The Synthesis and Study of Redox-Rich, Amido-Bridged Cu₂N₂ Dicopper Complexes Citation Harkins, Seth Beebe (2006) The Synthesis and Study of Redox-Rich, Amido-Bridged Cu₂N₂ Dicopper Complexes. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/1X2Q-3P72. https://resolver.caltech.edu/CaltechETD:etd-08132005-093856 Abstract A Cu 2 N 2 diamond core structure supported by an [SNS] - ligand exhibits a fully reversible one-electron redox process between a reduced Cu I Cu I , {(SNS)Cu} 2 , and a class III delocalized Cu 1.5 Cu 1.5 state, {(SNS)Cu} 2 ][B(C 6 H 3 (CF 3 ) 2 ) 4 ] ([SNS] - bis(2-t-butylsulfanylphenyl)amide). The Cu···Cu distance compresses appreciably (~0.13 Å) upon oxidation; a metal-metal distance of 2.4724(4) Å is observed in the mixed-valence molecule that is nearly identical to the dicopper Cu A site found in cytochrome c oxidase. The rate of electron self-exchange k s ) between the Cu I Cu I and the Cu 1.5 Cu 1.5 complexes was estimated to be ≥ 10 7 M -1 s -1 by 1 H NMR line-broadening analysis. The unusually large magnitude of k s reflects the minimal structural reorganization that accompanies Cu I Cu I ↔ Cu 1.5 Cu 1.5 interchange. A second generation of {(PNP)Cu I } 2 dimer supported by a [PNP] - ligand also has been investigated ([PNP] - = bis(2-(diisobutylphosphino)phenyl)amide). The highly emissive {(PNP)Cu I } 2 is characterized by a long-lived excited state (τ > 10 μs) with an unusually high quantum yield (φ > 0.65) at ambient temperature. Removal of an electron from the {(PNP)Cu I } 2 dimer yields a nearly isostructural, Cu 1.5 Cu 1.5 complex {(PNP)Cu} 2 ][B(C 6 H 3 (CF 3 ) 2 ) 4 ]. With a highly reducing excited state reduction potential (~ -3.2 V vs. Fc + /Fc) as well as the availability of two reversible redox processes, these bimetallic copper systems may be interesting candidates for photochemically driven two-electron redox transformations. Studies of Cu 2 N 2 diamond core complexes supported by the [ t Bu 2 -PNP] - ligand revealed that the dicopper complex {( t Bu 2 -PNP)Cu} 2 can not only be oxidized by one electron to {( t Bu 2 -PNP)Cu 1.5 } 2 ][B(C 6 H 3 (CF 3 ) 2 ) 4 ], but also by two-electrons to {( t Bu 2 -PNP)Cu} 2 ][SbF 6 ] 2 ([ t Bu 2 -PNP] - = bis(2-diisobutylphoshino-4- t butylphenyl)amide). These Cu 2 N 2 complexes show remarkably low structural reorganization for all oxidation states as evidenced by the solid-state molecular-structures. Based on these studies of [{( t Bu 2 -PNP)Cu} 2 ][SbF 6 ] 2 , we propose a formulation of one Cu I and one paramagnetic Cu III nuclei in compressed-tetrahedral environments in the Cu 2 N 2 core. Spectroscopic, redox, and magnetic data are consistent with a highly covalent M 2 N 2 core supported by a rigid ligand scaffold. These complexes are excellent mimics of the entatic state found in bimetallic copper proteins. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: amido-ligand; coordination chemistry; copper(III); diamond core; dicopper; highly emissive; luminescence; redox active Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Peters, Jonas C. Thesis Committee: Tirrell, David A. (chair) Gray, Harry B. Bercaw, John E. Peters, Jonas C. Defense Date: 27 June 2005 Record Number: CaltechETD:etd-08132005-093856 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-08132005-093856 DOI: 10.7907/1X2Q-3P72 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 3109 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 16 Aug 2005 Last Modified: 08 Nov 2023 00:44 Thesis Files Preview PDF (Title-ch1-ch2-ch3.pdf) - Final Version See Usage Policy. 2MB Preview PDF (Ch4.pdf) - Final Version See Usage Policy. 981kB Preview PDF (AppendixA.pdf) - Final Version See Usage Policy. 402kB Preview PDF (AppendixB.pdf) - Final Version See Usage Policy. 873kB Repository Staff Only: item control page

A-1 Appendix A Facial coordination of a pincer-like amido complex of platinum(IV) generated by photoisomerization† † Adapted from Harkins, S. B.; Peters, J. C.; Facial coordination of a pincer-like amido complex of platinum(IV) generated by photoisomerization. Inorg. Chem., to be submitted for publication. A-2 Introduction A growing number of mono-anionic chelating pincer-like amido ligands which contain a diaryl amide functionality {(LNL)¯} are currently being explored. 1, 2 , 3 , 4 , 5 Prior to this manuscript, these metal complexes have been described exclusively with meridinal ligand coordination. Our group and others have described cationic Pt complexes supported by (LNL)¯ ligands that have been shown to mediate C-H bond activation of benzene.b,d It is well documented that PtIV intermediates are involved in the catalytic oxidation of alkanes in Shilov-like processes using PtII catalysts. 6,7 Although in related chelating nitrogen ligand complexes of PtII, oxidation to PtIV with alkyl halides is readily achieved, the discrete oxidative addition of a substrate to an {(LNL)PtII} complex has yet to be demonstrated. 8 This manuscript describes the oxidative addition of MeI and subsequent photoisomerization and protonation of (BQA)PtMe, 1a {BQA = bis(8quinolinyl)amine}. We have previously reported the synthesis and characterization of 1a, and its protonation with triflic acid to afford the neutral complex (BQA)Pt(OTf).b This triflato complex undergoes base-promoted C-H activation of benzene at 150 °C to generate (BQA)PtPh (1b) in high yield.b Subsequently, we have found that addition of a catalytic amount of B(C6F5)3 (0.2 equivalents) to 1a in benzene quantitatively generates 1b in 5 h at 70 ºC. The focus of our current work has been to examine the intriguing behavior of (BQA)PtIV complexes, and we have been able to induce a facile ligand isomerization which affords an unprecedented sp3-hybridized diarylamido moiety. A-3 Results and Discussion Figure 1: Synthesis of 1b, 2, 3, and 4. Conditions: (i) 0.2 equiv. B(C6F5)3, C6H6, 5 h, 20 ºC; (ii) 10 equiv. MeI, acetone, 70 °C, 18 h; (iii) iodide dissociation; (iv) methyl scrambling; (v) acetone, 100 W incandescent light, 48 h; (vi) HBF4 etherate, CH2Cl2, 2h. Addition of MeI to 1a in acetone results in the formation of (mer-BQA)PtMe2I, 2, as a purple solid in very good yield (Figure 1). This reaction proceeds sluggishly at 25 °C in comparison to most oxidative additions reported for organoplatinum(II) complexes with nitrogen-donor ligands and must be heated at 70º C for 18 h to reach completion. Complex 2 has λmax= 534 nm (ε 15 500 M-1cm-1) which is slightly shifted to lower energy from 1a, which has λmax= 527 nm (ε 12 000 M-1cm-1).b There are three resonances in the 1H NMR which exhibit strong 195Pt-H coupling: the 2-position proton of the quinoline ring, 8.71 ppm (3JPtH =41 Hz); the equatorial methyl group, 1.67 (2JPtH = 60 Hz); and the axial methyl group, 1.43 (2JPtH = 70 Hz). These initial assignments were made based on literature values for the 2JPtH coupling constants 9 and were confirmed by a 1D-NOE 1H NMR experiment. The NOE experiment was performed by saturation A-4 (0.5 s mixing time) of the 2-position proton (8.71 ppm) of the quinoline ring. Coupling was observed to the equatorial methyl protons (1.67 ppm) and no coupling to the axial methyl protons (1.43 ppm) was observed. The solid-state structure of 2 was determined by X-ray diffraction (Figure 2), confirming the cis relationship between the methyl groups. Figure 2: Solid-state molecular structures (50% ellipsoids) for 2, 3, and 4 (H atoms and BF4 anion omitted). Table 1: Bond Lengths and Angles for (mer-BQA)PtMe2I (2), (fac-BQA)PtMe2I (3), [H(fac-BQA)PtMe2I][BF4] (4) 2 3 4 Pt-N1 2.008(4) Å 2.137(3) Å 2.152(4) Å Pt-N2 2.033(3) Å 2.036(3) Å 2.135(6) Å Pt-N3 2.017(4) Å 2.146(3) Å −b Pt-I 2.783(1) Å 2.630(2) Å 2.598(1) Å Pt-C19{10} a 2.088(5) Å 2.052(3) Å 2.152(4) Å Pt-C20 2.102(5) Å 2.045(3) Å − b N2-Pt-I 90.16(9)° 177.73(7)° 177.95(15)° N2-Pt-N1 81.47(16)° 81.06(10)° 80.62(11)° N2-Pt-N3 81.82(16)° 80.53(10)° − b N1-Pt-N3 163.22(17)° 97.50(10)° 94.67(12)° C19{10}-PtC20{10} a 87.15(17)° 90.55(12)° 87.08(16)° C8-N2-C17{8'} c 131.93(42)° 114.82(25)° 114.02(17)° a: C17, C19 and C20 corresponds to 2 and 3, C8' and C10 corresponds to 4. b: values are omitted A-5 because they are generated by the symmetry transformation symmetry transformation {x,-y+1/2,z} {x,-y+1/2,z}; c: C8' generated by a The oxidative addition was also performed using d3-MeI under the same conditions, and a 1:1 mixture of axial and equatorial CD3 products was observed, as confirmed by both 1H and 2H NMR experiments. This result is indicative of iodide dissociation in solution which facilitates mixing of the labeled methyl group between the axial and equatorial positions (iii and iv, Figure 1). The 1:1 ratio is expected in this case because both alkyl groups (CH3, CD3) are of comparable size and there is no steric preference. These findings are consistent with reports by Puddephatt and co-workers, where analogous results were observed for {cis-1-(N=CHC6H4)-2-(N=CHC6H5)C6H10}PtMe2I. During our study of 2, we encountered a new product which seemed to grow in over time. It was found that photolysis of 2 in acetone resulted quantitatively in the ligand isomerization product fac-BQAPt(Me)2I, 3 (Figure 1). 10 We initially hypothesized that the new isomer was an octahedral PtIV complex where the methyl groups had migrated from a cis to a trans conformation and that the iodo group migrates trans to the amide of the mer-BQA ligand. The 1H NMR spectrum displays a single methyl resonance at 1.25 ppm (2JPtH = 71 Hz) which integrates to 6H and exhibits no NOE to the 2-position protons of the quinoline rings. The solid-state molecular structure of 3 (Figure 2) was determined by X-ray diffraction and revealed that the initial model was inaccurate. The structure of 3 is in fact an octahedral PtIV complex where the BQA ligand is coordinated facially 11 and the two methyl groups are each trans to the neutral quinoline nitrogen ligands and cis to each other (90.55(12)°). The iodo ligand is trans to the amido-nitrogen and the Pt-I bond has shortened by 0.15 Å which is reflective of an increased bonding interaction (Table 1). This is likely a consequence of decreased electron density at Pt resulting from diminished A-6 lone-pair donation from the pyramidalized amide-N. The most interesting structural aspect is that the amide is now pyramidalized with sp3-hydridization, characterized by the contraction of C8-N2-C17 angle from 131.93(42)° in 2 to 114.82(25)° in 3. This geometry for nitrogen has not been reported previously for transition metal diaryl amide complexes. 12 Complex 3 is red in color, rather than purple as observed for 2 (Figure 3). It is probable that this results from disruption of the amide-nitrogen lone pair donation to the metal center caused when the ligand assumes a folded geometry. To the best of our knowledge an anionic tridentate ligand that is able to isomerize from mer to fac in an octahedral complex has not been reported. Tridentate, monoanionic ligands have been reported that bind to Pt to form either square planar PtII complexes or octahedral PtIV complexes with facial geometries. 13,14 The (BQA)PtIV system is particularly intriguing because the reaction is photolytically driven. Shaw and coworkers have reported the octahedral complex L'Ru(CO)Cl2 {L' = ((Ph2P)CH2(tBu)C=N- )2}which undergoes a related mer to fac photoisomerization. However, (BQA)PtMe2I and L'Ru(CO)Cl2 are largely dissimilar because the latter involves isomerization resulting from the photolytic dissociation of CO. 15 Shaw’s system also does not present the highly pyramidalized geometry about the central nitrogen. Figure 3: Absorption spectra of 2, 3 and 4 in CH2Cl2 A-7 The tetrahedral geometry at the amide nitrogen suggests that the lone pair on nitrogen donates to Pt to a much lesser extent in 3 than 2 which is also indicated by the drastic change in the Pt-amide MLCT band from λmax= 534 nm in 2 to λmax= 422 nm in 3. This charge transfer assignment is supported by empirical observation that addition of HBF4 etherate to 3 in CH2Cl2 results in the immediate bleach of the absorption band at 422 nm while preserving the lower energy band at 510 nm (Figure 3). The protonated complex 4 (Figure 1) was isolated in near quantitative yield (97%) as a rose colored solid. By 1H NMR spectroscopy a distinct downfield resonance assigned to N-H is now present (10.33 ppm) and integrates to one proton and exhibits easily discernable 195Pt satellites (2JPtH = 21 Hz). The observation of a single methyl resonance (1.54 ppm) with integration of 6H and unchanged 195Pt coupling (2JPtH = 70 Hz) also indicates that this protonation has not significantly effected the geometry of the complex. A 1D-NOE (1.0 s mixing time) experiment with saturation of the N-H proton resonance (10.33 ppm) indicated coupling to both the 8-position proton of the aryl ring (8.78 ppm) as well as to the methyl resonance (1.54 ppm). A new, broad absorption in the IR spectrum at ν = 3127 cm-1 attributed to an N-H vibration is consistent with the assignment of 4. The absolute structure of 4 was determined by X-ray diffraction (Figure 3). Notably, the Pt-amide and Pt- methyl bond lengths each elongate by 0.1 Å. Additionally, H11, the proton on N2, is hydrogen bonded to the BF4¯ counter ion (N2−F1: 2.821(7) Å). This is also manifested in the broadened N-H stretch in the solid and solution IR spectrum of 4. Neutral diaryl amine adducts of transition metals are exceedingly rare and have only been reported as a part of chelating ligands. The Pt-N2 bond of 4 is the shortest of this type to be reported. For comparison unchelated diphenylamine adducts of AlCl3 16 and A-8 GaMe2Cl 17 have been reported which exhibit Caryl-N-Caryl bond angles of 110.5(4)º and 105.2(2)º, respectively. These angles are somewhat more contracted than the comparable C8-N2-C8' angle of 3 (114.82(25)°) and 4 (114.02(17)º). A possible mechanism for this reaction proceeds through an intermediate whereby the iodide ligand must first dissociate. A photon absorbed by this five-coordinate intermediate then destabilizes the metal-amide bond allowing for relaxation of the structure to accommodate a more sp3-hybridized geometry about the amide nitrogen. This folded geometry is then trapped by recoordination of the iodide. This postulated mechanism is supported by the observation that the iodide ligand readily dissociates in solution, allowing for the scrambling of the methyl groups. The isomerization is also severely attenuated when photolyzed in a saturated NaI solution in acetone, which would be expected to inhibit iodide dissociation. Shortening of the Pt-I bond distance in 3 compared to 2, with no appreciable change in the Pt-N2 bond distance, supports the idea of a trapped structure resulting from a stronger Pt-I interaction. In conclusion, we have presented the first example of an octahedral PtIV complex containing a diaryl amide moiety. The observation that the (BQA)PtIV system undergoes an isomerization which results in a facially capped metal complex could lead to the design of novel C-H bond functionalization pathways via proton shuttling followed by reductive elimination in a catalytic cycle. Research in our laboratory into the unique chemical bonding of chelating diaryl amide ligands to metals is ongoing. A-9 Experimental Section General: All manipulations were carried out using standard Schlenk or glove box techniques under a dinitrogen atmosphere. Unless otherwise noted, solvents were deoxygenated and dried thoroughly by sparging with N2 gas followed by passage through an activated alumina column. Non-halogenated solvents were typically tested with a standard purple solution of sodium benzophenone ketyl in tetrahydrofuran to confirm effective oxygen and moisture removal. The preparation of (BQA)PtMe was previously reported.1b Other reagents were purchased from commercial vendors and used without further purification. Elemental analyses were performed by Desert Analytics, Tucson, Az. A Varian Mercury-300 NMR spectrometer or a Varian Inova-500 NMR spectrometer was used to record 1H, 2H, 13C, and 19F NMR spectra. 1D-NOE experiments were conducted using the Varian GLIDE software package. 1H and 13C NMR chemical shifts were referenced to residual solvent. 19 F NMR chemical shifts were referenced to an external C6F6 sample with a chemical shift of –165 ppm. Deuterated solvents were purchased from Cambridge Isotope Labs and were degassed and dried over activated 3 Å molecular sieves prior to use. UV-vis spectra were collected on a Cary 50 spectrophotometer in CH2Cl2 with 1 cm quartz cuvettes. IR measurements were obtained with a KBr solution cell using a Bio-Rad Excalibur FTS 3000 spectrometer controlled by Bio-Rad Merlin Software (v. 2.97) set at 4 cm-1 resolution. X-ray diffraction studies were carried out at the Beckman Institute Crystallographic Facility on a Brüker Smart 1000 CCD diffractometer and solved using SHELX v. 6.14. Synthesis of (mer-BQA)Pt(CH3)2I / (mer-BQA)Pt(CH3)(CD3)I, 2: A solution of (BQA)PtMe (400 mg, 0.833 mmol), CH3I (103 μL, 1.67 mmol), and acetone (50 mL) were combined in a glass reaction bomb, sealed with its Teflon stopcock, and heated to A-10 70 °C for 18 h in the absence of light. The purple solution was evaporated to dryness under reduced pressure, and washed with petroleum ether (2 x 40 mL). Drying in vacuo of the powdery-purple solid (470 mg, 92%) yield a spectroscopically pure product. X-ray quality crystals were obtained by vapor diffusion of petroleum ether into acetone. Reactions with CD3I proceeded under identical conditions. Characterization data for the all protio form: 1H NMR (CD2Cl2, 300 MHz, 25 °C): δ 8.71 (m, 2JPtH = 41 Hz, 2H), 8.30 (d, 2H), 7.77 (d, 2H), 7.66 (t, 2H), 7.48 (m, 2H), 7.16 (d, 2H), 1.67 (s, 2JPtH = 60 Hz, 3H), 1.43 (s, 2JPtH = 70 Hz, 3H). 13C NMR (CD2Cl2, 126 MHz, 25 °C): δ 155.1, 146.7, 140.3, 139.4, 133.4, 130.9, 122.3, 115.7, 115.6, 15.2 (1JPtC = 550 Hz), 1.89 (1JPtC = 582 Hz). UV-vis (nm (ε M-1 cm-1), CH2Cl2): 292 (30000), 300 (27100), sh 396 (2900), 388 (3400), 534 (15500). IR (cm-1, KBr): 3053 (vw), 2902 (w), 1582 (m), 1564 (s), 1496 (s), 1463 (s), 1400 (s), 1225 (w), 1180 (m), 1134 (m), 813 (m), 770 (m), 736 (m). Anal. Calcd. for C20H18IN3Pt: C, 38.60; H, 2.92; N, 6.75. Found: C, 38.76; H, 3.15, N, 6.51. Synthesis of (fac-BQA)Pt(CH3)2I / (fac-BQA)Pt(CH3)(CD3)I, 3: A solution of 2 (196 mg, 0.315 mmol) in acetone (20 mL) was added to a 200 mL reaction bomb, sealed with its Teflon stopcock, and placed 2 in. underneath a 100 watt incandescent light bulb for 48 h, with periodic agitation. During this period, the solution went from purple to red in color. The solvent was removed under reduced pressure affording a red solid which was washed with petroleum ether and dried in vacuo (196 mg, >99%). Quantitative conversion with was observed by NMR spectroscopy. Reactions (mer- BQA)Pt(CH3)(CD3)I proceeded under identical conditions. Characterization data for the all protio form: 1H NMR (CD2Cl2, 300 MHz, 25 °C): δ 9.80 (m, 3JPtH = 17 Hz, 2H), 8.25 (m, 2H), 7.96 (d, 2H), 7.58-7.47 (m, 4H), 7.40 (d, 2H), 1.25 (s {Pt-CH3}, 2JPtH = 71 Hz, 6H). 13C NMR (CD2Cl2, 126 MHz, 25 °C): δ 159.7, 150.2, 147.1, 138.3, 131.1, 129.1, A-11 123.4, 122.7, 121.2, -8.0 (1JPtC = 602 Hz). UV-vis (nm (ε M-1 cm-1), CH2Cl2): 284 (15900), sh 298 (10600), sh 325 (4000), 422 (4700), sh 502 (1300). IR (cm-1, KBr): 3051 (w), 2962 (w), 2897 (m), 1575 (m), 1565 (m), 1497 (s), 1458 (s), 1374 (s), 1308 (m), 1256 (m), 1224 (m), 1113 (m), 1056 (w), 1031 (w), 841 (w), 810 (m), 772 (m). Anal. Calcd. for C20H18IN3Pt: C, 38.60; H, 2.92; N, 6.75. Found: C, 38.81; H, 3.18, N, 6.50. Synthesis of [H(fac-BQA)Pt(CH3)2I][BF4], 4. A 54 wt% solution of HBF4 in ether (19 μL, 0.137 mmol) was added to a solution of 4 (81 mg, 0.130 mmol) in CH2Cl2 (7 mL) in a glass scintillation vial. The reaction was stirred for 2 h at 25 °C and the product was precipitated by addition of petroleum ether (13 mL). The rose colored solid (90 mg, 97%) was collected on a fritted glass funnel and dried in vacuo. The product was found to be pure by 1H NMR spectroscopy and micro analysis. Crystals for X-ray diffraction were obtained by diffusion of petroleum ether in to a concentrated solution of 4 in THF. 1 H NMR (CD2Cl2, 300 MHz, 25 °C): δ 10.33 (s {N-H}, 2JPtH = 21 Hz, 1H), 9.90 (m, 3JPtH =17 Hz, 2H), 8.78 (d, 2H), 8.50 (m, 2H), 7.96 (m, 2H), 7.83 (t, 2H), 7.74 (m, 2H) 1.54 (s {Pt-CH3}, 2JPtH = 70 Hz, 6H). 13C NMR (CD2Cl2, 126 MHz, 25 °C): δ 152.0, 144.8, 143.7, 139.9, 131.5, 130.5, 130.1, 129.9, 125.0, -12.41 (1JPtC = 514 Hz). 19F NMR (CD2Cl2, 282 MHz, 25°C): δ -147.4 (d, 1JBF = 15 Hz). UV-vis (nm (ε M-1 cm-1), CH2Cl2): 302 (12900), 317 (10000), sh 240 (1700), 510 (630). IR (cm-1, KBr): 3077 (br, N-H), 2978 (w), 2903 (w), 1510 (s), 1467 (m), 1398 (m), 1358 (m), 1221 (w), 1081 (s br, B-F), 835 (m), 798 (m), 769 (m) 734 (w). IR (cm-1, CH2Cl2 sol. in KBr): 3127 (br, N-H), 2910 (w), 1591 (w), 1567 (w), 1513 1467 (w), 1359 (m), 1080 (s br), 1000 (m), 835 (m). Anal. Calcd. for C20H19BF4IN3Pt: C, 33.82; H, 2.70; N, 5.92. Found: C, 33.84; H, 2.46, N, 5.73. A.4. References Cited A-12 1 (a) Peters, J. C.; Harkins, S. B.; Brown, S. D.; Day, M. W. Inorg. Chem. 2001, 40, 5083. (b) Harkins, S. B.; Peters, J. C. Organometallics 2002, 21, 1753. (c) Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 2885. d) Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 2030. 2 (a) Ozerov, O. V.; Guo, C.; Papkov, V. A.; Foxman, B. M. J. Am. Chem. Soc. 2004, 126, 4792. (b) Ozeroz, O. V.; Pink, M.; Watson, L. A.; Caulton, K. G. J. Am. Chem. Soc. 2004, 126, 2105. 3 (a) Liang, L.-C.; Lin, J.-M.; Hung, C.-H. Organometallics 2003, 15, 3007. (b) Huang, M-H; Liang, L-C. Organometallics 2004, 23, 2813. (c) Liang, L-C; Chien, P-S; Huang, M-H. Organometallics 2005, 24, 353. (d) Liang, L-C; Lin, J.-M.; Lee, W.-Y. Chem. Commun. 2005, AA. 4 Bailey, B. C.; Huffman, J. C.; Mindiola, D. J.; Weng, W.; Ozerov, O. V. Organometallics 2005, 24, 1390. 5 Winter, A. M.; Eichele, K.; Mack, H.-G.; Potuznik, S.; Mayer, H. A.; Kaska, W. C.; J. Organomet. Chem. 2003, 682, 149. 6 Shilov, A. E.; Shul'pin, G. B. Chem. Rev. 1997, 97, 2879. 7 (a) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1996, 118, 5961. (b) Stahl, S.; Labinger, J. A.; Bercaw, . E. Angew. Chem. Int. Ed. 1998, 37, 2181. 8 Rendina, L. M.; Puddephatt, R. J. Chem. Rev. 1997, 97, 1735; and references therein. 9 Baar, C. R.; Jenkins, H. A; Vittal, J. J.; Yap, G. P. A; Puddephatt, R. J. Organometallics 1998, 17, 2805. 10 Thermolysis of 2 in acetone (70 ºC) in the absence of light afforded no trace of 3 after 24 h. A-13 11 The angle of intersection of the planes defined by the two quinoline arms is 97.5°. 12 All angles around nitrogen <117º. Cambridge Structural Database; Cambridge University: Cambridge, England (accessed April 2005). 13 Canty, A. J.; Patel, J.; Skelton, B. W.; White, A. H. J. Organomet. Chem. 2000, 195- 199. 14 Hartshorn, C.; Steel, P.J. Organometallics 1998, 17, 3487. 15 Shaw, B. L.; Ike, U. U.; Perera, S. D.; Thorton-Pett, M. Inorg. Chim. Acta. 1998, 279, 95. 16 Burford, N.; Hicks, R. G.; Royan, B. W.; Borecka, B.; Cameron, T. S. Acta. Cryst. C. 1991, C47, 1066. 17 Niemeyer, M.; Goodwin, T. J.; Risbud, S. H.; Power, P. P. Chem. Mater. 1996, 8, 2745.  B-1 Appendix B Undiscussed solid-state molecular structures B-2 N N N Cu N N N (BQA)2Cu ................................................................................................................. B-4 Ad S S Cu N N S Ad Ad Cu S Ad {(AdSNS)Cu}2 ...........................................................................................................B-5 PMe3 S Cu PMe3 N S (PMe3)2Cu(SNS) ....................................................................................................... B-6 N N N S Cu S N (SNS)CuN3 ...............................................................................................................B-7 S SO2 SO2 N N Cu S N ((TosN)2-SNS)Cu ......................................................................................................B-8 t Bu t Bu i Pr i Pr i N iPr i Pr Cu i P Pr t t N P Pr Pr P Cu i i P Pr i Pr Bu {( Bu2- PNP)Cu}2 ................................................................................................... B-9 t i i i Bu Bu i Bu Bu Bu P P N Ni Hg Ni N P i Bu i Bu i Bu P i Bu {(PNP)Ni}2Hg .......................................................................................................... B-10 P Ni P N N t Bu t Bu O O O K O O O N [(PNP´)Ni-CH2CH(CH3)2][(crypt)K]....................................................................... B-11 98(2) 0.71073 23.875(6) 21.522(5) 25.877(7) 100(2) 0.71073 14.0498(9) 29.936(2) 13.4935(8) 90 104.211(1) 90 λ (Å) a (Å) b (Å) c (Å) α (º) β (º) γ (º) 90 90 73.818(1) 79.295(1) 82.890(1) 16.210(1) 9.8845(8) 9.7826(8) 0.71073 98(2) 560.22 C26H44CuN P2S2 Page B-6 5501.5(6) 13297(6) 1474.8(2) V (Å ) Crystal Monoclinic Orthorhombic Triclinic System Space Cc (#9) Pbca (#61) P-1 (#2) Group 8 2 Z 8 Dcalcd 1.459 1.350 1.262 (g/cm3) 0.833 1.017 1.005 μ (mm-1) 0.0254 R1 0.0473 0.0842 0.0667 wR2 0.0707 0.1836 (I>2σ(I)) a R1 = Σ ||Fo|-|Fc|| / Σ |Fo|, wR2 = {Σ[w(Fo2-Fc2)2]/Σ[w(Fo2)2]}1/2 3 1128.65 604.15 90 C64H76Cu2N2S4 C36H24CuN6 Page B-5 Chemical Formula Formula Weight T (K) Page B-4 3482.4(13) Monoclinic P2(1)/c (#14) 4 1.424 0.909 0.0513 0.0765 Monoclinic P2(1)/n (#14) 2 1.440 1.265 0.0320 0.0656 90 109.19(2) 90 16.544(3) 20.300(4) 10.979(3) 0.71073 2076.7(2) 90 91.2610(10) 90 9.0191(6) 25.869(2) 8.9030(6) 0.71073 100(2) 746.47 499.09 98(2) C34H40CuN3 O4S4 Page B-8 C20H26CuN4S2 Page B-7 Table B.1: X-ray parameters for assorted copper and nickel amide complexes. a B-3 0.820 0.0514 0.0888 1.222 8 P 2(1)/c (#14) Monoclinic 12529(2) 90 94.282(1) 90 38.483(3) 14.334(1) 22.778(2) 0.71073 100(2) 1152.45 C64H104Cu2 N2P4 Page B-9 3.506 0.0335 0.0665 1.442 4 C 2/c (#15) Monoclinic 5672.7(3) 90 93.390(1) 90 16.7667(5) 16.7095(4) 20.2832(5) 0.71073 100(2) 1231.17 C56H88HgN2 Ni2P4 Page B-10 0.501 0.0470 0.0796 1.177 4 P 2(1)/c (#14) Monoclinic 5886(5) 90 90 102.17(3) 20.600(9) 21.98(1) 13.297(5) 0.71073 100(2) 1043.09 [C36H60NP2Ni] [C18H36KN2O6] Page B-11 B-3 B-4 (BQA)2Cu Bond Lengths and Angles Cu1−N1 2.070(2) Å Cu1−N3 2.053(2) Å Cu1−N2 1.969(2) Å Cu1−N5 2.002(2) Å Cu1−N6 2.482(2) Å Cu1−N4 2.359(2) Å N1−Cu1−N3 N2−Cu1−N5 N1−Cu1−N2 N3−Cu1−N2 N3−Cu1−N5 N1−Cu1−N5 N N N Cu 159.76(8)° 171.12(9)° 80.50(9)° 80.60(8)° 96.58(8)° 103.17(8)° N N N N6 N5 Cu1 N1 N2 N3 N4 B-5 {(AdSNS)Cu}2 Bond Lengths and Angles Cu1−Cu2 2.6184(12) Å N1−Cu1 2.119(5) Å N1−Cu2 2.137(4) Å N2−Cu1 2.132(4) Å N2−Cu2 2.128(5) Å Cu1−S4 2.263(2) Å Cu1−S3 2.267(2) Å Cu2−S1 2.267(2) Å Cu2−S5 2.265(2) Å N1−Cu1−N2 N1−Cu2−N2 Cu1−N1−Cu2 Cu1−N2−Cu2 S3−Cu1−S4 S1−Cu2−S5 104.35(18)° 103.87(18)° 75.92(16)° 75.85(16)° 152.95(7)° 153.54(8)° Ad S S Cu Ad S N Cu S Ad S3 S1 N1 Cu1 N2 S4 N Ad Cu2 S5 B-6 (PMe3)2Cu(SNS) Bond Lengths and Angles Cu1−N1 2.043(1) Å Cu1−S1 2.393(1) Å Cu1−P1 2.275(2) Å Cu1−P2 2.229(1) Å Cu1−S2 3.536(1) Å N1−Cu1−S1 N1−Cu1−P1 N1−Cu1−P2 P2−Cu1−P1 P1−Cu1−S1 S1−Cu1−P2 PMe3 S N 85.28(3)° 111.26(3)° 114.85(4)° 118.39(2)° 95.58(1)° 126.15(1)° S P1 S2 P2 S1 Cu Cu1 N1 PMe3 B-7 (SNS)CuN3 Bond Lengths and Angles Cu−N1 1.911(2) Å Cu−S1 2.342(1) Å Cu−S2 2.370(1) Å Cu−N2 1.930(2) Å N2−N3 1.200(2) Å N3−N4 1.154(2) Å S1−Cu−N1 N1−Cu−S2 S2−Cu−N2 N2−Cu−S1 Cu−N2−N3 N2−N3−N4 S1−Cu−S2 N1−Cu−N2 N N N S Cu S N 86.14(5)° 85.88(5)° 93.28(5)° 96.17(5)° 123.81(14)° 177.01(19)° 162.45(2)° 174.26(7)° S2 N1 N2 Cu N3 N4 S1 B-8 ((TosN)2-SNS)Cu Bond Lengths and Angles Cu−N1 1.898(18) Å Cu−N2 2.017(3) Å Cu−N3 1.903(18) Å N1−S1 1.649(12) Å N3−S2 1.652(10) Å N1−Cu−N2 N2−Cu−N3 N3−Cu−N1 Cu−N1−S1 Cu−N3−S2 102.87(12)° 103.66(12)° 153.47(14)° 120.06(19)° 118.36(17)° S SO2 SO2 N N Cu N S2 N3 N2 Cu N1 S1 S B-9 {(tBu2-iPrPNP)Cu}2 Bond Lengths and Angles Cu1A-Cu2A 2.7027(6) Å Cu1A−N1A 2.143(3) Å Cu1A −N2A 2.171(2) Å Cu2A −N1A 2.186(2) Å Cu2A −N2A 2.169(3) Å Cu1A−P1A 2.237(1) Å Cu1A −P3A 2.215(1) Å Cu2A −P2A 2.215(1) Å Cu2A −P4A 2.221(1) Å N1A−Cu1A−N2A N1A−Cu2A−N2A Cu1A −N1A −Cu2A Cu1A −N2A −Cu2A P1A− Cu1A −P3A P2A− Cu2A −P4A t Bu t i Pr Bu i Pr P i Pr P Cu 103.50(9)° 102.17(9)° 77.26(8)° 77.04(8)° 128.82(4)° 137.75(4)° i N i P i Pr Pr Bu t Bu P2A Cu2A N1A P3A P1A Pr i t P4A Pr Cu N P Pr i Cu1A N2A B-10 {(PNP)Ni}2Hg Bond Lengths and Angles Ni1−P1 2.135(1) Å Ni1−N1 1.969(2) Å Ni2−P2 2.141(1) Å Ni2−N2 1.969(2) Å Ni2−Hg 2.509(0) Å Ni1−Hg 2.514(0) Å P1−Ni1−N1 P1−Ni1−P1 P2−Ni2−N2 P2−Ni2−P2 N1−Ni1−Hg Ni1−Hg−Ni2 Hg−Ni2−N2 i Bu i i i Bu Bu Bu P P N 88.63(1)° 177.25(1)° 88.25(1)° 176.49(1)° 180.00(3)° 180.0° 180.00(3)° Ni Hg Ni N P i Bu i Bu i Bu P i Bu P1 P2 N1 N2 Ni2 P2 Hg Ni1 P1 B-11 [(PNP´)Ni-CH2CH(CH3)2][(crypt)K] Bond Lengths and Angles N1−Ni 1.960(6) P1−Ni 2.142(24) P2−Ni 2.201(15) C33−Ni 1.964(12) N1−Ni−P1 P1−Ni−C33 Ni−P2−C33 P2−Ni−N1 N1−Ni−C33 P1−Ni−P2 86.35(6)° 95.18(7)° 40.27(5)° 87.79(6)° 170.52(9)° 162.54(4)° Ni P N t t Bu N O O O K O O O Bu N P2 N1 P1 P Ni C33 O1 N2 O3 O2 K O4 O6 N3  IV - 1 Chapter IV Redox behavior of copper and zinc dimers supported by bis(2-diisobutylphoshino-4-tbutylphenyl)amido ligands† † Adapted from Harkins, S. B.; Peters, J. C.; Amido-bridged Cu2N2 diamond core complexes in three oxidation states. J. Am. Chem. Soc., to be submitted for publication. IV - 2 IV.A. Introduction Research concerning the magnetic and spectroscopic properties of dicopper complexes is perhaps the most widely studied topic in bimetallic transition metal chemistry. 1 This interest derives from their role in enzymatic processes or for their potential use as photosensitizers, organic light emitting devices (OLEDs), and molecular magnets. 