Coordinating Tectons 4: Coordination Chemistry of ... - ACS Publications

Article pubs.acs.org/Organometallics

Coordinating Tectons 4: Coordination Chemistry of the 4,5Diazafluoren-9-yl Moiety as a Metallo-Ligand for Allenylidene Complexes Phil A. Schauer,*,† Brian W. Skelton,‡ and George A. Koutsantonis* School of Chemistry and Biochemistry, M313, University of Western Australia, Crawley, Western Australia, 6009, Australia S Supporting Information *

ABSTRACT: We describe how the 4,5-diazafluoren-9-yl moiety has been utilized in the construction of multinuclear complexes incorporating a ruthenium(II) allenylidene functionality. The coordination chemistry of diazafluorenylterminated allenylidene complexes is limited by the sensitivity (instability) of the allenylidene moiety under a variety of synthetic conditions. In contrast the κ2-N,N′-coordination of the diazafluorenyl propargylic alcohol (alkynol) to a metal center prior to allenylidene formation provides a facile route toward the synthesis of multinuclear allenylidene coordination complexes. Our synthetic attempts and successes are discussed in combination with spectroscopic and electronic characterization of the latter cases.

■

fluoren-9-yl})]PF6 (M = Ru, Os; PP = dmpe, dppm, dppe; not all combinations),18 and in this investigation we describe efforts to synthesize multinuclear complexes incorporating 1, Scheme 1. Recent developments in metallocumulene chemistry have been reported.29 Contemporary with our investigations examples of the formation of bimetallic complexes by reaction of metal precursors with the closely related allenylidene

INTRODUCTION There exists a fundamental interest in the properties of multimetallic molecular complexes, to afford both understanding and applications in catalysis,1−3 molecular electronics,4−6 photochemistry,7,8 and combinations thereof.9−12 Of particular interest are those complexes in which the metal sites cooperate to yield synergistic properties largely absent in mononuclear analogues. To this end, molecules in which multiple metal sites are connected by conjugated organic bridges are of particular interest, as such bridging moieties are expected to foster increased degrees of interaction between the metal sites. The synergistic effects of the metal sites may then be readily tuned via synthetic modifications to the bridging moiety or to the coordination sphere of the pendant metal sites or by changing the identity of the metal atom(s).13 The prevalent examples of such complexes are those bearing symmetrical (oligo)ethynylene14,15 or polypyridyl bridging ligands.16,17 It is our goal to synthesize discrete molecular components that can be assembled into multinuclear complexes via a coordination chemistry methodology, i.e., coordinating tectons.18−20 Specifically we seek also to develop these tectons as asymmetric components in order to provide facile synthetic access to heterometallic complexes, and thus more readily elucidate and tune the effects of a particular metal site on the molecular properties of the complex as a whole. Our lab and others have reported on transition metal σ-alkynyl complexes derived from 5-ethynyl-2,2′-bipyridine and the successful κ2N,N′-complexation of additional d- and f-block metals to the pendant diimine moiety.19,21−28 We have previously reported the synthesis and characterization of the complexes [MCl(PP)2(CC{4,5-diaza© XXXX American Chemical Society

Scheme 1. Synthesis of the Allenylidene Complex 1

Received: July 1, 2015

A

DOI: 10.1021/acs.organomet.5b00569 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics complexes [RuCl(16-TMC)(CCC{2-py}2)]PF6 and [RuCl(dppe)2(CC{4,5-diazafluoren-9-yl})]OTf have appeared.30,31 However, our analogous reactions of 1 with metal precursors failed to afford isolable products. Instead, we were able to exploit the tritopic nature of the 9-hydroxy-9ethynyl-4,5-diazafluorene proligand, L1, to afford multimetallic complexes by coordination of the diimine prior to formation of the allenylidene moiety. The synthesis and characterization of several unique N,N′-coordinated metallo-ligands and the resulting multimetallic allenylidene complexes are reported herein.

■

intense red-orange. The reaction mixture was taken to dryness by rotary evaporation, redissolved in water (500 mL), and loaded onto a column of CM-Sephadex (60 mm × 600 mm). Elution was carried out with 0.01 M increments of NaCl(aq) to 0.1 M concentration, then 0.05 M increments from 0.1 to 0.4 M. A total of eight distinct bands were collected, with the third fraction (eluted at 0.04 M) identified as the desired product. This fraction was treated with a saturated aqueous solution of NH4PF6, extracted into CH2Cl2 (4 × 150 mL), dried over MgSO4, and taken to dryness by rotary evaporation. Recrystallization from MeCN/iPrOH afforded the pure compound as deep red-orange crystals (415 mg, 0.455 mmol, 78.4%). Anal. Calcd for C33H26F12N6O2P2Ru·H2O: C, 42.64; H, 2.82; N, 9.04. Found: C, 42.71; H, 2.85; N, 8.76. 1H NMR (600 MHz, CD3CN): δ (ppm) 8.53−8.45 (m, 4H, Hbpy), 8.16 (d, 2H, H1), 8.14−8.03 (m, 6H, Hbpy), 7.88−7.85 (m, 2H, Hbpy), 7.57 (d, 2H, H3), 7.52−7.46 (m, 4H, H2 and Hbpy), 7.42−7.39 (m, 2H, Hbpy), 5.48 and 5.47 (s, 1H, OH), 3.01 (s, 1H, Hα). 13C{1H} NMR (150 MHz, CD2Cl2): δ (ppm) 161.2 (m, C4′), 152.9 (m, C3), 142.5 (m, C1′), 134.1 (m, C1), 129.6 (m, C2), 80.4 (s, C9), 77.3 (s, Cβ), 75.2 (s, Cα), 159.0−158.3 (m, Cbpy quaternary), 153.8, 153.7−153.5, 138.9, 138.7, 128.7, 125.05−125.02, 124.99−124.97 (m, Cbpy). IR (cm−1): νH2O 3645 (w), νOH 3513 (br, w), νCC−H 3278 (w), νCC 2124 (w) and 1982 (w), 1740 (w), 1624 (w), 1604 (mw), 1570 (w), 1419 (s), 1243 (w), 1230 (w), 1162 (w), 1072 (w), υPF6− 840 (vs), 761 (m), 740 (m), 730 (m). ESI+-MS (MeCN): m/z 311 (100%, [M − (PF6−)2]2+), 767 (14%, [M − PF6−]+). FAB+-MS (MeCN): m/z 766.6 (23%, [M − PF6−]+). UV−vis (MeCN) λ (nm) [ε × 104 M−1 cm−1]: 208 [4.17], 244 [1.96], 254sh [1.65], 288 [6.16], 446 [1.23]. The remaining fractions obtained from chromatography were subjected to an identical workup procedure, although solid samples were obtained only in trace amounts not readily amenable to analysis. [Ru(bpy){κ2-N,N′-(L1)}2](PF6)2 (3). [RuCl2(η6-p-cymene)]2 (106 mg, 0.173 mmol) and bpy (51 mg, 0.33 mmol) were dissolved in N2purged EtOH (10 mL), and the solution was stirred anaerobically at ambient temperature for 21/2 h, over which time the solution took on a golden color. This solution was then added to a suspension of L1 (145 mg, 0.698 mmol) in N2-purged water (20 mL), and the combined mixture refluxed for 16 h. On cooling to ambient temperature, the bright red solution was filtered through a pad of Celite and taken to dryness by rotary evaporation. The red-orange residue was then chromatographed on a SiO2 column, eluting first with MeCN to remove excess organics, then a mixture of MeCN/H2O/ saturated aqueous KNO3 in a 190:10:1 ratio to yield a major yellow band. This fraction was concentrated to ca. 10 mL by rotary evaporation, and the addition of saturated aqueous KPF6 (10 mL) yielded the pure complex as bright red-orange microcrystals collected by filtration (145 mg). The remaining supernatant was extracted with CH2Cl2 (5 × 20 mL), dried over MgSO4, and evaporated to dryness, yielding a bright orange powder that was spectroscopically indistinguishable from the precipitated material (104 mg). Combined yield: 249 mg (0.258 mmol, 74.6%). Crystals suitable for single-crystal X-ray crystallography were grown by vapor diffusion of Et2O into a saturated Me2CO solution of the complex. Anal. Calcd for C36H24F12N6O2P2Ru·H2O: C, 44.05; H, 2.67; N, 8.56. Found: C, 44.26; H, 2.96; N, 8.20. 1H NMR (500 MHz, CD3CN): δ (ppm) 8.52−8.44 (m, 2H), 8.31−8.26 (m, 1H), 8.22−8.11 (m, 5H), 8.10− 8.03 (m, 2H), 7.91−7.87 (m, 1H), 7.85−7.81 (m, 1H), 7.62−7.57 (m, 2H), 7.57−7.43 (m, 6H), 5.61−5.52 (m, 2H, OH), 3.02−2.99 (m, 2H, Hα). 13C{1H} NMR (125 MHz, CD3CN): δ (ppm) 162.26, 162.25,

