1983.105.2625. (b) Geiger,D.K.;Lee,Y.L.;Scheidt,W.R. J.Am.Chem.Soc, 1984,106,6339. 116. Konig.E. Magnetic Properties of Coordination and Organometallic Transition Metal  Compounds.Landolt-Bornstein.New Series,Vol.8,Hellwege,K.-H.;Hellwege,A.M.,Eds., Springer-Verlag.Berlin, 1976. 117. Earnshaw,A.;Figgis,B.N.;Lewis,J.;Peacock,R.D. J.Chem.Soc, 1961,3132. 118. Horrocks,Jr.,W.D. in NMR of Paramagnetic Molecules. Principles and Applications,LaMar.G.N.; Horrocks,Jr.,W.D.;Holm,R.H.,Eds..Academic Press.New york,1983,pp.86-215. 119. Latimer.W.M. The Oxidation State of the Elements and their Potentials in Aqueous Solution, Prentice-Hall.New York,1938,pp.94-95. 120. Reference 77, pp. 1042-3. 121. Reference 118, Chapter 2. 122. Thackray,D.C;Ariel,S.;Leung,T.W.;Menon,K.;James,B.R.;Trotter,J. Can.J.Chem., 1986,64,2440. 123. Castro.C.E.;Anderson,S.E. Inorg.Chem., 1985,24,1113, and references therein. 124. Scheidt,W.R.;Lee,Y.L.;Tamai,S.;Hatano,K. y.Am.CAm.Soc, 1983,105,778. 125. Morishima,I.;Takamuki,Y.;Shiro,Y. J.Am.Chem.Soc. 1984.106.7666. 126. Brown,G.M.;Hopf,F.R.;Ferguson,J.A.,Meyer,T.J.,Whitten,D.G. J.Am.Chem.Soc, 1973,95,5939. 127. Bottomley,L.A.;01son,L.;Kadish,K.M. in Electrochemical and Spectroelectrochemical Studies of  Biological Redox Components.Advances in Chemistry Series,Vol. 210,Kadish.K.M.,Ed.,American 229 Chemical Society .Washington, 1982,pp.279-311. 128. Bottomley,L.A.;Kadish,K.M. Inorg.Chem., 1977,16,711. 128. Phillippi,M.A.;Shimomura,E.T.;Goff,H.M. Inorg.Chem., 1981,20,1322. 130. Reference 127, Chapter 24. 131. Moyer,B.A.;Sipe,B.K.;Meyer,T.J/norg.CAe/w.,1981,20,1475. 132. Ellis,P.E.,Jr.;Lyons,J.E.y.CAem.Soc.,CAm.Commun., 1989,1189. 230 Appendix 1. Information needed for the use of the NMR simulation program DNMR3. A typical input data deck used to simulate the curves described in Chapter 3 is shown below: line input 1 2 3 4 5 6 7 8 0011201 Ru(OEP)PPh 3 + PPh 3 0.200 0.70 0101 0.00 0.180 0.820 18.00 182.00 200.00 1.50 34.00 These lines correspond to the following input parameters: line 1: NRUN,N,NE,MU,NMES (shown here as 0011201). The 001 refers to run number (eg. ran 1 must be input as 001); N is the number of nuclei, (= 1); NE is the number of chemical configurations (2); MU is for the mutual exchange case (0 indicated that it was not used); NMES refers to the number of sets of magnetically equivalent nuclei (1). line 2: TEXT (e.g. title). line 3: CHEMICAL SHIFT for the first chemical configuration (18.00 Hz), line 4: CHEMICAL SHIFT for the second chemical configuration (182.00 Hz). line 5: NRS, LPUNCH, LPLOT, FR1, FR2, SCALE, HEIGHT (shown here as 0101 0.00 200.00 1.50 34.00). NRS is the number of sets of rate constants (01); LPUNCH allows die output to be obtained as punched-cards (0 represents no punched-card output); LPLOT gives the option to generate a plot of the calculated data (1 represents plot output is desired); FR1 is the left