2,3 The ability of dicopper complexes to mediate multi-electron transfer processes and/or small molecule activation is a related concept. In order to realize such chemical transformations using bimetallic copper/ligand systems, complexes which can support two-electron redox processes need to be developed. Further, it is likely that for systems of this nature to be efficient, or even possible, the bimetallic copper complexes would have to resist major structural reorganization during electron-transfer processes. This idea, so termed the “entatic state,” has been found repeatedly in nature as mechanisms of small molecule activation and electron transfer in metalloproteins are probed. 4 An excellent example of this phenomenon is in the metalloenzyme nitrous oxide reductase(N2OR). 5 This dimeric protein catalyzes the two electron, reduction of N2O to N2 and H2O, and is facilitated though a synergistic process involving a tetranuclear Cu4S functional site (CuZ) and a Cu2S2 electron transfer site (CuA). In another example, particulate methane monooxygenase (pMMO) catalyzes the two electron oxidation of methane to methanol using a combination of mono- and dicopper cofactors. 6 Interestingly, the copper-catalyzed chemical transformations observed in N2OR and pMMO have never been demonstrated in small molecule copper chemistry, but both are thought to involve dicopper sites which minimize structural reorganization. IV - 3 Typically, the redox states of dicopper systems have been investigated in the context of understanding enzymatic processes. The dicopper site of N2OR and cytochrome c oxidase are excellent illustrations of this.5b, 7 , 8 Because of interest in understanding the CuA cofactor, numerous mixed-valence dicopper complexes have been synthesized over the last 30 years in an effort to probe the intramolecular electron transfer properties of the two copper nuclei. 9 , 10 , 11 , 12 Hundreds more dicopper(II,II) complexes have been prepared in a variety of studies, and have been found to primarily occur in square-planar or trigonal bipyramidal coordination geometries. While the field of dicopper chemistry has been extensively explored for a variety of oxidation states and gives rise to various types of magnetic interactions, a single dicopper system which can be isolated in each of three oxidation states has not been reported. Over fifteen years ago, Holm and co-workers first demonstrated this concept of low structural reorganization in a small molecule across three oxidation states with a bimetallic iron system. 13 In that system, an Fe2(OR)2 diamond-core complex was found in isostructural environments for the FeIIFeII, FeIIFeIII, and FeIIIFeIII oxidation states where the Fe-centers were constrained in octahedral geometries. Similar systems containing Cu have not been previously described, likely due to the tendency of the copper nuclei to distort and bind additional ligands upon oxidation thus hampering the reversibility of multi-electron redox processes. Luminescent, polymetallic copper complexes are also well known and though they have been studied for many years on a variety of ligand platforms, advances continue to be made. 14,15 They are almost exclusively cationic in nature and exhibit emissive lifetimes of ~100 ns – 10 μs. 16, 17 Distortion of the typical tetrahedral geometry of CuI towards square-planar upon photoexcitation requires a considerable amount of IV - 4 reorganization energy and is problematic when attempting to design highly emissive complexes. 18 As a result of this geometry change in the excited state, quenching due to exciplex formation is also common. Systems where structural reorganization in the excited state has been minimized and where the Cu centers are sterically protected from solvent coordination, not surprisingly exhibit the highest quantum yields. 19 Despite the vast number of complexes that have been prepared in order to study both rapid electron transfer and luminescence phenomena, a single system which exhibits both these characteristics through low-structural reorganization has yet to be reported. With this concepts in mind, a novel family of Cu2N2 complexes have been developed. Initially work was focused on the synthesis and characterization of a dinuclear, thioether-supported Cu2N2 complex {(SNS)CuI}2 ([SNS]− = bis(2-tert- butylsulfanylphenyl)amide). 20 This compound is characterized by a Cu2N2 diamond core configuration in which two d10 Cu centers are bridged by two diaryl-amido linkages. Oxidation of {(SNS)CuI}2 affords the class III, 21 fully-delocalized dicopper cation, [{(SNS)Cu1.5}2]+. A key feature of this intial system was the observation of very minimal structural reorganiztion between oxidation states. However, upon oxidation of {(SNS)CuI}2 to [{(SNS)Cu1.5}2]+, a ~0.13 Å contraction in the Cu···Cu distance from 2.599 Å to 2.472 Å was observed. 22 It is important to note that these minor perturbations to the core atoms occur in the absence of additional ligand coordination or significant topological reorganization, thus functionally modeling the electron transfer process of the CuA site., As expected from the small structural reorganization upon oxidation from {(SNS)CuI}2 to [{(SNS)Cu1.5}2]+, the electron self-exchange process between the two redox forms is very fast and a rate constant of ks > 107 M-1s-1 was estimated. A higher IV - 5 potential irreversible wave which was speculated to be an unstable dicopper(II,II) species was also observed. Building on these studies, our current efforts have been devoted to stabilizing the second oxidation process of a Cu2N2 diamond core complex. The problem was addressed by replacing the tbutyl thioether chelates with more reducing diisobutylphosphine groups. This change was found to be both sterically and electronically beneficial, affording the Cu2N2 complex {(PNP)Cu}2 (4.3) which does exhibit two reversible one-electron redox transitions by cyclic voltammetry. In the current manuscript, this chemistry has been elaborated by the preparation of a family of bimetallic copper and zinc complexes supported by [PNP]¯ type ligands (PNP = bis(2-(diisobutylphoshino)-phenyl)amide, 4.1; t Bu2-PNP = bis(2-diisobutylphoshino-4-tbutylphenyl)amide, 4.4) (Figure IV.1). These molecules are characterized by amido-bridged Cu2N2 diamond core geometries, where the chelating phosphines of one ligand coordinate to two different copper nuclei. The fully-reduced dicopper complexes, 4.3 and 4.5 are exceptionally luminescent at room temperature. One electron chemical oxidation of either 4.3 or 4.5 results in the fully delocalized class III mixed valence Cu1.52N2 diamond core complexes, 4.8 and 4.10, respectively. Though the two electron chemical oxidation of 4.3 results in decomposition, further oxidation of 4.5 results in the paramagnetic, dicationic dimer 4.11. The solid-state molecular structures of the bimetallic complexes have been determined and are presented along with spectroscopic and magnetic studies. Interestingly, structural, spectroscopic and magnetic studies of 4.11 also provides evidence for a resonance stabilized paramagnetic CuIII center in a pseudo-tetrahedral geometry. IV - 6 R i i R Bu P N N Bu i P P Bu i i Bu M1 i [X]n Bu M2 i P Bu 4.2: M 1= Li 1 M 2= Li 4.3: M = Cu M = Cu 4.5: M 1= Cu I M 2= Cu I R= tBu 4.6a: M 1= Zn II M 2= Zn II R= tBu X= PF 6 n= 2 4.6b: M 1= Zn II M 2= Zn II R= tBu X= OTf n= 2 M = Cu n= 1 Bu 4.8: M 1= Cu 1.5 M 2= Cu 1.5 R= H X= BArF n= 1 i 4.10a: M 1= Cu 1.5 M 2= Cu 1.5 R= tBu X= BArF n= 1 R 2 M = Zn II R= H 4.7: R I I t Bu 1 2 R= H I R= Bu X= PF 6 4.10b: M 1= Cu 1.5 M 2= Cu 1.5 R= tBu X= SbF 6 n= 1 4.11: M 1= Cu I M 2= Cu III R= tBu X= SbF 6 n= 2 Figure IV.1: Representative dimer of M2N2 complexes (BArF = [B(C6H3(CF3)2)4]). To our knowledge, the presented system is the first example of a dicopper system which can stabilize two successive one electron oxidations without deviating from a pseudo-tetrahedral geometry around the Cu centers. The collective characterization data presented herein all support the assignment of a highly covalent Cu2N2 structure, which through electronic delocalization is able to minimize structural reorganization across different electronic environments, thus mimicking the behavior of dicopper active sites in enzymes. IV.B. Results and Discussion IV.B.1. Diamagnetic complexes 4.1-4.7 Preparation of bimetallic, d10 complexes supported by PNP-type ligands. The ligand (PNP)H, 4.1, was prepared in good yield by metathesis of 2,2'difluorophenylamine with two equivalents of (iBu)2PLi in THF. Following quenching of the reaction with methanol and work-up, 4.1 was afforded as a pale green oil. In order to develop a useful metathesis reagent for metallation, 4.1 was deprotonated with nBuLi in IV - 7 petroleum ether which afforded the lithium dimer, {(PNP)Li}2 (4.2) as a light yellow powder. Dimeric 4.2 is analogous to the previously reported {(BQA)Li}2, 23 and does not form (PNP)Li(solvent)2-type complexes 24 when prepared in hydrocarbon solvents. Metathesis of 4.2 with (Me2S)CuBr in benzene immediately forms the dicopper complex 4.3 in nearly quantitative yield with concomitant loss of LiBr. Complex 4.3 is bright yellow and is characterized by a single resonance in its 31P NMR spectrum (-33.9 ppm) and exceptional luminescence in fluid solution at room temperature as we have previously communicated. i Bu i P i Bu Bu i P Bu H N BuLi 4.2 Pet. Ether CuBr•SMe2 4.3 Ether 4.1 i Bu i P i Bu Bu i P Bu H N t Mesityl Cu Pet. Ether 4.4 Bu t Bu 1) ZnMeCl, THF 2) TlPF 6, Δ 1) KH, THF 2) Zn(OTf) 2, Δ 4.5 + 4.6a 4.5 THF, Δ 4.6a 4.6b 4.7 Figure IV.2: Synthesis of complexes 4.2, 4.3, 4.5, 4.6, and 4.7. The current work describes the synthesis of a family of dimeric complexes where the ligand has been derivatized with a tbutyl group in the 4-position of the aryl ring. This IV - 8 substitution was accomplished following methodology recently published by Ozerov and co-workers for the synthesis of PNP-type ligands. 25 In our preparation, di(2-bromo-4t butylphenyl)amine was treated with three equivalents of nbutyl lithium in diethyl ether which lithiates the 2-positions of the aryl rings and deprotonates the amine. Subsequent addition of three equivalents of diisobutylchlorophosphine to the reaction mixture affords the tris(phosphine) substituted product after 36 h. The P-N bond can be readily cleaved with anhydrous HCl in ether which affords the desired ligand 4.4 as a white solid following work-up. Experimentally it has been determined that it is necessary to use ≥ 3 equivalents of diisobutylchlorophosphine to achieve the desired product. The first two equivalents metathesize with the amide and one of the aryl ring positions and do not undergo further rearrangement. This is in contrast to the synthesis reported by Ozerov, where diisopropylchlorophosphine reportedly undergoes the same substitution, but then migrates from the amide to the other aryl ring positions. This process does not occur in our system even with prolonged reaction times and elevated temperatures. We attribute this phenomenon to the decreased steric interaction between the diisobutylphosphine groups of the amide and the aryl ring compared to those interactions between diisopropylphosphine groups. In order to avoid the synthesis of lithiated 4.4 because of difficulties with isolation, a different route to metallation was pursued. Metathesis of amine ligands with mesityl copper is a well established route to achieve amido-copper complexes. 26 Effective metallation of 4.4 was achieved with mesityl copper in petroleum ether over a period of 12 h which afforded the copper dimer, {(tBu2-PNP)Cu}2 (4.5) in high yield upon workup. Complex 4.5 is characterized as a luminescent yellow solid that is extremely soluble IV - 9 in solvents such as petroleum ether and hexamethyldisiloxane but negligibly soluble in acetonitrile. A single resonance in the 31P NMR spectrum of 4.5 is observable at -35.3 ppm and is shifted upfield of 4.3, consistent with the more electron-rich phosphorus nuclei. Despite the presence of eight isobutyl and four tbutyl groups per dimeric molecule, the complex is highly crystalline allowing the growth of X-ray quality crystals by slow evaporation of a concentrated solution of 4.5 in THF. To enhance our understanding of the spectroscopic and electrochemical characteristics of 4.3 and 4.5, the Zn2N2 dimer was synthesized. This was accomplished by addition of a solution of MeZnCl in THF to a solution of 4.4, which after heating for 4 h was treated with TlPF6. Heating was continued for 16 h and after work-up [{(tBu2PNP)Zn}2][PF6]2 (4.6a) was isolated in modest yield as a white solid. Complex 4.6a is poorly soluble in THF once isolated, but can be crystallized from methylene chloride/petroleum ether. Because X-ray quality crystals of 4.6a could not be attained, the analogous triflate ([OTf]¯) salt was synthesized. Reaction of 4.4 with KH in THF followed by addition of Zn(OTf)2 and refluxing for 24 h afforded 4.6b which was found to be spectroscopically identical to 4.6a. Crystals suitable for X-ray diffraction were obtained by cooling a concentrated solution of 4.6b in methylene chloride/petroleum ether to -35 °C. The mixed-metal CuZnN2 dimer was also prepared. Conveniently, combination of 4.6a and 4.5 in THF quantitatively results in [{(tBu2-PNP)Cu}{(tBu2-PNP)Zn}][PF6], 4.7, after refluxing for 18 h in THF as determined by 31P NMR. The product was isolated as X-ray quality pale yellow crystals from THF/petroleum ether. The mixed metal formulation of crystalline 4.7 was assigned based on two resonances in the 31P NMR IV - 10 which are shifted slightly from the starting materials and are approximately equal in integration. The resonance in the 31P NMR spectrum at -31.8 ppm is assigned to the Cu coordinated phosphine atoms, which is broadened relative to the resonance at -40.0 ppm assigned to the Zn coordinated phosphines as a result of the large quadropolar moment of copper. 27 X-ray diffraction studies. To examine the structural impact of metal substitution on the dimer conformation, the solid-state molecular structures of the d10 complexes 4.2, 4.5, 4.6 and 4.7 have been solved and are detailed in Table IV.1. The X-ray diffraction study of 4.3 has been previously reported and the relevant bond lengths and angle measurements are also included in Table IV.1. Characteristic to all five complexes is the four-component, M2N2 diamond core structural arrangement formed by two metal atoms and two bridging amide nitrogen atoms. The metal atoms are additionally supported by two phosphine ligands which gives rise to the compressed tetrahedral geometry about each copper center and to the overall D2 molecular symmetry. In these closed-shell species, the intermetallic distances range from 2.4680(27)Å to 2.8826(3)Å and support N-M-N angles from 107.23(9)° to 92.93(6)° which is illustrative of the plasticity which can be supported while maintaining the dimeric conformation. The structure of 4.2 was determined and from this the similarity to {(BQA)Li}2 was confirmed. A short Li(1)···Li(2) distance of 2.4680(27) Å is found, which results from the short M-N bond distances of average 2.07 ± 0.01 Å as a result of the high ionic character of the bonds. This is the shortest intermetallic distance in the presented series, though it is somewhat longer than the Li(1)···Li(2) distance reported for {(BQA)Li}2 M1 N2 N1 M2 P3 P2 4.2 a 2.4680(27) 2.074(2) 2.062(2) 2.066(2) 2.087(2) 2.5126(19) 2.5309(19) 2.5437(19) 2.5333(19) 107.23(9) 106.57(9) 73.18(8) 73.01(8) 139.50(8) 140.47(8) 4.3 b 2.6245(8) 2.127(4) 2.191(4) 2.179(4) 2.219(4) 2.2173(13) 2.2339(13) 2.2241(13) 2.2235(13) 107.20(14) 104.44(14) 75.10(12) 73.03(12) 132.93(5) 137.69(5) 4.5 b 2.7283(7) 2.164(3) 2.139(3) 2.135(3) 2.173(3) 2.2326(13) 2.2439(13) 2.2236(13) 2.2139(13) 101.35(11) 101.19(11) 78.77(10) 78.52(10) 136.36(4) 139.27(5) Notes a: M(1), M(2) = Li; b: M(1), M(2) = Cu; c: M(1), M(2) = Zn; d: M(1) = Cu, M(2) = Zn M(1)-M(2) M(1)-N(1) M(1)-N(2) M(2)-N(1) M(2)-N(2) M(1)-P(1) M(1)-P(4) M(2)-P(2) M(2)-P(3) N(1)-M(1)-N(2) N(1)-M(2)-N(2) M(1)-N(1)-M(2) M(1)-N(2)-M(2) P(1)-M(1)-P(4) P(2)-M(2)-P(3) P4 P1 4.6b c 2.8826(3) 2.0931(16) 2.0933(16) 2.0805(16) 2.1071(17) 2.3250(5) 2.3350(6) 2.3515(5) 2.3457(5) 92.97(6) 92.93(6) 87.37(6) 86.67(6) 139.31(2) 141.92(2) 4.7 d 2.7518(5) 2.114(2) 2.159(2) 2.112(2) 2.121(2) 2.2851(7) 2.2744(7) 2.3033(7) 2.3080(8) 98.58(9) 99.84(9) 81.25(8) 80.02(7) 139.77(3) 141.08(3) Table IV.1: Core structural representations (50% probability ellipsoids), selected bond lengths (Å) and angles (deg.) for 4.2, 4.3, 4.5, 4.6b, and 4.7 IV - 11 IV-11 IV - 12 (2.395(5)). This robust structural configuration also gives rise to the thermal stability of 4.2 which is not typically expected for lithium amides. The structure of 4.3 is very analogous to 4.2, though a longer intermetallic distance of 2.6245(8) Å and longer M-N bond lengths, 2.18 ± 0.04 Å (av.), are observed. Shortened M-P bonds of 2.23 ± 0.01 Å reflect the increased covalent character of copper compared to that of lithium. Interestingly, four-coordinate LiI (0.59 Å) and CuI (0.60 Å) are reported to have the same ionic radii 28 and would not inherently suggest an increased bonding radius when they have the same charge. We attribute the overall core expansion to the repulsion between anti-bonding Cu-Cu d-orbitals interaction which exists in 4.3, but is absent between the lithium atoms of 4.2. While complexes 4.2 and 4.3 were the subject of our initial inquiry, the major aspect of the current work will focus on those metal complexes supported by the tbutyl substituted ligand 4.4. The solid-state molecular structures of 4.5 is very similar to that of 4.3 in that the Cu-N/Cu-P differ only slightly: 2.15 ± 0.02 Å/2.23 ± 0.01 Å for 4.5 and 2.18 ± 0.04 Å/2.23 ± 0.01 Å for 4.3 on average. The key differences lie in the N-M-N angles which are compressed by 4.6° in 4.5 relative to 4.3; this results in the 0.1 Å expansion of the Cu(1)···Cu(2) distance. This expansion is likely due to the increased electron density at the Cu2N2 core which creates more repulsion between the Cu d- and N p-orbitals. The increase in electron density is expected with the addition of electron donating tbutyl groups to the aryl rings. These structural changes are considered to be purely an electronic consequence, as no steric interactions associated with tbutyl group are evident from the crystal structure. IV - 13 The dizinc complex, 4.6b, adopts the same diamond core configuration as the Li and Cu dimers described above. This isostructural relationship was expected because fourcoordinate ZnII is also reported to have the same ionic radius as CuI (0.60 Å) and is isoelectronic with 4.5. Dimer 4.6b exhibits the longest M···M distance (2.8826(3) Å) of the M2N2 systems studied thus far. Shortening of the M-N bonds by ~0.06 Å and elongation of the M-P bonds by ~0.11 Å in 4.6b compared to 4.5 is also observed. This substantial core expansion is reasonably attributed to the inherent cationic repulsion between the Zn2+ ions. It is also likely that this repulsion allows for the formation 4.7 by redistribution between 4.6b with 4.5. In other respects the structure of 4.6 is generally unremarkable and is effectively isostructural with 4.5. X-ray diffraction study of the CuZnN2 dimer 4.7 indicates similarities to both of its homo-dimer parents 4.5 and 4.6b. The presence of a single PF6 anion in the asymmetric unit is in agreement with the monocationic, mixed metal formulation of 4.7. The internuclear Cu···Zn distance of 2.7518(5) Å is intermediate in length between the 2.7283(7) Å intermetallic distance of 4.5 and the 2.8826(3) Å distance observed in 4.6b. Crystallographically, the copper and zinc atoms are indistinguishable from each other given their nuclear similarity. For the purpose of structural refinement the Cu and Zn atoms have been assigned as indicated in Table IV.1, which was based on bond length similarities between the structure of 4.7 and those of 4.5 and 6. No comparable internal differences in the M-N bond lengths could be correlated as being N-Cu or N-Zn in parentage, consistent with the notion of a large degree of electronic delocalization across the M2N2 diamond core. However, differences in the M-P bond lengths did prove to be diagnostic. The M(1)-P bond lengths were found to be ~1% shorter than the M(2)-P bond IV - 14 lengths on average suggesting that M(1) is primarily Cu in the crystal lattice, and M(2) is primarily Zn. However, given the structural homogeneity between the two metal sites, it is impossible to know the precise percentage of each individual metal at either site. The crystal structure does, however, confirm the presumed isostructual identity of 4.7 to the other systems studied. Optical Spectroscopy. The absorption spectra of 4.2 in benzene and 4.3 in THF solution are shown in Figure IV.3 (top). Complex 4.2 exhibits two prominent absorption bands with λmax= 303 nm (ε 18 200 M-1cm-1) and 359 nm (ε 17 200 M-1cm-1), and shoulders at 398 nm and 414 nm which gives rise to the pale yellow color. Complex 4.3 has an intense charge-transfer band with λmax= 350 nm (ε 42 000 M-1cm-1), which is not as prominent in the spectrum of 4.2. Peaks with λmax= 423 nm (ε 4400 M-1cm-1) and a shoulder at ~446 nm have been assigned as the MLCT transitions which are typical for CuI complexes and result in the intense yellow color. The absorption spectra for 4.5, 4.6a, and 4.7 were also collected and plotted in Figure IV.3 (bottom). The Zn2N2 dimer, 4.6, absorbs strongly at 302 nm (ε 30 500 M-1 cm-1) with shoulders at 327 nm and 344 nm. As expected, 4.6 appears essentially unchanged from the ligand (4.4) absorption spectra which exhibits a maxima at 302 nm (ε 18 700 M1 cm-1) and a shoulder at 334 nm. Conversely, the Cu2N2 dimer, 4.5, has a much richer spectroscopic signature characterized by ligand absorption bands at 295 nm (ε 20 500 M-1 cm-1) and 318 nm (ε 21 100 M-1 cm-1) and a more intense band at 352 nm (ε 33 300 M-1 cm-1) with a shoulder at 383 nm. The spectrum of 4.5 also has two characteristic MLCT bands at 433 (ε 4600 M-1 cm-1) and 459 (ε 3800 M-1 cm-1), which appear at slightly lower energy of the corresponding bands in 4.3 as result of increased electron density at the metal center. IV - 15 The intense band at 302 nm is assigned to the metal-metal 3d→4p transition based on the assignments made by Che and Miskowski on related luminescent CuI-CuI dimers. 29 The related band is also present in the spectrum of 4.3 at 350nm which is notably shifted to lower energy of 4.5 by ~4500 cm-1. This assignment is supported by the fact that this transition is not observed in 4.7 which has absorption maxima associated with ligand-based transitions at 290 nm (ε 21 600 M-1 cm-1), 306 nm (ε 23 400 M-1 cm-1), and 350 nm (sh). A broad MLCT transition centered at 396 nm (ε 7400 M-1 cm-1) is also evident though it shifted to higher energy than the MLCT bands observed in the spectrum of 4.5. 4 4.3 3 2 4.2 e / 104 M-1cm-1 1 4.5 4.6 3 2 1 4.7 280 320 360 400 440 480 520 wavelength / nm Figure IV.3: Absorption spectra. Top: 4.2 (blue) in benzene and 4.3 (black) in THF. Bottom: 4.5 (black), 4.6 (red) and 4.7 (blue) in THF. IV - 16 Electrochemistry. The initial motivation for this study was spurred by the observation of a pseudo-reversible CuIICuII/Cu1.5Cu1.5 wave in an {(SNS)Cu}2 copper dimer at Epa= +0.57 V (0.35 M [TBA][PF6] in THF). It was hypothesized that by moving from thioether chelates to stronger donating phosphine ligands that this higher potential redox couple would become fully reversible. The electrochemistry of complexes 4.3, 4.5, 4.6, and 4.7 was studied by cyclic voltammetry and the traces are plotted in Figure IV.4. 30 The dicopper dimers, 4.3 and 4.5, display two cleanly reversible one electron redox processes. The low potential processes at −0.405 V and −0.490 V, for 4.3 and 4.5 respectively, is assigned to the Cu1.5Cu1.5/CuICuI transition based on previous studies. As a result of switching from thioether to more strongly reducing phosphine chelates, the potential has been substantially shifted to lower potential from the {(SNS)Cu}2 Cu1.5Cu1.5/CuICuI redox couple (E½ = −0.250 mV). Higher potential waves were observed at +0.295 V and +0.210 V for 4.3 and 4.5 respectively. 4.3 4.5 4.7 0.75 0.50 0.25 0.00 -0.25 -0.50 + Potential (V vs. Fc /Fc) -0.75 -1.00 Figure IV.4: Cyclic Voltammetry of 4.3 (black), 4.5 (blue), and 4.7 (red) in 0.35 M [TBA][PF6] in THF IV - 17 As a result of installing an electron-donating, tbutyl group to the aryl ring, both redox processes have been shifted to lower potential by 85 mV in 4.5 relative to 4.3. This is consistent with the copper centers being more electron rich and therefore harder to reduce. Interestingly, both redox processes are shifted together, suggesting that the redox processes are directly linked to redox processes involving one or both of the copper nuclei rather than simply oxidation of the ligand. Given that both redox processes are now observable on the electrochemical time-scale the electronic structure of the double oxidized species is clearly of interest. With complexes 4.6a and 4.7 in hand their cyclic voltammetry was also probed. The dizinc dimer, 4.4, showed no electrochemical events in the window (CH2Cl2) from +0.80 to -2.00 V vs. Fc+/Fc, which is typical for Zn complexes. This observation suggests strongly against the possibility of a pure ligand radical corresponding to the higher potential wave of either 4.3 or 4.5. The mixed-metal dimer, 4.7, shows no redox processes at potentials similar to those of the Cu1.5Cu1.5/CuICuI transitions, though a single reversible wave at +0.320 V is apparent. This electrochemical event is shifted 110 mV positive of the high potential wave of 4.5. As both 4.5 and 4.7 are isostructural, bimetallic d10 systems, it is reasonable to consider that they are resulting from the same electronic transition and the shifting of 4.7 to higher potential is associated with the cationic charge which is not present in 4.5. Emission Spectroscopy. The emission properties of complexes 4.2─4.5 have been examined in fluid solution. Complexes 4.3 and 4.5 have similar emission spectra centered at 500 nm and 511 nm, respectively, with negligible changes in the full-width at halfmax. Given the relatively small change in the Hammett parameter (σ), by switching IV - 18 from H- to Bu- on the aryl ring, the red-shifted maximum is considerable 31 thus t presenting the possibility for tuning the emissive properties. The lifetime (τ) and quantum yield (φ) of 4.5 were determined (τ = 11.4 μs and φ = 0.58) comparable to 4.3 (τ =10.9 μs, φ = 0.67). We also attempted to study the photophysics of complexes 4.6a and 4.7. The dizinc complex, 4.6a, is negligibly emissive and difficult to distinguish from trace impurities. However, complex 4.7 did have a weak, but resolvable, emission profile in dichloromethane centered at 455 nm (φ ~ 7 x 10-4) with a lifetime of τ ≤ 500 ps which is reminiscent of other mononuclear copper luminophores., From this data, it is evident that the dicopper interaction is essential in order to achieve long-lived excited states on this particular molecular scaffold. A reasonable explanation is that the dicopper complexes 4.3 and 4.5 are better able to stabilize the excited state by delocalizing the unpaired electron across the Cu2N2 core. This is based on the suggestions that the excited state molecules are similar to the one-electron oxidized species, 4.8 and 4.10. Whether this excited state delocalization necessitates a direct Cu-Cu bond, or a cuprophilic interaction, is yet to be established. We do suggest however that the intense emission observed for 4.3 and 4.5, but not for 4.6 and 4.7, results from excitation into the 3d→4p or 3d→4s transition owing to the uniqueness of the Cu2N2 system. Interestingly, the lithium dimer, 4.2, has a uncommonly large quantum yield (φ = 0.45) with a emission decay that is 1000 times faster (τ ~ 4 ns) than that observed for 4.3. Also, the emission spectrum is broader and the nonradiative decay rate is ~1 x 104 times larger than 4.3 and 4.5 which suggests that different electronic states are involved. A IV - 19 possible explanation for the high quantum yield is that lithium atoms constrict the organic framework in a rigid environment similar to other organic fluorophores. 4.2 400 450 4.3 4.5 500 550 wavelength / nm 600 650 700 Figure IV.5: Emission spectrum of complexes 4.2, 4.3, and 4.5 (intensity is not reflective of quantum yield). IV.B.2. Paramagnetic complexes 4.8-4.11 Oxidative Synthesis of 4.8-4.11. Based on the electrochemical studies of the d10 complexes and our prior success in isolating the one-electron oxidation products of Cu2N2 complexes supported by the [SNS]− ligand, the chemical oxidations of 4.3 and 4.5 were pursued. Addition of one equivalent of [Cp2Fe][B(C6H3(CF3)2)4] to 4.3 in dichloromethane immediately induces a color change from brilliant yellow to deep red and affords the one electron oxidized product, 4.8, as red-brown prisms following workup and recrystallization from diethyl ether/petroleum ether. The paramagnetic complex IV - 20 4.8 was found to contain a 1:1 ratio of dimer to counterion by elemental analysis consistent with the mixed-valence assignment. The two electron oxidation of 4.3 was also attempted. Reaction of 4.3 with two equivalents [NO][BF4] in either dichloromethane or chlorobenzene resulted in the darkening of the yellow solution to red, consistent the intermediate formation of the mono-oxidized complex 4.8. 