EXPERIMENTAL SECTION

General Considerations. Samples for infrared spectroscopy were prepared as a pellet (KBr) or Nujol mull (NaCl plates), and UV−vis spectra were obtained from solutions in quartz cuvettes. 1H, 13C{1H}, and 31P{1H} nuclear magnetic resonance spectra are referenced with respect to residual solvent signals32 or an external capillary of 85% H3PO4 for 31P NMR spectra. Mass spectra were acquired employing the fast-atom-bombardment (FAB) or electron-impact (EI) techniques. Elemental analyses were performed by Microanalytical Services, Research School of Chemistry, Australian National University, Canberra, Australia. Electrochemical measurements were performed with a 1 mm diameter Pt or GC disk working electrode, 1 mm diameter Pt counter electrode, and Ag/AgCl pseudoreference minielectrode. Solutions contained 0.1 M [nBu4N]PF6 and 5−10 mM analyte in MeCN or CH2Cl2 and were purged and maintained under an atmosphere of argon. Unless otherwise noted, scan rates for reported potentials were 100 mV s−1 and referenced to an internal [FcH]/[FcH]+ couple (E1/2(MeCN) = +0.40 V vs SCE, E1/2(CH2Cl2) = +0.46 V vs SCE).33 Column chromatography was performed using silica gel (230−400 mesh ASTM, Merck) as the stationary phase. Solvents for chromatography and general workup procedures were distilled prior to use, while solvents for Schlenk reactions were dried and purified by appropriate means34 prior to distillation and storage under an atmosphere of high-purity argon. Unless otherwise stated, all reactions were performed under an atmosphere of high-purity argon utilizing standard Schlenk techniques. We have previously reported the synthesis of compounds L1 and 1.18 The compounds cis-[RuCl2(dppm)2],35 [PdCl2(NCPh)2],36 [RuCl2(η6-p-cymene)]2,37 [RuCl2(bpy)2]·2H2O,38 and [RuCl2(CO)2]n39 were synthesized by literature procedures. All remaining reactants and reagents were purchased from commercial suppliers and used as received. Certain 13C{1H} NMR spectroscopic resonances are reported to two decimal places. Though not standard practice to imply this level of precision, the intent is to differentiate visibly unique resonances. Microanalytical data for complexes 2−4 are formulated with waters of hydration/crystallization, although absolute confirmation via 1H NMR spectroscopy was not possible on account of the residual H2O/ HOD signals persistent in deuterated NMR solvents (indeed, even with rigorous drying of solvent, resonances of the hydroxyl proton rarely integrated perfectly due to H/D exchange). As previously observed for the diazafluorenyl-allenylidenes,18 crystallization with CH2Cl2 is common, and this is reflected in the microanalytical data for complexes 6 and 7. Microanalytical data for complexes 4 and 6 admittedly lie outside the 0.4% range expected, with plausible impurities clearly evident in NMR spectra, although the data are nevertheless provided to illustrate the best values obtained to date; the identity of all complexes is confirmed by additional means. Crystallographic experimental details are provided in the Supporting Information, with crystal and refinement data collected in Table S1. Pertinent bond lengths and angles are included in the figure captions. [Ru(bpy)2{κ2-N,N′-(L1)}](PF6)2 (2). [RuCl2(bpy)2]·2H2O (302 mg, 0.580 mmol) and L1 (132 mg, 0.632 mmol) were combined in absolute ethanol (15 mL) and refluxed under argon for 5 h, during which time the deep purple color of the solution was replaced with an B


Article

Organometallics

(m, 2H), 8.45−8.41 (m, 2H), 8.28−8.25 (m, 2H), 8.20−8.14 (m, 2H), 8.08−8.02 (m, 2H), 7.90−7.86 (m, 2H), 7.74−7.70 (m, 2H), 7.51− 7.44 (m, 6H), 7.40−7.28 (m, 20H), 7.20−7.05 (m, 26H). 13C{1H} NMR (150 MHz, CD2Cl2): δ (ppm) 328.4 (m, Cα), 274.5 (m, Cβ), 166.6 (s, C4′), 151.9 (s, C3), 137.7 (m, Cγ), 133.4 (s, C1′), 130.7 (s, C1), 129.2 (s, C2), 42.6 (m, PCH2P). Cbpy and CPh: 158.4, 157.5, 153.2, 153.1, 138.6, 138.3, 132.9 (m), 132.8 (m), 132.1, 131.6, 131.6, 130.0 (m), 129.7 (m), 129.0 (m), 128.7, 127.9, 124.4, 124.4. 31P{1H} NMR (243 MHz, CD2Cl2): δ (ppm) −18.1 (m, PCH2P), −143.9 (sept, PF6−). IR (cm−1): νCCC 1910 (br, ms), νPF6− 840 (vs). ESI+-MS (Me2CO): m/z 1799 (12%, [M(PF6)2]+), 755 (100%, [M + H]2+), 503 (16%, [M]3+). UV−vis (MeCN) λ (nm) [ε × 104 M−1 cm−1]: 194 [16.4], 232sh [7.33], 286 [8.36], 456 [3.06]. Reaction of cis-[RuCl2(dppm)2] with 3. A bright orange solution of 3 (109 mg, 0.113 mmol) and cis-[RuCl2(dppm)2] (205 mg, 0.218 mmol) in CH2Cl2 (125 mL) was treated with NaPF6 (96 mg, 0.57 mmol) and stirred at ambient temperature for 23 days, during which time a dark red-orange color developed. The solution was filtered through Celite, concentrated to a minimum volume on a rotary evaporator, and then precipitated by dropwise addition to rapidly stirring Et2O (300 mL). The solid was subsequently chromatographed on silica utilizing 5% saturated KPF6/MeCN solution, with recrystallization of the main red-orange fraction from CH2Cl2/EtOH to afford the product (151 mg). 31P{1H} NMR (243 MHz, CD3CN): δ (ppm) −16 to −18 (br m, PCH2P), −143.21 (sept, PF6−). IR (cm−1): νCCC 1910 (br, ms), νPF6− 836 (vs). ES+-MS (CH2Cl2): m/z 1370.9 (9%, [(1)2Ru(bpy) (PF6)2]2+), 1296.0 (17%, [(1)2Ru(bpy) (PF6)]2+), 1224.7 (25%, [(1)2Ru(bpy)]2+), 865.1 (19%, [(4)2Ru(bpy) (PF6)]3+), 816.4 (100%, [(1)2Ru(bpy)]3+), 611.9 (51%, [(1)2Ru(bpy)]4+). Reaction of cis-[RuCl2(dppm)2] with 4. A bright orange solution of 4 (47 mg, 0.060 mmol) and cis-[RuCl2(dppm)2] (157 mg, 0.167 mmol) in CH2Cl2 (125 mL) was treated with NaPF6 (164 mg, 0.978 mmol) and stirred at ambient temperature for 26 days. Workup as for the above reaction with 3 resulted in the isolation of a red-orange microcrystalline product (86 mg). 31P{1H} NMR (243 MHz, (CD3)2CO)): δ (ppm) −16.5 to −17.5 (br m, PCH2P), −142.69 (sept, PF6−). IR (cm−1): νCCC 1888 (br, ms), νPF6− 841 (vs). ES+-MS (CH2Cl2): m/z 1179.1 (15%, [(1)3Ru(PF6)]3+), 1129.1 (12%, [(1)3Ru]3+), 847.8 (18%, [(1)3Ru]4+). [RuCl(η6-p-cymene){κ2-N,N′-(L1)}]PF6 (5). A flask was charged with [RuCl2(η6-p-cymene)]2 (74 mg, 0.12 mmol) and L1 (52 mg, 0.25