plot frequency in Hz (0.00); FR2 is the right plot frequency in Hz (200.00); SCALE allows the scale to be varied in units of mm/Hz (1.50); and HEIGHT allows the plots to be scaled to any height in mm to a maximum of 240 mm. line 6: POP(l), POP(2). Represents the populations of each of the nuclei having a different chemical configuration, normalized to a total of 1 (0.180 and 0.820). line 7: T 2 refers to the transverse relaxation time in seconds (0.20). line 8: RC represents the rate constants used for the run (0.70). 231 Since DNMR3 is a general program for the simulation of observed NMR signals, the input deck shown will need significant modification before utilization to simulate cases other than an exchange involving just two chemical configurations and one nuclei. The method for determining the position and population of each signal involved in die exchange mechanism at each temperature has already been described in Chapter 3. The T 2 value used for the simulation of each spectrum was taken to be the same as that for the solvent, as previously discussed. The program DNMR3 utilizes these parameters and a rate constant to calculate a function which yields an intensity vs. frequency plot (i.e. an NMR spectrum).64 Values for the rate constants were chosen on a 'best guess' basis until a reasonable fit was obtained, as estimated by comparing the simulated and actual spectra. The uncertainty in a rate constant was found to be +_ 5% at the 80% level of confidence. Once diree rate constants at three different temperatures were estimated, an Eyring plot was used to obtain other approximate rate constants corresponding to the odier temperamres. These rate constants were then adjusted (a small correction) individually to obtain a good fit of die simulated and experimental NMR spectra. Finally, these rate constants were then used in a second Eyring plot, and the activation parameters reported in Chapter 3 were calculated from the slope and the intercept of this plot using a least-squares fit program. In all cases, it was found that the uncertainty in the slope of the second Eyring plot was less that +_ 3% at the 9 5 % uncertainty level, and so estimates of the uncertainties in the activation parameters A H J and A S | of +_ 10% and +_ 2 0 % (at the 9 5 % imcertainty level), respectively, were presented with these data reported in Chapter 3. Several sources of experimental error contribute to the imcertainty limits of the activation parameters: (1) while the temperature measurement of the system was calibrated using a sample of neat methanol (this technique is described in the user manual for the Varian XL-300 NMR spectrometer), and found to have an error limit of _+ 1 K, mere are obvious differences such as sample size and concentration, and the type of NMR tube used between the standard sample used to calibrate the temperature and the actual sample