32 Within minutes, the solution became blue in color and in approximately 30 min had converted to an inky violet-colored homogeneous solution. Isolation and characterization of the product, 4.9, proved arduous. The violet solid, which results from removal of the solvent in vacuo, is only compatible with halogenated solvents as it is insoluble in non-polar solvents such as ether and benzene, and is decomposed by any donor solvents (THF 33 , CH3CN, or trace H2O). Product isolation was ultimately achieved by recrystallization of 4.9 from chlorobenzene/petroleum ether which afforded green crystals that are purple as an amorphous powder. Microanalysis indicated 1:1:1 relationship of [4.1]−:Cu:BF4 which was consistent with the desired doubly oxidized dimeric product of 4.3. The 1H and 31P NMR spectra of crystalline 4.9 were indicative of a diamagnetic product which was completely dissimilar from the spectrum of 4.3. The 19F NMR spectrum of 4.9 contained a broad bimodal signal from the BF4 counterion and is indicative of fluxional interaction with the Cu nuclei and is also evident from the multiple broad signals observed in the 1H and 31P NMR spectra. The solid-state molecular structure of 4.9 was determined by X-ray diffraction. The crystallized product, which is the majority (> 90%) based on absorption spectroscopy, is actually a ligand-fused decomposition product which results presumably from intermolecular radical coupling (Figure IV.6). In the asymmetric unit (triclinic, P−1) IV - 21 there are two unconnected copper containing half-molecules which are related by inversion through the newly formed C=C bond between the aryl rings. Two BF4 counterions, and two chlorobenzene solvent molecules were also found in the asymmetric unit, neither of which are associated with the copper atoms in the solid state (closest contacts: Cl⋅⋅⋅Cu 9.269(9) Å; F⋅⋅⋅Cu 11.896(3) Å). The bond order assigned in the schematic of 4.9 (Figure IV.6) is based the crystallographically determined bond lengths and on an assignment of a CuI oxidation state based on its coordination geometry and molecular diamagnetism. This high degree of π-conjugation gives rise to intense absorption band at 613 nm (ε 16 250 M-1cm-1) and results in the inky violet color. Molecules that display this type of conjugation have been previously reported by Hammond and are spectroscopically similar to 4.9. 34 From these studies, 4.9 can be described as a CuI cation coordinated by a neutral η3-ligand in a t-shaped geometry which explains its facile decomposition in coordinating solvents. The presence of the chlorobenzene solvent molecules in the crystals lattice also explains why crystallization was only accomplished in this solvent. We theorize that 4.9 results from an intermolecular coupling between two [{(PNP)Cu}2]2+ with reductive loss of H2, which results from radical character at the 4-position of the aryl rings. Despite evidence for the isostructural two electron oxidation product of 4.3 as an intermediate in the synthesis of 4.9 based on the color transitions and the reversibility of the electrochemical redox processes, further attempts to isolate the dimeric intermediate were not pursued. IV - 22 Figure IV.6: Top, Decomposition of 4.3 upon two electron oxidation. Bottom, Solid-state molecular structure (50% probability ellipsoids) of 4.9 (isobutyl groups, solvent molecules, and anions omitted for clarity). Selected bond lengths (Å) and angles (deg.): Cu(1)-N(1), 2.108(6); Cu(1)-P(1), 2.212(2); Cu(1)-P(2), 2.212(2); N(1)-Cu(1)-P(1), 86.29(18); N(1)-Cu(1)-P(2), 86.07(18); P(1)-Cu(1)-P(2), 167.55(8). The observation of 4.9 was the initial impetus for the synthesis of ligand 4.4, which was designed to prevent the observed radical coupling mechanism. It has been welldocumented that installing tbutyl groups on aryl rings of semiquinonates can help stabilize resonance radicals at those positions. 35 Reaction of 4.5 with [Cp2Fe][B(C6H3(CF3)2)4] in ether or AgSbF6 in THF rapidly afforded the one electron oxidized [{(tBu2-PNP)Cu1.5}2]+ complexes 4.10a and 4.10b, respectively. Both complexes were obtain as red-brown crystalline solids in good yield upon recrystallization. Cation 4.10a formed much larger crystals than 4.10b presumably IV - 23 because of a more matched cation/anion packing. Microanalysis of both 4.10a and 4.10b found a 1:1 cation:anion ratio which is consistent with mixed valence assignment. Addition of two equivalents of [NO][SbF6] to 4.5 dissolved in dichloromethane immediately resulted in the characteristic red color of 4.10b, which progressed to deep blue over the course of 10 min. It is important to note that this is the identical reaction profile to what was observed for the early stages of two electron oxidation of 4.3. In this case the inky violet color of the decomposition complex 4.9 was not observed even after stirring for several hours. Crude isolation of the two-electron oxidation product, 4.11, as a blue solid was accomplished by removal of the solvent in vacuo. Multiple recrystallizations of 4.11 from dichloromethane/petroleum ether at -35 ºC afforded analytically pure material in modest yield. Microanalysis indicated a precise 1:2 ratio of dimer to SbF6, thus confirming the two fold oxidation of 4.5 and the inherent dicationic nature of 4.11. Similar to 4.9, 4.11 is largely incompatible with coordinating solvents and is insoluble in hydrocarbons. Complex 4.11 is paramagnetic and characterization by NMR is not possible due to the large quadaropolar moment of the copper nuclei. The paramagnetic behavior of 4.11 was initially surprising because of the presumed proximity of the copper nuclei in the dimer. With the short Cu⋅⋅⋅Cu distances afforded by this ligand scaffold and the ability of the Cu2N2 core to delocalize electron density, antiferromagnetic coupling between unpaired spins and thus room temperature diamagnetism was expected, and not observed. Crystallographic Studies. The solid-state molecular structures of all oxidized species were determined by X-ray diffraction and the core representations with relevant bond lengths and angles are shown in Table IV.2. The mono-oxidized species 4.8 and IV - 24 4.10a both exhibit pronounced contractions in the Cu(1)⋅⋅⋅Cu(2) distance of 0.08 Å and 0.20 Å respectively, compared to that observed in the CuICuI complexes 4.3 and 4.5 (Table IV.1). Comparable Cu-P bond lengths of 2.22-2.27 Å exist in both structures of 4.8 and 4.10a and contain P-Cu-P bond angles of 133.4-140.8º. In comparing 4.8 and 4.10a to 4.3 and 4.5, it was found that the Cu-P bonds contract by 0.12 Å on average while P-Cu-P angles are essentially unchanged. Table IV.2: Core structural representations (50% probability ellipsoids), selected bond lengths (Å) and angles (deg.) for 4.8, 4.10a, and 4.11 P1 Cu1 P4 P2 N1 N2 Cu2 P3 M(1)-M(2) M(1)-N(1) M(1)-N(2) M(1)-P(1) M(1)-P(4) M(2)-N(1) M(2)-N(2) M(2)-P(2) M(2)-P(3) N(1)-M(1)-N(2) N(1)-M(2)-N(2) M(1)-N(1)-M(2) M(1)-N(2)-M(2) P(1)-M(1)-P(4) P(2)-M(2)-P(3) 4.8 2.5406(6) 2.0567(19) 2.0596(16) 2.2657(6) 2.2470(6) 2.3627(19) 2.0931(17) 2.2531(6) 2.2219(6) 113.37(7) 100.89(6) 69.89(6) 75.55(6) 133.36(2) 140.77(2) 4.10a 2.5264(6) 2.148(3) 2.206(3) 2.2574(9) 2.2594(10) 2.061(3) 2.059(3) 2.2624(9) 2.2580(9) 102.03(10) 110.50(10) 73.75(8) 72.56(8) 136.13(4) 133.91(4) 4.11 2.5245(9) 2.332(3) 2.452(3) 2.2278(8) 2.2244(8) 2.002(3) 1.976(2) 2.2766(9) 2.2836(8) 95.19(8) 125.31(10) 70.83(8) 68.60(8) 149.89(3) 127.94(3) More remarkable structural contortions are observable in the Cu2N2 diamond core construction. Unlike the highly symmetric Cu2N2 diamond core observed for [{(SNS)Cu1.5}]+, larger deviations are observed in the systems presented here. Cation 4.8 contains three Cu-N bonds of ~2.07 Å and one of 2.3627(19) Å which facilitate the IV - 25 contracted Cu(1)⋅⋅⋅Cu(2) distance of 2.5406(6) Å. Conversely, the solid-state structure of 4.10a does not exhibit the same elongated Cu-N bond, but it still achieves a highly contracted Cu(1)⋅⋅⋅Cu(2) distance of 2.5264(6) Å. In this case, Cu(1) has Cu-N bond distances of 2.18 Å (ave.) and Cu(2) has Cu-N bond distances of 2.06 Å (av.) suggesting the possibility for electronic inequivalence between the copper centers. Despite the structural aberrations which are observable between 4.8 and 4.10a, the overall trend of the Cu2N2 core contracting upon oxidation remains consistent, though the contraction from 4.5 to 4.10a is more than twice that of 4.3 to 4.8. Finally, the deep blue crystalline complex, 4.11, was studied by X-ray diffraction. Immediately upon inspection of the solid-state structure significant structural deviation from the mixed-valence complex 4.10a is apparent. The first feature of interest is that the Cu(1)⋅⋅⋅Cu(2) distance of 2.5245(9) Å in 4.11 is unchanged from that of 4.10a. While this bond length and overall ligand super-structure is unperturbed, significant inequivalence in the bonding environment at the Cu sites has resulted. At Cu(1), bonds to N(1) and N(2) of 2.332(3) Å and 2.452(3) Å, respectively, are observed and based on length suggest dative interactions between Cu(1) with both N atoms. 36 Cu(2), on the other hand, has comparatively short bonds to N(1) and N(2) of 2.002(3) Å and 1.976(2) Å indicative of metal-amide bonds and may be indicative of electronic inequivalence between the two Cu nuclei. The comparable Cu(1)-N(1)-Cu(2) and Cu(1)-N(1)-Cu(2) bond angles which result are illustrative of the conserved molecular symmetry despite changes in bonding at the individual copper nuclei. Changes to the electronic structure at the individual Cu nuclei is also evidenced by lengthening of the Cu-P bonds at Cu(2) to ~2.28 Å and shortening of the Cu-P bonds at Cu(1) to ~2.23 Å. 37 IV - 26 Magnetic Studies The X-band EPR spectrum of the mixed-valence complexes 4.8 and 4.10a were obtained in 2-methyltetrahydrofuran at 3.7 K and are shown with their corresponding simulations in Figure IV.7. The data obtained are consistent with a class III delocalized mixed-valence species. The observed EPR signal of 4.8, clearly displays a seven-line hyperfine coupling due to the interaction of a single unpaired electron delocalized over two Cu (S=3/2) nuclei. The hyperfine coupling is most clearly resolved in the gz tensor, and spectrum can be adequately simulated with the following parameters: gx = 1.996, gy = 2.020, gz = 2.095, Ax = 9.5⋅10-4 cm-1, Ay = 26⋅10-4 cm-1, Az = 28⋅10-4 cm-1. As with previous amido-dicopper system we have studied, the gz tensor is located at a lower g-value than is typical of class III delocalized systems, and makes the precise hyperfine coupling assignment difficult in the absence of higher frequency (Q-band) experiments. Super-hyperfine coupling to the bridging nitrogen nuclei of the Cu2N2 core or to the phosphine chelates are also not discernable from these experiments. Cationic 4.10a exhibits a very isotropic signal with poorly defined hyperfine coupling; these characteristics persisted regardless of the solvent compositions, temperature (3.7-100 K), or counter ion (4.10b) which were tried. The spectrum does however bear gross similarity to the spectrum of 4.8 and can be simulated using substantially large linewidths. The parameters used for the simulation shown in Figure IV.7 are gx = 1.987, gy = 2.025, gz = 2.098, Ax = 10.5⋅10-4 cm-1, Ay = 20⋅10-4 cm-1, Az = 25⋅10-4 cm-1, which are very consistent with those obtained for the simulation of 4.8. Based on these experiments we conclude that despite the odd structural distortions observed in the solid state, significant electronic delocalization in the Cu2N2 core persists in both molecules and that the class III assignment is fully consistent with data. IV - 27 300 312 324 336 348 360 Field / mT Figure IV.7: EPR spectrum (X-band) at 3.7 K. Top: 4.8 2-Me-THF, experimental spectrum (black) and simulation (gray). Bottom: 4.10b in 2-Me-THF, experimental spectrum (black) and simulation (gray). EPR studies of 4.11 were conducted on the polycrystalline solid and the frozen dichloromethane:toluene (~20:1) glass. Both samples contained intense signals over the temperature range of 3.7-154 K. In both samples half-field absorptions were detected for the forbidden transition (165 mT) at 3.7 K confirming an S=1 state for 4.11 (Figure IV.8). The easily resolvable signal for the integer spin systems at X-band frequency must correspond to a system with very small zero-field splitting (D < 0.3 cm-1) of the triplet state. The spectrum of 4.11 in frozen dichloromethane:toluene glass (Figure IV.8, bottom) exhibits a broad, nearly axial signal consistent with a paramagnetic metal center. The very weak half-field signal is evident at 165 mT (g = 4.02) which is >1000 times smaller in amplitude than the major absorption feature at g = 1.998. The temperature dependence of the sample in a frozen dicholoromethane glass was examined. The IV - 28 integration of the microwave absorption at constant power (20 μW) was found to increase exponentially with decreasing temperature over the range 3.7−154 K which is characteristic of a ground state paramagnet. X-band EPR studies of a polycrystallinepowder sample over a temperature range of 3.7-100 K exhibited an axial signal at g = 2.006 with very weak half-field signal at 167 mT (g = 4.020) (Figure IV.8, top). No resolvable EPR signals characteristic of mononuclear CuII were evident in these studies. 38 140 300 160 mT 180 310 200 320 330 Field / mT 340 350 360 Figure IV.8: EPR spectrum (X-band). Top: polycrystalline solid sample of 4.11 (9.378 GHz, 6.377 μW). Bottom: frozen glass 4.11 in CH2Cl2 / toluene (9.339 GHz, 6.377 μW). Bottom inset: weak half-field signal in frozen glass (9.373 GHz, 6.393 mW). Magnetic susceptibility studies of 4.10b and 4.11 were conducted in an effort to better understand the electronic structure of these two dimers. For complex 4.10b, fitting of χT as a function of T was achieved using the spin-only formula derived from the Curie law with an added term for the temperature independent paramagnetism (χTIP) using values of g = 1.851(3) and χTIP = 0.001960(8) (eq. IV.1). IV - 29 χT = (Ng β / 4k) + χTIP·T 2 2 (IV.1) The relatively large value for the χTIP 39 may be attributable to population of low-lying excited states as has been previously reported by Vila and Gray in their studies of the mixed-valent, CuA site of CcO. 40 From this data, 4.10b was determined to have a μeff value which slopes linearly from 1.58-2.17 Bohr magnetons (BM) (3 K – 300 K) consistent with the expected doublet ground state of the cation. A value of 1.63 BM was determined for the room-temperature solution magnetic moment of 4.10b in dichloromethane by the method of Evans 41 and corresponds reasonably well with the values determined by SQUID magnetometry on the solid state material. The magnet susceptibility of 4.11 was also probed by SQUID magnetometry and as was found by EPR the magnitude of χm increased exponentially with decreasing temperature as would be expected for a ground state triplet (Figure IV.9). Plotting of χmT as a function of T reveals a linear relationship where the value of χmT slopes from a minimum of 0.29 emu⋅T⋅mol-1 at 3K to 0.447 at 300 K emu⋅T⋅mol-1. From this data, values of μeff can be calculated and were found to increase steadily from 1.52 BM at 3 K to 1.90 BM at 300 K indicating an increasing effective magnetic moment with increasing temperature. Though the magnetic susceptibility would seem wholly consistent with 4.11 being a simple doublet spin system, analogous to 4.10b, the elemental analysis of the SQUID sample indicates that two [SbF6]¯ anions must be present, which indicates the two-fold oxidation of 4.5. Moreover, the susceptibility data is consistent with a single localized paramagnetic center and is likely not reflective of a system containing two exchange-coupled doublets as is classically observed for formal dicopper(II,II) complexes. The solution magnetic moment was also determined in this case by the IV - 30 Evans’ Method and afforded a value for the room temperature moment of μeff = 1.69 BM in dichloromethane. Figure IV.9: Magnetic susceptibility study of 4.11. (●) χ vs. T; (○) χΤ vs. T Absorption spectroscopy of the oxidized species. The absorption spectra for mixed valence complexes 4.8 and 4.10b are shown in Figure IV.10. Both spectra are rich in electronic transitions which potentially derive from metal→ligand, ligand→metal, d→d, and intervalence charge transfer (IVCT) processes resulting from a vacancy in the singly occupied molecular orbital (SOMO). Gaussian fitting of the spectra affords values which are illustrative of the close relationship of the electronic structures of 4.8 and 4.10b (Table IV.3). The same transitions are observed for each complex varying only slightly in the energy of the transition. Notably the intervalence charge transfer bands (IVCT) are easily recognizable from the spectra with λmax = 6068 cm-1 and 5687 cm-1 for 4.8 and 4.10b, respectively. This IVCT absorption band is characteristic of the delocalized IV - 31 mixed-valence state of diatomically bridged Cu1.5Cu1.5 systems. The absorption with λmax = 31 700 cm-1 and shoulder at ~26 500 cm-1 results from ligand centered π→π* transitions which are perturbed to lower energy as a result of coordination to copper which is evident by comparison to the absorption spectra of d10 complexes 4.3, 4.5, and 4.6. 25 15 3 ε / 10 cm -1 20 10 5 0 35 30 25 20 15 3 wave numbers / 10 cm 10 5 -1 Figure IV.10: Absorption Spectra of 4.8 (red), 4.10b (black), and 4.11 (blue) in CD2Cl2. The transitions #3 and #4 (Table IV.3) occur at higher energy in 4.8 than 4.10b, but the trend reverses for transitions #5-#7. We speculate that transitions #3 and #4 result IV - 32 from MLCT transitions by promoting an electron from the highest occupied orbitals to unoccupied ligand antibonding orbitals. Similarly, transitions #5-#7 could correspond to LMCT or d-d transitions from the lower filled orbitals to the SOMO. The intervalence charge transfer bands, #1 and #2, can be described in the same fashion, which is completely consistent the molecular geometries determined by X-ray crystallography. The longer Cu(1)-Cu(2) distance of 4.8 as compared to 4.10b would obviously result in a diminished ability to exchange electrons in a direct Cu-Cu exchange regime and therefore should result in an increased energy of the charge transfer transition between valence states. This observation is also supportive of the proposed direct Cu-Cu interaction in the d10 complexes 4.3 and 4.5. Table IV.3: Absorption bands determined by Gaussian fitting of the spectra in Figure IV.10 4.8 4.10b 4.11 -1 -1 -1 -1 -1 -1 -1 # cm ε (M cm ) cm ε (M cm ) cm ε (M-1cm-1) 1 5516(15) 2248(242) 5327(2) 3180(93) 7782(22) 335(9) 2 6755(78) 2799(143) 6514(30) 2875(49) 16 105(1) 10 671(36) 3 10 065(14) 1416(25) 10381(7) 1232(12) 16 858(11) 2546(26) 4 12 012(27) 867(52) 12 337(14) 956(14) 18 522(51) 863(19) 5 16 152(70) 323(49) 16 062(23) 484(14) 28 184(10) 8953(229) 6 19 133(28) 967(47) 19 101(14) 1485(15) 30 082(61) 13 525(145) 7 22 831(13) 1595(55) 22 367(9) 2380(12) 8 26 754(16) 2514(112) 26 115(8) 2885(23) 9 31 403(5) 4986(64) 31 351(3) 5654(22) The spectra of the characteristically blue compound 4.11 is plotted in Figure IV.10 and exhibits similar charge transfer bands in the visible region between 21 000 and 13 000 cm-1 as is observed in 4.10b. Gaussian fitting of the spectra of 4.11 finds major visible transitions centered at 16 105(1) cm-1 (10 671(35) M-1cm-1) and 16 858(11) cm-1 (2546(26) M-1cm-1). The intensity of the absorption at 16105 and narrow line width (350 cm-1) immediately suggests the possible presence of an organic or main group radical cation in the system. 42 IV - 33 A broad, low ε band is also present in the NIR position of the spectrum which is characteristic of a d-d transition. This band assignment is suggested to be the transition from the highest energy filled d-orbital to the highest energy SOMO of 4.11. Absorption bands corresponding to d-d transition may also occur in complex 4.10b; however, it is not distinguishable here because of the intense IVCT bands. Strong ligand-based absorptions at 28 200 cm-1 with a shoulder at ~26 500 cm-1 are roughly consistent with the CuICuI complex, 4.5, possibly suggesting that one of the Cu centers is closed shell. Further studies are needed in order to rigorously assign the visible bands to the appropriate ligand field transitions. Ar2 N P Cu II I P Ar2 N P Cu I Cu II P Cu I II P N Ar2 P P N Ar2 P P Ar2 N P P Ar2 N P III Cu II P Cu I N Ar 2 Cu III P P IV Cu I N Ar2 P Figure IV.11: Possible resonance contributors to the electronic structure of 11 {•NAr2+ = delocalized, ligand center cation radical}. Discussion of the electronic localization in 4.11. Given the experimental data presented herein, four reasonable possibilities exist for the localized electronic structure of the dicationic dimer, 4.11. The distinction of a “localized” electronic structure is made because in reality the cationic charge is delocalized across the entire molecular species, and the molecule is also likely to be resonance stabilized. From the electrochemical studies, an initial assignment of two CuII nuclei was postulated based on the assumption IV - 34 of a symmetric, non-redox active ligand platform (Figure IV.11, I). In fact, the featureless cyclic voltammetry profile observed for the Zn dimer, 4.7, in both THF and dichloromethane suggests the ligand is not redox active in the absence of Cu within the desired potential window. However the intense X-band EPR signal of 4.11, which is indicative of a very small zero field splitting (<0.3 cm-1), 43 is not typical of dicopper(II,II) complexes with dinuclear bridging units. 44 The magnetic susceptibility data is not consistent with an exchange coupled triplet spin system. Exchange-coupling would have been expected in this case given the close molecular proximity of the two CuII nuclei. The plot of χT as a function of T was experimentally determined and can be crudely fit using the Bleany-Bowers exchanged coupled model (eq. IV.2) with values for g = 1.238(1) J = -0.028(26) cm-1, and χTIP = 5.32(2) x10-4 emu·mol-1. However, given the large uncertainty in J, and the unreasonably low g value, the presented data would appear largely inconsistent with a simple S=1 exchange coupled model. χT = (2Ng2β2 / k) · [3 + e-J/kT]-1 + χTIP·T (IV.2) Upon inspection of the Cu2N2 core geometry of 4.11 afforded by X-ray crystallography, structural nonequivalence between the two copper sites is immediately apparent. This structural data then demands that electronic asymmetry must also exist in 4.11, and so I can be eliminated from consideration as a major resonance contributor. For the same reason, II is also eliminated as a major contributor to the electronic structure (•NAr2+ denotes a ligand centered cation radical which is not restricted to localization on the N atoms). If two ligand centered radicals were present in the structure, as shown in II, there would be no apparent need for nonequivalence between the copper centers. IV - 35 Moreover, the radical character should also be manifest in the electrochemical studies of 4.6. Resonance contributors III and IV, however, are both consistent with the asymmetry observed for 4.11 by X-ray crystallography. The apparent dative Cu(1)-N interactions (~2.4 Å) are assigned to a formally cationic CuI center which is closed shell. The two shorter Cu(2)-N bond distances (~2.0 Å) are clearly covalent in nature and consistent with the bond lengths observed previously for copper amides. Based on this, an assignment of a neutral CuII (III) or a cationic CuIII (IV) center are reasonable. In order to balance the oxidation state of III, a delocalized cation radical must also be present in the ligand. Based on literature precedent for the disproportionation of 2 CuII → CuI + CuIII, we speculate the 4.11 forms as a result of rapid intramolecular disproportionation of a dicopper(II,II) complex. 45 From the data presented in the current work, a conclusive assignment of whether III or IV is the dominant resonance contributor cannot be established. However this electronic structure problem is not new in the field of bioinorganic chemistry and has been addressed for many years in the studies of galactose oxidase and ribonucleotide reductase. 46 In these cases, conclusive assignments were only attained by the use of X-ray absorption near-edge structure (XANES) analysis. Given the available data, the best description of the electronic structure of 4.11 is IV, though it is likely resonance stabilized through III. This formulation presents the opportunity to study a d8 CuIII (S=1) nucleus in a compressed tetrahedral environment, something that has not been previously characterized. Previously, S=1 CuIII systems have been observed exclusively in octahedral geometries. 47,48 Though pseudo-tetrahedral examples of Cu III IV - 36 have not been reported, a large volume of literature does exist for isoelectronic NiII complexes in distorted tetrahedral environments with S=1 ground states. 49 The formulation of IV is supported by the magnetic data. Though the EPR spectrum of 11 has not been effectively simulated, the experimental spectrum is reasonable for a system with two unpaired electrons localized on a single Cu center (spin 3/2) with a small zero-field splitting parameter. Further, the lack of prominent ferromagnetic or antiferromagnetic exchange coupling from the SQUID data supports the postulate of spin localization. The magnetic moment observed in both solid (SQUID) and solution (Evans’ method) samples is significantly lower than the spin only value for two unpaired electron (2.8 BM), and is likely the result of considerable spin-orbit coupling or population of lowlying singlet excited states. 50,51 The magnitude of 1.90 BM for μeff-300K of 4.11 seems more consistent with a doublet spin system than a triplet, and is remarkably similar to 4.10b. The possibility of the SQUID sample in fact being 4.10b and not 4.11 is not possible, because the EPR and optical spectroscopy of the sample used for the SQUID experiment were not characteristic of 4.11, and showed no signs of degradation. Further, the microanalysis of the sample of 4.11 used for the SQUID is precisely consistent with the dicationic formulation of 4.11, which along with the X-ray diffraction study definitively establishes the two electron oxidation of 4.5. Subnormal magnetic moments have been observed previously for paramagnetic CuIII. Weighardt and co-workers reported a high-spin, octahedral CuIIIS6 complex which is reported to have a μeff = 2.4 BM at 20 K. IV - 37 IV.C. Summary and Future Directions In this manuscript we have presented a molecular dicopper system which can stabilize multiple oxidation states with minimal structural reorganization and no additional ligand coordination to the Cu centers. The spectroscopic, magnetic, and redox studies described herein are indicative of the highly covalent Cu2N2 diamond core architecture of the presented metal complexes. An important aspect of this system is that the ligand superstructure remains externally rigid while allowing minor adjustments within the metalligand core to accommodate a variety of metals and oxidation states. We characterize this system as a demonstration of the “entatic state” which is embodied by the polymetallic cofactors that mediate multi-electron redox processes at or near enzymatic functional sites. Particularly, this manuscript details the synthesis and study of dicopper complexes (4.3 and 4.5) which exhibit highly efficient luminescence and can sustain two reversible one electron oxidations {4.5→4.10→4.11}. Unique to this dicopper system is that each of these oxidized species can be isolated and structurally characterized, and so we are able to observe that there is minimal structural reorganization of the molecule and no change to the metal coordination number. The unexpected assignment of a putative pseudo-tetrahedral CuIII/CuI dimer further serves to illustrate the varied electronic states which the ligand can support. The covalent nature of the system is also evident in the exceptionally high quantum yields and long emission lifetimes which are observed for the dicopper systems, but not for the Zn2N2 (4.6) or CuZnN2 (4.7) complexes. In this manuscript, we also present one of the few systems where the one-electron oxidation products of a highly luminescent copper complex (4.8 and 4.10) can be isolated and are structurally unperturbed. This presents an ideal opportunity for the acquisition and IV - 38 comparison of an excited-state molecular structure to both the ground-state of the luminophore and the one-electron oxidized complex. The analogous dizinc and heterobimetallic copper-zinc molecular complexes were also been prepared and aided in the understanding of the spectroscopic and redox characteristics of 4.5. Future work involving this molecular system will address the electronic structure assignment of 4.11 and potential analogs. The utilization of more advanced spectroscopic methods coupled with appropriate theoretical studies will allow for a more accurate description of the electronic distribution of 4.11. Previously, XANES, XMCD, and resonance Raman techniques have been utilized to deconvolute similar electronic structure problems and will be pursued in due course. We also intend to address the excited state reactivity of 4.3 and 4.5. Previously, we have reported the time-resolved diffusion-limited emission quenching of 4.3 by 2,6,-dichloroquinone. In preliminary studies, we have found the intriguing result that by saturating a THF solution of 4.5 with N2O, 97% of the emission is quenched. Prolonged irradiation of 4.5 under N2O with visible light (> 375 nm) results in total consumption of the starting material 4.5, and a mixture of undetermined products. Reaction in the absence of light was not observed. Regardless of the product composition this is the first reported example of a molecular copper complex showing any such activity with N2O which presumably is facilitated by the highly reducing excited state of 4.5. These results are particularly interesting in the context of understanding the mechanism of the tetranuclear CuZ functional site in N2OR and it is our hope that direct evidence for the intermediacy Cu-[NNO] adducts in this decomposition will be attainable. IV - 39 IV.D. Experimental Section General. All manipulations were carried out using standard Schlenk or glove-box techniques under a dinitrogen atmosphere. Unless otherwise noted, solvents were deoxygenated and dried by thorough sparging with N2 gas followed by passage through an activated alumina column. Non-halogenated solvents were tested with a standard purple solution of sodium benzophenone ketyl in tetrahydrofuran to confirm effective oxygen and moisture removal. All reagents were purchased from commercial vendors and used without further purification unless otherwise stated. Mesityl copper, 2, 2´difluorodiphenylamine, diethylphosphoramidous dichloride, 52 4,4´-di-tbutyldiphenyl amine, 53 and [Cp2Fe][B(C6H3(CF3)2)4] 54 were prepared according to literature procedures. Elemental analyses were performed by Desert Analytics, Tucson, AZ. Deuterated solvents were purchased from Cambridge Isotope Laboratories, Inc. and degassed and dried over activated 3 Å molecular sieves prior to use. A Varian Mercury300 or INOVA-500 NMR spectrometer was used to record 1H, 13C, 19F and 31P NMR spectra at ambient temperature. 