161.73, 161.67, 159.46, 159.49, 155.31, 155.28, 155.04, 154.99, 153.97, 153.90, 153.60, 153.54, 142.50, 142.46, 142.38, 142.27, 138.85, 134.34, 134.29, 134.03, 129.68, 129.18, 128.29, 128.25, 128.22, 124.84, 80.50, 80.47, 77.45, 75.10, 75.04. IR (cm−1): νH2O 3636 (m), νOH 3514 (m), νCC−H 3284 (ms), νCC 2120 (w) and 1969 (br, w), 1740 (w), 1717 (w), 1625 (w), 1604 (m), 1449 (vs), 1420 (vs), νPF6− 844 (vs). UV−vis (MeCN) λ (nm) [ε × 104 M−1 cm−1]: 206 [4.93], 245 [1.12], 290 [4.82], 444 [1.13]. [Ru{κ2-N,N′-(L1)}3]Cl2 (4). A mixture of “RuCl3·xH2O” (73 mg, 0.28 mmol for x = 3) and L1 (223 mg, 1.07 mmol) was dissolved in 5:1 EtOH/H2O (30 mL), and the solution sparged with N2 for 15 min, then refluxed under argon for 21 h. On cooling to ambient temperature, the dark red-orange solution was filtered and solvent removed under reduced pressure. The residue was redissolved in H2O (50 mL) and washed with CH2Cl2 (4 × 100 mL) and Et2O (2 × 50 mL). Solvent was removed from the aqueous solution by rotary evaporation to yield a deep orange oil. Repeated rotary evaporation from anhydrous MeCN yielded the hydrated product as a bright red crystalline film (203 mg) of 80+% purity. Anal. Calcd for C39H30Cl2N6O6Ru·3H2O: C, 55.06; H, 3.55; N, 9.88. Found: C, 54.39; H, 3.59; N, 9.83. 1H NMR (600 MHz, D2O): δ (ppm) 8.25− 8.20 (m, 2H, H1), 8.12−8.06 and 8.00−7.94 (m, 2H, H3), 7.61−7.52 (m, 2H, H2), 3.20−3.17 (m, 1H, Hα). 13C{1H} NMR (150 MHz, D2O): δ (ppm) 162.7 (m, C4′), 154.6 (m, C3), 140.8 (m, C1′), 134.0 (s, C1), 129.0 (m, C2), 79.5 (m, Cβ), 76.9 (m, C9), 75.8 (m, Cα). IR (cm−1): νH2O 3600−3000 (br, s), νCC 2114, (w) and 1980 (vs), 1717 (w), 1627 (m), 1602 (m), 1570 (w), 1419 (vs), 1287 (ms), 1229 (s), 1082 (s), 812 (s), 801 (s), 740 (vs), 700 (s). UV−vis (MeCN) λ (nm) [ε × 104 M−1 cm−1]: 204 [5.37], 248 [1.11], 298 [4.05], 438 [0.857]. [Ru{κ2-N,N′-(L1)}3](Cl) (PF6) (4b). A solution of 4 (54 mg, 0.064 mmol) in H2O (0.5 mL) was treated with saturated aqueous KPF6 (2 mL) and allowed to stand overnight. The bright orange precipitate was collected under argon and dried at 100 °C in vacuo for 60 h to yield the mixed chloride−hexafluorophosphate salt (61 mg, 0.063 mmol). Crystals suitable for single-crystal X-ray crystallography were obtained by slow diffusion of Et2O into a dilute Me2CO solution of the complex. IR (cm−1): νH2O 3640 (w), νOH 3518 (w), νCC−H 3288 (m), νCC 2123 (w) and 1987 (m), 1717 (w), 1628 (w), 1603 (w), 1572 (w), 1421 (vs), 1230 (ms), 1081 (ms), υPF6− 845 (vs), 810 (s), 801 (s), 740 (s). FAB+-MS (Me2CO): m/z 869.2 (37%, [Ru(L1)3(PF6)]+). ESI+-MS (Me2CO): m/z 363.0 (100%, [Ru(L1)3]2+). [(1)-κ2-N,N′-Ru(bpy)2](PF6)3 (6). A bright orange solution of 2 (53 mg, 0.058 mmol) and cis-[RuCl2(dppm)2] (52 mg, 0.055 mmol) in

mmol), then dissolved in CH2Cl2 (40 mL) to give a dark green suspension. This mixture was immediately treated with solid KPF6 (52 mg, 0.28 mmol) to yield a very dark orange-green suspension, which was stirred at ambient temperature for an hour. The reaction mixture was then filtered aerobically through Celite to yield a bright orange solution, from which solvent was removed under reduced pressure to give a brown oil. The oil was extracted into Me2CO (100 mL) and filtered, and the solution reduced in volume to a minimum (4 mL) and then precipitated by addition to rapidly stirring Et2O (250 mL). The precipitate was collected by filtration, washed with Et2O, and then airdried to yield the product as a bright orange powder (114 mg, 0.184 mmol, 75%). Anal. Calcd for C23H22ClF6N2OPRu: C, 44.28; H, 3.55; N, 4.49. Found: C, 44.01; H, 3.94; N, 4.37. 1H NMR (600 MHz, (CD3)2CO): δ (ppm) 9.29−9.26 (m, 2H, H3), 8.42−8.38 (m, 2H, H1), 7.88−7.82 (m, 2H, H2), 7.16 and 6.98 (s, 1H, OH), 6.32−6.28 (m, 2H, HD), 6.10−6.07 (m, 2H, HC), 3.35 and 3.26 (s, 1H, Hα), 3.00−2.91 (m, 1H, HF), 2.30 and 2.29 (s, 3H, HA), 1.25−1.22 (m,

CH2Cl2 (100 mL) was treated with NaPF6 (15 mg, 0.086 mmol) and refluxed for 22 h, during which time a dark red-orange color developed. The solution was cooled to room temperature, filtered through Celite, concentrated to a minimum volume under reduced pressure, and then precipitated by dropwise addition to rapidly stirring Et2O (120 mL). The solid was collected by filtration and dried under vacuum at 50 °C overnight to yield the product as a deep orange powder (96 mg) of 90+% purity. Anal. Calcd for C83H66ClF18N6P7Ru2· CH2Cl2: C, 49.73; H, 3.38; N, 4.14. Found: C, 49.88; H, 3.43; N, 4.67. 1 H NMR (600 MHz, CD2Cl2): δ (ppm) 7.64−7.61 (m, 2H, H3), 7.05−7.01 (m, 2H, H2), 5.66−5.62 (m, 2H, H1), 5.62−5.55 and 5.34−5.27 (m, obscured CHDCl2, PCH2P), Hbpy and HPh: 8.50−8.45 C


Article

Organometallics 6H, HG). 13C{1H} NMR (150 MHz, (CD3)2CO): δ (ppm) 160.5 and 160.2 (s, C4′), 154.2 and 154.0 (s, C3), 142.9 and 142.8 (s, C1′), 135.9 and 135.8 (s, C1), 129.6 and 129.5 (s, C2), 105.7 and 105.6 (s, HE), 101.6 and 101.5 (s, HB), 84.0 and 83.8 (s, HD), 82.64 and 82.57 (s, HC), 81.04 and 80.97 (s, Cα), 77.5 (s, Cγ), 75.3 and 75.1 (s, Cβ), 32.0 (s, CF), 22.32 and 22.27 (s, CG), 18.80 and 18.78 (s, CA). IR (cm−1): νCC−H and νOH 3629 (br, w) and 3550−3150 (br, w) and 3276 (w), νCC 2119 (w), 1630 (w), 1605 (m), 1576 (w), 1423 (ms), 1291 (w), 1279 (w), 1230 (mw), 1161 (br, w), 1081 (br, m), 1037 (w), 1017 (w), νPF6− 843, 740 (ms), 699 (w). FAB+-MS (thf): m/z 479.4 (100%, [M − PF6−]+). UV−vis (MeCN) λ (nm) [ε × 104 M−1 cm−1]: 194 [3.01], 206 [3.05], 316 [1.18], 332 [1.20]. [RuCl(η6-p-cymene){κ2-N,N′-(1)}](PF6)2 (7). A bright yellow suspension of 5 (40 mg, 0.065 mmol), cis-[RuCl2(dppm)2] (62 mg, 0.066 mmol), and NaPF6 (15 mg, 0.087 mmol) in CH2Cl2 (100 mL) was refluxed for 24 h, during which time the solution became a very dark orange. On cooling to ambient temperature, the solution was aerobically filtered through Celite and solvent removed under reduced pressure. The residue was redissolved in a minimum of CH2Cl2 (2 mL) and precipitated by dropwise addition to rapidly stirring Et2O (100 mL). The precipitate was collected by filtration and air-dried to yield the product as a dark red-orange powder (102 mg, 0.062 mmol, 94%). Anal. Calcd for C73H64Cl2F12N2P6Ru2·11/2CH2Cl2: C, 50.17; H, 3.79; N, 1.57. Found: C, 50.18; H, 4.01; N, 1.70. 1H NMR (600 MHz, CD2Cl2): δ (ppm) 9.01 (d, 2H, H3), 7.38 (m, 8H, ortho), 7.37 (m, 4H, para), 7.32 (m, 4H, para), 7.24 (m, 8H, meta), 7.22 (d, 2H, H2), 7.17 (m, 8H, ortho), 7.12 (m, 8H, meta), 5.99 (d, 2H, HD), 5.82 (d, 2H, HC), 5.64 (d, 2H, H1), 5.64−5.57 (m, 2H, PCH2P), 5.35−5.31 (m, 2H, PCH2P), 2.94 (sept, 1H, HF), 2.29 (s, 3H, HA), 1.28 (d, 6H, HG). 13C{1H} NMR (150 MHz, CD2Cl2): δ (ppm) 328.1 (pent, 2JPC = 14 Hz, Cα), 272.5 (t, 3JPC = 4 Hz, Cβ), 166.2 (s, C4′), 153.0 (m, C3), 137.2 (t, 4JPC = 1 Hz, Cγ), 133.5 (s, C1′), 132.9 (m, ortho), 132.7 (m, ortho), 132.4 (s, para), 132.0 (s, C1), 131.6 (s, para), 129.90 (m, ipso), 129.90 (s, meta), 129.0 (s, meta), 128.9 (m, ipso), 128.7 (s, C2), 104.5 (s, CE), 101.2 (s, CB), 83.7 (s, CD), 81.6 (s, CC), 42.3 (pent, 1 JPC = 13 Hz, PCH2P), 31.7 (s, CF), 22.2 (s, CG), 18.8 (s, CA). 31 1 P{ H} NMR (243 MHz, CD2Cl2): δ (ppm) −17.8 (br s, dppm), −143.8 (sept, PF6−). IR (cm−1): νCCC 1913 (br, m), 1603 (m), 1584 (w), 1572 (w), 1505 (w), 1436 (s), 1414 (m), 1309 (w), 1190 (w), 1161 (w), 1096 (ms), 1026 (w), 999 (w), νPF6− 837 (vs), 738 (w), 694 (w), 615 (w). FAB+-MS (Me2CO): m/z 1511 (11%, [M − PF6−]+), 1367 (45%, [M − PF6− + H+]+). UV−vis (MeCN) λ (nm) [ε × 104 M−1 cm−1]: 194 [16.3], 276 [4.61], 448 [1.69], 522sh [0.970].