studied. Large errors would originate from the 232 temperature measurement if the sample were not equilibrated at each temperature for a sufficiently long enough period of time (20 minutes); (2) shimming the magnet improperly was found to be a critical factor in determining the shapes of the lines recorded at each temperature. A technique which proved useful was to adjust the lock shims until the lock signal stabilized for a five-minute time period, and this procedure sometimes required 20-30 minutes of fine adjustment before the sample was run; (3) the comparison of the simulated and measured spectra was done by simply superposing one spectrum over another on a table illuminated via the underneath surface, a source of some additional error. 233 Appendix 2; Data obtained from the titration of Ru(OEP)CO by PPh 3 in C 7 H 8 as a function of  Temperature.-Temperature = 293 K. [Ru] = 7.40 x 10"6 M, Cell Volume = 3.00 mL, [PPh 3] s t o c k = 0.0867 M. Volume of PPh 3, (fth) 0.0 1.0 3.0 10.0 20.0 60.0 80.0 Absorbance (393 nm) 1.860 1.663 1.416 0.965 0.734 0.490 0.460 K = 1.47 x IO"4 M"1, slope = 0.77 [Ru] = 5.74 x 10"6 M, Cell Volume = 3.00 mL, [PPh 3] s t o c k = 0.0867 M. Volume of PPh 3, (/*L) 0.0 1.0 3.0 15.0 30.0 50.0 90.0 Absorbance (393 nm) 1.442 1.306 1.082 0.632 0.477 0.404 0.343 K = 1.57 x 10^ M"1, slope = 0.83 [Ru] = 3.87 x IO"6 M, Cell Volume = 3.00 mL, [PPh 3] s l o c k = 0.0867 M. . Volume of PPh 3, (u-L) 0.0 Absorbance (393 nm) 0.9972 K = 1.22 x IO"4 M"1, slope = 0.82 Temperature = 303 K. [Ru] = 6.57 x IO"6 M, Cell Volume Volume of PPh 3, (M L ) 0.0 Absorbance (393 nm) 1.650 K = 1.33 x IO"4 M"1, slope = 0.78 [Ru] = 2.47 x 10"6 M, Cell Volume Volume of PPh 3, (fiL) 0.0 Absorbance (393 nm) 0.621 K = 1.16 x I f f 4 M'1, slope = 0.78 1.0 3.0 15.0 30.0 50.0 90.0 0.878 0.756 0.522 0.328 0.253 0.217 = 4.00 mL, [PPh 3] s l o c k = 0.0867 M. 1.0 3.0 10.0 30.0 55.0 80.0 1.520 1.325 0.974 0.656 0.526 0.475 = 4.00 mL, [PPh 3] s t o c k = 0.0867 M. 1.0 3.0 10.0 30.0 55.0 80.0 0.566 0.486 0.366 0.259 0.216 0.202 Temperature = 313 K. [Ru] = 5.93 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3] s t o c k = 0.0867 M. Volumeof PPh 3, (/*L) 0.0 1.0 3.0 10.0 20.0 60.0 80.0 Absorbance (393 nm) 1.310 1.224 1.073 0.785 0.534 0.430 0.393 K = 1.39 x IO"4 M"1, slope = 0.76 [Ru] = 7.87 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3 ] s l o c k = 0.0867 M. Volumeof PPh 3, ( M L) 0.0 1.0 3.0 10.0 20.0 60.0 80.0 Absorbance (407 nm) 0.421 0.527 0.676 0.933 1.157 1.240 1.260 K = 1.40 x IO"4 M"1, slope = 0.81 Temperature = 323 K [Ru] = 2.85 x 10"6 M, Cell Volume = 4.00 mL, rPPh 3 ] s t o c k = 0.0599 M. Volumeof PPh 3, ( M L) 0.0 1.0 3.0 10.0 30.0 55.0 80.0 100.0 120.0 Absorbance (393 nm) 0.716 0.685 0.625 0.476 0.328 0.253 0.208 0.187 K = 1.52 x I f f 4 M"1, slope = 0.77 234 Appendix 2 continued. [Ru] = 3.77 x 1CT6 M, Cell Volume = 4.00 mL, [PPh 3 ] s t o c k = 0.0599 M. Volumeof PPh 3, (/*L) 0.0 1.0 3.0 10.0 30.0 55.0 80.0 100.0 120.0 Absorbance (393 nm) 0.948 0.910 0.833 0.655 0.447 0.356 - - 0.288 K = 1.59 x 10"* M 1 , slope = 0.77 [Ru] = 2.03 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3 ] s t o c k = 0.0599 M. Volume of PPh 3, (/xL) 0.0 1.0 3.0 10.0 30.0 55.0 80.0 100.0 120.0 Absorbance (393 nm) 0.509 0.481 0.432 0.342 0.235 0.188 - 0.145 K = 1.50 x 10"4 M 1 , slope = 0.91 Uncertainty in temperature +.0.1K. No correction for the amount of coordinated PPh 3 was made for [PPh 3] f because the uncertainty limits for each set of data at one temperature were overlapping with those obtained at the other temperatures, and because the [PPh 3] correction was small in comparison with the large scatter in the equilibrium constant. 235 Appendix 3: Data obtained from the titration of Ru(OEP)CO by PPh 3 in CDC1 3 as a  function temperature.8-Temperature = 293 K [Ru] = 4.10 x 10"6 M , Cell Volume = 4.00 mL, [PPh 3 ] s t o c k = 0,742 M. Volume of PPh 3, (/xL)* 0.0 1.0 6.0 16.0 40.0 65.0 90.0 Absorbance (393 nm) 1.029 0.822 0.694 0.535 0.363 0.282 0.238 K = 2.13 10"3 M" 1 , slope = 0.69 [Ru] = 2.81 x IO"6 M, Cell Volume = 4.00 mL, [PPh;,],^ = 0.742 M . Volume of PPh 3, ( M L ) 0.0 2.0 10.0 30.0 55.0 80.0 120.0 Absorbance (393 nm) 0.707 0.633 0.461 0.287 0.208 0.174 0.141 K = 1.97 x 10"3 M"1, slope = 0.79 [Ru] = 6.75 IO"6 M, Cell Volume = 5.00 mL, [PPh 3 ] s l o c k = 0.676 M Volume of PPh 3, ( M L ) 0.0 2.0 8.0 30.0 55.0 80.0 150.0 Absorbance (393 nm) 1.696 1.580 1.236 0.811 0.617 0.513 0.390 K = 1.89 x 10"3 M"1, slope = 0.83 Temperature = 303 K [Ru] = 7.01 x 10"6 M, Cell Volume = 5.00 mL, [PPh 3] s t o c k = 0.742 M Volumeof PPh 3, (M L) 0.0 2.0 10.0 30.0 55.0 80.0 150.0 Absorbance (393 nm) 1.762 1.625 1.272 0.859 0.651 0.545 0.409 K = 2.14 x 10"3 M"1, slope = 0.85 [Ru] = 3.10 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3] s t o c k = 0.742 M Volume of PPh 3, (»L) 0.0 2.0 8.0 30.0 55.0 80.0 150.0 Absorbance (393 nm) 0.778 0.703 0.543 0.315 0.229 0.194 0.137 K = 2.21 x 10"3 M"1, slope = 0.85 [Ru] = 2.50 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3 ] s t o c k = 0.847 M Volume of PPh 3, (M L) 0.0 2.0 8.0 20.0 45.0 200.0 Absorbance (393 nm) 0.342 0.311 0.249 0.188 0.142 0.105 K = 2.40 x 10"3 M"1, slope = 0.87 Temperature = 313 K [Ru] = 7.36 x IO"6 M , Cell Volume = 5.00 mL, [PPh 3 ] s l o c k = 0.824 M. Volume of PPh 3, (;tL) 0.0 2.0 8.0 20.0 45.0 70.0 150.0 Absorbance (393 nm) 1.847 1.715 1.439 1.120 0.827 0.695 0.517 K = 2.51 x 10"3 M"1, slope = 0.88 [Ru] = 5.63 x IO"6 M, Cell Volume = 4.00 mL, [PPh 3 ] s l o c k = 0.824 M. Volume of PPh 3, (uL) 0.0 2.0 8.0 20.0 45.0 70.0 150.0 Absorbance (393 nm) 1.415 1.286 1.030 0.789 0.549 0.455 0.346 K = 2.51 x 10"3 M" 1 , slope = 0.86 * includes 30 A»L of a 0.059 M PPh 3 stock solution. 