1H chemical shifts were referenced to residual solvent. GC-MS data was obtained by injecting a dichloromethane solution into an Agilent 6890 GC equipped with an Agilent 5973 mass selective detector (EI). High-resolution EI mass spectroscopy was carried out by the Caltech Chemistry Mass Spectral Facility using a JEOL JMS600. UV-vis measurements were taken on a Cary 500 UV/Vis/NIR Spectrophotometer or a Cary 50 UV/Vis Spectrophotometer using a 1.0 cm or a 0.1 cm quartz cell with a Teflon stopper. X-ray diffraction studies were carried out in the Beckman Institute Crystallographic Facility on a Bruker Smart 1000 CCD diffractometer. IV - 40 Magnetic Measurements. Measurements were recorded using a Quantum Designs SQUID magnetometer running MPMSR2 software (Magnetic Property Measurement System Revision 2). Data were recorded at 5000 G. Samples were suspended in the magnetometer in a clear plastic straw sealed under nitrogen with Lilly No. 4 gel caps. Loaded samples were centered within the magnetometer using the DC centering scan at 35 K and 5000 G. Data were acquired at 3–30 K (one data point every 1 K) and 30–300 K (one data point every 5 K). The magnetic susceptibility was adjusted for diamagnetic contributions using the constitutive corrections of Pascal’s constants. 55 The molar magnetic susceptibility (χm) was calculated by converting the calculated magnetic susceptibility (χ) obtained from the magnetometer to a molar susceptibility (using the multiplication factor {(molecular weight)/[(sample weight)*(field strength)]}). CurieWeiss behavior was verified by a plot of χm-1 versus T. Effective magnetic moments were calculated using equation IV.3. Solution magnetic moments were measured by the Evans method. μeff = sqrt(7.997 χm T) (IV.3) EPR Measurements. X-band EPR spectra were obtained on a Bruker EMX spectrometer (controlled by Bruker Win EPR Software v. 3.0) equipped with a rectangular cavity working in the TE102 mode. Variable temperature measurements were conducted with an Oxford continuous-flow helium cryostat (temperature range 3.6–300 K). Accurate frequency values were provided by a frequency counter built into the microwave bridge. Solution spectra were acquired in 2-methyltetrahydrofurn or 20:1 dichloromethane:toluene solution. Sample preparation was performed under a dinitrogen atmosphere in an EPR tube equipped with a ground glass joint. IV - 41 Electrochemistry. Electrochemical measurements were carried out in a glove-box under a dinitrogen atmosphere in a one-compartment cell using a BAS model 100/W electrochemical analyzer. A platinum electrode and platinum wire were used as the working and auxiliary electrodes, respectively. The reference electrode was Ag/AgNO3 in THF or CH2Cl2. Solutions of electrolyte (0.35 M tetra-n-butylammonium hexafluorophosphate in THF or 0.1 M tetra-n-butylammonium hexafluorophosphate in CH2Cl2) and analyte were prepared in a glove-box. Lifetime measurements. A solution of analyte in either tetrahydrofuran or dicholoromethane was prepared in a nitrogen filled glovebox. The quartz cuvettes (1 cm pathlength) were charged with this solution, sparged briefly with argon, and sealed with a ground-glass stopper. The absorption spectra were acquired both before and after measurements to ensure the sample was not photodegrading. Luminescence lifetime measurements were carried out using one of the configurations listed below (A-C) below and specified in Table IV.4. Configuration A. Fluorescence decay kinetics measurements were carried out as previously described 56 using 8 ns pulses (at a repetition rate of 10 Hz) from a Nd:YAG laser pumped OPO (Quanta Ray Pro, Spectra Physics). The luminescence was dispersed through a monochromator (Instruments SA DH-10) onto a photomultiplier tube (PMT) (Hamamatsu R928). The PMT current was amplified and recorded with a transient digitizer (Tektronix). Measurements were performed with λex = 460, λem= 510 nm at 298 K. A 500 nm low-pass filter was placed in front of the PMT in order to eliminate noise due to scattered laser light. The emission decay was averaged over 50 laser pulses and fit to an exponential function from which kobs was determined. IV - 42 Configuration B. Fluorescence decay kinetics measurements were carried out as previously described 57 using the third harmonic of a regeneratively amplified, modelocked Nd:YAG laser (355 nm) for excitation, and a picosecond streak camera (C5680, Hamamatsu Photonics, Hamamatsu City, Japan) for detection. The fluorescence was selected with both long- and short-pass cutoff filters (450 nm < λobs < 500 nm, 7; 400 nm < λobs < 450 nm; 6a). Configuration C. Fluorescence decay kinetics measurements were carried out as previously described 58 using the second harmonic (435 nm) of a regeneratively amplified femtosecond Ti:sapphire laser (Spectra-Physics) as an excitation source, and a picosecond streak camera (C5680, Hamamatsu Photonics, Hamamatsu City, Japan) in the photon-counting mode for detection. Emission was selected by an interference filter (λ = 488 ± 5 nm). Table IV.4: Data for Excited State Lifetime Measurements. Solution Config. kobs (s-1) Lifetime (τ, 1/ kobs) (μs) C k1 (75%) 3.74E08 τ 1 2.7E–03 150 μM 4.2 in pet. ether k2 (25%) 1.25E08 τ 2 8.0E–03 a – 9.14E+04 10.9a 10 μM 4.3 in THF A 8.72E+04 11.4 10 μM 4.5 in THF B ─ ─ 10 μM 4.6 in CH2Cl2 B ≤ 5E–04 ≥ 2E+09 10 μM 4.7 in CH2Cl2 (a) Value taken from: Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 2030. Quantum yield experiments. Emission spectra were recorded on a Spex Fluorolog-2 spectro-fluorometer. A solution of analyte or reference in either benzene or tetrahydrofuran was prepared in a nitrogen filled glovebox. Cuvettes (1 cm path) were charged with this solution, sparged briefly with argon, and sealed with a greased groundglass stopper. The absorption spectra were acquired both before and after fluorescence IV - 43 measurements to ensure the sample was not degrading. Fluorescent measurements were performed at the specified wavelength and corrected for detector response after equilibration to 298 K. The area under the curve of the emission spectrum was determined using standard trapezoidal integration methods. Quantum yields (Table IV.5) were then calculated by the methods described by Demas and Crosby 59 using eqn. IV.4 . Q = (QR)(I / IR)(ODR / OD) (n2 / nR2) Q: QR: I: I R: ODR: OD: n: nR: (IV.4) quantum yield of the sample. quantum yield of reference. integrated intensity of analyte. integrated intensity of reference. optical density of the reference in absorption units. optical density of the analyte in absorption units. index of refraction of the solvent in which the analyte was dissolved. index of refraction of the solvent in which the reference was dissolved. Table IV.5: Data for Quantum Yield Measurements. Solution λex OD (meas.) I (meas.) Q (cacld.) Quantum Yield of 4.2 0.819 2.048E09 0.68 a 30 μM 4.3 in benzene 340 nm 1.021 2.184E09 50 μM 4.2 in benzene 340 nm 0.58 Quantum Yield of 4.5 10 μM 4.3 in THF 8.2 μM 4.5 in THF 440 nm 440 nm 3.807E-02 3.786E-02 Quantum Yield of 4.7 C20H24N2O2 in 1.0 N H2SO4b,c 72 μM 4.7 in CH2Cl2 370 nm 370 nm 0.2742 0.2703 N2O Experiment quenching 8.2 μM 4.5 in THF / no N2O 8.2 μM 4.5 in THF / sat. N2O 440 nm 440 nm 4.008E-02 3.924E-02 6.547E08 0.68a 5.504E08 0.57 4.237E09 0.55 5.073E06 7.47E-04 5.091E08 0.68a 1.611E07 0.022 (a) Value taken from: Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 2030. (b) C20H24N2O2 = quinine. (c) Ref. 59. Synthesis of Lithium Diisobutylphosphide. In a 500 mL Erlenmeyer flask diisobutylphosphine (25.0 g, 0.171 mol) was dissolved in 200 mL of petroleum ether and cooled to -80 ºC, at which time at 1.6 M solution of nbutyl lithium in hexane (107 mL, IV - 44 0.171 mol) was added over 20 min. The reaction was then stirred at ambient temperature for 24 h, concentrated in vacuo to ca. 50 mL, and the white solids were then collected on a sintered-glass frit. Washing of the solids with petroleum ether afforded a single phosphorous containing product (21.1 g, 81%) by 31P NMR upon drying. 31 P{1H} NMR (121.5 MHz, THF): -91.2. Synthesis of Bis(2-(diisobutylphosphino)phenyl)amine, (PNP)H (4.1). In a 250 mL sealable reaction bomb, a 1.6 M nbutyl lithium solution (7.90 mL, 12.6 mmol) in hexanes was added dropwise to a solution of bis(2-fluorophenyl)amine (2.46 g, 12.0 mmol) in THF (20 mL). After stirring for 15 min, the solution was concentrated in vacuo to remove the majority of the reaction volatiles. A solution of lithium diisobutylphosphide (5.47 g, 36 mmol) in THF (40 mL) was then added and the vessel was sealed with a Teflon plug. The reaction was heated at 45 ºC for 4 days and was monitored by 19F{1H} NMR to confirm the complete disappearance of the aryl fluoride resonance. The reaction was then quenched with methanol (5 mL) and the solution became yellow in color. Petroleum ether (50 mL) was added and the mixture was filtered twice through Celite to remove solids. Removal of the solvent in vacuo afforded an orange oil which was diluted in petroleum ether (30 mL) and filtered through two plugs of silica in a 60 mL sintered-glass frit. Evaporation of the solvent under reduced pressure afforded a spectroscopically pure, pale green oil (4.10 g, 75%). 1 H NMR (300.1 MHz, CDCl3): δ 8.03 (t, 1H), 7.52 (m, 2H), 7.33 (m, 2H), 7.26 (t, 2H), 7.00 (t, 2H), 1.73 (m, 12H), 1.08 (d, 12), 1.03 (d, 12H). 13C{1H} NMR (75.5 MHz, CDCl3): δ 147.8, 131.8, 129.4, 128.0, 121.0, 119.3, 116.7, 39.4, 26.6, 24.7, 24.3. 31P{1H} NMR (121.5 MHz, CDCl3): δ -54.5. UV-vis (benzene, nm(M-1cm-1)): 302(18,700), 334 IV - 45 sh. FAB+ MS: Calcd. for C28H45NP2: 457.3027. Found: 458.3122 [M+H], 400.2226 [Mi Bu], 312.1904 [M-(iBu2P)]. Synthesis of {(PNP)Li}2 (4.2): At ambient temperature a 1.6 M nbutyl lithium solution (4.0 mL, 6.4 mmol) in hexanes was added dropwise to a solution of 4.1 (2.63 g, 5.76 mmol) in petroleum ether (50 mL) over 15 min. The reaction was stirred for 30 min, at which time a solid began to precipitate. The solution was concentrated to ca. 30 mL and cooled to -30 °C for 12 h. The resulting solids were collected on a sintered-glass frit as a fine pale yellow powder and dried thoroughly (2.24 g, 84 %). Crystals suitable for Xray diffraction were obtained by cooling a concentrated solution of 4.2 in petroleum ether to –35 ºC. 1 H NMR (499.9 MHz, C6D6): δ 7.21 (m, 2 H), 7.06 (m, 4 H), 6.71 (m, 2 H), 1.8-0.6 (br m, 36 H). 13C{1H} NMR (125.7 MHz, C6D6): δ 170.6, 132.4, 131.2, 128.7, 128.1, 118.6, 42.3, 36.2, 26.3, 25.3. 31P{1H} NMR (121.5 MHz, C6D6): δ -49.4 (q, 61 Hz). UV-vis (benzene, nm(M-1cm-1)): 303(18,200), 359(17200), 398 sh, 414 sh. Anal. calcd. for C56H88Li2N2P4: C, 72.55; H, 9.57; N, 3.02. Found: C, 73.21; H, 9.43; N, 3.14. Synthesis of {(PNP)Cu}2 (4.3): A solution of 4.2 (1.00 g, 2.16 mmol) in diethyl ether (20 mL) was added to a slurry of CuBr⋅S(CH3)2 (0.466 g, 2.27 mmol) in diethyl ether (30 mL) and stirred for 12 h. The solvent was removed in vacuo and the resulting yellow solids were dissolved in petroleum ether (50 ml) and filtered to remove insoluble materials. Removal of petroleum ether and drying the yellow solids in vacuo afforded analytically pure material (1.04 g, 92%). Crystals suitable for X-ray diffraction were obtained by slow-evaporation of a petroleum ether solution of 4.3. IV - 46 1 H NMR (499.9 MHz, CD2Cl2): δ 7.20 (m, 4H), 6.93 (m, 4H), 6.71 (br d, 4H), 6.61 (t, 4H), 1.81 (m, 4H), 1.58 (m, 8H), 1.39 (m, 8H), 1.29 (m, 4H), 0.99 (d, 12H), 0.73 (d, 12H), 0.70 (d, 12H), 0.60 (d, 12H). 13C{1H} NMR (125.7 MHz, CD2Cl2): δ 169.4, 131.8, 130.5, 128.7, 125.2, 117.8, 40.2, 36.4, 26.3, 26.2, 26.0, 25.7, 25.4, 25.1. 31P{1H} NMR (121.5 MHz, C6D6): δ -33.9. UV-vis (THF, nm(M-1cm-1)): 294 (sh), 311 (17000), 350 (42,000), 387 sh, 423 (4400), 446 sh. Anal. calcd. for C56H88Cu2N2P4: C, 64.65; H, 8.53; N, 2.69. Found: C, 64.54; H, 8.25; N, 2.62. Synthesis of diisobutylchlorophosphine: Neat diethylphosphoramidous dichloride (58.0 g, 0.333 moles) was added to a 2.0 M solution of isobutyl magnesium chloride in diethyl ether (350 mL) at 0 ºC over a period of 30 min. Following addition, the solution was stirred at ambient temperature for 1 h and the crude 31P NMR indicated that the starting material had been consumed and diisobutylchlorophosphine (31P NMR: 47.9 ppm, (Et)2NP(iBu)2) was the only phosphorus containing product. The solution was again cooled to 0 ºC and a 2.0 M solution of anhydrous HCl in ether (350 mL) was added via cannula while stirring vigorously. A considerable amount of solid precipitated over a 2 h period and the supernatant was isolated by cannula filtration. The solids were extracted with 200 mL of diethyl ether, and the solvent was removed by fractional distillation. The crude viscous oil was purified by fractional vacuum distillation (26–29 ºC at 0.005 Torr) followed by removal of the residual diethyl ether by prolonged exposure to vacuum at -10 ºC with stirring. The product was isolated as a spectroscopically pure, colorless oil (38.6 g, 65 %) which exhibited a single resonance by 31 P NMR spectroscopy consistent with previously published results. 60 1 IV - 47 H NMR (300 MHz, C6D6): δ 1.87 (m, 2H), 1.78 (br m, 2H), 1.24 (br m, 2H), 0.90 (br s, 12H). 31P{1H} NMR (121 MHz, C6D6): δ 109.9. Synthesis of di(2-bromo-4-tbutylphenyl)amine: In air, neat Br2 (3.6 mL, 0.071 mol.) at ~5 ºC was added dropwise to a slurry of 4,4´-di-tbutylphenylamine (10.0 g, 0.0356 mol) in acetic acid at ~16 ºC. Following addition, the solution was stirred at ambient temperature for 2 hours and a dilute solution of Na2S2O4 (500 mL) was added and the resulting solution was stirred for 15 min. The solids were collected on a frit and washed with H2O (3 x 200 mL). The crude product was purified by crystallization at –20 ºC from methanol/chloroform as a white solid (12.35 g, 80 %). 1 H NMR (300 MHz, CDCl3): δ 7.59 (m, 2H), 7.23 (m, 4H), 6.30 (br s, 1H) 1.32 (s, 18H). 13 C{1H} NMR (75.5 MHz, CDCl3): 145.9, 137.9, 130.2, 125.3, 117.8, 114.1, 34.5, 31.5. GC-MS(ES) 439 (M), 424 (M-CH3). Synthesis of bis(2-diisobutylphoshino-4-tbutylphenyl)amine, (tBu2-PNP)H (4.4): A 1.6 M solution of nbutyl lithium in hexane (26 mL) was added dropwise to a solution of di(2-bromo-4-t-butylphenyl)amine (6.0 g, 13.7 mmol) in diethyl ether (100 mL) at –70 ºC with stirring. The solution immediately became yellow in color and was stirred at ambient temperature for 4 h at which time a precipitate had formed. The reaction mixture was again cooled to –70 ºC, at which time diisobutylchlorophosphine (7.67 g, 41.7 mmol) was added as a 1:1 solution with diethyl ether and the reaction was allowed to warm to room temperature After 36 hours at ambient temperature a large amount of precipitate had formed, and was removed by filtration through Celite® after diluting the reaction solution with diethyl ether (100 mL). The filtrate was treated with a 1.0 M solution of anhydrous HCl in ether (2 x 50 mL) which immediately resulted in the precipitation of IV - 48 the product as an HCl salt. The solids were collected on a frit and washed with petroleum ether (3 x 50 mL). These solids were suspended in a 4:1 mixture of THF:CH3CN and excess (> 4 equiv.) NaOMe was added and the mixture was stirred for 2 h. The solvent was removed in vacuo. The sticky solids were extracted with petroleum ether (50 mL) and filtered through Celite®. The solvent was removed from the filtrate under reduced pressure. The resultant tacky translucent solid was stirred in CH3CN for 1 h, affording a tractable white solid (5.35 g, 68%) which was pure by NMR spectroscopy. 1 H NMR (300 MHz, CD2Cl2): δ 7.66 (t, 1H), 7.46 (m, 2H), 7.16 (m, 4H), 1.34 (s, 18H), 1.00 (d, 6H), 0.94 (d, 6H). 13C{1H} NMR (75.5 MHz, CD2Cl2): δ 145.9, 143.5, 128.8, 127.4, 126.7, 116.6, 39.6, 34.9, 31.9, 27.1, 24.9, 24.6. 31P{1H} NMR (121 MHz, CD2Cl2): δ –53.1. Anal. calcd. for C36H61NP2: C, 75.88; H, 10.79; N, 2.46; Found: C, 75.35; H, 10.20; N, 2.81. UV-vis (THF, nm(M-1cm-1)): 303 (20700), 344 sh. Synthesis of {(tBu2-PNP)Cu}2 (4.5): Mesityl-Cu (128 mg, 0.702 mmol) and 4.4 (400 mg, 0.702 mmol) were dissolved in petroleum ether and stirred for 12 hr at ambient temperature. The emissive yellow solution was filtered through a glass micro-filter and the solvent was in vacuo. The bright yellow solid was stirred in CH3CN for 1 h in the dark and collected on a frit. Drying of the solids in vacuo afforded a finely divided analytically pure yellow solid (400 mg, 90%). Crystals suitable for X-ray diffraction were obtained by slow evaporation of the solvent from a solution of 4.5 in THF. 1 H NMR (300 MHz, C6D6): δ 7.43 (m, 4H), 7.02 (m, 8H), 2.12 (n, 8H), 1.84 - 1.50 (m, 16H) 1.34 (s, 36H), 1.15 (d, 12), 0.89 (dd, 24), 0.76 (d, 12H). 13C{1H} NMR (75.5 MHz, C6D6): δ 168.1, 139.5, 128.5, 128.2, 127.1, 125.4, 40.0, 36.3, 34.6, 32.3, 26.7, 26.5, 25.9, 25.8. 31P{1H} NMR (121 MHz, C6D6): δ –35.3. Anal. calcd. for C72H120Cu2N2P4: C, IV - 49 68.38; H, 9.56; N, 2.21; Found: C, 68.55; H, 9.38; N, 2.61. UV-vis (THF, nm(M-1cm-1)): 295 (20 500), 318 (21 100), 357 (33 300), 383 sh, 433 (4600), 459 (3800). Synthesis of [{(tBu2-PNP)Zn}2][PF6]2 (4.6a): In a 20 mL reaction vial, a 2.0 M solution of CH3ZnCl in THF (68.0 mg, 0.132 mmol) was added to a solution of 4.4 (75 mg, 0.13 mmol) in THF (5 mL). The solution was heated to 75 °C for 4 h. TlPF6 (46.1 mg, 0.132 mmol) was added and the solution was heated for 12 h longer. The solvent was removed in vacuo and the resultant solids were washed with petroleum ether (2 x 5 mL). The solids were then extracted into 1:1 dichloromethane:THF (2 mL) and layered with petroleum ether (10 mL) which afforded analytically pure, colorless crystals (59 mg, 57%) after 2 days. 1 H NMR (300 MHz, CD2Cl2): δ 7.62 (m, 4H), 7.55 (dd, 4H), 6.68 (td, 4H), 2.27 (m, 4H), 2.05(m, 4H), 1.9–1.4 (m, 16H), 1.36 (s, 36H), 1.06 (d, 12H), 0.91 (d, 12H), 0.83 (d, 12H), 0.75 (d, 12H). 13C{1H} NMR (125.6 MHz, CD2Cl2): δ 160.2, 149.2, 133.3, 130.1, 127.6, 116.0, 35.3, 33.5, 32.6, 31.5, 25.5–24.2 (m, -CH3). 31P{1H} NMR (121 MHz, CD2Cl2): δ –36.6 (s), -143.0 (sept, 730 Hz). Anal. calcd. for C72H120F6N2P6Zn2: C, 55.49; H, 7.76; N, 180. Found: C, 54.91; H, 7.56; N, 1.77. UV-vis (THF, nm(M-1cm-1)): 302 (30 500), 334 sh. Synthesis of [{(tBu2-PNP)Zn}2][OTf]2 (4.6b): A solution of 4.4 (469 mg, 0.823) in THF (15 ml) was treated with excess (>5 equiv.) potassium hydride at which time the solution turned yellow and effervescence was observed. The reaction was stirred for 3 h and filtered to remove the unreacted potassium hydride. The filtrate was added to a sealable reaction bomb. Zn(OTf)2 (300 mg, 0.823 mmol) was added as a solid and the bomb flask was sealed with a Teflon® plug. The reaction was heated to 75 ºC for 24 h at IV - 50 which time all of the Zn(OTf)2 had dissolved. The solvent was removed under reduced pressure and the residual off-white solids were washed with petroleum ether (2 x 10 mL) and dried thoroughly in vacuo. The white solids were dissolved in CH2Cl2 (8 mL), filtered through a glass micro-filter, and layered with petroleum ether (10 mL), affording analytically pure colorless crystals (422 mg, 70%) after 2 days. 1H ,13C NMR, and UVvis were as reported for 4.6a. 31 P{1H} NMR (121 MHz, CD2Cl2): δ –36.6 (s). Anal. calcd for C74H120F6N2O6P4S2Zn2: C, 56.06; H, 7.84; N, 1.82; Found: C, 56.38; H, 7.33; N, 1.95. Synthesis of [{(tBu2-PNP)Cu}{(tBu2-PNP)Zn}][PF6] (4.7): Quantities of 4.5 (89 mg, 0.064 mmol) and 4.6 (100 mg, 0.064) were combined in a 20 mL reaction vial with THF (10 mL), sealed with a Teflon cap, and stirred at 75 °C for 14 h. The pale yellow solution was evaporated to dryness under reduced pressure and the resultant pale yellow solids were washed with petroleum ether (20 mL). The solids were then redissolved in CH2Cl2, the solution filtered, and the solvent removed from the filtrate in vacuo which afforded spectroscopically pure product (173 mg, 96%). Analytically pure crystals suitable for Xray diffraction were obtained by vapor diffusion of petroleum ether into a concentrated solution of 4.7 in THF. 1 H NMR (500 MHz, CD2Cl2): δ 7.46 (m, 2H), 7.31 (m, 4H), 2.65 (m, 2H), 6.76 (m, 2H), 6.54 (m, 2H), 2.1-1.4 (m, 19 H), 1.04 (d, 12H), 0.94 (m, 5H), 0.86 (d, 6H), 0.83 (d, 6H), 0.80 (d, 6H), 0.76 (d, 6H), 0.71 (d, 6H), 0.62 (d, 6H). 13C NMR (125.7 MHz,, CD2Cl2): δ 164.6, 162.4, 145.7, 143.4, 131.3, 129.9, 129.1, 128.4, 127.2, 125.3, 117.6, 117.4, 38.7, 36.4, 35.6, 34.9, 34.7, 32.6, 31.8, 31.7, 26.4, 26.3, 26.0, 25.8, 25.6, 25.3, 25.1, 24.9. 31 P{1H} NMR (121 MHz, CD2Cl2): δ –31.8 (s, P-Cu), -40.0 (s, P-Zn), -143.0 (sept, 730 IV - 51 Hz). Anal. calcd for C72H120CuF6N2P5Zn: C, 61.26; H, 8.57; N, 1.98. Found: C, 61.48; H, 8.73; N, 1.73. UV-vis (THF, nm(M-1cm-1)): 290 (21 600), 306 (23 400), 350 sh, 396 (7400). Synthesis of [{(PNP)Cu}2][B(C6H3(CF3)2)4], 4.8: Complex 4.5 (150 mg, 0.144 mmol) and [Cp2Fe][B(C6H3(CF3)2)4] (144 mg, 0.137 mmol) were combined in a 20 mL reaction vial and dissolved in CH2Cl2 (10 mL). The reaction immediately turned redbrown and stirring was continued for 30 min at which time the solvent was removed in vacuo. The resultant solids were washed with petroleum ether (2 x 10 mL), extracted into diethyl ether (7 mL), filtered through a glass micro-filter, and layered with petroleum ether. Analytically pure red crystals (203 mg, 78%) suitable for X-ray diffraction were obtained after two successive recrystallizations from diethyl ether/petroleum ether at –35 ºC. Anal. calcd. for C88H100BCu2F24N2P4 C, 55.53; H, 5.30; N, 1.47; Found: C, 55.73; H, 5.21; N, 1.39. UV-vis (CD2Cl2, nm(M-1cm-1)): 260 (sh), 309 (14900), 359 (sh), 416 (2930), 529 (1090), 583 (sh), 661 (sh), 818 (1520), 981 (2053), 1684 (4500). Synthesis of [Cu(PNP)=(PNP)Cu][BF4]2 (4.9): NOBF4 (22.5 g, 0.192 mmol) and 4.3 (100 mg, 0.096 mmol) were combined in chlorobenzene and stirred for 18 h, after which the solution had become inky purple in color. The solution was filtered through a glass micro-filter and layered with petroleum ether. Upon standing for 4 days, analytically pure, X-ray quality crystals were obtained (86 mg, 74%). The room temperature NMR of the crystalline material is poorly resolved which is likely due to a fluxional interaction of the BF4 counterion and the 16 e¯ Cu centers. This is evident from very broad 19F NMR resonances rather than the sharp signal that is expected for an uncoordinated anion. 1 IV - 52 H NMR (300 MHz, CD2Cl2): δ 8.7 (br d), 8.51 (br d), 8.2 (br d), 7.8-7.3 (m), 6.93 (s), 2.45-1.55 (m), 1.11 (d), 1.03 (d), 0.98-0.92 (m). 19F NMR (282 MHz, CD2Cl2): δ -108.0 (br s), -125.5 (br s). 31P{1H} NMR (121 MHz, CD2Cl2): δ -27–-31 (m), -32.8 (s). Anal. calcd. for C56H86B2Cu2F8N2P4 C, 55.41; H, 7.31; N, 2.31. Found: C, 55.01; H, 7.38; N, B 2.26. UV-vis (CD2Cl2, nm(M-1cm-1)): 293 (7370), 327 (sh), 570 (sh), 613 (16250). Synthesis of [{(tBu2-PNP)Cu}2][B(C6H3(CF3)2)4] (4.10a): Complex 4.5 (134 mg, 0.106 mmol) and [Cp2Fe][B(C6H3(CF3)2)4] (111 mg, 0.106 mmol) were combined in a 20 mL reaction vial and dissolved in diethyl ether (10 mL). The reaction immediately turned red-brown and stirring was continued for 12 h after which time the solvent was removed in vacuo. The resultant solids were washed with petroleum ether (2 x 10 mL), extracted with diethyl ether (7 mL), filtered through a glass micro-filter, and the filtrate layered with petroleum ether. Analytically pure red crystals (226 mg, 88%) suitable for X-ray diffraction were obtained after two days at –35 ºC. 19 F{1H} NMR (282 MHz, CD2Cl2): δ -60.2. Anal. Calcd. for C104H132BCu2F24N2P4: C, 58.70; H, 6.25; N, 1.32. Found: C, 58.34; H, 6.14; N, 1.41. Synthesis of [{(tBu2-PNP)Cu}2][SbF6] (4.10b): Complex 4.5 (165 mg, 0.131 mmol) and AgSbF6 (44.8 mg, 0.131 mmol) were combined in a 20 mL reaction vial and dissolved in THF (10 mL). The reaction immediately turned red-brown and stirring was continued for 12 h, after which time the solution was filtered through a glass micro-filter to remove the Ag0 byproduct. The solvent was removed from the filtrate under reduced pressure, and the resultant red solids extracted into CH2Cl2 (4 mL), filtered and layered with petroleum ether. After two days, analytically pure red crystals (159 mg, 81%) were obtained. Anal. calcd. for C72H120Cu2F6N2P4Sb C, 57.63; H, 8.06; N, 1.87; Found: C, IV - 53 57.64; H, 8.03; N, 2.17. Evans’ Method (CD2Cl2, 298 K): 1.63 BM. UV-vis (CD2Cl2, nm(M-1cm-1)): 260 (sh), 312 (22 360), 377 (sh), 430 (3810), 510 (2280), 586 (sh), 639 (sh), 804 (1510), 941 (940), 1788 (5490). Synthesis of [{(tBu2-PNP)Cu}2][SbF6]2 (4.11): Complex 4.5 (100 mg, 0.079 mmol) and NOSbF6 (42 mg, 0.18 mmol) were combined in a 20 mL reaction vial and dissolved in CH2Cl2 (10 mL). Upon dissolution, solution immediately became red-brown in color then became blue after 10 min. After an additional 3 h of stirring the solution was concentrated to 3 mL, filtered through a glass micro-filter and layered with petroleum ether (15 ml). Cooling the mixture to -35 ºC for 4 days caused a deep blue solid to precipitate. Two successive recrystallizations of 4.11 from petroleum ether/CH2Cl2 at -35 ºC afforded an analytically pure, blue microcrystalline solid (65 mg, 47%) suitable for X-ray diffraction studies. Modest yields result from multiple recrystallizations to insure purity and are not reflective of a low conversion to product. Anal. calcd. for C72H120Cu2F12N2P4Sb2: C, 49.81; H, 6.97; N, 1.61; Found: C, 49.41; H, 6.96; N, 1.76. Magnetic susceptibility (CD2Cl2, 298 K): 1.69 BM. UV-vis (CD2Cl2, nm(M-1cm-1)): 294 (16 500), 321 (sh), 346 (21700), 426 (sh), 510 (sh), 574 (sh), 620 (12 900), 1297 (300). IV - 54 IV.E. References Cited 1 (a) Kato, M., Johnassen, H. B.; Fanning, J. C.; Chem. Rev. 1964, 64, 99. (b) Kato, M.; Muto, Y.; Coord. Chem Rev. 1988, 92, 45. (c) Kahn, O.; Angew. Chem. Int. Ed. Eng. 1985, 24, 834. 2 (a) Bioinorganic Chemistry of Copper; Karlin, K. D.; Tyeklar, Z.; Eds.; Chapman & Hall: New York, N.Y., 1993. (b) Bioinorganic Chemistry; Bertini, I.; Gray, H. B.; Lippard, S. J.; Valentine, J. S. University Science Books: Mill Valley, CA, 1994. 3 (a) Elliott, C. M.; Pichot, F.; Bloom, C. J.; Rider, L. S. J. Am. Chem. Soc. 1998, 120, 6781. (b) Bargossi, C.; Fiorini, M. C.; Montalti, M.; Prodi, L.; Zaccheroni, N. Coord. Chem. Rev. 2000, 208, 17. (c) de Silva, A. P.; Fox, D. B.; Huxley, A. J. M.; Moody, T. S. Coord. Chem. Rev. 2000, 205, 41. (d) Drummond, T. G.; Hill, M. G.; Barton, J. K. Nat. Biotech. 2003, 21, 1192. (e) Sutin, N.; Creutz, C. Pure & Appl. Chem. 1980, 52, 2717. 4 (a)Vallee, B. L.; Williams, R. J. P. Proc. Natl. Acad. Sci. U.S.A. 1968, 59, 498. (b) Williams, R. J. P. Inorg. Chim. Acta Rev. 1971, 5, 137. (c) Williams, R. J. P. Eur. J. Biochem. 1995, 234, 363. (d) Hathaway, B. J.; Billing, D. E. Coord. Chem. Rev. 1970, 5, 143. (e) Hathaway, B. J. Coord. Chem. Rev. 1981, 35, 211. 5 (a) Chen, P.; Gorelsky, S. I.; Ghosh, S.; Solomon, E. I.; Angew. Chem. Int. Ed. 2004, 43, 4132. (b) Rosenzweig, A. C. Nat. Struct. Biol. 2000, 7, 169. (c) Zumft, W.G. Microbiol. Mol. Biol. Rev. 1997, 61, 533. 6 (a) Lieberman, R. L.; Rosenzweig, A. C. Nature 2005, 434, 177. (b) Lieberman, R. L.; Shrestha, D. B.; Doan, P. E.; Hoffman, B. M.; Stemmler, T. L.; Rosenzweig, A. C. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 3820. IV - 55 7 (a) Ferguson-Miller, S.; Babcock, G. T. Chem. Rev. 1996, 96, 2889. (b) Gamelin, D. R.; Randall, D. W.; Hay, M. T.; Houser, R. P.; Mulder, T. C.; Canters, G. W.; de Vries, S.; Tolman, W. B.; Lu, Y.; Solomon, E. I. J. Am. Chem. Soc. 1998, 120, 5246. (c) George, S. D.; Metz, M.; Szilagyi, R. K.; Wang, H. X.; Cramer, S. P.; Lu, Y.; Tolman, W. B.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 2001, 123, 5757. (d) Ramirez, B. E.; Malmstrom, B. G.; Winkler, J. R.; Gray, H. B. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 11949. 8 (a) Kroneck, P. M. H.; Antholine, W. E.; Riester, J.; Zumft, W. G. FEBS Lett. 1989, 248, 212. (b) Kroneck, P. M. H.; Antholine, W. E.; Riester, J.; Zumft, W. G. FEBS Lett. 1988, 242, 70. (c) Antholine, W. E.; Kastrau, D. H. W.; Steffens, G. C. M.; Buse, G.; Sumft, W. G.; Kroneck, P. M. H. Eur. J. Biochem. 1992, 209, 875. 9 (a) Houser, R. P.; Young, V. G., Jr.; Tolman, W. B. J. Am. Chem. Soc., 1996, 118, 2101. (b) Blackburn, N. J.; deVries, S.; Barr, M. E.; Houser, R. P.; Tolman, W. B.; Sanders, D.; Fee, J. A. J. Am. Chem. Soc. 1997, 119, 6135. (c) Hagadorn, J. R.; Zahn, T. I.; Que, L., Jr.; Tolman, W. B. J. C. S. Dalton Trans. 2003, 1790. 10 Al-Obaidi, A.; Baranovič, G.; Coyle, J.; Coates, C. G.; McGarvey, J. J.; McKee, V.; Nelson, J. Inorg. Chem. 1998, 37, 3567. 11 (a) LeCloux, D. D.; Davydov, R.; Lippard, S. J. Inorg. Chem. 1998, 37, 6814. (b) LeCloux, D. D.; Davydov, R.; Lippard, S. J. J. Am. Chem. Soc. 1998, 120, 6810. (c) He, C.; Lippard, S. J. Inorg. Chem. 2000, 39, 5225. 