We shifted our attention toward an alternate, complementary paradigm for the synthesis of multinuclear coordination compounds incorporating the diazafluorenyl allenylidene moiety: the coordination of a metal to the diimine moiety prior to formation of the allenylidene functional group. Coordination Chemistry of 9-Hydroxy-9-ethynyl-4,5diazafluorene (L1). L1 is a tritopic ligand whereby the alcohol, alkyne, and diimine functional groups provide potential reaction sites for a metal precursor. Complexes [PdCl2(NCR)2] (R = Me, Ph) readily substitute the nitrile ligands for bipyridyl derivatives to yield [PdCl2(N∩N)] complexes,43 which may react further in the presence of AgX halide abstracting agents to afford [Pd(N∩N)2]X2 (X = OTf−, OTs−, PF6−).44 However, the reaction of PdCl2(NCPh)2 with L1 under standard reaction conditions rapidly afforded an orange precipitate. The product showed marginal solubility only in DMSO, solutions of which rapidly turned green with accompanying deposition of an intractable black solid. The solid-state IR spectrum of the orange solid showed no absorptions attributable to nitrile moieties or to the hydroxyl or terminal alkyne functionality of L1. The bands at 1738 and 1716 cm−1 plus a medium-strong absorption at 1602 cm−1 are not readily assigned within the context of the reactants or targeted coordination complex, but the energy of these vibrations is within the appropriate range for Pd(0) π-alkyne complexes (e.g., [(dippe)Pd(η2-HCCR)], R = H, 1619 cm−1; R = Me, 1756 cm−1; R = Ph, 1720 cm−1; dippe = 1,2-bis{(diisopropyl)phosphine}ethane).45 A number of competing reaction pathways must be considered as possibilities given the rich alkyne chemistry of palladium and particularly that of “PdCl2” toward oligomerization and cyclization reactions.46−48 Attempts to crystallize complexes of L1 with a variety of simple metal salts (MxXy·nH2O; M = Cr−Zn; X = Cl−, SO42−, ClO4−; not all combinations) in aqueous solution or appropriate precursors in nonaqueous media (e.g., [Co(dmso)4](ClO4)2, [Cu(NCMe)4]PF6) were completely unsuccessful. Pursuing the synthesis of “traditional” ruthenium(II) diimine complexes analogous to the ubiquitous [Ru(bpy)3]2+ finally afforded success. It is relevant to note the particular examples of related diazafluorenyl complexes, namely, [Ru(4,5-diazafluoren9-one) n (bpy) 3−n ](PF 6 ) 2 (n = 1−3) 49,50 and [Ru(4,5diazafluorene)n(bpy)3−n](PF6)2 (n = 1 or 3),51 synthesized under standard conditions.41,42,52 Synthesis of [Ru(bpy)3−n{κ2-N,N′-(L1)}n]X2 (n = 1−3) Derivatives. The complex [Ru(bpy)2{κ2-N,N′-(L1)}](PF6)2 (2) was readily prepared through the reaction of RuCl2(bpy)2 and L1 in refluxing ethanol, with the complex isolated in 78% overall yield after chromatographic purification. To access the corresponding bis-L1 complex [Ru(bpy){κ2-N,N′-(L1)}2](PF6)2 (3) by an analogous synthetic procedure, we first attempted (unsuccessfully) to prepare RuCl2{κ2-N,N′-(L1)}2 through reaction of L1 with traditional precursors such as RuCl3·xH2O or [RuCl2(dmso)4] (Scheme 2). In the latter case, there was significant spectroscopic and spectrometric evidence for decomposition of the alkynol moiety to afford complexes instead incorporating the diazaf luorenone moiety (see below). Instead, treating an ethanolic solution of [RuCl2(η6-pcymene)]2 with bpy to form the intermediate complex [RuCl(η6-p-cymene)(bpy)]Cl, followed by refluxing with an aqueous suspension of L1, yielded the target complex [Ru(bpy)(L1)2](PF6)2 (3) as a diastereomeric mixture in 75% overall yield. The tris-complex [Ru(L1)3]Cl2 (4) was readily isolated through the reaction of L1 with RuCl3·xH2O in

■

RESULTS AND DISCUSSION Coordination Chemistry of [RuCl(dppm)2(4,5,-diazafluoren-9-yl)]PF6 (1). Complexation to cation 1 is hampered by the need to overcome the Coulombic repulsion experienced by the incoming metal center. This also needs to be balanced against the inherent reactivity of the cumulene C3 chain40 and the often harsher conditions required to labilize the existing ligands of the incoming metal complex to be coordinated at the diimine functionality of 1.41 Nevertheless, the examples of [RuCl(16-TMC)CCC(2-py)2{MLn}]+ (MLn = Ru(acac)2 and ZnCl2)31 and [RuCl(dppe)2CCC(4,5diazafluoren-9-yl){MLn}]m+ (MLn = ReCl(CO)3; Ru(bpy)2; Ru(dtbpy)2)30 provide evidence that such coordination is feasible.42 We have investigated the coordination properties of 1 in reactions with FeCl2, [Cu(NCMe)4]PF6, [RuCl2(dmso)4], and [Rh(μ-Cl)(COD)]2 but these reactions gave quantitative recovery of starting material under mild conditions or else intractable mixtures under a variety of more forcing conditions. Similarly, the reaction of 1 with metal precursors such as ReCl(CO)5, M(CO)6 (M = Cr, Mo), [RuCl2(CO)2]n, PdCl2(NCR)2 (R = Me, Ph), and PtCl2(SMe2)2 also gave intractable mixtures, under a variety of conditions. D


Article

Organometallics Scheme 2. Synthesis of the Metallo-Alkynol Precursors 2−4a

rationale accounting for both the (over)consumption of L1 and presence of carbonyl involves the formation and decomposition of an allenylidene, vinylidene, or acetylide complex stabilized by a derivative of “[RuXn(N∩N)]” or “[Ru(N∩N)2]”, whose reactions with oxygen give carbonyl cation byproducts.29,55 The decomposition of these postulated organometallic intermediates could also result in the formation of the ketone 4,5diazafluoren-9-one (L2), the presence of which is implicated in mass spectra of the residues of the reactions of L1 with RuCl3· xH2O or RuCl2(dmso)4. Infrared spectra of these crude reaction mixtures also show two strong absorptions at 1739 and 1719 cm−1, broadly consistent with inclusion of this ketone (cf. νCO 1716 cm−1 L2; 1741 cm−1 [Ru(bpy)2(L2)](PF6)2).50,56 The presence of L2 is not readily explained by other mechanisms, with the decomposition of propargylic alcohols to the parent ketone generally requiring highly caustic reaction media, such as saturated aqueous KOH or heated Al2O3.57 We postulate that the presence of water in the synthetic conditions hinders competitive reactivity of the propargylic alcohol functionality, through intermolecular hydrogen bonding, thereby facilitating κ2-N,N′-coordination of the diimine moiety to the ruthenium center. Stereochemistry of the Tris(diimine)ruthenium(II) Complexes. Complexes of L1 may exhibit numerous isomers on coordination due to the asymmetry of the alkynol moiety. These isomers are defined by the relationship between the alkynyl and hydroxyl functional groups of adjacent diimine ligands. As such, octahedral tris(diimine) complexes incorporating two or three L1 ligands may exist in three and two isomeric forms, respectively, as depicted in Figure 1 (in addition to the Δ and Λ optical isomers).