236 Appendix 3 continued [Ru] = 3.94 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3 ] s l o c k = 0.824 M. Volume of PPh 3, (>L) 0.0 2.0 8.0 20.0 45.0 70.0 150.0 Absorbance (393 nm) 0.990 0.899 0.715 0.531 0.379 0.308 0.230 K = 2.52 x lO" 3 NT1, slope = 0.86 Temperature = 323 K [Ru] = 7.15 x 10'6 M, Cell Volume = 4.00 mL, [ P P h j ] ^ = 0.690 M. Volume of PPh 3, (/xL) 0.0 2.0 8.0 20.0 45.0 200.0 Absorbance (393 nm) 1.796 1.665 1.384 1.058 0.756 0.412 K = 2.93 x 10"3 M"1, slope = 0.93 [Ru] = 3.29 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3] s t C ) c k = 0.690 M. Volume of PPh 3, (fth) 0.0 2.0 8.0 20.0 45.0 200.0 Absorbance (393 nm) 0.826 0.765 0.633 0.489 0.359 0.206 K = 3.15 x 10-3M-', slope = 1.01 [Ru] = 7.04 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3] s t o c k = 0.690 M. Volume of PPh 3, ( M L) 0.0 2.0 8.0 20.0 45.0 200.0 Absorbance (393 nm) 1.769 1.640 1.366 1.045 0.748 0.418 K = 2.89 x 10"3 M->, slope = 0.92 Uncertainty in temperature _+ 0.1 K. No correction for the amount of coordinated PPh 3 for the calculation of [PPh 3] f was made because die correction was less than 3 %. 237 Appendix 4: Data obtained from the titration of Ru(OEP)PPh 3 by PPh 3 in C 7D 8 as a function of  temperature.-Temperature = 293 K [Ru] = 7.43 x 10"6 M , Cell Volume = 4.00 mL, [PPh 3 ] s t o c k = 0.0422 M. Volume of PPh 3, (uh) 0.0 0.5 1.0 1.5 2.0 25.0 40.0 Absorbance (395 nm) 0.769 0.590 0.514 0.458 0.427 0.298 0.289 Absorbance (422 nm) 1.166 1.443 1.557 1.636 1.683 1.857 1.866 K (at 395 nm) = 8.87 x 10"6 NT1, slope = 0.96 K (at 422 nm) = 8.04 x 10"6 M"1, slope = 0.95 [Ru] = 6.34 10"6 M , Cell Volume = 5.00 mL, [PPh 3] s t o c k = 0.676 M Volumeof PPh 3, (/xL) 0.0 0.5 1.0 1.5 2.0 25.0 40.0 Absorbance (395 nm) 0.699 0.526 0.453 0.412 0.384 0.267 0.259 Absorbance (422 nm) 0.926 1.199 1.320 1.372 1.418 1.583 1.593 K (at 395 nm) = 8.20 x 10"6 M"1, slope = 1.01 K (at 422 nm) = 7.52 x 10"6 M"1, slope = 1.00 Temperature = 303 K [Ru] = 5.97 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3 ] s l o c k = 0.0422 M. Volume of PPh 3, (fiL) 0.0 0.5 1.0 1.5 2.0 25.0 40.0 Absorbance (395 nm) 0.785 0.622 0.539 0.490 0.446 0.268 0.257 Absorbance (422 nm) 0.675 0.930 1.077 1.136 1.219 1.485 1.500 K (at 395 nm) = 1.20 x 10'5 M"1, slope = 1.01 K (at 422 nm) = 1.15 x 10"5 M"1, slope = 0.97 [Ru] = 6.52 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3] s t o c k = 0.0422 M. Volumeof PPh 3, ( M L) 0.0 0.5 1.0 1.5 2.0 25.0 40.0 Absorbance (395 nm) 0.847 0.700 0.608 0.539 0.489 0.310 0.299 Absorbance (422 nm) 0.741 0.994 1.171 1.291 1.378 1.628 1.639 K (at 395 nm) = 1.26 x 10"5 M"1, slope = 0.85 K (at 422 nm) = 1.09 x 10"5 M"1, slope = 0.76 Temperature = 313 K [Ru] = 2.93 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3] s t o c k = 0.0422 M. Volumeof PPh 3, (/iL) 0.0 0.5 1.0 1.5 2.0 25.0 40.0 Absorbance (395 nm) 0.606 0.521 0.474 0.439 0.393 0.245 0.233 Absorbance (422 nm) 0.211 