12 Gupta, R.; Zhang, Z. H.; Powell, D.; Hendrich, M. P., Borovik, A. S. Inorg. Chem., 2002, 41, 5100. IV - 56 13 Snyder, B. S.; Patterson, G. S.; Abrahamson, A. J.; Holm, R. H.; J. Am. Chem. Soc. 1989, 111, 5214. 14 Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 2030. 15 (a) Glazer, E. C.; Magde, D.; Tor, Y.; J. Am. Chem. Soc. 2005, 127, 4190. (b) Browne, W. R.; O'Boyle, N. M.; Henry, W.; Guckian, A. L.; Horn, S.; Fett, T.; O'Connor, C. M.; Duati, M.; De Cola, L.; Coates, C. G.; Ronayne, K. L.; McGarvey, J. J.; Vos, J. G.; J. Am. Chem. Soc. 2005, 127, 1229. (c) Zhou, Z.; Fahrni, C. J.; J. Am. Chem. Soc. 2004, 126, 8862. 16 Ford, P. C.; Cariati, E.; Bourassa, J. Chem. Rev. 1999, 99, 3625. 17 McMillin, D. R.; McNett, K. M. Chem. Rev. 1998, 98, 1201. 18 Roundhill, D. M. Photochemistry and Photophysics of Metal Complexes; Fackler, J. P., Jr.; In Modern Inorganic Chemistry; Plenum Press: New York, NY, 1994; pp 49 & 165. 19 Cuttell, D. G.; Kuang, S.-M.; Fanwick, P. E. McMillin, D. R.; Walton, R. A. J. Am. Chem. Soc. 2002, 124, 6. 20 Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 2885. 21 (a) Robin, M. B.; Day, P. Adv. Chem. Phys., 1967, 10, 247. (b) Day, P. Endeavour 1970, 29, 45. 22 {(SNS)CuI}2: CCDC 234450; [{(SNS)Cu1.5}2]+: CCDC 234451; Cambridge Structural Database; Cambridge, England (accessed May 28, 2005). 23 Peters, J. C.; Harkins, S. B.; Brown, S. D.; Day, M. W. Inorg. Chem. 2001, 40, 5083. 24 Liang, L.-C.; Lin, J.-M.; Hung, C.-H. Organometallics 2003, 15, 3007. 25 Ozerov, O. V.; Guo, C.; Papkov, V. A.; Foxman, B. M.; J. Am. Chem. Soc. 2004, 126, 4792. IV - 57 26 Tsuda, T.; Yazawa, T.; Watanabe, K.; Fuji, T.; Saegusa, T. J. Org. Chem. 1981, 46, 192 27 Drago, R. S. Physical Methods for Chemists, 2nd ed.; Surfside Scientific Publishers: Gainesville, FL, 1992. 28 Shannon, R. D.; Acta. Cryst. 1976, A32, 751. 29 Che, C.-M.; Mao, Z.; Miskowski, V. M.; Tse, M.-C.; Chan, C.-K.; Cheung, K.-K.; Philips, D. L.; Leug, K.-H. Angew. Chem. Int. Ed. 2000, 39, 4084. 30 Parameters for electrochemical experiments: Pt working electrode, AgNO3/Ag non- aqueous reference, Pt wire auxillary, 0.35 M tetrabutylammonium hexafluoroborate in THF, 200 mV/sec. It was necessary to switch to a Pt working electrode from glassy carbon to attenuate the formation of surface adsorbed species due to the highly non-polar nature of the analyte. 31 Smith, M. B.; March, J. March’s Advanced Organic Chemistry, 5th Ed. John Wiley & Sons, Inc.: New York, NY, 2001. 32 Complexes 4.9 and 4.11 can also prepared by the one electron oxidation of 4.8 or 4.10b, however, we find the two electron oxidation route from 4.3 or 4.5 to be preferable. 33 THF was found to be suitable for electrochemistry which is attributed to modest stability of the dimeric, doubly oxidized product of 4.3 prior to decomposition to 4.9. 34 Seidal, B.; Hammond, G. S.; J. Org. Chem. 1963, 28, 3280. 35 (a) Chaudhuri, P.; Hess, M.; Müller, J.; Hildenbrand, K.; Bill, E.; Weyhermüller, T.; Wieghardt, K. J. Am. Chem. Soc. 1999, 121, 9599. (b) Shultz, D. A; Vostrikova, K. E.; Bodnar, S. H.; Koo, H.-J.; Whangbo, M.-H.; Kirk, M. L.; Depperman, E. C.; Kampf, J. W. J. Am. Chem. Soc. 2003, 125, 1607. (c) Bill, E.; Müller, J.; Weyhermüller, T.; IV - 58 Wieghardt, K. Inorg. Chem.; 1999, 38, 5795. (d) Min, K. S.; Weyhermüller, T.; Bothe, E.; Wieghardt, K. Inorg. Chem.; 2004, 43, 5795. 36 Harkins, S. B.; Peters, J. C.; Facial Coordination of a Pincer-like Amido Complex of Platinum(IV) Generated by Photoisomerization. Inorg. Chem. to be submitted for publication. 37 The structural inequivalency between the copper centers was also found in the crystal structure of the bis-[PF6]¯ salt of 4.11. Unpublished results. 38 Discreet mononuclear CuII complexes supported by ligands 4.1 or 4.4 have not been prepared because of the reducing nature of the phosphine ligands. However, characteristic four-line hyperfine splitting patterns characteristic of a CuII are easily discernable in the previously described (SNS)CuIICl complex. 39 Magnetochemistry; Carlin, R. L.; Springer-Verlag: Heidelberg, Germany, 1986. 40 Fernandez, C. O.; Cricco, J. A., Slutter, C. E.; Richards, J. H.; Gray, H. B.; Vila, A. J. J. Am. Chem. Soc.; 2001, 123, 11678. 41 (a) Sur, S. K. J. Magn. Reson. 1989, 82, 169. (b) Evans, D. F. J. Chem. Soc. 1959, 2003. 42 Electron Paramagnetic Resonance : Elementary Theory and Practical Applications, Weil, J. A.; Bolton, J. R.; Wertz, J. E.; Wiley Interscience: New York, N.Y., 1994. 43 Molecular Magnetism, Kahn, O.; VCH Publishers, Inc.: New York, N.Y., 1993. 44 Electron Paramagnetic Resonance of Exchange Coupled Systems, Bencini, A.; Gatteschi, D.; Springer-Verlag: Berlin, Germany, 1990. 45 Ribas, X.; Jackson, D. A.; Donnadieu, B.; Mahia, J.; Parella, T.; Xifra, R.; Hedman, B.; Hodgson, K. O.; Llobet, A.; Stack, T. D. P. Angew. Chem., Int. Ed. 2002, 41, 2991. IV - 59 46 (a)Whittaker, J. W. Chem. Rev. 2003, 103, 2347. (b) DuBois, J. L.; Mukherjee, P.; Stack, T.D.P.; Hedman, B.; Solomon, E. I.; Hodgson, K. O. J. Am. Chem. Soc. 2000, 122, 5775. 47 Krebs, C.; Glaser, T.; Bill., E.; Weyhermüller, T.; Meyer-Klaucke, W.; Wieghardt, K.; Angew. Chem. Int. Ed. 1999, 38, 359. 48 a) Klemm, W.; Huss, E.; Z. Anorg. Chem. 1949, 259, 221.; b) Hoppe, R.; Angew. Chem. 1950, 62, 339. 49 (a) Gerloch, M.; Slade, R.C.; J. Chem. Soc. A. 1969, 1012. (b) Advanced Inorganic Chemistry, Cotton, F. A.; Wilkinson, G.; John Wiley & Sons, New York, N.Y., 1988. 50 Bates, C. A.; Proc. Phys. Soc., 1962, 79, 69. 51 Neese, F.; J. Chem. Phys. 2005, 122, 34107. 52 Perich, J. W.; Johns, R. B. Synthesis 1988, 2, 142. 53 Zhao, H.; Tanjutco, C.; Thayumanavan, S. Tet. Lett. 2001, 42, 4421. 54 Chảvez, I.; et al. J. Organomet. Chem. 2000, 601, 126. 55 O’Connor, C. J.; Prog. Inorg. Chem. 1982, 29, 203. 56 Wenger, O. S.; Henling, L. M.; Day, M. W.; Winkler, J. R.; Gray, H. B.; Inorg. Chem., 2004, 43, 2043. 57 Lyubovitsky, J. G.; Gray, H. B.; Winkler, J. R.; J. Am. Chem. Soc. 2002, 124, 5481. 58 Lee, J. C.; Engman, K. C.; Tezcan, F. A.; Gray, H. B.;Winkler, J. R.; Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 14778. 59 Demas, J. N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991. 60 Rabinowitz, R.; Pellon, J.; J. Org. Chem. 1961, 26, 1961.  The Synthesis and Study of Redox-Rich, AmidoBridged Cu2N2 Dicopper Complexes Thesis by Seth Beebe Harkins In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 2006 (Defended June 27, 2005) ii © 2006 Seth Beebe Harkins All Rights Reserved iii Acknowledgment First and foremost, I must thank my parents and family, for they have taught me more life lessons than I can count and have given me unconditional love and support. In particular, I thank my father for teaching me the life lesson that “A job worth doing, is a job worth doing right.” As trite as it may seem, it is a lesson that has transcended to my chemical endeavors and has encouraged excellence in my scientific pursuits. To my mother, my first instructor in the chemical sciences, I thank you for your candor and insistence that no matter what you do in life, to make sure that it is what makes you most happy. With this paradigm in mind, it is my pleasure to state that I am still elated every day I have the opportunity to walk into the laboratory and make something new. I thank Professor Ayusman Sen for giving me the opportunity to work in his lab as an undergraduate. It was this chance to be involved with chemistry at the forefront of innovation that made it clear to me that inorganic chemistry was to be my field of choice. I thank him for encouraging me, in the face of adversity, to purse graduate study at the very best institution possible. From the Sen Lab, I would also like to thank then graduate students, Dr. Jim Pawlow, Dr. Shahid Murtuza, and Dr. April D. Hennis Marchetti, for putting up with me as an eager undergraduate researcher and always helping me in spite of the sacrifice to their own productivity. I would also like to acknowledge John D. Higgins, III for being a role model to me in what I wanted from my life as a synthetic chemist. I have never met anyone who made it so cool to be a nerd as John. I wish to acknowledge my fellow “idiot troops” in the Peters Lab for adding levity to the research environment, happily serving as a sounding board for my chemical ramblings, and providing lots of well-needed proofreading. In particular, I would like to iv thank the irreplaceable combination of Dr. Steven D. Brown and Dr. Theodore A. Betley for many a diversion from work, endless reality checks, and their essential thoughts on chemistry. I have also really appreciated my box-mate, Dr. Cora E. MacBeth, for constant conversation, input on my project, “decorating” the ceiling of the dry-box, and not getting too irritated with me when I destroy the atmosphere with methylene chloride. With the uttermost sincerity, I express my gratitude to my adviser Professor Jonas C. Peters for giving me the latitude to always chase my ideas, even when he didn’t quite know why I was doing it at the time. He had the foresight to create a near perfect laboratory environment for inorganic synthesis and then to stock it with a unique blend of graduate students and post-docs that produced science worthy of his tenure. I am happy to have been part of the legacy. Moreover (that one is for you Jonas), I would like to thank my committee, Professors John D. Bercaw, Harry B. Gray, and David A. Tirrell for critical evaluation and suggestions for my project, as well as thoughtful discussions of my propositions. I would like to acknowledge the Institute, the Division of Chemistry and Chemical Engineering, and the supporting cast of faculty, scientists, and staff members, which is not limited to Bruce Brunschwig, Michael Day, Angelo Di Bilio, Dian Buchness, Rick Gerhart, Kathleen Hand, Larry Henling, Jay R. Winkler, and the members of the Gray and Bercaw research groups. Finally, I thank my loving wife Jennifer for giving me a reason to go home every night. You are the love of my life and you have made me a better person and a better scientist over the past five years. I also thank you for your tireless assistance with the numerous laser experiments presented herein; without you they would not have been possible. I look forward to starting the next stage of my life with you. v Abstract A Cu2N2 diamond core structure supported by an [SNS]¯ ligand exhibits a fully reversible one-electron redox process between a reduced CuICuI, {(SNS)Cu}2, and a class III delocalized Cu1.5Cu1.5 state, [{(SNS)Cu}2][B(C6H3(CF3)2)4] ([SNS] ¯ = bis(2- t-butylsulfanylphenyl)amide). The Cu⋅⋅⋅Cu distance compresses appreciably (~0.13 Å) upon oxidation; a metal-metal distance of 2.4724(4) Å is observed in the mixed-valence molecule that is nearly identical to the dicopper CuA site found in cytochrome c oxidase. The rate of electron self-exchange (ks) between the CuICuI and the Cu1.5Cu1.5 complexes was estimated to be ≥ 107 M−1s−1 by 1H NMR line-broadening analysis. The unusually large magnitude of ks reflects the minimal structural reorganization that accompanies CuICuI ↔ Cu1.5Cu1.5 interchange. A second generation of {(PNP)CuI}2 dimer supported by a [PNP]¯ ligand also has been investigated ([PNP]¯ = bis(2-(diisobutylphosphino)phenyl)amide). The highly emissive {(PNP)CuI}2 is characterized by a long-lived excited state (τ > 10 μs) with an unusually high quantum yield (φ > 0.65) at ambient temperature. Removal of an electron from the {(PNP)CuI}2 dimer yields a nearly isostructural, Cu1.5Cu1.5 complex [{(PNP)Cu}2][B(C6H3(CF3)2)4]. With a highly reducing excited state reduction potential (~ −3.2 V vs. Fc+/Fc) as well as the availability of two reversible redox processes, these bimetallic copper systems may be interesting candidates for photochemically driven twoelectron redox transformations. Studies of Cu2N2 diamond core complexes supported by the [tBu2-PNP]¯ ligand revealed that the dicopper complex {(tBu2-PNP)Cu}2 can not only be oxidized by one electron to [{(tBu2-PNP)Cu1.5}2][B(C6H3(CF3)2)4], but also by two-electrons to [{(tBu2- PNP)Cu}2][SbF6]2 t ([ Bu2-PNP]¯ = vi bis(2-diisobutylphoshino-4-tbutylphenyl)amide). These Cu2N2 complexes show remarkably low structural reorganization for all oxidation states as evidenced by the solid-state molecular-structures. Based on these studies of [{(tBu2-PNP)Cu}2][SbF6]2, we propose a formulation of one CuI and one paramagnetic CuIII nuclei in compressed-tetrahedral environments in the Cu2N2 core. Spectroscopic, redox, and magnetic data are consistent with a highly covalent M2N2 core supported by a rigid ligand scaffold. These complexes are excellent mimics of the entatic state found in bimetallic copper proteins. vii Table of Contents Acknowledgements.................................................................................................... iii Abstract...................................................................................................................... v Table of Contents....................................................................................................... vii List of Tables and Figure........................................................................................... ix Nomenclature............................................................................................................. xii Chapter I. Late Transition Metal Complexes of Chelating Amido Ligands I.A. Thesis Overview and Motivation ..........................................................I - 2 I.B. Summary of Group 10 Metal Studies.................................................... I - 8 I.C. References Cited.................................................................................... I - 14 Chapter II. Electrochemical Studies of Dicopper Complexes Supported by Bis(2-tertbutylsulfanylphenyl)amido Ligands II.A. Introduction.......................................................................................... II - 2 II.B. Results and Discussion......................................................................... II - 4 II.C. Experimental......................................................................................... II - 19 II.D. References Cited...................................................................................II - 27 Chapter III. Luminescence Studies of Dicopper(I) Complexes Supported by Bis(2-diisobutylphosphinophenyl)amido Ligands III.A. Introduction......................................................................................... III - 2 III.B. Results and Discussion........................................................................ III - 3 III.C. Experimental....................................................................................... III - 12 III.D. References Cited................................................................................. III - 20 viii Chapter IV. Redox Behavior of Copper and Zinc Dimers Supported by Bis(2-diisobutylphoshino-4-t-butylphenyl)amido Ligands IV.A. Introduction......................................................................................... IV - 2 IV.B. Results and Discussion ...................................................................... IV - 6 IV.B.1. Diamagnetic Complexes 4.1 – 4.7........................................ IV - 6 IV.B.2. Paramagnetic Complexes 4.8 – 4.11..................................... IV - 19 IV.C. Summary and Future Directions......................................................... IV - 37 IV.D. Experimental....................................................................................... IV - 39 IV.E. References Cited................................................................................ IV - 54 Appendix A. Facial Coordination of a Pincer-Like Amido Complex of Platinum(IV) Generated by Photoisomerization Introduction.................................................................................................... A - 2 Results and Discussion.................................................................................. A - 3 Experimental.................................................................................................. A - 9 References Cited............................................................................................ A - 12 Appendix B. Undiscussed Solid-State Molecular Structures ix List of Figures, Graphs, and Tables Chapter I Figure I.1. Generalized M2N2 diamond-core complex.............................................. I - 2 Figure I.2. Schematic of cytochrome c oxidase......................................................... I - 4 Figure I.3. Schematic summary of Cu2N2 core contraction....................................... I - 5 Figure I.4. Schematic representation of {(PNP)CuI}2............................................... I - 6 Figure I.5. Generalized ligand synthesis via Pd0 catalyzed aryl amination............... I - 9 Figure I.6. Displacement ellipsoid (50%) representation of [Li][BQA].................... I - 10 Figure I.7. C-H bond activation................................................................................. I - 12 Chapter II Figure II.1. Molecular representations of 2.2 and 2.3................................................ II - 6 Figure II.2. Solid state molecular structure of 2.4..................................................... II - 7 Figure II.3. X-band EPR spectrum (9.40 GHz) of 2.4............................................... II - 7 Figure II.4. Cyclic voltammetry of 2.2...................................................................... II - 9 Figure II.5. Electronic absorption spectrum of 2.2, 2.3, and 2.4.............................. II - 10 Figure II.6. DFT minimized structure and contour plot of 2.3.................................. II - 12 Figure II.7. X-band EPR spectrum (9.38 GHz) of 2.3............................................... II - 14 Graph II.1. Plots of line width and mole fraction vs. chemical shift......................... II - 15 Graph II.2. SQUID magnetization study of 2.3......................................................... II - 21 Scheme II.1................................................................................................................ II - 4 Table II.1. X-ray diffraction data for 2.2, 2.3, and 2.4.............................................. II - 20 Table II.2. Line-broadening data, concentrations, and mole fractions...................... II - 22 x Chapter III Figure III.1. Molecular and space filling representation of 3.2................................. III - 3 Figure III.2. Synthesis of 3.1, {[3.1][Li]}2, and 3.2................................................... III - 4 Figure III.3. Cyclic voltammetry of 3.2..................................................................... III - 5 Figure III.4. Absorption, emission, and excitation spectrum of 3.2.......................... III - 6 Figure III.5. Low temperature emission of 3.2......................................................... III - 7 Figure III.6. Geometry optimization and electronic structure calculation for 3.2.... III - 10 Figure III.7. Diagram of the photophysical and redox properties of 3.2.................. III - 11 Graph III.1. Time-resolved emission quenching of 3.2............................................ III - 8 Graph III.2. Fit of the excited state decay with residuals.......................................... III - 18 Table III.1. Comparison of DFT and X-ray structural parameters for 3.2................III - 9 Table III.2. Data for quantum yield measurements................................................... III - 17 Table III.3. Data for excited state lifetime measurements......................................... III - 18 Table III.4. Time-resolved emission quenching measurements................................ III - 19 Chapter IV Figure IV.1. Representative dimer of M2N2 complexes............................................ IV - 6 Figure IV.2. Synthesis of complexes 4.2, 4.3, 4.5, 4.6, and 4.7................................ IV - 7 Figure IV.3. Absorption spectra................................................................................IV - 15 Figure IV.4. Cyclic voltammetry............................................................................... IV - 16 Figure IV.5. Emission spectrum............................................................................... IV - 19 Figure IV.6. Decomposition of 4.3 upon two electron oxidation.............................. IV - 22 Figure IV.7. EPR spectrum (X-band) of 4.8 and 4.10b ........................................... IV - 27 Figure IV.8. EPR spectrum (X-band) of 4.11........................................................... IV - 28 xi Figure IV.9. Magnetic susceptibility study of 4.11................................................... IV - 30 Figure IV.10. Absorption spectra of 4.8, 4.10b, and 4.11......................................... IV - 31 Figure IV.11. Possible resonance contributors to the electronic structure of 11....... IV - 33 Table IV.1. Core structural representations for 4.2, 4.3, 4.5, 4.6, and 4.7................ IV - 11 Table IV.2. Core structural representations for 4.8, 4.10a, and 4.11........................ IV - 24 Table IV.3. Absorption bands determined by Gaussian fitting of the spectra........... IV - 32 Table IV.4. Data for excited state lifetime measurements......................................... IV - 42 Table IV.5. Data for quantum yield measurements................................................... IV - 43 Appendix A Figure 1. Synthesis of 2, 3, and 4............................................................................... A - 3 Figure 2. Solid-state molecular structures (50% ellipsoids) for 2, 3, and 4............... A - 4 Figure 3. Absorption spectra of 2, 3 and 4 in CH2Cl2................................................A - 6 Table 1. Bond lengths and angles for 2, 3, and 4....................................................... A - 4 xii Nomenclature 1 { H} hydrogen-1 decoupled ° degrees in measure of angles °C degrees Celsius 1 H 13 C 19 F 31 hydrogen-1 carbon-13 fluorine-19 P phosphorus-31 Å Angstrom, 10 Anal. Calcd elemental analysis calculated Ar general aryl group av average Ax EPR hyperfine coupling where X is the nucleus coupling to the unpaired electron (also sometimes abbreviated Ax) BArF tetrakis(3,5-trifluoromethylphenyl)borate B3LYP Becke three-parameter functional with Lee-YangParr correlation functional BM Bohr magnetons br broad BQA bis(8-quinolinyl)amido Bu butyl Calcd calculated CCD charge coupled device -10 m cm -1 cm 3 xiii centimeter(s) inverse centimeters or wavenumbers cm cubic centimeters cont. continued d doublet DC direct current deg degrees in measure of angles DFT density functional theory E an atom or functional group forming a metal-ligand multiple bond EPR electron paramagnetic resonance Eq. equation equiv. equivalents ESI/MS electrospray ionization mass spectrometry Et ethyl fac facial coordination g gram G gauss GC/MS gas chromatography mass spectrometry GHz gigahertz g isotropic g-factor h hour(s) H applied magnetic field HOMO highest occupied molecular orbital iso xiv Hz hertz I nuclear spin of atom n i n Pr iso-propyl IR infrared K degrees in Kelvin Kcal kilocalories kHz kilohertz L dative ligand for a transition metal LACVP Los Alamos core valence potential LFT ligand field theory LUMO lowest unoccupied molecular orbital m multiplet M general metal Me methyl Mes mesityl mg milligram(s) MHz megahertz, one million Hertz min minute(s) mL milliliter(s) mmol millimoles MO molecular orbital mol moles ms millisecond(s) MS xv mass spectrometry mT millitesla(s) mV millivolt(s) mW milliwatt(s) NA not applicable n Bu n-butyl near-IR near-infrared nm namometer(s) NMR nuclear magnetic resonance OTf trifluoromethanesoulphonate p- para position on an aryl ring Ph phenyl PMe3 trimethyl phosphine PNP bis(2-(diisobutylphosphino)phenyl)amide t Bu2-PNP bis(2-diisobutylphoshino-4-tbutylphenyl)amide ppm parts per million q quartet R general alkyl or aryl substituent s second(s) S spin SNS bis(2-tertbutylsulfanylphenyl)amide SOMO singly occupied molecular orbital SQUID superconducting quantum interference device T xvi temperature TBA tetrabutylammonium t Bu tert-butyl THF tetrahydrofuran TMS trimethylsilyl UV-vis ultraviolet-visible X monoanionic atom or group, such as a halide or thiolate XRD X-ray diffraction δ delta, chemical shift ε extinction coefficient in M cm λ wavelength -1 λ max -1 wavelength of maximum absorption µ absorption coefficient (X-ray diffraction) µ-A bridging atom µ Bohr magnetons µ effective magnetic moment, measured in Bohr B eff magnetons µL microliter(s) ν frequency θ Weiss constant χ magnetic susceptibility χ m molar magnetic susceptibility I-1 Chapter I Late Transition Metal Complexes of Chelating Amido Ligands† † Text taken in part from Peters, J. C.; Harkins, S. B.; Brown, S. D.; Day, M. W. Inorg. Chem. 2001, 40, 5083. Harkins, S. B.; Peters, J. C. Organometallics 2002, 21, 1753. Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 2885. Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 2030. I-2 I.A. Thesis Overview and Motivation Here is presented the first preparation of amido-bridged Cu2N2 dicopper coordination complexes and the subsequent study of their rich spectroscopic and redox properties. These unique metal dimers are supported by chelating diaryl amido ligands that give rise to M2N2 diamond core configurations with overall D2 symmetry (Figure I.1). As a result of this thesis work, we have demonstrated that these ligand scaffolds allow for the metal centers to reside in a protein like “entatic state,” which facilitates low structural reorganization upon oxidation or photoexcitation of the d10d10 metal complexes. Vallee and Williams have described the entatic state as the constrained environment of an enzymatic active site which allows for specific activity, while minimizing the reorganization energy needed to reach the transition state. 1 This characteristic gives rise to a multitude of interesting consequences which include rapid inter- and intra-molecular electron transfer phenomena, intense photoluminescence, and stability in multiple oxidation states. All of these processes occur without changes to the coordination number or structural distortion of the overall molecule. As a whole, these advances have added a new dimension to the field of dicopper chemistry and laid the ground work for many further studies in the general area of late transition metal chemistry supported by amide ligands. L L N M M N L L Figure I.1. Generalized M2N2 diamond-core complex. {M = Li, Cu, Zn; L = N, O, P, S}. I-3 Dicopper electron transfer (ET) cofactors in biology are now well established, yet the topic of how the site is able to minimize structural reorganization and maintain its coordination geometry during this process is still of interest. 2, 3 , 4 For example, a collection of spectroscopic and, more recently, structural data has revealed that cytochrome c oxidase accomplishes enhanced ET rates into and out of a buried CuA site by virtue of a highly covalent, thiolate-bridged Cu2S2 diamond-core structure (Figure I.2). 5, 6 , 7 Amongst the various low molecular weight systems that have been developed to model aspects of this CuA site, 8, 9 , 10 , 11 perhaps the most structurally relevant to the enzyme is Tolman’s thiolate-bridged, mixed-valence {LN3SCu1.5}2+ diamond-core complex.8a This complex features an EPR signal consistent with a fully-delocalized, class III, diamond- core description,5b,8a as is observed for the resting state form of the CuA site.5b,6 Despite this similarity, the Cu···Cu distance in the Tolman model system is much larger than that observed in CuA, and the complex does not reproduce the reversible CuICuI to Cu1.5Cu1.5 redox behavior observed for the enzyme. Several dicopper systems that diverge from the diamond core structural motif do show reversible redox behavior for a Cu1.5Cu1.5 mixed-valence state.10c,12 For example, a family of dicopper azacryptates that successfully bring the two copper centers into much closer proximity (~ 2.35–2.4Å) has been developed,12a and many of these can undergo rapid electron self-exchange (ks) between their oxidized and reduced forms (~ 105 M-1 s-1).12b I-4 S HN N Cu S O Cu N NH S Me Cu Cu = ~ 2.5 Å Figure I.2. Left, Schematic of cytochrome c oxidase from Thermus thermophilius ba3;7 right, Core representation of the CuA site. Diamond core copper systems that reversibly model both the reduced CuICuI and the delocalized Cu1.5Cu1.5 redox states observed in CuA had yet to be developed. 