Conditions: (i) Ru(bpy)2Cl2, Δ EtOH, 78%; (ii) [RuCl(η6-pcymene)(bpy)]+, Δ EtOH/H2O, 75%; (iii) RuCl3·xH2O, Δ EtOH/ H2O, 86%. a

1:5 H2O/EtOH, affording the hydrated product [Ru(L1)3]Cl2· 3H2O (4) in ca. 85% overall yield as a mixture of diastereomers, although unidentifiable impurities (10−15%) were always evident through 1H and 13C NMR spectroscopy (Scheme 2). Treating a solution of 4 with aqueous KPF6 in an attempt to metathesize the chloride ligands leads to the isolation of an extremely hygroscopic mixed chloride−hexafluorophosphate complex, [Ru(L1)3](Cl)(PF6) (4b), whereas the dichloride salt 4 is not hygroscopic and thus more readily manipulated. The presence of water appears essential to the syntheses of both 3 and 4, whereby reactions in 95% or “absolute” ethanol afford a complex mixture of products as probed by both infrared and 1H NMR spectroscopy. In particular the observation of peaks consistent with monocationic carbonyl complexes [Ru(N∩N)2(CO)m(Cl)n]+ (m = 1, 2; n = 0−2) in mass spectra of the intractable reaction mixtures is salient. The presence of free carbon monoxide under the synthetic conditions employed (cf. synthesis via N,N-dimethylformamide reduction of RuCl3·xH2O) is unlikely.53,54 These carbonyl complexes are not observed as fragment ions in mass spectra of the pure, isolated complexes 3, 4, and 4b prepared via syntheses in which water was present and in the case of 4 remain persistent even when an excess of L1 is utilized. A possible

Figure 1. Schematic representation of the various isomers of the bisand tris-complexes. No ready system of nomenclature is applicable to these isomers, which will henceforth be distinguished by the labels depicted herein. E


Article

Organometallics

crystal lattice. The extended lattice structure shows sheets of the cations separated by sheets of the hexafluorophosphate anions, and, apart from some relatively short contacts between the bpy ligand and PF6− anion (C−H···F−P, 2.21 Å), as well as a significant intermolecular contact of the propargylic alcohol functionality (CCH···F−P, 2.68 Å), there are no other major inter- or intramolecular interactions of the cation. Various attempts to grow single crystals of the tris(alkynol) complex 4 as the dichloride salt were unsuccessful, although suitable single crystals were obtained on metathesis of the anion, giving the mixed chloride−hexafluorophosphate salt 4b (Figure 3). The

The resolution of such isomeric forms is by no means trivial; take for example the series of complexes [Ru(5-R-bpy)3]2+ (R = Me, Et, Pr, CH2iPr, CH2tBu, CO2Me), for which isomeric separation was achieved after extensive cation-exchange chromatography utilizing aqueous sodium hexanoate as eluent.58 In these complexes the different symmetry of the mer and fac isomers is expressed by differing hydrophobic interactions with the hexanoate anions, allowing the isomers to be (painstakingly) resolved. The geometric relationship between isomers in this case is expressed in a pseudolinear fashion, whereas for complexes of L1 the only geometrical differences at the alkynol moiety are essentially expressed in a spherical fashion; consequently the difference in molecular dipole moment between isomers is marginal in the latter case compared to the former. Although it may be conceptually possible to exploit some other feature of the alkynol moiety (e.g., chelation of cations between two neighboring hydroxyl or alkynyl groups), the development of such a technique for the resolution of diastereomers was not considered vital to this investigation. These complexes were synthesized as metalloalkynol precursors for allenylidene complexes, and the stereoisomerism at the sp3 “alkynol carbon” is lost on transformation to a sp2 allenylidene carbon. The solid-state structures of the tris(diimine)ruthenium complexes were investigated by means of single-crystal X-ray crystallography. Only a weakly diffracting sample of 2 was obtained (Figure S1), and a directional anisotropy in two of the pyridyl rings leads to a disorder that precludes a detailed structural analysis. Crystals of the bis(alkynol) complex 3 were more readily acquired, but again failed to diffract well despite numerous well-formed crystal samples being obtained. The structure obtained was exclusively that of the Λ enantiomer (Figure 2), with the hydroxyl and alkynyl groups at both coordinated diazafluorenyl ligands disordered over two sites (half-occupancy). This disorder is indicative of the three possible isomers being randomly distributed throughout the

Figure 3. Depiction of the cation present in [Ru(L1)3](Cl)(PF6)· 3H2O, 4b, projected approximately along the pseudo-3-fold axis (solvent atoms omitted for clarity, displacement ellipsoids at the 50% probability level). Selected bond distances (Å) and angles (deg): Ru− N (av) 2.101(5); C−Cα(Cn−Cn1) 1.46(2), 1.31(2), 1.41(1) (n = 1, 2, 3) Cα−Cβ(Cn1−Cn2) 1.19(2), 1.23(2), 1.14(2) (n = 1, 2, 3); Cn− On 1.38(2), 1.40(1), 1.37(1) (n = 1, 2, 3); C−Cα−Cβ(Cn−Cn1− Cn2): 174(1), 177(2), 179(1) (n = 1, 2, 3).

complex crystallizes exclusively as the Δ enantiomer of the ASYM isomer. In contrast to 3 no significant intermolecular interactions are evident between the cations, although short contacts between one of the diazafluorenyl ligands and the PF6− anion (C−H···F−P, 2.452 Å)59 and between the chloride anion and two hydroxyl groups of different cations (O−H···Cl, 2.31 Å, 2.20 Å)60 are evident. As a consequence of the disorder present in the structures of 2 and 3, it is somewhat difficult to make significant comparisons between the series of three complexes, although close scrutiny does develop some generalized observations. The complexes show a distorted octahedral bonding arrangement about the ruthenium center as expected on account of the asymmetry imposed by the coordinated diazafluorenyl ligand. Bond lengths and angles between ruthenium and the diimine ligands are comparable to literature analogues,41 although with both parameters larger for coordinated L1 than for bipyridine as a consequence of the increased bite angle of the diazafluorenyl ligand. While the free ligand and the complex 3 maintain a planar geometry across the diazafluorenyl moiety, the central five-membered rings of 4 are somewhat puckered, with the angle between planes of the individual pyridyl rings varying at 4.2(3)°, 5.8(3)°, and 9.7(3)° in each of the three chelated ligands.

Figure 2. Structure of the cation of [Ru(bpy)(L1)2](PF6)2, 3, projected approximately along the pseudo-3-fold axis. Only one component of each of the disordered atoms is shown (displacement ellipsoids at the 20% probability level). Selected bond distances (Å) and angles (deg): Ru−Ndaf 2.076(6), 2.109(6); Ru−Nbpy 2.033(6); C−Cα(C11) 1.51(2); Cα−Cβ (C11−C12) 1.21(2); C−O 1.27(2) Å; C−Cα−Cβ 160(4). F


Article

Organometallics

Table 1. Selected Infrared (Nujol Mull), UV−Vis (MeCN), and Electrochemical Parameters (0.1 M nBu4NPF6 in MeCN, 0.1 V s−1 on Pt Electrode vs Internal [FcH]/[FcH]+) of the Tris(diimine)ruthenium complexes νH2O

νOH

νCC−H cm

L1 3542, 3259, 3235, 3104 2 3645 3513 3278 3 3636 3514 3284 4 3600−3000 br 4b 3640 3518 3288 [Ru(bpy)3](PF6)2c [Ru(bpy)2(L2)](PF6)2c [Ru(bpy)(L2)2](PF6)2c [Ru(L2)3](PF6)2c a

νCC

λ [abs]

−1

nm [ε × 10 M 4

2102 2124, 2120, 2114, 2123,

1982 1969 1980 1987

226 208 206 204

[1.00] [4.17] [4.93] [5.37]

266 244 245 248 240 240 236 233

[0.787] [1.96] [1.12] [1.11] [3.0] [3.9] [5.1] [6.0]

−1

304 288 290 298 285 285 288 298

−1

cm ] [1.52] [6.16] [4.82] [4.05] [7.8] [6.6] [3.9] [4.0]

314 446 444 438 450 439 427 419

[1.93] [1.23] [1.13] [0.857] [1.4] [1.5] [1.4] [1.4]

λcalcb

E1/2 or Ep

nm

V

449 440

0.97 1.04

−1.79a −1.78a

−2.00a −2.06a

434

1.10 0.88 0.99 1.12 1.24

−1.76a −1.72 −1.74 −1.74 −1.84

−1.98a −1.92 −1.94 −1.92 −1.95

−2.13a −2.17 −2.18 −2.18 −2.05

Irreversible process. bλcalc = {1010/([Eox − Ered] × 8.0656)}. cReference 50.