0.338 0.422 0.466 0.499 0.729 0.736 K (at 395 nm) = 1.76 x 10"5 M"1, slope = 0.93 K (at 422 nm) = 1.67 x 10"5 M" 1, slope = 1.03 [Ru] = 8.94 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3 ] s t o c k = 0.0422 M. Volumeof PPh 3, (/iL) 0.0 0.5 1.0 1.5 2.0 25.0 40.0 Absorbance (395 nm) 1.371 1.154 0.997 0.887 0.791 0.396 0.373 K (at 395 nm) = 1.62 x 10'5 M"1, slope = 0.87 238 Appendix 4 continued Temperature = 323 K [Ru] = 6.82 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3] s t o c k = 0.0422 M. Volume of PPh 3, (^L) 0.0 1.0 2.0 3.0 4.0 25.0 40.0 Absorbance (395 nm) 1.134 0.071 0.758 0.680 0.623 0.350 0.327 Absorbance (422 nm) 0.507 0.844 1.059 1.194 1.285 1.665 1.713 K (at 395 nm) = 2.47 x 10"5 M"1, slope = 0.93 K (at 422 nm) = 2.53 x 10"5 M"1, slope = 0.95 [Ru] = 4.20 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3 ] s t o c k = 0.0422 M. Volumeof PPh 3, (/xL) 0.0 0.5 1.0 1.5 2.0 25.0 40.0 Absorbance (395 nm) 0.814 0.706 0.658 0.606 0.564 0.315 0.298 Absorbance (422 nm) 0.259 0.443 0.507 0.591 0.639 1.019 1.054 K (at 395 nm) = 2.33 x 10"5 M"1, slope = 1.07 K (at 422 nm) = 2.40 x 10"5 M"1, slope = 1.20 [Ru] = 4.49 x 10"6 M, Cell Volume = 4.00 mL, [PPh 3] s l o c k = 0.0422 M Volume of PPh 3, (/xL) 0.0 0.5 1.0 1.5 2.0 Absorbance (395 nm) 0.804 0.688 0.618 0.536 0.499 Absorbance (422 nm) 0.264 0.442 0.553 0.694 0.742 K (at 395 nm) = 2.56 x 10"5 M"1, slope = 1.18 K (at 422 nm) = 2.30 x 10"5 M"1, slope = 1.11 - Uncertainty in temperature .+ 0.1 K. [PPh 3] f, the free [PPh 3] in solution was corrected for partial coordination of die added PPh 3 to Ru(OEP)PPh 3 using [PPh 3] f = [PPh 3] t o l. [(A-A 0)/(A 0-Aj)][Ru], as described in Chapter 3. 25.0 40.0 0.261 0.231 1.108 1.129 239 Appendix 5. Raw data for the reaction of CH 2C1 2 with rRu(OEP)l 2 at 293 K, Entry 1 [Ru 2] = 0.17 mM [CH 2C1 2] = 15.6 M Time.s ^540 run 0 ~0.235 100 0.256 200 0.265 300 0.272 400 0.277 500 0.284 600 0.287 700 0.290 800 0.293 900 0.295 1000 0.296 Entry 2 [Ru 2] = 0.17 mM [CH 2C1 2] = 15.6 M Time, s y^ 516 nm 0 ~0.212 100 0.221 200 0.224 300 0.226 400 0.227 500 0.233 600 0.234 700 0.235 800 0.237 900 0.237 1000 0.237 Entry 3 [Ru 2] = 0.19 mM [CH 2C1 2] = 15.6 M Time, s ^ 540 nm 0 ~ 0.281 100 0.296 200 0.306 300 0.311 400 0.316 500 0.321 600 0.324 700 0.326 800 0.328 900 0.329 1000 0.331 Entry 4 [Ru 2] = 0.19 mM [CH 2C1 2] = 15.6 M Time, s ^ 516 nm 0 ~~ 0.265 100 0.268 200 0.271 300 0.273 400 0.275 500 0.277 600 0.278 700 0.279 800 0.280 900 0.280 1000 0.280 Entry 5 Entry 6 Entry 7* Entry 8 [Ru 2] = 0.28 mM [Ru 2] = 0.37 mM [Ru 2] = 0.39 mM [Ru 2] = 0.46 mM [CH 2C1 2] = 15.6 M [CH 2C1 2] = = 15.6M [CH 2C1 2] = 15.6 M [CH 2C1 2] = 15.6 M Time.s y^ 540 nm Time, s ^540 nm Time, s A 5 4 0 n m Time.s A 5 4 0 n m 0 ^0.390 0 ~0.334 0 0.522 0 0.542 84 0.429 260 0.373 84 0.565 200 0.618 168 0.455 400 0.409 168 0.576 400 0.659 252 0.475 540 0.432 252 0.620 600 