13 Ligand systems that would help to minimize structural reorganization between these two redox forms and hold the two copper centers of a diamond core in close enough proximity to reproduce the short Cu-Cu bond distance observed in CuA (~2.5 Å) 14 might serve as simplified functional models for the redox behavior of CuA.5b,c,d One avenue into this regime using small molecule design is to replace the thiolate bridging units inherent to the Cu2S2 diamond core of CuA with amido bridging units. Such a strategy should slide the copper centers closer together so as to better model the distance observed in CuA while still maintaining a Cu2X2 diamond core motif. Moreover, it was thought that such an approach might shed some light on whether the thiolate moiety plays a unique role in facilitating rapid electron transfer chemistry by bringing the two copper centers of a diamond core into strong electronic communication. I-5 In Chapter II, this topic has been addressed through the synthesis and study of a Cu2N2 diamond core system supported by the tridentate amido ligand, bis(2tertbutylsulfanylphenyl)amido (Figure II.3). These dicopper systems were found to exhibit a fully reversible and very facile one electron redox process between a Cu1Cu1 and a class III delocalized Cu1.5Cu1.5 form. Solid-state structural snapshots of both redox forms reveal (i) very short Cu···Cu distances akin to those in CuA, and (ii) minimal structural reorganization upon oxidation from the reduced Cu1Cu1 to the delocalized Cu1.5Cu1.5 state. These features facilitate very facile rates of electron self-exchange between the reduced and one electron oxidized forms of the system (≥ 107 M-1 s-1). To our knowledge, this is the fastest electron self-exchange rate reported for low molecular weight copper complexes.3a,12b A distinct Cu···Cu compression accompanies the one electron oxidation process and may reflect the onset of direct Cu-Cu electronic exchange. Using cyclic voltammetry, this reversible one electron oxidation was found to occur at -250 mV vs. Fc+/Fc in THF. A marginally-reversible wave at higher potential (Epa ≈ +570 mV), was also observed. We speculated this is an unstable dicopper(II,II) species. mixed-valence t Bu t S S S Cu N N N S t t Bu S S S Cu Cu compression Cu Cu N N S t t Cu1 Cu1 2.5989(3) Å Bu Bu Bu Bu S Bu Cu S t Cu1.5 Cu1.5 2.4724(4) Å t Bu Diamond core structures display minimal reorganization upon reversible 1-electron redox process Figure I.3. Schematic summary of Cu2N2 core contraction upon oxidation. I-6 Optimistic at the possibility of obtaining a system that could mediate two sequential one-electron oxidations with nominal change to the coordination sphere, we began to explore the effects of ligand modification. It was conjectured that by switching from thioether to more reducing alkyl phosphine chelates, an electron-rich Cu2N2 complexes would result. By shifting the dicopper redox potential more negative, it was thought that perhaps both oxidation processes would be stabilized. This was accomplished by using bis(2-diisobutylphosphinophenyl)amido, [PNP]¯, ligands. Similar Cu2N2 complexes were afforded (Figure I.4) with this new ligand and now two reversible one electron processes were observed by cyclic voltammetry. Surprisingly, we found that {(PNP)Cu}2 complex was an exceptional luminophore (Figure I.4). The photophysical properties of this complex are the topic of Chapter III. Notably, {(PNP)CuI}2 is characterized by a longlived excited state (τ > 10 μs) with an unusually high quantum yield (φ > 0.65) at ambient temperature. Figure I.4. Left, Schematic representation of {(PNP)CuI}2; right, Photo of {(PNP)CuI}2 in petroleum ether solution in room light and upon irradiation with 365 nm light. I-7 Photoluminescent polypyridine-supported CuI systems have garnered much attention due to their promise as low cost organic light emitting devices. 15 However, the tendency of CuI complexes to display weak emission and short-lived excited states is problematic. 16 McMillin has described these weaknesses in the current systems, 17 and his group recently reported a mononuclear complex, [Cu(dmp)(POP)]+ (dmp = 2,9-dimethyl1,10-phenanthroline; POP = bis[2-(diphenylphosphino)phenyl]ether)]), that exhibits both a relatively high quantum yield and a long-lived excited-state as compared to other polypyridine-Cu(I) systems. 18 Incorporation of a bulky, bis(phosphine) chelate appears to create a rigid environment around the copper center which (i) suppresses solvent-induced exciplex formation, and (ii) limits problematic ligand dissociation from the excited state. It is likely that through the same property of low structural reorganization energy that gives rise to rapid electron transfer between the oxidized and reduced forms of the {(SNS)Cu}2 system, highly conserved photon emission is now observed. We have found these PNP supported Cu dimers are among the most emissive inorganic complexes that have been reported and postulate that the intense emission may result from a direct Cu↔Cu interaction. Returning to the redox chemistry of the Cu2N2 species, the chemical oxidation of {(PNP)CuI}2 was evaluated with the intention of developing photolytically driven twoelectron chemistry. Chapter IV details the multielectron redox processes of the PNP dicopper complexes. Through targeted ligand modification, dicopper complexes in three oxidation states with pseudo-tetrahedral geometries are described in terms of their solidstate molecular structures, spectral and magnetic properties. The preparations of the isoelectronic dizinc and mixed copper-zinc dimers are also discussed. These latter I-8 complexes serve to highlight the unique character of the Cu↔Cu interaction to function of the system. Based on our studies, it is evident that much of this stabilization results from delocalization across the entire molecular scaffold. I.B. Summary of Group 10 metal studies Our initial work in the exploration of late transition metal amide chemistry focused on the Group 10 transition metals. In order to build new molecular systems relevant to our goals of exploring small-molecule activation and uncovering novel metal-ligand coordination environments, robust anionic chelating ligands were pursued. Initially complexes based on the monoanionic chelating ligand bis(8-quinolinyl)amine, (BQA)H, and its structural analogues were targeted. The synthesis of (BQA)H had been first reported by Nielsen in 1964, 19 and the first reported metallation followed in 1986, when Petersen published the crystal structure of the square planar complex, 20 (BQA)CuIICl. The literature preparation of (BQA)H was achieved by condensation of 8hydroxyquinoline and 8-aminoquinoline in a Bucherer-type reaction.7,8 Yields were of less than 10% after one week of reflux which was unacceptable for producing multi-gram quantities of the ligand. Thus, a catalytic cross-coupling strategy was targeted. 21 Pdcoupling of commercially available 8-aminoquinoline and 8-bromoquinoline was achieved under typical reaction conditions (Pd2(dba)3, toluene, 110 °C, 3 days) with NaOtBu as the base and rac-BINAP as the cocatalyst. 22 Recrystallization of the crude product afforded orange crystalline solid, typically in yields in excess of 85%. This strategy enabled the generation of a readily prepared family of useful, monoanionic amido ligands, Figure I.5. Although hard amido ligands can lead to undesirable reduction and/or degradation of late metal complexes, the use of chelating I-9 amido ligands has been shown to circumvent such problems. 23 Our desire to explore the divalent Group 10 chemistry of (BQA)H was motivated, in part, by the following rationale: Square planar, monoalkyl complexes of ligand (BQA)H should adopt a coordination geometry in which the alkyl group is forced to occupy a coordination site trans to the amido nitrogen donor group. Typically, divalent group 9 and 10 complexes containing two strongly trans-influencing ligands adopt coordination geometries that place these two ligands cis to one another. 24 The BQA ligand discriminates against such a cis preference, and our hope was that the resulting complexes would prove to be reactive at the site trans to the amido nitrogen donor. In this regard, BQA-type ligands are conceptually related to the popular family of anionic “pincer” ligands, a class of ligands that now affords a rich reaction chemistry amongst metals from groups 8, 9 and 10. 25 Moreover, it was hoped the aryl organic groups attached to the amide would serve to “soften” otherwise hard ligands and allow for generation of electron-rich organometallic complexes. R1 Ar NH2 + Ar Br Pd0 Ar Ar N Ar [Li] H BuLi N Ar = R2 P R S 2 R R O SO3 N R Ar N N Figure I.5. Generalized ligand synthesis via Pd0 catalyzed aryl amination. Deprotonation of (BQA)H with nBuLi in toluene affords the lithium amide complex {(BQA)Li}2, whose dimeric solid-state crystal structure is shown.22 Crystals of {(BQA)Li}2 were obtained and the dimeric structure was established by an X-ray diffraction study (Figure I.6). This Li2N2 complex has D2 molecular symmetry with C2 I - 10 axis running through the two lithium atoms in the plane of the parallelepiped formed by N2A, LiB, N2B and LiA (Figure I.6). The dimeric conformation is favored in the absence of a donor solvent and is robust by virtue of each donor arm of the [BQA]¯ ligand coordinating to a different lithium center. While the synthesis of {(BQA)Li}2 provided us with a useful metathesis reagent for our Group 10 metal studies, it also gave us the first inkling of the potential capacity of this genre of chelating amide to support bridging M2N2 diamond core complexes (vide supra). N3A N3B LiA C2 axis N2B N2A LiB N1B N1A Figure I.6. Displacement ellipsoid (50%) representation of [Li][BQA]. Selected bond distances (Å) and angles (deg) are as follows: LiA-N(1A), 2.024(4); LiA-N(2A), 2.079(4); LiA-N(2B), 2.063(4); LiA-N(3B), 1.990(4); LiA-LiB, 2.395(5); LiB-N(1B), 1.958(4); LiB-N(2B), 2.051(4); LiB-N(2A), 2.018(4); LiB-N(3A), 1.999(4); N(2B)-LiAN(2A), 107.15(18); N(2A)-LiA-LiB, 53.05(13); N(2B)-LiA-LiB, 54.16(13); N(2B)-LiAN(2A), 107.15(18); N(1A)-LiA-N(2A), 82.50(15); N(3B)-LiA-N(2A), 103.80(17); C(8A)-N(2A)-C(17A),120.37(17). Entry into the Group 10 chemistry of (BQA)H was effected by both protolytic and metathetical strategies. The divalent chloride complexes (BQA)PtCl, (BQA)PdCl, and I - 11 (BQA)NiCl were prepared and fully characterized.22 An X-ray structural study for each of these three complexes shows them to be well-defined, square planar complexes in which the auxiliary BQA ligand binds in the expected planar, η3-fashion. The (BQA)PtMe and (BQA)PtPh complexes were also synthesized and isolated. 26 Both were prepared in excellent yield by transmetallation of {(BQA)Li}2 with (COD)PtMeCl and (COD)PtPhCl, respectively, to afford purple microcrystalline solids. Platinum satellites are easily observed in the 1H NMR for both (BQA)PtMe and (BQA)PtPh confirming the attachment of the R group to the Pt center. Our primary interest in developing these systems was to study the ability of pincerlike amido complexes of platinum to undergo intermolecular C-H bond activation processes. This process is in analogy to the first step of a Shilov-type reaction, in which methane is selectively oxidized to methanol. 27,28 The first step of this mechanism is hypothesized to proceed by a d8, square planar [PtIICl4]2- reacting with methane, resulting in the formation of a square planar PtII-CH3 complex with HCl as the formal by-product; water formally acts as a Bronsted base. By analogy, we sought to examine the activation of benzene by (BQA)PtCl, another robust, d8, square planar, PtII-Cl complex. A suspension of (BQA)PtCl in benzene was heated to 150 °C in a sealed reactor, at which time the complex became fully soluble, for several days, and showed no reactivity. A tertiary, sterically encumbering, alkyl amine base, NEtiPr2, was then added to examine whether a soluble base would be able to promote reactivity. Again no new Pt products were detected, nor was any of the expected conjugate acid, [HNEtiPr2][Cl], evident by 1 H NMR. I - 12 N N Pt Cl N 1 NiPr2Et 150 °C, 36 h no reaction N 1 NiPr2Et N N Pt OTf 150 °C, 36 h N Pt N + [HNiPr2Et][OTf] N Figure I.7. C-H bond activation by (BQA)PtIICl and (BQA)PtIIOTf complexes. In order to gain entry to a C-H activation system, neutral (BQA)PtOTf was selected because the triflate ligand was anticipated to be much more labile than the corresponding chloride.26 (BQA)Pt-OTf was also quite insoluble in benzene, however upon heating to 150°C, a red homogeneous solution developed and, after several days, no reactivity was observed. When one equivalent of NEtiPr2 was added and heating was resumed at 150°C, the expected conjugate acid product [HNEtiPr2][OTf] became visible by 1H NMR after a few hours (Figure I.7). Further heating for 36 h resulted in consumption of the starting material and a high yield (> 90%) of (BQA)PtPh with a stoichiometric equivalent of [HNEtiPr2][OTf]. (BQA)PtPh was isolated by column chromatography (70% yield) and identified by 1H-NMR and FAB-MS. At lower temperature (~ 100°C), only 25% conversion to the phenyl product is observed after 5 days, suggesting a very high thermodynamic barrier to the reaction. This barrier is not difficult to reconcile based on the observation of the microscopic reverse in which (BQA)PtOTf is instantly formed upon addition of a stoichiometric amount of triflic acid to (BQA)PtPh at ambient temperature. The (BQA)PtMe complex was also found to undergo oxidation by MeI which afforded the octahedral PtIV complex (BQA)Pt(Me2)I. I - 13 The synthesis and intriguing photoisomerization of the (BQA)PtIV complex from meridial to facial ligand coordination complexes is described in Appendix A. While many intriguing aspects of Group 10 amide chemistry remain poised for exploration, this path of inquiry was ultimately set aside in order to develop the Group 11 systems described herein. I - 14 I.C. References Cited 1 a) Vallee, B. L.; Williams, R. J. P. Proc. Natl. Acad. Sci. U.S.A. 1968, 59, 498. b) Rorabacher, D. B. Chem. Rev. 2004, 104, 651. 2 Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265. 3 (a) Ambundo, E. A.; Yu, Q.; Ochrymowycz, L. A.; Rohrbacher, D. B. Inorg. Chem. 2003, 42, 5267. (b) Kyritsis, P.; Dennison, C.; Ingeldew, W. J.; McFarlane, W.; Sykes, A. G. Inorg. Chem. 1995, 34, 5370. (c) Groenveld, C. M.; Canters, G. W. J. Biol. Chem. 1988, 263, 167. (d) Groenveld, C. M.; Dahlin, S.; Reinhammer, B.; Canters, G. W. J. Am. Chem. Soc. 1987, 109, 3247. 4 (a) Solomon, E. I.; LaCroix, L. B.; Randall, D. W. Pure Appl. Chem. 1998, 70, 799. (b) Randall, D. W.; Gamelin, D. R.; LaCroix, L. B.; Solomon, E. I. J. Biol. Inorg. Chem. 2000, 5, 16. 5 (a) Ferguson-Miller, S.; Babcock, G. T. Chem. Rev. 1996, 96, 2889. (b) Gamelin, D. R.; Randall, D. W.; Hay, M. T.; Houser, R. P.; Mulder, T. C.; Canters, G. W.; de Vries, S.; Tolman, W. B.; Lu, Y.; Solomon, E. I. J. Am. Chem. Soc. 1998, 120, 5246. (c) George, S. D.; Metz, M.; Szilagyi, R. K.; Wang, H. X.; Cramer, S. P.; Lu, Y.; Tolman, W. B.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 2001, 123, 5757. (d) Ramirez, B. E.; Malmstrom, B. G.; Winkler, J. R.; Gray, H. B. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 11949. 6 (a) Kroneck, P. M. H.; Antholine, W. E.; Riester, J.; Zumft, W. G. FEBS Lett. 1989, 248, 212. (b) Kroneck, P. M. H.; Antholine, W. E.; Riester, J.; Zumft, W. G. FEBS Lett. 1988, 242, 70. (c) Antholine, W. E.; Kastrau, D. H. W.; Steffens, G. C. M.; Buse, G.; Sumft, W. G.; Kroneck, P. M. H. Eur. J. Biochem. 1992, 209, 875. I - 15 7 Williams, P. A.; Blackburn, N. J.; Sanders, D.; Bellamy, H.; Stura, E. A.; Fee, J. A.; McRee, D. E. Nat. Struct. Biol. 1999, 6, 509. 8 (a) Houser, R. P.; Young, Jr., V. G.; Tolman, W. B. J. Am. Chem. Soc., 1996, 118, 2101. (b) Blackburn, N. J.; deVries, S.; Barr, M. E.; Houser, R. P.; Tolman, W. B.; Sanders, D.; Fee, J. A. J. Am. Chem. Soc. 1997, 119, 6135. (c) Hagadorn, J. R.; Zahn, T. I.; Que Jr., L.; Tolman, W. B. J. C. S. Dalton Trans. 2003, 1790. 9 Al-Obaidi, A.; Baranovič, G.; Coyle, J.; Coates, C. G.; McGarvey, J. J.; McKee, V.; Nelson, J. Inorg. Chem. 1998, 37, 3567. 10 (a) LeCloux, D. D.; Davydov, R.; Lippard, S. J. Inorg. Chem. 1998, 37, 6814. (b) LeCloux, D. D.; Davydov, R.; Lippard, S. J. J. Am. Chem. Soc. 1998, 120, 6810. (c) He, C.; Lippard, S. J. Inorg. Chem. 2000, 39, 5225. 11 Gupta, R.; Zhang, Z. H.; Powell, D.; Hendrich, M. P.; Borovik, A. S. Inorg. Chem., 2002, 41, 5100. 12 (a) Nelson J.; McKee, V.; Morgan, G. G. Prog. Inorg. Chem. 1998, 47, 167. (b) Coyle, J. L.; Elias, H.; Herlinger, E.; Lange, J.; Nelson, J. J. Biol. Inorg. Chem. 2001, 6, 285. 13 Dicopper systems structurally distinct from the Cu2X2 diamond core motif have in certain cases shown a reversible Cu1Cu1/Cu1.5Cu1.5 redox process. See, for example, 10c and references therein. 14 (a) Blackburn, N. J.; Barr, M. E.; Woodruff, W. H.; van der Oost, J.; de Vries, S. Biochemistry 1994, 33, 10401. (b) Williams, M.; Lapplalainen, P.; Kelly, M.; SauerEriksson, E.; Saraste, M. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 11955. I - 16 15 (a) Cui, Ji.; Huang, Q.; Veinot, J. G. C.; Yan, H. Marks, T. J.; Adv. Mater. 2002, 14, 565. (b) Zhang, Q.; Zhou, Q.; Cheng, Y.; Wang, L.; Ma, D.; Jing, X.; Wang, F. Adv. Mater. 2004, 16, 432. 15 Ford, P. C.; Cariati, E.; Bourassa, J. Chem Rev. 1999, 99, 3625. 16 McMillin, D. R.; McNett, K. M. Chem. Rev. 1998, 98, 1201. 18 Cuttell, D. G.; Kuang, S.-M.; Fanwick, P. E.; McMillin, D. R.; Walton, R. A. J. Am. Chem. Soc. 2002, 124, 6. 19 Jenson, K. A.; Nielson, P. H. Acta Chem. Scand. 1964, 18, 1. 20 Puzas, J. P.; Nakon, R.; Pertersen, J. L. Inorg. Chem. 1986, 25, 3837. 21 (a) Tomori, H.; Sadighi, J. P.; Buchwald, S. L. J. Org. Chem. 2000, 65, 1158. (b) Wolfe, J. P.; Tomori, H.; Sadighi, J. P.; Yin, J. J.; Buchwald, S. L. J. Org. Chem. 2000, 65, 1158. (c) Alcazar-Roman, L. M.; Hartwig, J. F.; Rheingold, A. L.; Liable-Sands, L. M.; Guzei, I. A. J. Am. Chem. Soc. 2000, 122, 4618. (d) Hartwig, J. F. Accounts Chem. Res. 1998, 31, 852. (e) Hamamm, B. C.; Hartwig, J. F. J. Am. Chem. Soc. 1998, 120, 12706. 22 Peters, J. C.; Harkins, S. B.; Brown, S. D.; Day, M. W. Inorg. Chem. 2001, 40, 5083. 23 For some relevant lead references see: a) Fryzuk, M. D.; Macneil, P. A.; Rettig, S. J.; Secco, A. S.; Trotter, J. Organometallics 1982, 1, 918. b) Fryzuk, M. D.; Leznoff, D. B.; Thompson, R. C.; Rettig, S. J. J. Am. Chem. Soc. 1998, 120, 10126. c) Deacon, G. B.; Gatehouse, B. M.; Grayson, I. L.; Nesbit, M. C. Polyhedron 1984, 3, 753. d) Buxton, D. P.; Deacon, G. B.; Gatehouse, B. M.; Grayson, I. L.; Wright, P. J. Acta Crystallogr., Sect. C 1985, 41, 1049. e) Dori, Z.; Eisenberg, R.; Steifel, E. I.; Gray, H. B. J. Am. Chem. Soc. 1970, 92, 1506. f) Kawamoto, T.; Nagasawa, I.; Kuma, H.; Kushi, Y. Inorg. Chim. Acta I - 17 1997, 265, 163. g) Endres, H.; Keller, J.; Poveda, A. Z. Naturforsch., Teil B 1977, 32, 131. h) Sacco, A.; Vasapollo, G.; Nobile, C. F.; Piergiovanni, A.; Pellinghelli, M. A.; Lanfranchi, M. J. Organomet. Chem. 1988, 356, 397. 24 a) Vila, J. M.; Pereira, M. T.; Ortigueira, J. M.; Lata, D.; Torres, M. L.; Fernandez, J. J.; Fernandez, A.; Adams, H. J. Organomet. Chem. 1998, 566, 93. b) Crespo, M.; Solans, X.; Font-Bardia, M. Polyhedron 1998, 17, 3927. c) Gandelman, M.; Vigalok, A.; Shimon, L. J. W.; Milstein, D. Organometallics 1997, 16, 3981. 25 a) Liu, F. C.; Pak, E. B.; Singh, B.; Jensen, C. M.; Goldman, A. S. J. Am. Chem. Soc. 1999, 121, 4086. b) Sundermann, A.; Uzan, O.; Milstein, D.; Martin, J. M. L. J. Am. Chem. Soc. 2000, 122, 7095. c) Dani, P.; Karlen, T.; Gossage, R. A.; Gladiali, S.; van Koten, C. Angew. Chem. Int. Ed. 2000, 39, 743. 26 Harkins, S. B.; Peters, J. C. Organometallics 2002, 21, 1753. 27 Shilov, A. E.; Shul'pin, G. B. Chem. Rev. 1997, 97, 2879. 28 a) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1996, 118, 5961. b) Stahl, S.; Labinger, J. A.; Bercaw, J. E. Angew. Chem. Int. Ed. 1998, 37, 2181. II - 1 Chapter II Electrochemical Studies of Dicopper Complexes Supported by Bis(2-tertbutylsulfanylphenyl)amido Ligands † † Adapted from Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 2885. II - 2 II.A. Introduction Biological electron transfer agents that feature copper are most typically mononuclear in nature, as observed in the well known “blue copper” family of proteins. 1, 2 , 3 Such systems and relevant small molecule model complexes have been studied intensively to gain an appreciation of how the inner coordination geometry of a copper ion dictates its ability to mediate rapid electron transfer. 4 While the issues that dictate electron transfer rates in copper proteins are an ongoing topic of concern,4a,5 it is widely accepted that structural reorganization needs to be minimized to achieve rapid rates. 6 Higher nuclearity copper-based electron transfer agents, while less common, are now well established. For example, a collection of spectroscopic and more recent structural data has revealed that cytochrome c oxidase accomplishes enhanced electron transfer (ET) into and out of a buried CuA site by virtue of a highly covalent, thiolate-bridged Cu2S2 diamond core structure type. 7,8 Amongst the various low molecular weight systems that have been developed to model aspects of this CuA site, 9, 10 , 11 , 12 perhaps the most structurally relevant concerns Tolman’s thiolate-bridged, mixed-valence {LN3SCu1.5}2+ diamond core complex.9a This complex features an EPR signal consistent with a fully delocalized class III diamond core description, 7b,9a as observed for the resting state form of CuA.7b,8 Despite this similarity, the Cu···Cu distance in the Tolman model system is much larger than that observed in CuA, and the complex does not reproduce the reversible Cu1Cu1 to Cu1.5Cu1.5 redox behavior of CuA. Several dicopper systems that diverge from the diamond core structural motif do show reversible redox behavior for a Cu1.5Cu1.5 mixed-valence state.11c,13 For example, a family of dicopper azacryptates has been developed that successfully brings two copper centers constrained by the II - 3 azacryptate cage into much closer proximity (~ 2.35 – 2.4 Ǻ).13a Some of these complexes, which are characterized by a direct metal-metal σ bond that is quite distinct from the bonding situation in CuA, nonetheless exhibit very rapid electron self-exchange (ks) between their oxidized and reduced forms (~ 105 M-1 s-1).13b Yet to be developed are diamond core copper systems that can reversibly model both the reduced CuICuI and the delocalized Cu1.5Cu1.5 redox states observed in CuA. 14 Ligand systems that would help to minimize structural reorganization between these two redox forms, and hold the two copper centers of a diamond core in close enough proximity to reproduce the short Cu-Cu bond distance observed in CuA (~ 2.5 Å), 15 might serve as simplified functional models for the redox behavior of CuA.7b,c,d One avenue into this regime using small molecule design is to replace the thiolate bridging units inherent to the Cu2S2 diamond core of CuA with amido bridging units. Such a strategy should slide the copper centers closer together so as to better model the distance observed in CuA while still maintaining a Cu2X2 diamond core motif. Moreover, such an approach might shed some light on whether thiolate plays a unique role in facilitating rapid electron transfer chemistry by bringing two copper centers of a diamond core into strong electronic communication. Herein we describe a Cu2N2 diamond core system supported by sulfur-rich, tridentate amido ligands that exhibits a fully reversible and very facile 1electron redox process between a Cu1Cu1 and a class III delocalized Cu1.5Cu1.5 form. Solid-state structural snapshots of both redox forms are described that reveal (i) very short Cu···Cu distances akin to those in CuA, and (ii) minimal structural reorganization upon oxidation from the reduced Cu1Cu1 to the delocalized Cu1.5Cu1.5 state. These features facilitate very facile rates of electron self-exchange between the reduced and one II - 4 electron oxidized forms of the system (≥ 107 M-1 s-1), rates which to our knowledge are faster than those measured for other low molecular weight copper complexes.4a,13b A distinct Cu···Cu compression accompanies the 1-electron oxidation process that may reflect the onset of direct Cu-Cu electronic exchange. II.B. Results and Discussion Scheme II.1 S NH S S Pd2(dba)3 DPPF S 1.6 M n-BuLi NH NaOtBu toluene, 100oC 18 h Li N S pet. ether 25oC S Br (2.1) CuCl2 THF 25oC, 18 h t Bu S S 2.1 CuBr•Me2S Cu N N C6H6 25oC, 4h t Bu S Cu t Bu Na/Hg THF 6h -NaCl N S S Cu Cl S t (2.2) Bu (2.4) Palladium(0) cross-coupling of 2-tert-butylsulfanyl bromobenzene with 2-tertbutylsulfanyl aniline afforded bis(2-tert-butylsulfanylphenyl)amine in 87% isolated yield. 16,17 Addition of nBuLi to the purified amine provided its lithium salt, abbreviated as [SNS][Li] (2.1) (Scheme II.1). Access to a diamagnetic dicopper(1,1) complex was accomplished by reaction of 2.1 with CuBr·Me2S in benzene. Analytically pure yellow II - 5 needles were obtained by recrystallization of the crude product from dichloromethane, and XRD analysis established the dimeric species {[SNS][Cu]}2 (2.2) shown in Figure II.1. Addition of a stoichiometric amount of [Cp2Fe][B(3,5-(CF3)2C6H3)4] to 2.2 produced the dinuclear complex [{[SNS][Cu]}2][B(3,5-(CF3)2C6H3)4] (2.3), which exhibits a temperature-independent magnetic moment between 10 and 295 K (μeff = 1.54 μB at 75 K (SQUID)). Analytically pure 2.3 was isolated as red-brown crystals by crystallization from a petroleum ether/diethyl ether mixture. Employing a Cu(II) source provided access to a mononuclear copper(II) complex supported by the [SNS] ligand scaffold. This was accomplished by stirring a slurry of 2.1 and CuCl2 in THF for 18 h, after which time the square-planar copper complex [SNS]CuCl (2.4) was isolated as a deep green solid. X-ray quality crystals were obtained by recrystallization from THF/petroleum ether at –30 ºC and the solid-state structure of 2.4 is shown in Figure II.2. Of note from the structure are the orientation of the tert-butyl groups, which are directed towards opposite faces of the square plane, and the canted nature of the aryl rings. These features provide the complex with molecular C2 symmetry. The Cu-N bond distance is 1.915(2) Å, and the Cu-S1 and Cu-S2 bond distances are 2.3476(7) Å and 2.3465(7) Å, respectively. Few amido complexes of copper(II) have been prepared for comparison of the Cu-N bond distance. Perhaps the most relevant complex is [BQA]CuCl (BQA = bis(8-quinolinyl)amido), 18 which features a Cu-N(amido) distance of 1.935(2) Å. The Cu-S distances in 2.4 appear to be typical for Cu(II) based upon an analysis of data stored in the Cambridge Structural Database (CSD, avg. = 2.38 Å). 19 The EPR spectrum of 2.4, which is shown in Figure II.3, confirms its expected S = 1/2 spin state. The reduction of 2.4 by 1.05 equivalents of Na/Hg amalgam in THF provides an alternative synthesis of the dicopper complex 2.2. II - 6 Figure II.1. Molecular representations (top) of 2.2 and 2.3 (BArF4 = B(3,5(CF3)2C6H3))4. Displacement ellipsoid representations (50%) of the core atoms of 2.2 (bottom left) and 2.3 (bottom right), and graphical overlay (center; 2.2 in red; 2.3 in blue). Selected bond lengths (Å) and angles (º): For 2.2 vs. 2.3: Cu1-Cu2, 2.5989(3) vs. 2.4724(4); Cu1-N1, 2.1149(14) vs. 2.0887(16); Cu1-N2, 2.1303(14) vs. 2.1011(17); Cu2N1, 2.0850(14) vs. 2.0641(16); Cu2-N2, 2.1395(14) vs. 2.0568(16); Cu1-S2, 2.2730(5) vs. 2.2805(6); Cu1-S3, 2.2854(5) vs. 2.2805(6); Cu2-S4, 2.2735(5) vs. 2.2797(6); Cu2S1, 2.2853(5) vs. 2.2834(6); Cu2-N1-Cu1, 76.45(5) vs. 73.07(5); Cu1-N2-Cu2, 74.99(5) vs. 72.96(5); S2-Cu1-S3, 152.921(19) vs. 150.50(2); S4-Cu2-S1, 153.233(19) vs. 150.71(2); N1-Cu1-N2, 103.88(5) vs. 105.53(6); N2-Cu2-N1, 104.59(5) vs. 108.04(5). II - 7 Figure II.2: Solid state molecular structure of 2.4 (50% displacement ellipsoids). The protons have been removed for clarity. Selected bond lengths (Å) and angles (º): Cu-N, 1.915(2); Cu-Cl, 2.2068(7); Cu-S1, 2.3476(7); Cu-S2, 2.3465(7); N-Cu-Cl, 173.44(7); S1-Cu-S2, 167.35(3); N-Cu-S1, 86.02(6); N-Cu-S2, 86.37(6); S1-Cu-Cl, 95.89(2); S2Cu-Cl, 92.80(2). 270 290 310 Field / mT 330 350 Figure II.3: X-band EPR spectrum (9.40 GHz) of 2.4 at 77 K in 2methyltetrahydrofuran. II - 8 The solid-state structures of complexes 2.2 and 2.3 are unique and of central interest in this paper. A common, dinuclear Cu2N2 diamond core structure type featuring anionic amido (NR2-) bridging units is observed for each complex (Figure II.1). The fourcoordinate copper centers of both 2.2 and 2.3 are quite distorted from idealized tetrahedra and are perhaps better described as cis-divacant octahedra. In this latter formulation, the thioether donors occupy the two axial sites (avg. S-Cu-S angle: 153º for 2.2; 150.5º for 2.3) and the amido bridges the two equatorial sites (avg. Cu-N-Cu angle: 75.8º for 2.2; 73.0º for 2.3). The two sulfur donors common to each amido bridge bind such that each coordinates a different copper center, thereby providing idealized D2 symmetry to each system. The distinct ligand binding motifs between structures 2.2 and 2.3 and the square planar complex 2.4 underscore the coordinative flexibility of the [SNS]- ligand. The high structural similarity between the reduced and oxidized forms of 2.2 and 2.3, clearly evident from the graphical overlay of their cores that is shown in Figure II.1, is to our knowledge unprecedented in a dicopper diamond core model system. Close inspection of the bond distances of the core atoms of 2.2 and 2.3 does expose several noteworthy differences. Foremost amongst these is the marked shortening of the Cu···Cu distance in 2.3 (2.472 Å) by comparison to that of 2.2 (2.599 Å), a net change of ~ 0.13 Å. The short Cu···Cu distance observed for 2.3 is nearly identical to that observed for the mixed-valence form of CuA in cytochrome c oxidase. The Cu-N bond distances in 2.3 are on average slightly shorter than in 2.2 (avg. for 2.3 = 2.08 Å; avg. for 2.2 = 2.12 Å), and the Cu-N-Cu angles become slightly more acute (by approximately 2º) as the copper centers slide closer together. Interestingly, despite the contraction of the Cu2N2 II - 9 core, the Cu-S bond distances remain effectively unchanged (the average of the Cu-S distances for both 2.2 and 2.3 is 2.28 Å). Complex 2.2 exhibits a fully reversible redox process at –390 mV in CH2Cl2 (Fc+/Fc, 0.30 M [nBu4N][PF6], 50 mV/s) that we assign as the Cu1.5Cu1.5/Cu1Cu1 redox couple (Figure II.4). 