Spectroscopic and Electrochemical Characterization of the Tris(diimine)ruthenium(II) Metallo-Alkynol Complexes. The tris(diimine)ruthenium(II) complexes exhibit complex 1H and 13C{1H} NMR spectra owing to the complete asymmetry of the compounds and, in the case of 3 and 4, due to overlapping resonances of the multiple isomers present in solution. The bis(bipyridyl)-mono(alkynol) complex 2 exhibits individual resonances for every carbon nucleus in the 13C{1H} NMR spectrum, the propargylic alcohol functionality of the coordinated diazafluorenyl ligand L1 imposing a unique environment on each of the six pyridyl rings. From both 1H and 13C{1H} NMR spectra it is evident that both isomers of the tris(alkynol) complex 4 (Figure 1) are present in solution, most notably evinced by resonances attributable to the alkynyl functional group. Four resonances of equivalent intensity appear at 3.20−3.17 ppm in the 1H NMR spectrum, and with the SYM and ASYM isomers giving rise to one and three unique resonances, respectively, one may infer a SYM:ASYM ratio of approximately 1:3. In the case of the mono(bipyridyl)bis(alkynol) complex 3 it was once again evident from 1H and 13 C{1H} NMR spectra that a mixture of isomers was present. Through standard 2D correlation experiments it was nevertheless possible to differentiate the resonances of the coordinated ligands and to assign the symmetry-related resonances of the diazafluorenyl ligand. Infrared spectra of the complexes are very similar across the series, the successive substitution of bpy for L1 at the tris(diimine)ruthenium core promoting a marginal decrease in energy of the νCC−H and νOH absorptions (Table 1). Waters of crystallization are also evident in each of the complexes (νHOH ca. 3640 cm−1), and in the case of [Ru(L1)3]Cl2 (4) sufficient water was entrained in the sample that absorptions of relevant functional groups were obscured. These groups gave assignable bands in the spectrum of the mixed-anion derivative [Ru(L1)3](Cl)(PF6),(4b), which was more readily dehydrated. Two absorptions appear in the spectral region anticipated for the νCC stretching mode, the weaker absorption at ca. 2120 cm−1 consistent with expectations for a low-energy shift of the νCC absorption on coordination to the ruthenium center. The stronger absorption at ca. 1980 cm−1 is not readily assigned, with the energy appropriate for ruthenium hydride,61,62 acetylide,63 allenylidene,64,65 or carbonyl53 complexes, none of which were detected by additional spectroscopic techniques. The absorption spectra of the new complexes closely resemble those of [Ru(bpy)3]2+ (Figure 4), and the trend in substitution of bpy for L1 ligands is consistent with that

Figure 4. UV−vis absorption spectra of the tris(diimine)ruthenium complexes (MeCN solution at ambient temperature).

observed in the analogous series of complexes [Ru(bpy)3−n(L)n]2+ (L = L2 or 9-(ethylenedioxo)-4,5-diazafluorene, n = 1−3; L = 4,5-diazafluorene, n = 1 or 3).50,51 By analogy to the detailed assignment of transitions in [Ru(bpy)3]2+, the most intense bands centered at ca. 200 and 300 nm are readily interpreted as intraligand π* ← π transitions of the diimine ligands, while the bands at ca. 250 and 450 nm are presumably of π* ← dRu MLCT character.66,67 The disappearance of the absorption centered at 250 nm on successive substitution of bpy for L1 is coincident with the emergence of an absorption centered at 320 nm. The bathochromic shift is attributed to a lower energy LUMO+n (n > 0) of L1 as a consequence of greater conjugation in the more planar diazafluorenyl moiety relative to bpy. The corresponding decrease in intensity of the visible-region MLCT transitions is interpreted as a reflection of a lower transition dipole in the excited state, due to the alkynol moiety. Despite the rather close correlation of electronic transitions between the series of complexes and related tris(diimine)ruthenium complexes, the electrochemical properties of the new complexes differ markedly on account of the coordinated alkynol ligand. The complex [Ru(bpy)3]2+ is characterized in cyclic voltammetry studies by a single quasi-reversible anodic process assigned to the RuII/RuIII oxidation couple and three quasi-reversible cathodic processes corresponding to successive reduction of each of the bipyridine ligands. Related complexes such as [Ru(bpy)3−n(L2)n]2+ (n = 1−3) show a similar series of processes, plus n additional cathodic redox couples due to the ketone functionality.50 Each of the new complexes 2−4 similarly exhibits a quasi-reversible anodic process consistent G


Article

Organometallics with assignment as the RuII/RuIII oxidation couple, the potential of which moves to more positive values with successive replacement of bpy by L1 at the ruthenium center (Table 1). The cathodic, ligand-centered reductions are less well-defined, only partially reversible under the voltammetric conditions employed. For all three complexes, the difference in energy between the potential of the first oxidation and first major reduction correlates well with the energy of the MLCT transition observed in UV−vis spectra. Applying the simplified form of Dodsworth and Lever’s equation (λcalc in Table 1),68

This latter feature would nevertheless be consistent with the irreversible reduction of three diimine ligands as anticipated. Additionally, an anodic feature at ca. 1.3 V is evident on glassy carbon that is not readily interpreted. Repeated potential cycling leads to a significant deterioration in current response for the mono- and tris(alkynol) complexes 2 and 4, whereas solutions of the bis(alkynol) complex 3 evolve new redox features at more positive potentials (relative to the existing couples in a fresh solution; see Figures S2 and S3 in the Supporting Information). There is no evidence of precipitation or electrode deposition in any cases. Complexes 3 and 4 also exhibit markedly different redox behavior on exchanging platinum for glassy carbon working electrodes as observed for 2. The sensitivity of the redox response to the nature of the electrode, in addition to changes consequent of potential cycling, and in comparison to analogous [Ru(N∩N)3]2+ complexes, leads us to believe that the underlying redox attributes of the complexes 2−4 are analogous to [Ru(bpy)3]2+, yet the voltammetric response is complicated by irreversible chemical reactions of the alkynol functionality coupled to electrochemical processes (i.e., an “EC” mechanism).69 Metallo-Alkynol Precursors toward Stepwise Substitution after Allenylidene Formation. The complexes 2−4 described above allow for the formation of allenylidene complexes incorporating a predetermined Ru(N∩N)n core. As an alternative we sought to investigate the synthesis of a ruthenium center that could bind to the allenylidene moiety and allow further stepwise substitution to incorporate additional ligands, thereby simplifying the construction of trisheteroleptic complexes. Initial attempts utilized [RuCl2(CO)2]n as there is ready precedent for the facile construction of mono-, bis-, and tris(diimine) complexes by selective and successive bridge-cleavage, halide-abstraction, and decarbonylation reactions.52 The reaction of [RuCl2(CO)2]n with L1 under standard conditions readily yields the product [RuCl2(CO)2{κ2-N,N′(L1)}] as a relatively intractable gum-like white solid in low, 33% yield (see Supporting Information). Dimeric [RuCl2(η6-p-cymene)]2 similarly undergoes bridgecleavage reactions with N∩N ligands to yield cationic [RuCl(η6-p-cymene)(N∩N)]+ complexes, to which additional chelating ligands may be introduced by successive thermolytic displacement of the arene and chloride ligands.70−73 Reaction between L1 and [RuCl(η6-p-cymene)(μ-Cl)]2 readily affords the chelated complex [RuCl(η6-p-cymene){κ2N,N′-(L1)}]+ (5) in good yield under mild reaction conditions (Scheme 3). The crude reaction product is shown by NMR spectroscopy to be a mixture of two very closely related species. Most readily observed in the 1H NMR spectrum, resonances attributable to the aromatic diazafluorenyl protons and the pcymene protons are almost coincident, while only the hydroxyl and alkynyl protons of the coordinated ligand L1 show a significant difference in chemical shift between the two species present (Figure S4, Supporting Information). The two species did not interconvert, before decomposition, either thermally or photochemically, and their identities are likely the two isomers depicted in Scheme 3. Slow crystallization from THF/hexanes yields crystals corresponding to isomer Z-5 (Figure 6). The Ru−N bond lengths and N−Ru−N bite angle in the structure of 5 are marginally larger than in bpy- and phen-ligated analogues,74−77 due to the constrained geometry imposed by the fused-ring structure of L1, while structural parameters within the

hvMLCT = Eox − Ered

shows excellent correlation to the experimental values and consequently affirms the origin of the oxidation (RuII/RuIII HOMO) and reduction (bpy or L1 LUMO) processes witnessed. However, taking 2 as a case example, the two major reduction processes at a platinum working electrode (−1.79 V, −2.00 V) are preceded by two additional redox couples: a welldefined couple at −1.04 V with an intensity only one-quarter that of the remaining processes and a broad, ill-defined process centered at ca. −1.48 V (Figure 5a). In contrast, these two processes are barely discernible in experiments performed at a glassy-carbon working electrode, which instead introduces a third major reductive process (ca. 1.71 V), which merges with the two “existing” diimine couples to form a poorly defined, irreversible feature between −1.74 and −1.68 V (Figure 5b).