0.689 336 0.487 680 0.450 336 0.640 800 0.711 420 0.502 820 0.469 420 0.654 1000 0.730 504 0.509 960 0.484 504 0.670 1200 0.748 588 0.515 1100 0.465 588 0.681 1400 0.762 672 0.519 1240 0.505 672 0.693 1600 0.772 00 0.575 1380 0.514 756 0.702 1800 0.783 1520 0.522 840 0.711 2000 0.791 1660 0.529 924 0.720 1800 0.535 1008 0.726 2220 0.552 1092 0.736 00 0.585 1176 0.740 00 0.794 * These data were used for plot shown in Figure 4.1 lb in Section 4.2.e. 240 Appendix 5 continued. Entry 9 [Ru 2] = 0.46 mM [CH 2C1 2] = 15.6 M Time.s ^516 nm 0 ~0.484 200 0.513 400 0.529 600 0.540 800 0.548 1000 0.554 1200 0.561 1400 0.566 1600 0.571 1800 0.575 2000 0.578 Entry 13 [Ru 2] = 0.92 mM [CH 2C1 2] = 15.6 M Time.s ^516 nm 0 1.285 300 1.294 600 1.301 900 1.308 1200 1.315 1500 1.319 1800 1.326 2100 1.332 2400 1.338 2700 1.342 3000 1.349 3300 1.354 3600 1.358 00 1.687 Entry 10 [Ru 2] = 0.57 mM [CH 2C1 2] = 15.6 M Time, s y^ 540 nm 0 ~0.670 420 0.780 720 0.849 1020 0.904 1320 0.934 1620 0.972 1920 0.994 2220 1.028 2520 1.045 2820 1.056 00 1.108 Entry 14 [Ru 2] = 1.25 mM [CH 2C1 2] = 15.6 M Time, s ^540 nm 0 ~0.934 720 1.156 1320 1.249 1920 1.308 2520 1.351 3120 1.383 3720 1.417 4320 1.436 4920 1.459 5520 1.475 6120 1.494 6720 1.503 7320 1.542 Entry 11 [Ru 2] = 0.91 mM [CH 2C1 2] = 15.6 M Time, s A 540 nm 0 ~ 0.931 40 1.050 340 1.197 640 1.284 940 1.341 1240 1.383 1540 1.417 1840 1.444 2140 1.469 2440 1.495 2740 1.517 3040 1.531 00 1.575 Entry 15 [Ru 2] = 4.43 mM [CH 2C1 2] = 15.6 M Time, s A 540 nm 0 0.412 300 0.437 600 0.453 1800 0.479 3000 0.497 4200 0.517 5400 0.531 6600 0.562 7800 0.574 9000 0.588 10200 0.603 11400 0.618 12600 0.631 13800 0.640 00 0.668 Entry 12 [Ru 2] = 0.92 mM [CH 2C1 2] = 15.6 M Time, s y^ 540 nm 0 1.420 300 1.441 600 1.459 900 1.475 1200 1.490 1500 1.506 1800 1.520 2100 1.531 2400 1.546 2700 1.557 3000 1.568 3300 1.580 3600 1.593 00 1.895 Entry 16 [Ru 2] = 0.61 mM [CH 2C1 2] = 8.87 M Time, s ^ 540 nm 0 ~ 0.840 600 0.855 1200 0.907 1800 0.921 2400 0.939 3000 0.958 3600 0.968 4200 0.981 4800 0.990 5400 1.000 6000 1.009 00 1.053 241 Entry 17 Entry 18 Entry 19 Entry 20 [Ru 2] = 0.73 mM [CH 2C1 2] = 1.97 M Time.s ^540 nm 0 ~b.794 1800 0.822 3600 0.843 5400 0.861 7200 0.875 9000 0.886 10800 0.896 12600 0.903 14400 0.910 16200 0.915 18000 0.920 00 0.922 [Ru 2] = 0.73 mM [CH 2C1 2] = 1.97 M Time, s y^ 516 nm 0 _0.810 1800 0.818 3600 0.825 5400 0.832 7200 0.836 9000 0.840 10800 0.843 12600 0.846 14400 0.848 16200 0.851 18000 0.851 00 0.853 [Ru 2] = 1.20 mM [CH 2C1 2] = 0.9 M Time, s y^ 540 nm 0 ~~ 1.307 600 1.369 1800 1.392 3000 1.415 4200 1.448 5400 1.482 6600 1.524 7800 1.566 9000 1.609 00 1.985 [Ru 2] = 1.20 mM [CH 2C1 2] = 0.9 M Time, s y^ 516 nm 0 ~ 1.297 600 1.342 1800 1.356 3000 1.372 4200 1.395 5400 1.423 6600 1.451 7800 1.482 9000 1.512 00 2.073 242 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059668/manifest

Comment

Related Items