20 The redox process is also reversible in THF solution at -250 mV (Fc+/Fc, 0.35 M [nBu4N][PF6], 50 mV/s). An irreversible redox process 21 is observed at much higher potential in CH2Cl2 (Epa = +560 mV) and in THF (Epa ~ +570 mV) that most likely reflects oxidation to an unstable dicopper(II,II) species. The large difference in the anodic potentials (880 mV in CH2Cl2) between these waves suggests the possibility of strong stabilization of the C1.5Cu1.5 form due to class III delocalization. EPR data presented below supports this assertion. Figure II.4: Cyclic voltammetry of 2.2 referenced vs. Fc+/Fc in CH2Cl2 (0.10 M [nBu4N][PF6]. Left, 2.2 (80 mV/s); right, 2.2 in CH2Cl2 (50 mV/s). The optical spectrum of yellow 2.2 (Figure II.5) was recorded in the range between 24 000 and 4000 cm-1 and shows a single, fairly intense absorption at 23 000 cm-1 (ε = 3900 M-1cm-1). The optical spectrum of 2.3 (Figure II.5) in this range is much richer. While its spectrum is clearly unique from that obtained for CuA and related mixedvalence model complexes, it does contain several gross features that may be related to II - 10 some of the absorption features recorded for these other systems.7b For example, a low energy near-IR absorption at 8550 cm-1 (ε = 2050 M-1cm-1 in CD2Cl2) is observed for 2.3 that lies slightly to the blue of a band that has been assigned as the Ψ → Ψ * transition of {LN3SCu1.5}2+ (6760 cm-1 in solution).7b At slightly lower energy, an appreciably more intense absorption is observed (5240 cm-1; ε = 4700 M-1cm-1 in CD2Cl2) in the spectrum for 2.3 which appears to be unique from those spectra obtained for CuA and relevant model complexes. Another intriguing band of fairly low intensity is centered around 13400 cm-1 (εmax = 600 M-1cm-1). This feature appears to be a superposition of two uniquely defined absorptions. Several broad features are also evident at higher energy between 16000 and 24000 cm-1. The optical spectrum of square planar 2.4 shows two broad and relatively weak transitions, one centered at 15000 cm-1 (ε = 320 M-1cm-1) and one at 7800 cm-1 (ε = 280 M-1cm-1). Figure II.5: Electronic absorption spectrum of 2.2 (red), 2.3 (blue), and 2.4 (green) in methylene chloride-d2. II - 11 For comparison, the lowest energy band of CuA, which has been assigned as the Ψ→Ψ* transition, is observed at 13,400 cm-1.7b The significant energy difference in the Ψ→Ψ* transition between {LN3SCu1.5}2+ and CuA has been attributed to the significant difference in their respective Cu-Cu distances. The distance in {LN3SCu1.5}2+ is 2.92 Å, and that in CuA is ~ 2.5 Å.7b Whereas direct Cu-Cu π overlap is possible for CuA, no direct exchange is presumably operative in {LN3SCu1.5}2+ owing to its much larger internuclear distance. Given these data, it would certainly be of interest to determine the origin of the bands centered at ~ 13,400 cm-1 for 2.3, or those lower energy features centered around 8600 cm-1 and 5240 cm-1. In particular, it is of interest to determine which of these, if any, arises from a Ψ→Ψ* transition given that this latter complex features a Cu-Cu distance that is very similar (2.4724(4) Å) to that observed in CuA. A geometry optimization and electronic structure calculation of 2.3 using DFT (JAGUAR 5.0, B3LYP/LACVP**) was performed using the crystallographically determined X-ray coordinates as the initial geometry guess. 22 The calculation provided a theoretically determined structure whose geometry agreed remarkably well with the experimental structure. The electronic structure of 2.3 afforded by the DFT calculation suggests that the redox active SOMO (Figure II.6) contains significant orbital contributions from the four atoms of the Cu2N2 diamond core. The SOMO is antibonding with respect to each of the four Cu-N interactions and also the Cu-Cu interaction. This phase pattern is consistent with the observed Cu2N2 core expansion that occurs when this orbital becomes doubly occupied by reduction to the Cu1Cu1 species 2.2. The SOMO shown in Figure II.6 shows a phase relationship similar to that calculated for the SOMO II - 12 II - 13 of a thiolate-bridged dicopper site of CuA.7c The next fourteen highest energy occupied orbitals are also shown to enhance the orbital description. The X-band EPR spectrum of 2.3 was obtained in 2-methyltetrahydrofuran in the temperature range between 5 K and 80 K (Figure II.7). The data obtained are fully consistent with describing the system as a class III mixed-valence species. The EPR signal observed, which retains its structure from 5 K to 80 K, displays 7-line hyperfine coupling due to the dicopper core, as expected for a system featuring one unpaired electron that is coupled to two copper centers. While super-hyperfine coupling to the bridging nitrogen nuclei of the Cu2N2 core might be anticipated, such coupling is not readily discerned in the X-band spectra. We attempted to simulate the experimental EPR spectrum for 2.3 acquired at 5.2 K and simulated the spectrum. The difficulty with simulating the spectrum does not appear to be due to a trace copper(II) impurity: the spectrum shown proved highly reproducible using samples that had been independently synthesized and recrystallized. Moreover, the EPR spectrum of complex 2.4 shown in Figure II.3 is typical of copper(II) complexes supported by the [SNS]- ligand, and similar features are not present in the spectrum of 2.3. The three g tensors in the spectrum of 2.3 are poorly resolved (gx = 1.998, gy = 2.065, gz = 2.067, AxCu = 3 G, AyCu = 18.4 G, AzCu = 44.5 G), and this makes the simulation difficult. Acquisition of the EPR spectrum of 2.3 at higher field strength should help to resolve the spectrum. The experimental EPR data and our crude simulation do nonetheless support assignment of 2.3 as an S = ½ system coupled to a symmetric dicopper center. II - 14 5K 41 K 80 K simulation 285 305 325 Field / mT 345 365 Figure II.7: X-band EPR spectrum (9.38 GHz) of 2.3 at 80 K , 41 K, and 5 K in 2methyltetrahydrofuran (black). Simulated spectrum (blue) {gx = 1.998, gy = 2.065, gz = 2.067, AxCu = 3 G, AyCu = 18.4 G, AzCu = 44.5 G}. Given the small structural reorganization that occurs upon oxidation of 2.2 to 2.3, and the reversibility of the process, we reasoned that unusually rapid intermolecular electron transfer kinetics might be observable in the present system. Whereas electron transfer kinetics at biological copper sites and low molecular weight copper complexes have been the subject of many previous studies,4,13, 23 in part motivated by the rapid electron transfer rates obtained in type 1 copper sites of blue copper proteins,4b,c,d to our knowledge such studies have not included diamond core dicopper model systems. We therefore sought to measure, or at least approximate, the intermolecular self-exchange rate between 2.2 and 2.3 directly using the technique of 1H NMR line-broadening analysis. 24 II - 15 900 1 800 0.9 0.8 700 Line Width (Hz) 0.6 500 0.5 400 0.4 300 0.3 200 0.2 100 0 500 Mole Fraction 0.7 600 0.1 1000 1500 2000 2500 Chemical Shift (Hz) 3000 0 3500 Graph II.1: Plots of line width vs. chemical shift (‹) and mole fraction vs. chemical shift (z). A single and sharp resonance for the tert-butyl protons of 2.2 is observed at 1.23 ppm. For a pure sample of 2.2 its line width W2 (where line width indicates the full-width at half maximum for the Lorentzian-shaped line) is 1.3 ± 0.2 Hz. The 1H NMR spectrum of a pure sample of 2.3 features a very broad resonance centered at 6.79 ppm. This resonance is slightly convoluted by overlap with the aryl protons of the [B(3,5(CF3)2C6H3)4] counter-anion. Nonetheless its line width, W3, can be estimated as 807 ± 25 Hz. The addition of increasing amounts of cationic 2.3 to a solution of 2.2 in CD2Cl2 causes the observed resonance to broaden and to shift downfield. A plot of the chemical shift of this resonance as a function of the mole fraction of 2.2, χ2, reveals a linear correlation (Graph II.1) and therefore suggests that the rate of electron-exchange between 2.2 and 2.3 falls within the fast-exchange regime. If we assume a two-site exchange II - 16 system comprised of diamagnetic 2.2 and paramagnetic 2.3, and if the rate law for electron-transfer is first order with respect to each reactant concentration, then equation II.1 relating line widths and reactant lifetimes holds: 25,26 W23 = χ3W3 + χ2W2 + [χ3χ2{4π(δν)2ks-1C23-1}] (II.1) In this equation δν represents the contact shift between 2.2 and 2.3, ks represents the rate constant for electron self-exchange (M-1 s-1), and C23 represents the total molarity of the solution of 2.2 and 2.3. To obtain a reliable value for ks in the fast exchange region, it is necessary that W23 be considerably larger than the sum {χ3W3 + χ2W2}. When ks approaches very high values (>108 M-1 s-1), the third term [χ3χ2{4π(δν)2ks-1C23-1}] approaches zero and W23 converges to the sum {χ3W3 + χ2W2}. Within the error estimated for our line width measurements W2, W3, and W23, the magnitude of W23 is too close to the sum {χ3W3 + χ2W2} to accurately determine the magnitude of ks. In other words, small errors in our values for W2, W3, and W23 will dramatically affect the value predicted. Therefore the rate of the exchange process is too fast to measure accurately by the technique of NMR line-broadening analysis, and we must at present be satisfied with suggesting a conservative lower boundary limit for ks as ≥ 107 M-1 s-1. We attempted to slow the exchange rate down by diluting the total concentration of 2.2 and 2.3, but the line widths W23 obtained were still too close to the sum {χ3W3 + χ2W2} to provide accurate values for ks. The crude lower boundary we estimate for ks can be compared to that measured for Fc/[Fc][PF6] (Fc = ferrocene) by the same technique (ks = (4.3 ± 0.3) x 106 M-1 s-1 at 298 K in CD2Cl2).24c More interesting is to consider values that have been measured for CuA for comparison. While electron self-exchange rates are obviously not available for the buried CuA site of cytochrome c oxidase, meaningful values for electron II - 17 transfer rates into and out of CuA domains have been measured. Perhaps the most meaningful value for comparison to the present model system is that determined by Slutter et al. using flash photolysis to initiate electron transfer between excited (2,2’bipyridyl)ruthenium(II) (RuII(bpy)3*) and the soluble CuA domain from the cytochrome ba3 of Thermus thermophilus. 27 The value estimated in that study for the second-order ET rate constant from RuII(bpy)3* to the fully oxidized CuA site at high ionic strength was determined to be 2.2 x 108 M-1 s-1. Stopped flow studies with a soluble CuA domain and its natural partner cytochrome c550 from P. denitrificans provided rates of 1.5 x 106 M-1 s-1. 28 Electron self-exchange values determined for the oxidized and reduced forms of plastocyanin from Anabaena variabilis obtained by longitudinal NMR relaxation were 1.50 ± 0.13 x 105 M-1 s-1 at 298 K in pH 7.0 H2O. 29 Clearly, the lower boundary for ks we have proposed suggests our model system can achieve electron-transfer rates comparable to CuA. It is therefore of obvious interest to more accurately determine an ET value for the present Cu2N2 model system for direct comparison to a known value for a soluble CuA domain. Efforts are now underway to examine the flash photolysis methodology used by Slutter et. al. to determine whether this will be a viable strategy. To conclude, the amido-bridged Cu2N2 diamond core structures described herein provide an interesting structural departure from the thiolate-bridged diamond cores of biological CuA. Nonetheless, the {[SNS][Cu]}2+n (n = 0, 1) system is unique amongst dicopper diamond cores in its ability to functionally model both the reduced and the oneelectron oxidized mixed-valence states of biological CuA. Minimal structural reorganization of the [SNS]- ligand framework, and the short Cu-Cu distance it supports, are likely important factors governing the reversible redox behavior of {[SNS][Cu]}2+n. II - 18 Moreover, these same factors seem to provide access to a very facile Cu1Cu1↔Cu1.5Cu1.5 electron self-exchange process (≥107 M-1 s-1), the lower limit of whose magnitude is to our knowledge unprecedented in copper model complexes. These data collectively suggest that amido bridges are an attractive alternative to thiolate as a bridging unit for dicopper diamond core systems, especially if a facile electron transfer agent is desired. Many questions remain to be answered for the dicopper systems discussed herein, and future studies will probe the spectroscopy and electron-transfer kinetics of these new systems in more detail. Moreover, we will explore whether synthetic access to related Cu2N2 structure types using other [LNL]- ligands is possible so as to probe structure/function relationships in such systems more thoroughly. II - 19 II.C. Experimental General. All manipulations were carried out using standard Schlenk or glove-box techniques under a dinitrogen atmosphere. Unless otherwise noted, solvents were deoxygenated and dried by thorough sparging with N2 gas followed by passage through an activated alumina column. Non-halogenated solvents were tested with a standard purple solution of sodium benzophenone ketyl in tetrahydrofuran to confirm effective oxygen and moisture removal. All reagents were purchased from commercial vendors and used without further purification unless otherwise stated. 2-tert-butylsulfanyl aniline, 30 [Cp2Fe][B(3,5-(CF3)2C6H3)4], 31 and CuCl2·0.66 THF 32 were prepared according to literature procedures. Elemental analyses were performed by Desert Analytics, Tucson, AZ. Deuterated solvents were purchased from Cambridge Isotope Laboratories, Inc. and degassed and dried over activated 3 Å molecular sieves prior to use. A Varian Mercury-300 or INOVA-500 NMR spectrometer was used to record 1H, 13 C, 19F NMR spectra at ambient temperature. 1 H chemical shifts were referenced to residual solvent. Temperature calibration of the NMR probe was accomplished using an anhydrous methanol standard. Line shape analysis of experimental NMR was performed using the Varian 6.1c software package. GC-MS data was obtained by injecting a dichloromethane solution into a Agilent 6890 GC equipped with an Agilent 5973 mass selective detector (EI). High-resolution EI mass spectroscopy was carried out by the Caltech Chemistry Mass Spectral Facility using a JEOL JMS600. UV-vis measurements were taken on a Cary 500 UV/Vis/NIR Spectrophotometer using a 0.1 cm quartz cell with a Teflon stopper. IR measurements were obtained with a KBr solution cell using a Bio-Rad Excalibur FTS 3000 spectrometer controlled by Bio-Rad Merlin Software (v. II - 20 2.97) set at 4 cm resolution. X-ray diffraction studies were carried out in the Beckman -1 Institute Crystallographic Facility on a Bruker Smart 1000 CCD diffractometer. Table II.1 shows X-ray diffraction data for 2.2, 2.3, and 2.4. Table II.1: X-ray diffraction data for 2.2, 2.3, and 2.4.a 2.2 2.4 Chemical Formula C40H52CuN2S4 C72H64BCu2F24N4S4 · ½ (C4H10O) C20H26ClCuNS2 Formula Weight 816.206 1716.478 443.53 T (K) 98 100 96 λ (Å) 0.71073 0.71073 0.71073 a (Å) 10.2263(9) 10.3179(8) 17.0807(13) b (Å) 10.6226(9) 19.3779(14) 12.3173(10) c (Å) 19.4884(17) 19.9116(15) 19.5782(15) α (º) 94.017(2) 74.234(1) 90 β (º) 90.634(2) 83.222(1) 90 γ (º) 107.305(1) 85.603(1) 90 3 V (Å ) 2015.1(3) 3800.4(5) 4119.0(6) Space Group P-1 (#2) P-1 (#2) Pbca (#61) Z 2 2 8 Dcalcd (g/cm ) 1.345 1.500 1.430 -1 μ (mm ) 1.292 0.773 1.396 R1, wR2 (I>2σ(I)) 0.0303, 0.0660 0.0393, 0.0810 0.0381, 0.0694 3 a 2.3 R1 = Σ ||Fo|-|Fc|| / Σ |Fo|, wR2 = {Σ[w(Fo -Fc ) ]/Σ[w(Fo ) ]} 2 2 2 2 2 1/2 Magnetic Measurements. Measurements were recorded using a Quantum Designs SQUID magnetometer running MPMSR2 software (Magnetic Property Measurement System Revision 2). Data were recorded at 5000 G. Samples were suspended in the magnetometer in a clear plastic straw sealed under nitrogen with Lilly No. 4 gel caps. Loaded samples were centered within the magnetometer using the DC centering scan at 35 K and 5000 G. Data were acquired at 2 – 20 K (one data point every 2 K) and 20 – 295 K (one data point every 5 K). The magnetic susceptibility was adjusted for diamagnetic contributions using the constitutive corrections of Pascal’s constants. The II - 21 molar magnetic susceptibility (χm) was calculated by converting the calculated magnetic susceptibility (χ) obtained from the magnetometer to a molar susceptibility (using the multiplication factor {(molecular weight) / [sample weight)*(field strength)]}). CurieWeiss behavior was verified by a plot of χm-1 versus T. Graph II.2 shows the plot of the values for χΤ vs. T. Effective magnetic moments were calculated using equation 2.2. μeff = sqrt(7.997 χm T) (eq 2.2) 0.6 0.5 χT (emu*K) 0.4 0.3 0.2 0.1 0 0 50 100 150 200 250 300 350 Temp (deg. K) Graph II.2: SQUID Magnetization study of 2.3. EPR Measurements. X-band EPR spectra were obtained on a Bruker EMX spectrometer (controlled by Bruker Win EPR Software v. 3.0) equipped with a rectangular cavity working in the TE102 mode. Variable temperature measurements were conducted with an Oxford continuous-flow helium cryostat (temperature range 3.6 – 300 K). Accurate frequency values were provided by a frequency counter built into the microwave bridge. Solution spectra were acquired in toluene. Sample preparation was performed under a dinitrogen atmosphere in an EPR tube equipped with a ground glass joint. II - 22 Electrochemistry. Electrochemical measurements were carried out in a glove-box under a dinitrogen atmosphere in a one-compartment cell using a BAS model 100/W electrochemical analyzer. A glassy carbon electrode and platinum wire were used as the working and auxiliary electrodes, respectively. The reference electrode was Ag/AgNO3 in THF. Solutions (CH2Cl2) of electrolyte (0.30 M tetra-n-butylammonium hexafluorophosphate) and analyte were prepared in a glove-box. Electron self-exchange rate between 2.2 and 2.3: 1 H NMR spectra were acquired in CD2Cl2 on a Varian Inova Spectrometer operating at 499.852 MHz at 303 K. The data collection parameters were as follows: 0 s relaxation delay, 5.5 μs pulse width, and an 89988.8 Hz sweep width. The samples were weighed and dissolved in 1.0 mL of CD2Cl2, after which each sample tube was charged with ~0.5 mL of the solution. The sample concentrations, mole fractions, and experimental values for W are shown in Table II.2 below. The chemical shifts and line widths of the experimental spectra were obtained by Lorentzian line-fitting using the Varian 6.1c software package. Table II.2. Line-broadening data, concentrations, and mole fractions used to estimate the lower boundary of the self-exchange reaction rate between 2.2 and 2.3 in dichloromethane solution. See Graph II.1 for plots of (i) chemical shift versus mole fraction of 2.2, and (ii) W2, W3, and W23 versus χ2. 2.2 only 2.3 only Mix-A Mix-B Mix-C χ2.2 1 0 0.75 ± 0.01 0.59 ± 0.01 0.41 ± 0.01 χ2.3 0 1 0.25 ± 0.01 0.41 ± 0.01 0.59 ± 0.01 Total conc. (mM) - - 29 24 27 Number of scans 100 1000 1000 1000 1000 Chemical shift (Hz) 619 3396 1314 1766 2263 197 ± 15 314 ± 15 458 ± 15 W (Hz) 1.3 ± 0.2 807 ± 25 II - 23 Synthesis of 2-tert-butylsulfanyl bromobenzene. In air, 2-bromothiophenol (25 mL, 0.208 mol) was added dropwise to a vigorously stirred solution of tert-butyl alcohol (23 g, 0.312 mol), H2O (150 mL), and concentrated H2SO4 (200 ml) at -10ºC in air. Following the addition, the reaction was allowed to come to ambient temperature and stirring was continued for 18 h, at which time ether (50mL) was added to the reaction mixture and the organic phase was washed with 1M Na2CO3 solution (100 mL), water (3 x 100 mL), dried over Na2SO4, and the solvent removed in vacuo. The resultant cloudy oil was fractionally distilled (90ºC/0.01 mmHg) affording a colorless oil. 1 H NMR (300 MHz, CDCl3): δ 7.64 (m, 2H, Ar-H), 7.24 (t, 1H, Ar-H), 7.15 (t, 1H, Ar- H), 1.32 (s, 9 H, C(CH3)3). 13C NMR (126 MHz, CDCl3): δ 139.7, 134.4, 133.7, 132.8, 130.3, 127.5, 48.8, 31.5. GC-MS(M/z): 246/244[M], 190/188[M-(CH3CCH2)], 108, 109, 82, 69, 57. Synthesis of bis(2-tert-butylsulfanylphenyl)amine: A 200 mL reaction vessel equipped with a Teflon stopcock and stir bar was charged with Pd2(dba)3 (0.365 g, 0.399 mmol), bis(diphenylphosphino)ferrocene (DPPF) (0.442 g, 0.798 mmol), and toluene (30 mL) under a dinitrogen atmosphere. The resulting solution was allowed to stir for 5 min, after which time 2-tert-butylsulfanyl bromobenzene (9.77 g, 39.9 mmol), 2-tert-butylsulfanyl aniline (7.23 g, 39.9 mmol), and additional toluene (70 mL) were added. The subsequent addition of NaOtBu (5.37 g, 55.9 mmol) resulted in a brown solution that was stirred vigorously for 18 h at 100 °C. The solution was then allowed to cool and filtered through a silica plug that was then extracted with toluene to ensure complete removal of the desired product. Concentration of the collected extracts and removal of solvent yielded a II - 24 brown solid. These solid were extracted with hexanes and filtered. Purification by recrystallization from hexanes at –30 ºC afforded beige crystalline blocks (11.95 g, 87%). 1 H NMR (300 MHz, CD2Cl2): δ 8.30 (s, 1H, N-H), 7.52 (d, 2H, Ar-H), 7.43 (d, 2H, Ar- H), 7.25 (t, 2H, Ar-H), 6.83 (t, 2H, Ar-H), 1.33 (s, 18 H, C(CH3)3). 13C NMR (126 MHz, CDCl3): δ 145.8, 140.0, 130.2, 120.5, 120.0, 115.4, 47.9, 31.3. IR (KBr/CH2Cl2, cm-1): 2924, 1575, 1509, 1364, 1319, 1167, 1034. HR-EI MS: Calcd for C20H27NS2: 345.1585; Found: 345.1575. Synthesis of [Li][SNS], 2.1: In a 250 mL flask, bis(2-tert-butylsulfanylphenyl)amine (2.5 g, 7.25 mmol) was dissolved in petroleum ether (100 mL) and a 1.6 M solution of nbutyl lithium in hexanes (5.9 mL, 9.43 mmol) was added dropwise with stirring at ambient temperature. A pale yellow solid began precipitating immediately and stirring was continued for 20 min. The solids were collected on a sintered glass frit and washed with petroleum ether (3 x 30mL). A spectroscopically pure off-white powder (2.5 g, 98%) was obtained upon drying in vacuo. 1 H NMR (300 MHz, C6D6): δ 7.52 (d, 2H, Ar-H), 7.34 (d, 2H, Ar-H), 7.15 (t, 2H, Ar-H), 6.68 (t, 2H, Ar-H), 1.11 (s, 18 H, C(CH3)3). 13C NMR (126 MHz, C6D6): δ 164.7, 139.5, 131.0, 123.7, 121.7, 117.3, 47.4, 31.5. Synthesis of {[SNS]Cu}2, 2.2: A solution of 2.1 (800 mg, 2.28 mmol) in benzene (20 mL) was added dropwise with stirring to a suspension of CuBr·Me2S (469 mg, 2.28 mmol) in benzene (40 mL). The solution immediately became bright yellow in color and after 4 h the reaction mixture was filtered through a pad of Celite on a sintered glass frit and the solvent removed in vacuo. The yellow solids were washed with petroleum ether (3x50 mL) and dried thoroughly which afforded spectroscopically pure product (850 mg, II - 25 91%). This complex decomposes photolytically over time so efforts were made to minimize light exposure during synthesis and storage. Analytically pure product is obtained as fine yellow crystalline needles by cooling a methylene chloride solution at -35 ºC. 1 H NMR (300 MHz, CD2Cl2): δ 7.32 (d, 2H, Ar-H), 7.06 (t, 2H, Ar-H), 6.91 (d, 2H, Ar- H), 6.57 (t, 2H, Ar-H), 1.23 (s, 18 H, C(CH3)3). 13C NMR (126 MHz, CD2Cl2): δ 160.8, 136.3, 130.3, 122.9, 121.4, 116.3, 49.8, 31.1. UV/Vis/NIR (CD2Cl2, nm (M-1cm-1)): 426 (4000). IR (KBr pellet, cm-1): 3045 (w), 2959 (m), 2921 (w), 1573 (s), 1543 (m), 1451 (s), 1428 (s), 1362 (m), 1312 (s), 1268 (m), 1154 (s), 1123 (w), 1031 (m), 818 (w), 743 (s), 725 (m). Anal. Calcd. for C40H52Cu2N2S4: C, 58.86; H, 6.42; N, 3.43. Found: C, 58.46; H, 6.22; N, 3.28. Synthesis of [{(SNS)Cu}2][B(C6H3(CF3)2)4], 2.3: In a 20 mL reaction vessel equipped with a Teflon stirbar, 2.2 (150.0 mg, 0.184 mmol) was suspended in diethyl ether (15 mL) and [Cp2Fe][B(3,5-(CF3)2C6H3)4] (192.8 mg, 0.184 mmol) was added as a solid in one portion. The green reaction mixture gradually became red-brown in color and after 90 min, the solution was filtered through glass wool, and the filtrate was dried in vacuo. The solids were extracted with petroleum ether (3 x 30 mL) to remove the ferrocene byproduct, and drying under reduced pressure afforded the desired product as a burgundy solid (290 mg, 94%). Analytically pure material was obtained by recrystallization from layering a diethylether solution with petroleum ether. 19 F NMR (282 MHz, CD2Cl2): -60.19 (s, Ar-CF3). IR (KBr pellet, cm-1): 3293 (w), 3053 (w), 2968 (m), 1609 (m), 1578 (m), 1499, (m), 1454 (s), 1355 (s), 1279 (s), 1133 (br s), 887 (m), 835, (m), 756 (m), 756 (m), 714 (m), 670 (m). UV/Vis/NIR(CD2Cl2, nm (M- 1 II - 26 cm )): 426(2500), 527(1650), 716(590), 1165(2050), 1920(4800). SQUID (solid, -1 average 10-295 K): 1.52 B.M. (R2 = 0.9991). Anal. Calcd. for C72H64BCu2F24N2S4: C, 51.49; H, 3.84; N, 1.67. Found: C, 51.63; H, 3.80; N, 2.00. Synthesis of (SNS)CuCl, 2.4: In a 50 ml round bottom flask equipped with a Teflon stirbar, CuCl·0.66 THF (1.00 g, 5.50 mmol), was suspended in THF (30 mL) and 2.2 (1.93 g, 5.50 mmol) was added portionwise immediately affording a forest-green solution. The reaction mixture was stirred for 18 h at ambient temperature and the solvent was removed in vacuo. The resultant solid was triturated with benzene (2 x 10 mL), extracted into benzene, filtered through Celite and dried thoroughly under reduced pressure to afford 2.4 (2.30 g, 94%). Analytically pure material was obtained by recrystallization from THF/petroleum ether solution at -30 ºC. IR (KBr pellet, cm-1): 3056 (w), 2961 (w), 1573, (m), 1456 (s), 1431 (m), 1365 (m), 1321 (s), 1240 (w), 1155 (m), 1032 (w), 767 (w), 750 (m), 740 (m). UV/Vis/NIR (CD2Cl2, nm (M-1cm-1)): 606 (600), 1280 (280). Anal. Calcd. for C20H26ClCuNS2: C, 54.16; H, 5.91; N, 3.16. Found: C, 54.22; H, 5.84; N, 3.08. Reduction of 2.4 to 2.2: An amalgam of sodium (14.3 mg, 0.622 mmol) in mercury (11.1 g, 55.3 mmol) was prepared in a 20 mL reaction vessel and a solution of 2.4 (263 mg, 0.592 mmol) in THF (10 mL) was added. The reaction mixture was vigorously stirred for 6 h, filtered through Celite, and dried in vacuo. The yellow solid was extracted into benzene (20 ml), lyophilized, and washed with cold petroleum ether to afford spectroscopically pure 2.2 (188 mg, 78%). II - 27 II.D. References Cited 1 (a) Guss, J. M.; Freeman, H. C. J. Mol. Biol. 1983, 169, 521. (b) Gray, H. B.; Malmstrom, B. G.; Williams, R. J. P. J. Biol. Inorg. Chem. 2000, 5, 551. 2 Shibata, N.; Inoue, T.; Nagano, C.; Nishio, N.; Kohzuma, T.; Onodera, K.; Yoshizaki, F.; Sugimura, Y.; Kai, Y. J. Biol. Chem. 1999, 274, 4225. 3 Suzuki, S.; Kataoka, K.; Yamaguchi, K.; Inoue, T.; Kai, Y. Coord. Chem. Rev. 1999, 192, 245. 4 (a) Ambundo, E. A.; Yu, Q.; Ochrymowycz, L. A.; Rorabacher, D. B. Inorg. Chem. 2003, 42, 5267. (b) Kyritsis, P.; Dennison, C.; Ingeldew, W. J.; McFarlane, W.; Sykes, A. G. Inorg. Chem. 1995, 34, 5370. (c) Groenveld, C. M.; Canters, G. W. J. Biol. Chem. 1988, 263, 167. (d) Groenveld, C. M.; Dahlin, S.; Reinhammer, B.; Canters, G. W. J. Am. Chem. Soc. 1987, 109, 3247. 5 (a) Solomon, E. I.; LaCroix, L. B.; Randall, D. W. Pure Appl. Chem. 1998, 70, 799. (b) Randall, D. W.; Gamelin, D. R.; LaCroix, L. B.; Solomon, E. I. J. Biol. Inorg. Chem. 2000, 5, 16. 6 Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265. 7 (a) Ferguson-Miller, S.; Babcock, G. T. Chem. Rev. 1996, 96, 2889. (b) Gamelin, D. R.; Randall, D. W.; Hay, M. T.; Houser, R. P.; Mulder, T. C.; Canters, G. W.; de Vries, S.; Tolman, W. B.; Lu, Y.; Solomon, E. I. J. Am. Chem. Soc. 1998, 120, 5246. (c) George, S. D.; Metz, M.; Szilagyi, R. K.; Wang, H. X.; Cramer, S. P.; Lu, Y.; Tolman, W. B.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 2001, 123, 5757. (d) Ramirez, B. E.; Malmstrom, B. G.; Winkler, J. R.; Gray, H. B. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 11949. II - 28 8 (a) Kroneck, P. M. H.; Antholine, W. E.; Riester, J.; Zumft, W. G. FEBS Lett. 1989, 248, 212. (b) Kroneck, P. M. H.; Antholine, W. E.; Riester, J.; Zumft, W. G. FEBS Lett. 1988, 242, 70. (c) Antholine, W. E.; Kastrau, D. H. W.; Steffens, G. C. M.; Buse, G.; Sumft, W. G.; Kroneck, P. M. H. Eur. J. Biochem. 1992, 209, 875. 9 (a) Houser, R. P.; Young, Jr., V. G.; Tolman, W. B. J. Am. Chem. Soc., 1996, 118, 2101. (b) Blackburn, N. J.; deVries, S.; Barr, M. E.; Houser, R. P.; Tolman, W. B.; Sanders, D.; Fee, J. A. J. Am. Chem. Soc. 1997, 119, 6135. (c) Hagadorn, J. R.; Zahn, T. I.; Que Jr., L.; Tolman, W. B. J. C. S. Dalton Trans. 2003, 1790. 10 Al-Obaidi, A.; Baranovič, G.; Coyle, J.; Coates, C. G.; McGarvey, J. J.; McKee, V.; Nelson, J. Inorg. Chem. 1998, 37, 3567. 11 (a) LeCloux, D. D.; Davydov, R.; Lippard, S. J. Inorg. Chem. 1998, 37, 6814. (b) LeCloux, D. D.; Davydov, R.; Lippard, S. J. J. Am. Chem. Soc. 1998, 120, 6810. (c) He, C.; Lippard, S. J. Inorg. Chem. 2000, 39, 5225. 12 Gupta, R.; Zhang, Z. H.; Powell, D.; Hendrich, M. P., Borovik, A. S. Inorg. Chem., 2002, 41, 5100. 13 (a) Nelson J.; McKee, V.; Morgan, G. G. Prog. Inorg. Chem. 1998, 47, 167. (b) Coyle, J. L.; Elias, H.; Herlinger, E.; Lange, J.; Nelson, J. J. Biol. Inorg. Chem. 2001, 6, 285. 14 Dicopper systems structurally distinct from the Cu2X2 diamond core motif have in certain cases shown a reversible Cu1Cu1/Cu1.5Cu1.5 redox process. See, for example, 11c and references therein. 15 (a) Blackburn, N. J.; Barr, M. E.; Woodruff, W. H.; van der Oost, J.; de Vries, S. Biochemistry 1994, 33, 10401. (b) Williams, M.; Lapplalainen, P.; Kelly, M.; SauerEriksson, E.; Saraste, M. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 11955. II - 29 16 See related L2N- ligand [bis(8-quinolinyl)amido]Li: Peters, J. C.; Harkins, S. B.; Brown, S. D.; Day, M.W. Inorg. Chem. 2001, 40, 5083. 17 (a) Wolfe, J. P.; Tomori, H.; Sadighi, J. P.; Yin, J. J.; Buchwald, S. L. J. Org. Chem. 2000, 65, 1158. (b) Alcazar-Roman, L. M.; Hartwig, J. F.; Rheingold, A. L.; LiableSands, L. M.; Guzei, I. A. J. Am. Chem. Soc. 2000, 122, 4618. 18 Puzas, J. P.; Nakon, R.; Petersen, J. L. Inorg. Chem. 1986, 25, 3837. 19 Cambridge Structural Database; Cambridge University: Cambridge, England (accessed November 2003). 20 Scaling of the Cu1.5Cu1.5/Cu1Cu1 redox couple to NHE provides an approximate value of +550 mV, reasonably close to the CuIICuI/CuICuI couple measured for the CuA cofactor of Thermus thermophilus cytochrome ba3 in water (+240 mV).33 21 The second oxidation process in dichloromethane solvent suggests the process may be modestly reversible under such conditions. 