Figure 5. CV (upper trace, blue) and DPV (lower trace, red) of 2 on (a) platinum working electrode and (b) glassy-carbon working electrode (0.1 M nBu4NPF6/MeCN). Potentials referenced to internal [FcH]/[FcH]+ = 0 V (not shown). The region −0.5 to + 0.5 V, lacking any voltammetric features, is omitted for clarity. H


Article

Organometallics Scheme 3. Synthesis of Complex 5a

anions, although no further significant intermolecular contacts exist. The infrared spectrum of the product exhibits a sharp resonance at 3276 cm−1 and a broader resonance at ca. 3200 cm−1, attributable to νCC−H and νOH, with νCC at 2119 cm−1 shifted 17 cm−1 to higher energy relative to the free ligand L1. In addition those resonances attributable to νCC and νCN absorptions of the diazafluorenyl moiety, 1630− 1423 cm−1, are shifted to higher wavenumber relative to the free ligand, in agreement with coordination of the ruthenium moiety. The UV−vis spectrum of complex 5 shows a major, split absorption at 316 nm (1.18 × 104 M−1 cm−1) and 332 nm (1.20 × 104 M−1 cm−1), in addition to intense transitions at 194 nm (3.01 × 104 M−1 cm−1) and 206 nm (3.05 × 104 M−1 cm−1). The latter high-energy transitions are readily assigned to π* ← π intraligand transitions of the two aromatic ligands. Given that ligand L1 has an intense band at 314 nm, the bands observed at 316 and 332 nm in the spectra of 5 are most likely also due to perturbed π* ← π aborptions within this ligand. Interestingly there is no low-energy absorption corresponding to a πbpy ← dRu MLCT transition, in contrast to [RuCl(η6-pcymene)(bpy)]PF6 (428 nm)78 and Ru(bpy)n systems in general (ca. 450−500 nm).79 Considering the almost isoenergetic MLCT transitions of 2−4 compared to [Ru(bpy)3]2+, one would infer a similar frontier molecular orbital energy distribution (and thus low-energy transitions) for 5 compared to [RuCl(η6-p-cymene)(bpy)]+. The electrochemistry of [RuCl(η6-p-cymene)(N∩N)]+ complexes is typified by an initial two-electron reduction of the cationic species via an ECE mechanism, consisting of a oneelectron reduction of [M − Cl]+ to [M − Cl]•, which rapidly loses chloride to yield [M]•+, which is subsequently reduced to [M]0.78,80 This neutral species may then undergo an additional, irreversible reduction to [M]− at significantly more negative potential or a quasi-reversible two-electron oxidation to [M]2+. A second redox process at positive potential is ascribed to an irreversible one-electron oxidation corresponding to the RuII/ RuIII couple of the parent complex. The electrochemistry of 5 appears similar to that described above (Figure 7), with the exception that it was not possible to observe a RuII/RuIII oxidation due to limitations imposed by solvent and electrolyte. An irreversible reduction process at −1.43 V is consistent with the two-electron reduction of [M − Cl]+ to [M]0 + Cl− and is

a

The E/Z nomenclature is based on the relative positions of the chloride ligand and hydroxyl group.

Figure 6. Molecular representation of the structure of [RuCl(η6-pcymene){κ2-N,N′-(L1)}]PF6·1/2THF, compound 5 (solvent atoms omitted for clarity, displacement ellipsoids are at the 50% probability level). Selected bond lengths (Å) and angles (deg): Ru−Cl 2.4168(6), Ru−N(111) 2.169(2), Ru−N(121) 2.146(2), N(111)−N(121) 2.790(3), C−O 1.411(3), C−Cα 1.485(3), Cα−Cβ 1.173(5), N(111)−Ru−N(121) 80.56(7), C−Cα−Cβ 176.3(3).

coordinated diazafluorenyl moiety are comparable to the tris(diimine) analogues 2−4 described above. The extended lattice shows significant intermolecular contacts, with dimeric units of the cation forming through symmetric contacts between the hydroxyl proton and chloride atom of neighboring molecules (Ru−Cl···H−O 2.22 Å, cf. van der Waals H+Cl ∼2.9 Å). These dimeric units alternate with two disordered THF solvent molecules to form a sheet parallel to a sheet of PF6−

Figure 7. Cyclic voltammograms of complex 5 (Pt electrode, 0.1 M n Bu4NPF6/MeCN). Potential referenced to internal [FcH]/[FcH]+ = 0 V (present in depicted voltammograms). I


Article

Organometallics followed by another irreversible one-electron process at ca. −1.91 V observable only at rapid scan rates (ν > 0.4 V s−1). As the reduction of [M]0 to [M]− of [RuCl(η6-p-cymene)(bpy)]+ is ca. 1.4 V more negative than the preceding reduction to [M]0, the ca. 0.5 V difference between reduction potentials in the case of 5 make it unlikely that this is an analogous process. Bimetallic Complexes. Reactions of the Tris(diimine)ruthenium(II) Metallo-Alkynol Precursors to Form Allenylidene Complexes. The reactions of 2−4 as “metallo-alkynol” precursors to form multinuclear allenylidene coordination complexes were investigated, utilizing the standard Selegue protocol of dehydration and rearrangement of the alkynol functionality at a coordinatively unsaturated transition metal center.64 In contrast to the use of a neutral organic alkynol compound, cationic metallo-alkynol complexes may hinder allenylidene formation as a consequence of Coulombic repulsion between the (cationic) metal centers. Steric factors must also be taken into consideration in attempts to construct oligo(allenylidene) complexes, as already noted in the formation of pseudodendritic vinylidene- and allenylideneruthenium(II) complexes.81−84 For example, the reaction of a tris(alkynol) compound to form the derived tris(allenylidene) complex was rapid in the case of [RuCl(η6-p-cymene)(PCy3)]OTf but extremely slow for [RuCl(dppe)2]BF4.84 Given that the Coulombic repulsions in both these complexes are comparable, the decreased reactivity of the [RuCl(dppe)2]+ derivative is most readily attributable to the increased steric encumbrance imposed by the dppe ligands. The synthesis of [(1)Ru(bpy)2](PF6)3 was first pursued under conditions identical to the formation of 1, but the products were invariably complex mixtures of numerous species after two to three days. The major products identified by IR and 31P{1H} NMR spectroscopy were the carbonyl complex [RuCl(dppm)2(CO)]+ and the free allenylidene 1, and all attempts to separate these product mixtures further resulted in significant degradation. In contrast, the reaction of 2 with cis[RuCl2(dppm)2] in ref luxing CH2Cl2 afforded the target bimetallic complex 6 in reasonable yield and purity (Scheme 4). The formation of the bis- and tris-allenylidene complexes was successfully achieved through the reaction of 3 or 4 respectively with cis-[RuCl2(dppm)2] and NaPF6 under ambient conditions over the course of a month but contaminated by a substantial quantity of [RuCl(dppm) 2(CO)]+ impurity. Attempted purification by chromatographic techniques resulted in significant decomposition of the allenylidene complexes. Infrared spectra exhibit the characteristic νCCC stretching mode at 1910 and 1888 cm−1 for the bis- and tris-allenylidene complexes, respectively, with no additional absorptions attributable to a free alkyne or hydroxyl functional group evident. The most convincing evidence for the presence of the multimetallic allenylidene complexes is from positive-ion electrospray mass spectrometry, wherein spectra of both complexes exhibit tri- and tetracationic molecular ions with differing associations of the PF6− anion. Reduced or extended reaction times under ambient conditions did not appreciably decrease the number of impurities, whereas reaction of 3 or 4 with cis-[RuCl2(dppm)2] in refluxing CH2Cl2 (cf. synthesis of 6) produced a mixture with no evidence of any allenylidene product or unreacted starting materials. The reactions of 3 and 4 with five-coordinate [RuCl(dppm)2]PF6 and [RuCl(dppe)2]PF6 were also investigated, although this again led to intractable mixtures.