22 Geometry optimization and electronic structure calculation (JAGUAR 5.0, B3LYP/LACVP**) were performed on the complete structure of the cation of 3 assuming a doublet ground-state (no symmetry constraints applied; crystallographic coordinates of 3 were used as the HF initial guess). Jaguar 5.0, Schrodinger, LLC, Portland, Oregon, 2002. 23 Xie, B.; Elder, T.; Wilson, L. J.; Stanbury, D. M. Inorg. Chem. 1999, 38, 12. 24 (a) Jameson, D. L.; Anand, R. J. Chem. Ed. 2000, 77, 88. (b) Nielson, R. M.; McManis, G.; Golovin, M. N.; Weaver, M. J. J. Phys. Chem., 1988, 92, 3441. (c) Yang, E. S.; Chan, M. S.; Wahl, A.C. J. Phys. Chem. 1980, 84, 3094. (d) Soper, J. D.; Mayer, J. M. J. Am. Chem. Soc. 2003, 125, 12217. II - 30 25 Larsen, D. W.; Wahl, A. C. J. Phys. Chem. 1965, 43, 3765. 26 Chan, M. S.; Deroos, J. B.; Wahl, A. C. J. Phys. Chem. 1973, 77, 2163. 27 Slutter, C. E.; Langen, R.; Sanders, D.; Lawrence, S. M.; Wittung, P.; Di Bilio, A.; Hill, M. G.; Fee, J. A.; Richards, J. H.; Winkler, J. R.; Malmström, B. G. Inorg. Chim. Acta 1996, 243, 141. 28 Lappalainen, P.; Watmough, N. J.; Greenwood, C.; Saraste, M. Biochemistry 1995, 34, 5824. 29 Jensen, M. R.; Hansen, D. F.; Led, J. J. J. Am. Chem. Soc. 2002, 124, 4093. 30 Courtin, v. A.; von Tobel, H. R.; Auerbach, G. Helv. Chim. Acta. 1980, 63, 1412. 31 Chảvez, I.; et. al. J. Organomet. Chem. 2000, 601, 126. 32 So, J. H.; Boudjouk, P. Inorg. Chem. 1990, 29, 1592. The precise number of equivalents of THF contained in this starting material was determined by microanalysis. 33 Immoos, C.; Hill, M. G.; Sanders, D.; Fee, J. A.; Slutter, C. E.; Richards, J. H.; Gray, H. B. J. Biol. Inorg. Chem. 1996, 1, 529. III - 1 Chapter III Luminescence Studies of Dicopper(I) Complexes Supported by Bis(2-diisobutylphoshinophenyl)amido Ligands † † Adapted from Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 2030. III - 2 III.A. Introduction Photoluminescent complexes have garnered much attention due to their possible utilization in electrochemical devices, as sensors and biological imaging agents, and in solar energy conversion schemes. 1 In this context, polypyridine-supported Cu(I) systems show promising features including the availability of low-lying charge-transfer (CT) excited states and the relatively low cost of copper in comparison to other transition metal luminophores. However, their tendency to display weak emission and short-lived excited states is problematic. 2 McMillin has underscored these collective points, 3 and his group recently reported a fascinating mononuclear complex, [Cu(dmp)(POP)]+ (dmp = 2,9dimethyl-1,10-phenanthroline; POP = bis[2-(diphenylphosphino)phenyl]ether), that exhibits both a relatively high quantum yield and long-lived excited-state as compared to other polypyridine-Cu(I) systems. 4 Incorporation of a bulky, bis(phosphine) chelate appears to create a rigid environment around the copper center which (i) suppresses solvent-induced exciplex formation, and (ii) limits problematic ligand dissociation from the excited state. Herein we describe an amido-bridged bimetallic copper system, {(PNP)CuI}2 (3.2), derived from a chelating bis(phosphine)amide ligand, Figure III.1 ((PNP) = bis(2(diisobutylphosphino)phenyl)amide). This species is an exceptional luminophore in its own right. Aside from the absence of supporting polypyridine ligands, its combined quantum yield (φ > 0.65) and lifetime (τ > 10 μs), in combination with its dinuclear structure and redox behavior, are without precedent. The synthesis of 3.2 was motivated by our recent elucidation of the dinuclear, thioether-supported Cu2N2 complex {(SNS)CuI}2 ((SNS)− = bis(2-tert-butylsulfanylphenyl)amide). 5 {(SNS)CuI}2 exhibits a III - 3 reversible 1e oxidation to form a mixed-valence [{(SNS)Cu1.5}2]+ complex, and - electrochemical and XRD studies establish minimal structural reorganization between these two redox partners. While {(SNS)CuI}2 is negligibly emissive at 298 K, its phosphine congener 3.2 emits strongly in solution and in the solid-state when irradiated by visible light. Figure III.1: Left, Molecular representation of 3.2. Center, Space-filling representation of 3.2 from crystal coordinates. Right, Displacement ellipsoid representations (50%) of the core atoms of 3.2. III.B. Results and Discussion The required tridentate PNP-H ligand, 3.1, was synthesized by addition of nBuLi to bis(2,2'-difluorophenyl)amine in THF followed by the addition of lithium diisobutylphosphide. 6 Heating this mixture at 45 °C for 4 days and subsequent passage of the crude product through silica gel afforded amine 3.1 as a spectroscopically pure viscous oil (75% yield), Figure III.2. Deprotonation with nBuLi affords [PNP][Li], {[3.1]Li}2, in good yield. Related bis(phosphino)amido ligands were first introduced by Fryzuk and have received the attention of several groups more recently., 7 , 8 {[3.1]Li}2 reacts rapidly with CuBr·Me2S in diethyl ether to generate a luminescent yellow solution. III - 4 The neutral, diamagnetic copper complex 3.2 can be subsequently isolated in pure form by crystallization (92%). Alternatively, complex 3.2 can be generated in good yield by the addition of {(2,4,6-Me3C6H2)Cu} 9 to free amine 3.1, producing mesitylene upon metallation. i Bu i n F NH F 1) 1 eq. BuLi / THF 2) 3 eq [iBu2P][Li], 50°C P NH P 3) MeOH i i Bu i P Cu N i Bu i P Bu i CuBr•Me 2S Bu Bu Bu Bu (3.2) i P Li N i Bu i Bu i P ether -LiBr, -Me 2S i Bu i N Cu i P i Bu P Bu 1.6 M n BuLi pet. ether 25°C Ether -Mesitylene Bu i Bu (3.1) Cu i Bu P Bu N Li i P Bu i Bu Bu ({[3.1]Li}2) Figure III.2: Synthesis of 3.1, {[3.1][Li]}2, and 3.2. Complex 3.2 is characterized by a single resonance in its 31P NMR spectrum (−33.9 ppm) and XRD analysis establishes the dimeric structure represented in Figure III.1. A short Cu⋅⋅⋅Cu bond distance of 2.6245(8) Å is observed that is similar in length to the Cu⋅⋅⋅Cu distance reported for {(SNS)CuI}2. The P-Cu-P angles of 132.93(5)º and 137.69(5)º are significantly smaller than the S-Cu-S angles in {(SNS)CuI}2 (avg 153º). III - 5 As a result the copper centers are less severely distorted from a tetrahedral geometry than in {(SNS)CuI}2, which more closely approximates a cis-divacant octahedron at each copper site. These geometrical differences presumably arise from the different steric requirements of the thioether [SNS]− and diisobutylphosphine [PNP]− ligands. Figure III.3: Cyclic voltammetry of 3.2 referenced vs. Fc+/Fc in CH2Cl2 (0.10 M [nBu4N][PF6], 250 mV/sec). Electrochemical analysis of 3.2 in CH2Cl2 (Fc+/Fc, 0.3 M [nBu4N][PF6], 250 mV/sec, Fc = ferrocene) reveals two reversible waves, one centered at −550 mV and the other at 300 mV, Figure III.3. An irreversible wave is encountered at higher potential (Epa = 860 mV). 10 The event at −550 mV is assigned to a reversible Cu1.5Cu1.5/CuICuI redox process by analogy to the {(SNS)CuI}2 system. The Cu1.5Cu1.5/CuICuI event is cathodically shifted for 3.2 in comparison to {(SNS)CuI}2 (by ~ 160 mV) due to its stronger III - 6 phosphine donors. The second reversible wave observed for 3.2 (E½ = 300 mV) is noteworthy and is distinct from {(SNS)CuI}2, for which only an irreversible redox process is observed at similar potential (Epa= 560 mV). Incorporation of the phosphine donors appears to stabilize an unusual second oxidation event in 3.2, at least on the time scale of the electrochemical experiment (250 mV/s). While it is tempting to assign this second wave to a CuIICuII/Cu1.5Cu1.5 redox process, at this stage it is equally plausible to suggest that a ligand-centered oxidation process is operative. 11,12 Figure III.4: Left (A): Absorption spectra for 3.1 and 3.2 in cyclohexane. Right (B): Corrected emission spectrum (green, λex = 440 nm) and excitation spectrum (black, λem = 560 nm) of 3.2 in cyclohexane. Inset (C): Luminescence decay measurement of 3.2 in cyclohexane (laser pulse at t = 0 μs, λex = 440 nm, λem = 510 nm). The absorption spectra for ligand 3.1 and complex 3.2 are shown in Figure III.4(A), and the corrected emission and excitation spectrum for 3.2 (298 K in cyclohexane) is also shown (B). 13 The optical spectrum of 3.2 is typical for Cu(I). MLCT bands at 23,800 and −1 22,300 cm III - 7 give rise to its yellow color. Its emission spectrum, collected by excitation into its lowest energy absorption band (λex = 440 nm, ≈ 22,700 cm−1), shows a λmax at 20,000 cm−1. Its corresponding excitation profile is also shown (λem = 560 nm, ≈17,900 cm−1). Low temperature emission studies at 77 K of 3.2 were also attempted for a frozen methylcyclohexane glass and on a single crystal of 3.2 under He, Figure III.5. While some enhanced vibrational structure was observed by going to lower temperature, no conclusions regarding the nature of the excited state could be reached. Interestingly, a high energy emission band at 480 nm was found in the sample of 3.2 in a methylcyclohexane glass, but not for those measurements done on the single crystal. Figure III.5: Low temperature emission of 3.2. Black: solution of 3.2 in cyclohexane (298 K); Blue: single crystal of 3.2 under He(g) (77 K); Red: frozen glass solution of 3.2 in methylcyclohexane (77 K). The intensity of the emission from the excited state of 3.2, *3.2, is quite striking to the eye, even at room temperature in relatively polar donor solvents such as tetrahydrofuran (THF). This property is consistent with the unusually high quantum yield III - 8 we measured for 3.2 at 298 K: φ = 0.68(2) in cyclohexane and φ = 0.67(4) in THF. These quantum yields were determined by established methods using a fluorescein standard (φ = 0.90 in 0.1 N NaOH). 14 It was also of interest to determine the excited-state lifetime of *3.2, measured as 10.2(2) μs in cyclohexane (see inset in Figure III.4) and 10.9(4) μs in THF. These lifetimes were determined by a monoexponential fit to raw decay data collected at 510 nm upon excitation at 460 nm. Complex 3.2 is thus a remarkably efficient luminophore, with a lifetime similar to that McMillin has reported for mononuclear [Cu(dmp)(POP)]+ and a quantum yield that is approximately four times greater. We also note that diffusion-limited excited state electron transfer (kQ = 1.2 x 1010 M−1s−1) has been demonstrated by time-resolved quenching experiments using 2,6dichloroquinone (Graph III.1). 15 6.E+06 5.E+06 y = 1.205E+10x + 1.894E+05 R2 = 9.947E-01 -1 kobs (s ) 4.E+06 3.E+06 2.E+06 1.E+06 0.E+00 0.0E+00 8.0E-05 1.6E-04 2.4E-04 M 3.2E-04 4.0E-04 4.8E-04 Graph III.1: Time-resolved emission quenching of 3.2 with DCQ in THF. Emissive dinuclear copper systems have been reported previously, but these species typically feature much shorter excited-state lifetimes. Perhaps most structurally related to 3.2 is the neutral complex {[DPT]CuI}2 (DPT = 1,3-triphenyltriazine anion), which features a Cu···Cu distance of 2.451(8) Å. 16 This neutral d10-d10 system exhibits a III - 9 fluorescence maximum at 570 nm and a lifetime of only 2.23 ns at 77 K, 17a approximately three orders of magnitude shorter than 3.2. Other CuI2 luminophores that have been described are typically supported by polypyridine type ligands and have rather long Cu···Cu distances.b Quantum yields for these systems are small in magnitude by comparison to their well-studied, substituted Cu(phen)2+ analogues.b The unusual emission properties exhibited by 3.2 may be due to several factors. Foremost among these may be the relatively low structural reorganization between 3.2 and *3.2. This assertion is at least consistent with the relatively narrow full-width at halfmaximum of its emission band shown in Figure III.4 (B) (2400 cm−1), which can be converted to an estimate of the total reorganization energy λ = 2600 cm-1. Also, steric protection afforded by the bulky PNP ligand, in addition to the absence of a net cationic charge for 3.2, removes the possibility of anion binding and likely renders the excitedstate complex resistant to donor solvent ligation. Each of these factors can otherwise contribute to undesirable exciplex quenching. Lewis acidic Cu(I) cations, such as Cu(phen)2+ systems, suffer from exciplex quenching due to solvent and/or counter-anion binding in the excited state. The space-filling model of 3.2 shown in Figure III.1 reveals just how effectively the copper sites are shrouded by the surrounding phosphine ligand framework. Table III.1 Comparison of bond lengths and angles between the determined by X-ray diffraction for 3.2 and those calculated by DFT. Interatomic Distances (Å) Cu1-Cu2 Cu1-N1 Cu1-N2 Cu2-N1 Cu2-N2 X-ray DFT 2.6245(8) 2.127(4) 2.191(4) 2.179(4) 2.219(4) 2.762 2.208 2.248 2.239 2.341 Interatomic Angles (deg.) Cu1-N1-Cu2 Cu1-N2-Cu2 N1-Cu1-N2 N1-Cu2-N2 X-ray DFT 75.10(12) 73.03(12) 107.20(14) 104.44(14) 76.78 73.98 106.66 102.56 III - 10 Figure III.6: Geometry optimization and electronic structure calculation for 3.2 using DFT (JAGUAR: B3LYP/LACVP**). Contour plot (value = 0.0905, 0.088, 0.065) of the highest lying molecular orbitals are shown. A geometry optimization and electronic structure calculation of 3.2 using DFT (JAGUAR 5.0, B3LYP/LACVP**) was performed using the crystallographically determined X-ray coordinates as the initial guess for the geometry (Figure III.6). 18 The theoretically determined structure agrees well with that determined experimental by X-ray diffraction (Table III.1). The electronic structure of 3.2 afforded by the DFT calculation suggests that the redox active HOMO (Figure III.6) contains significant orbital contributions from the four atoms of the Cu2N2 diamond core. The HOMO is antibonding with respect to each of the four Cu-N interactions and also the Cu-Cu interaction. This orbital configuration is consistent with the {(SNS)Cu}2 system (Chapter II). The LUMO is found to be virtually devoid of metal character and almost completely associated with the aryl ring structure. This result could indicate that the excited state involves promotion of an electron from the Cu2N2 core to the outer ligand framework. As a final point of interest, we note that a value for E00 = 2.6 eV can be estimated from the intersection of the emission and excitation profiles of 3.2 (Figure III.7). III - 11 Subtracting this value from the reversible Cu1.5Cu1.5/Cu1Cu1 redox couple provides an estimated value of −3.2 V (vs. Fc+/Fc) for the excited state reduction potential of *3.2. This estimated potential is unusually low, and it is possible that *3.2 will prove to be a potent photoreductant/photosensitizer. 19 Indeed, given the presence of two reversible redox couples within this bimetallic copper system, there may be an opportunity to photochemically drive two-electron reaction processes. 20 *{(PNP)Cu} 2 λ max= 440 nm ε = 3600 h υ em -3.2 V τ =10.9(4) μs φ = 0.67 ± 5% E°° = 2.6 eV h υ abs {(PNP)Cu}2 -0.56 V [{(PNP)Cu}2 ]+ Figure III.7: Diagram of the photophysical and redox properties of 3.2. III - 12 III.C. Experimental General. All manipulations were carried out using standard Schlenk or glove-box techniques under a dinitrogen atmosphere. Unless otherwise noted, solvents were deoxygenated and dried by thorough sparging with N2 gas followed by passage through an activated alumina column. Non-halogenated solvents were tested with a standard purple solution of sodium benzophenone ketyl in tetrahydrofuran in order to confirm effective oxygen and moisture removal. Spectral grade THF was purchased from Aldrich and distilled from molten potassium prior to use. All reagents were purchased from commercial vendors and used without further purification unless otherwise stated. A volumetric solution of 0.1 N NaOH was purchased from J. T. Baker and used as received. [Cp2Fe][B(C6H3(CF3)2)4], 21 was prepared according to literature procedure. Elemental analyses were performed by Desert Analytics, Tucson, AZ. Deuterated solvents were purchased from Cambridge Isotope Laboratories, Inc., degassed, and dried over activated 3 Å molecular sieves prior to use. X-ray diffraction studies were carried out at the Beckman Institute Crystallographic Facility on a Brüker Smart 1000 CCD diffractometer and solved using SHELX v. 6.14. Electrochemistry. Electrochemical measurements were carried out in a glove-box under a dinitrogen atmosphere in a one-compartment cell using a BAS model 100/W electrochemical analyzer. A glassy carbon electrode and platinum wire were used as the working and auxiliary electrodes, respectively. The reference electrode was Ag/AgNO3 in THF. The ferrocene couple Fc+/Fc was used as an external reference. Solutions (THF) of electrolyte (0.35 M tetra-n-butylammonium hexafluorophosphate) and analyte were also prepared under an inert atmosphere. III - 13 Spectroscopic measurements. High-resolution EI mass spectroscopy was carried out by the Caltech Chemistry Mass Spectral Facility using a JEOL JMS600. A Varian Mercury-300 or INOVA-500 NMR spectrometer was used to record 1H, 13C, 19F, 31P NMR spectra at ambient temperature. 1H and 13C chemical shifts were referenced to the residual solvent peaks. 19F and 31P NMR chemical shifts were referenced to external hexafluorobenzene (δ = -165 ppm) and phosphoric acid (δ = 0 ppm), respectively. Emission spectra were recorded on a Spex Fluorolog-2 spectro-fluorometer. Excitation for the luminescence lifetime experiments employed 8 ns pulses (at a repetition rate of 10 Hz) from a Nd:YAG laser pumped OPO (Quanta Ray Pro, Spectra Physics).. The luminescence was dispersed through a monochromator (Instruments SA DH-10) onto a photomultiplier tube (PMT) (Hamamatsu R928). The PMT current was amplified and recorded with a transient digitizer (Tektronix). UV-vis measurements were taken on a Cary 50 UV/Vis Spectrophotometer using a 1 cm quartz cell or 500 UV/Vis/NIR Spectrophotometer using either a 2 cm or 1 cm quartz cell sealed with a Teflon stopper. Synthesis of Lithium Diisobutylphosphide. In a 500 mL Erlenmeyer flask diisobutylphosphine (25 g, 0.171 mol) was dissolved in 200 mL of petroleum ether and cooled to -80ºC, at which time at 1.6 M solution of nbutyl lithium in hexane (107 mL, 0.171 mol) was added over 20 min. The reaction was then stirred at ambient temperature for 24 h, concentrated in vacuo to ca. 50 mL, and the white solids were then collected on a sintered-glass frit. Washing of the solids with petroleum ether afforded a single phosphorous containing product (21.1 g, 81%) as by 31P NMR upon drying. 31 P{1H} NMR (121.5 MHz, THF): -91.2. III - 14 Synthesis of Bis(2-(diisobutylphosphino)phenyl)amine, 3.1. In a 250 mL sealable reaction bomb, a 1.6 M n butyl lithium solution (7.9 mL, 12.6 mmol) in hexanes was added dropwise to a solution of bis(2-fluorophenyl)amine (2.46 g, 12.0 mmol) in THF (20 mL). After stirring for 15 min, the solution was concentrated in vacuo to remove the majority of the reaction volatiles after which time a solution of lithium diisobutylphosphide (5.47 g, 36 mmol) in THF (40 mL) was added and the vessel was sealed with a Teflon plug. The reaction was heated at 45ºC for 4 days and was monitored by 19F NMR for the complete disappearance of the aryl fluoride resonance. The reaction was then quenched with methanol (5 mL) and the solution became yellow in color. Petroleum ether (50 mL) was added and the mixture was filtered twice through Celite to remove solids. Removal of the solvent in vacuo afforded an orange oil which was diluted in petroleum ether (30 mL) and flashed through two plugs of silica gel in a 60 mL sintered-glass frit. Evaporation of the solvent under reduced pressure afforded a spectroscopically pure, pale green oil (4.10 g, 75%). 1 H NMR(300.1 MHz, CDCl3): δ 8.03 (t, 1H), 7.52 (m, 2H), 7.33 (m, 2H), 7.26 (t, 2H), 7.00 (t, 2H), 1.73 (m, 12H), 1.08 (d, 12), 1.03 (d, 12H). 13C{1H} NMR(75.5 MHz, CDCl3): δ 147.8, 131.8, 129.4, 128.0, 121.0, 119.3, 116.7, 39.4, 26.6, 24.7, 24.3. 31P{1H} NMR (121.5 MHz, CDCl3): δ -54.5. UV-vis (benzene, nm(M-1cm-1)): 302 (18,700), sh 334(5900). FAB+ MS: calcd for C28H45NP2: 457.3027. Found: 458.3122 [M+H], 400.2226 [M-iBu], 312.1904 [M-(iBu2P)]. Synthesis of {[PNP]Li}2, {[3.1]Li}2. At ambient temperature a 1.6 M nbutyl lithium solution (4.0 mL, 6.4 mmol) in hexanes was added dropwise to a solution of 3.1 (2.63 g, 5.76 mmol) in petroleum ether (50 mL) over 15 min. The reaction was stirred for 30 min, III - 15 at which time a solid began to precipitate. The solution was concentrated to ca. 30 mL and cooled to -30 °C for 12 h. The resultant solids were collected on a sintered-glass frit as fine pale yellow powder, dried thoroughly (2.24 g, 84 %). 1 H NMR(499.9 MHz, C6D6): δ 7.21 (m, 2 H), 7.06 (m, 4 H), 6.71 (m, 2 H), 1.8-0.6 (br m, 36 H). 13 C{1H} NMR(125.7 MHz, C6D6): δ 170.6, 132.4, 131.2, 128.7, 128.1, 118.6, 42.3, 36.2, 26.3, 25.3. 31 P{1H} NMR (121.5MHz, C6D6): δ -49.4 (q, 61 Hz). UV-vis (cyclohexane, nm(M-1cm-1)): 301(18,700), 357(31,400), sh 395(10,200). Anal. Calcd. for C56H88Li2N2P4: C, 72.55; H, 9.57; N, 3.02. Found: C, 73.21; H, 9.43; N, 3.14. Synthesis of {(PNP)Cu}2, 3.2: A solution of {[3.1]Li}2 (1.0 g, 2.16 mmol) in diethyl ether (20 mL) was added to a slurry of CuBr⋅S(CH3)2 (0.466 g, 2.27 mmol) in ether (30 mL) and stirred for 12 h. The solvent was removed in vacuo and the resultant yellow solids were dissolved in petroleum ether (50 ml) and filtered to remove insoluble materials. Removal of petroleum ether and drying the yellow solids in vacuo afforded analytically pure material (1.04 g, 92%). Crystals suitable for X-ray diffraction were obtained both by slow-evaporation of a petroleum ether solution of 3.2. 1 H NMR(499.9 MHz, CD2Cl2): δ 7.20 (m, 4H), 6.93 (m, 4H), 6.71 (br d, 4H), 6.61 (t, 4H), 1.806 (m, 4H), 1.58 (m, 8H), 1.39 (m, 8H), 1.29 (m, 4H), 0.99 (d, 12H), 0.73 (d, 12H), 0.70 (d, 12H), 0.60 (d, 12H). 13C{1H} NMR(125.7 MHz, CD2Cl2): δ 169.4, 131.8, 130.5, 128.7, 125.2, 117.8, 40.2, 36.4, 26.3, 26.2, 26.0, 25.7, 25.4, 25.1. 31P{1H} NMR (121.5MHz, C6D6): δ -33.9. UV-vis (cyclohexane, nm(M-1cm-1)): 298(19,300), 314(19,900), 352(41,000), sh 387(14,300), 425(5100), sh 454(3400). Anal. calcd. for C56H88Cu2N2P4: C, 64.65; H, 8.53; N, 2.69. Found: C, 64.54; H, 8.25; N, 2.62. III - 16 Quantum yield experiments. A volumetric solution of 3.2 (10 μM) in either cyclohexane (n = 1.426) 22 or tetrahydrofuran (n = 1.407) was prepared in a nitrogen-filled glove-box. Three cuvettes (1 cm path) were charged with this solution, sparged briefly with argon, and sealed with a greased ground-glass stopper. The absorption spectra were acquired both before and after fluorescence measurements to ensure the sample was not degrading. A solution of fluorescein in an aqueous 0.1 N NaOH solution was prepared and sparged with argon, the concentration was adjusted such that the optical density (OD) at 440 nm was the same as that of the individual solutions of 3.2. Fluorescent measurements were performed with λex = 440 nm at 298 K and corrected for detector response. The area under the curve of the emission spectrum was determined using standard trapezoidal integration methods. Quantum yields (Table III.1) were then calculated by the methods described by Demas and Crosby 23 using equation III.1. Q = (QR)(I / IR)(ODR / OD)(n2 / nR2) (III.1) Q: quantum yield of the sample. QR: quantum yield of fluorescein in aqueous 0.1 N NaOH solution (QR = 0.9). I: integrated intensity of 3.2. I R: integrated intensity of fluorescein sample. ODR: optical density of the fluorescein sample in absorption units. OD: optical density of 3.2 in absorption units. n: index of refraction of the solvent in which 3.2 was dissolved. nR: index of refraction of 0.1N NaOH solution (nR = 1.3351), measured on a Bausch & Lomb refractometer. III - 17 Table III.2: Data for Quantum Yield Measurements. Solution OD (measured) I (measured) Q (calculated) Fluorescein in 0.1N NaOH 0.0398 2.4725E+09 10 μM 3.2 in cyclohexane 0.0352 1.5000E+09 0.70 10 μM 3.2 in cyclohexane 0.0369 1.5067E+09 0.67 10 μM 3.2 in cyclohexane 0.0408 1.6423E+09 0.67 Average 0.68 std. deviation 0.02 Q (calculated) Solution OD (measured) I (measured) Fluorescein in 0.1N NaOH 0.0119 6.5402E+08 10 μM 3.2 in tetrahydrofuran 0.0174 6.6049E+08 0.69 10 μM 3.2 in tetrahydrofuran 0.0193 6.6209E+08 0.62 10 μM 3.2 in tetrahydrofuran 0.0137 5.3344E+08 0.71 Average 0.67 std. deviation 0.04 Lifetime measurements. A solution of 3.2 (10 μM) in either cyclohexane or tetrahydrofuran was prepared in a nitrogen-filled glove-box. The cuvettes (1 cm path) were charged with this solution, sparged briefly with argon, and sealed with a greased ground-glass stopper. The absorption spectra were acquired both before and after the fluorescence measurements to ensure the sample was not degrading. Fluorescent measurements were performed with λex =460, λem= 510 nm at 298 K. A 500 nm lowpass filter was placed in front of the PMT in order to eliminate noise due to scattered laser light. The emission decay was averaged over 50 laser pulses and fit to an exponential function from which kobs was determined (see Table III.3). III - 18 Table III.3: Data for Excited State Lifetime Measurements. Entry 1 2 3 Solution 10 μM 3.2 in tetrahydrofuran 10 μM 3.2 in tetrahydrofuran 10 μM 3.2 in tetrahydrofuran kradiativea(average) knon-radiativeb (average) Entry 4 5 Lifetime (1/ kobs) (μs) 11.15 11.15 10.53 10.94 0.4 6.22E+04 2.93E+04 Solution c 10 μM 3.2 in cyclohexane 10 μM 3.2 in cyclohexane kradiativea (average) knon-radiativeb (average) kobs (s-1) 8.9683E+04 8.9656E+04 9.4984E+04 average std. deviation kobs (s-1) 9.6675E+04 9.9590E+04 average std. deviation Lifetime (1/ kobs) (μs) 10.34 10.04 10.19 0.2 6.58E+04 3.24E+04 (a) kradiative = (Quantum Yield)/(Lifetime); (b) knon-radiative = kobs - kradiative (c) indicates the representative fit shown below Graph III.2: Fit of the Excited State Decay with Residuals for Entry 4, Table III.3. III - 19 Time-resolved luminescent quenching experiments. A solution of 3.2 (20 μM) in tetrahydrofuran was prepared in a nitrogen-filled glove-box. Two cuvettes (1 cm path) were charged with 2 mL of this solution, sparged briefly with argon, and sealed with a rubber stopper. The initial emission decay of 2 in each cuvette was measured with λex = 440 nm and λem= 500 nm prior to the introduction of the quencher. A solution of 2,6dichloroquinone (DCQ) (10 mM) was then sequentially added to the cuvettes via syringe in volumes listed in Table III.4, and the emission decay was measured. The combined data from the two runs is plotted in Graph III.2, and a first order rate constant of 1.2 x 1010 M-1s-1 was determined for the emission quenching. Table III.4: Time-resolved Emission Quenching Measurements. Run 1 μL of DCQ solution (total) 0 4 20 40 80 μM of DCQ in sample 0 20 99 200 380 kobs (s-1) 2.1E+05 4.2E+05 1.3E+06 2.4E+06 4.7E+06 μM of DCQ in sample 0 50 150 240 340 kobs (s-1) 1.1E+05 9.7E+05 2.0E+06 3.2E+06 4.4E+06 Run 2 μL of DCQ solution (total) 0 10 30 50 70 III - 20 III.D. References Cited 1 (a) Elliott, C. M.; Pichot, F.; Bloom, C. J.; Rider, L. S. J. Am. Chem. Soc. 1998, 120, 6781. (b) Bargossi, C.; Fiorini, M. C.; Montalti, M.; Prodi, L.; Zaccheroni, N. Coord. Chem. Rev. 2000, 208, 17. (c) de Silva, A. P.; Fox, D. B.; Huxley, A. J. M.; Moody, T. S. Coord. Chem. Rev. 2000, 205, 41. (d) Drummond, T. G.; Hill, M. G.; Barton, J. K. Nat. Biotech. 2003, 21, 1192. (e) Sutin, N.; Creutz, C. Pure & Appl.Chem. 1980, 52, 2717. 2 Ford, P. C.; Cariati, E.; Bourassa, J. Chem Rev. 1999, 99, 3625. 3 McMillin, D. R.; McNett, K. M. Chem. Rev. 1998, 98, 1201. 4 Cuttell, D. G.; Kuang, S.-M.; Fanwick, P. E. McMillin, D. R.; Walton, R. A. J. Am. Chem. Soc. 2002, 124, 6. 5 Harkins, S. B.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 2885. 6 Liang, L.-C.; Lin, J.-M.; Hung, C.-H. Organometallics 2003, 15, 3007. 7 (a) Fryzuk, M. D.; MacNeil, P. A.; Rettig, S. J.; Secco, A. S.; Trotter, J. Organometallics 1982, 1, 918. (b) Fryzuk, M. D.; Leznoff, D. B.; Thompson, R. C.; Rettig, S. J. J. Am. Chem. Soc. 1998, 120, 10126. 8 (a) Ozerov, O. V.; Guo, C.; Papkov, V. A.; Foxman, B. M. J. Am. Chem. Soc. 2004, 126, 4792. (b) Ozeroz, O. V.; Pink, M.; Watson, L. A.; Caulton, K. G.; J. Am. Chem. Soc. 2004, 126, 2105. 9 Eriksson, H.; Håkansson, M. Organometallics 1997, 16, 4243. 10 Low current signal is also evident at ~ -235 mV. 11 Bill, E.; Müller, J.; Weyhermüller, T.; Wiegardt, K. Inorg. Chem. 1999, 38, 5792. 12 Synthetic studies are underway to isolate and more thoroughly characterize the mono- and dicationic forms of the {(PNP)Cu}2n+ (n = 0, 1, 2) system. III - 21 13 Luminescence measurements were collected in the Beckman Institute Laser Resource Center at the California Institute of Technology by previously published methods. Wenger, O. S.; Henling, L. M.; Day, M. W.; Winkler, J. R.; Gray, H. B. Inorg. Chem. 2004 , 43, 2043. 14 Demas, J. N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991. 15 Maruyama, M.; Kaizu, Y. J. Phys. Chem. 1995, 99, 6152. 16 Brown, I. D.; Dunitz, J. D. Acta Crystallogr. 1961, 14, 480. 17 (a) Harvey, P. D. Inorg. Chem. 1995, 34, 2019. (b) Balzani, V.; Juris, A.; Venturi, M.; Campagna, S.; Serroni, S. Chem. Rev. 1996, 96, 759. 18 Geometry optimization and electronic structure calculation (JAGUAR 5.0, B3LYP/LACVP**) were performed on the complete structure of the cation of 3 assuming a doublet ground-state (no symmetry constraints applied; crystallographic coordinates of 3 were used as the HF initial guess). Jaguar 5.0, Schrodinger, LLC, Portland, Oregon, 2002. 19 Roundhill, D. M. In Photochemistry and Photophysics of Metal Complexes; Fackler, J. P., Jr.; Modern Inorganic Chemistry; Plenum Press: New York, NY, 1994; p. 49-55 and 165-210. 20 (a) Heyduk, A, F.; Nocera, D. G. Science 2001, 293, 1639. (b) Gray, H. B.; Maverick, H. W. Science 1981, 214, 1201. 21 Chávez, I.; et al. J. Organomet. Chem. 2000, 601, 126. 22 Lide, D. R., Ed. CRC Handbook of Chemistry and Physics, 77th Edition; CRC Press: New York, NY, 1996. 23 Demas, J. N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991