Scheme 4. Synthesis of the Bimetallic Complex 6

Reaction of 5 to Form the Bimetallic Allenylidene Complex 7. Reaction of 5 with cis-[RuCl2(dppm)2] under analogous conditions for the formation of 6 readily gave the pure bimetallic complex 7 (Scheme 5). Notably there was no Scheme 5. Reaction of the Metallo-Alkynol 5 to Form the Bimetallic Allenylidene Complex 7

evidence of the free alkynol L1 nor allenylidene 1 in the crude reaction product, indicating that the [RuCl(η6-p-cymene)]+ fragment is not labilized during the alkynol to allenylidene rearrangement at the [RuCl(dppm)2]+ moiety. However, after a week in solution (under argon in a freezer), complex 7 shows almost complete decomposition. Characterization of the Bimetallic Allenylidene Complexes 6 and 7. The 1H and 13C{1H} NMR spectra of the J


Article

Organometallics Table 2. Selected Spectroscopic Data of the Bimetallic Allenylideneruthenium(II) Complexes 13

C{1H}

31

P{1H}

E1/2/Epe

−1

δ ppm 1a 6b 7b

IRc d

cm

Cα

Cβ

Cγ

dppm

νCCC

324.6 328.4 328.1

241.1 274.5 272.5

146.8 137.6 137.2

−15.16 −18.02 −17.78

1927 1910 1913

UV−vis

nm [ε × 10 M 4

232 [5.98] 194 [16.4] 194 [16.3]

274 [5.46] 286 [8.36] 276 [4.61]

−1

V −1

cm ]

454sh [1.59] 456 [3.06] 448 [1.69]

oxd

red1

red2

red3

0.98

−0.64 −0.32 −0.34

−1.51 −0.98 −1.01

−1.95 −2.03

510 [1.65] 522sh [0.970]

(CD3)2CO. bCD2Cl2. cNujol mull. dMeCN solution. ePt electrode, first-scan DPV, 0.1 V s−1, 0.1 M nBu4NPF6/CH2Cl2, referenced to internal [FcH]/[FcH]+ = 0 V.

a

Cγ ca. 10 ppm upfield relative to 1 (Cα remaining essentially unchanged). The UV−vis absorption spectrum of 1 is dominated by intense high-energy intraligand transitions of the diphosphine and diazafluorenyl moieties (Figure 8), in addition to some

bimetallic complex 7 (Table 2) closely resemble those of the metallo-alkynol precursor 5 with respect to resonances of the pendant p-cymene ligand at the coordinated ruthenium center, with only marginal changes in chemical shift observed. Resonances attributed to the diazafluorenyl moiety are also comparable to those observed in the free allenylidene complex 1, as are resonances associated with the methylene bridge and phenyl groups of the dppm ligand. Notably the presence of isomers evident in the 1H and 13C{1H} NMR spectra of 5 are not observed in the spectra of the allenylidene derivative 7, supporting the assignment of these geometric isomers for 5. NMR spectroscopy of the tris(diimine) allenylidene complex 6 is significantly more complicated with regard to aromatic resonances in both 1H and 13C NMR spectra (Figures S13 and S14). There appear to be “excess” aromatic resonances and, on the basis of 1H NMR integration, excess protons, which implicate the presence of an impurity. However, the additional resonances are not readily assigned to either the reactants nor “free” allenylidene 1. Furthermore, the presence of an impurity is not evident from UV−vis spectroscopy, mass spectrometry, or electrochemistry (see below), and microanalytical data for the complex are only marginally outside acceptable deviation. Although we are unable to conclusively affirm the reaction product to be perfectly pure, we posit that the anomalous NMR spectroscopic resonances may be (at least partially) a consequence of stereochemical effects of a putative isomer. For complex 6, a stereochemical effect is observed in the 31 1 P{ H} NMR spectrum, where the resonance associated with the phosphine ligand is observed as a complex multiplet (Figure S15). In the free allenylidene complex 1 the trans-octahedral dppm ligands provide a singlet resonance for the four phosphorus atoms as an AA′A″A‴ spin system in the 31 1 P{ H} NMR spectrum. In 6 interactions between phenyl rings of the dppm ligands and bpy ligands of the coordinated ruthenium center presumably destroy the chemical equivalence of the phosphorus nuclei, and an AA′BB′ spin system results. This same effect has previously been observed for the complexes [RuCl(dppm)2CCC(Me)(R)]+ (R = Nphenothiazine or N-iminostilbene),85 although a similar loss of symmetry is not reported for the dppe-ligated analogues of 6 reported by Pélerin et al.30 In 7 the coordinated [RuCl(η6-pcymene)]+ moiety is less sterically demanding than the tris(diimine) analogue, and only a singlet resonance is observed in the 31P{1H} NMR spectra, indicating that spatial interactions between the dppm and p-cymene ligands at the two different ruthenium centers are minimal. Although the resonances of the dppm and diazafluorenyl moieties observed in the 13C{1H} NMR spectra are essentially comparable between the free allenylidene 1 and the bimetallic complexes 6 and 7, the allenylidene carbon resonances are significantly different, with Cβ shifted ca. 30 ppm downfield and

Figure 8. UV−vis absorption spectra of the bimetallic allenylidene complexes (MeCN solution).

transitions of Ru−P charge transfer character, with the split absorption in the visible region attributed to metal-perturbed π* ← π ILCT transitions of the allene.18 The complexes 6 and 7 exhibit similar absorptions to 1 at wavelengths below 350 nm, with 6 exhibiting markedly increased intensity at ca. 285 nm as a consequence of additional 2,2′-bipyridyl ILCT transitions. As is evident in the visible region the dominant π*bpy ← dRu MLCT transitions of the Ru(N∩N)3 chromophore of 6 exhibit a minor bathochromic shift relative to the metallo-alkynol precursor 2 (λ 420, 446 nm 2; 420, 456 nm 6), and these transitions of 6 are consistent with a straightforward additive superposition of the relevant transitions of 1 and 2 (as opposed to any significant change in the electronic nature of the coincident transitions). In contrast, this absorption at ca. 450 nm appears essentially unchanged for 7 relative to 1. However, the lowest energy absorption centered at ca. 510 nm is strongly affected by coordination of an additional metal for both 6 and 7, exhibiting a significant reduction in intensity relative to 1, possibly concurrent with a degree of broadening. We plausibly attribute this to the underlying transition at this energy involving a molecular orbital that partially extends from the allene on to the diazafluorenyl moiety, with coordination of the latter affecting a minor perturbation of electronic character, which is expressed in a reduced transition dipole of the absorption, a hypothesis consistent with similarly minor absorption changes present at ca. 280 nm. Cyclic voltammetry studies of the bimetallic complexes indicate the electrochemical properties are dominated by the K


Organometallics

Article

■

CONCLUSIONS The formation of multinuclear coordination complexes incorporating the diazafluorenyl allenylidene moiety has been explored through a number of methods and has proven to be significantly more problematic than initially envisaged. Investigations consequently shifted toward an alternate, “reverse” synthetic strategy involving the formation of ruthenium complexes of the diazafluorenyl ligand L1, followed by subsequent formation of the allenylidene moiety from a metallo-alkynol. A series of alkynol-ruthenium complexes were successfully synthesized and characterized spectroscopically, and their structures were determined by means of single-crystal X-ray crystallography. In two instances these metallo-alkynol precursors reacted cleanly to form homobimetallic allenylidene complexes of 1 bearing coordinated “[Ru(bpy)2]” (6) and “[RuCl(η6-p-cymene)]” (7) groups, which were comprehensively characterized by spectroscopic means. In short, the significant differences observed via infrared spectroscopy, 13C NMR spectroscopy, and cyclic voltammetry implied that the changes in electronic character between the monometallic (1) and bimetallic (6 and 7) complexes were largely centered on the MCCC allenylidene moiety. UV−vis absorption spectroscopy indicated only a minor perturbation of the frontier molecular orbitals involved in the lowest energy transitions, which appear largely decoupled from the diimine-coordinated metal center. We therefore postulate that the entire molecular orbital manifold is moved to less positive energy on coordination of the diimine moiety, thereby approximately maintaining the same energy gap in the case of the UV−vis transitions and explaining the more readily observed reductive processes in cyclic voltammetry. The exact electronic correlation between the diimine- and carbenecoordinated metal centers is at present elusive and is the primary target of future studies.

allenylidene moiety. Relative to the free allenylidene complex 1, the two one-electron reductions of the allenic moiety for both 6 and 7 show a significant anodic shift, ca. 0.3 V for the first reduction and ca. 0.5 V for the second, with a concurrent 0.2 V decrease in the separation between potentials of the two reduction processes (Figure 9). A further, poorly reversible

Figure 9. Representative CVs of the bimetallic allenylidene complexes 6 and 7 compared to that of 1 (Pt electrode, 100 mV s−1, 0.1 M n Bu4NPF6/CH2Cl2, referenced to internal [FcH]/[FcH]+ = 0 V; FcH present for 1 and 7, absent for 6).

reduction process is then observed at ca. −2.03 V for 7 and ca. −1.95 V for 6, in addition to an anodic oxidation process at +0.98 V for 6 that is essentially coincident with the RuII/RuIII oxidation couple of the metallo-alkynol precursor 2. Cyclic voltammograms of the complexes in CH2Cl2 /nBu4 NPF6 electrolyte show no marked differences on repeated cycling or with differing scan rates. In MeCN/nBu4NPF6 electrolyte the processes due to the bimetallic complexes disappear with repeated cycling to be replaced by redox processes consistent with the free allenylidene complex 1, presumably indicating that the doubly reduced allenylidene complex is stable only toward dissociation of the diazafluorenyl-coordinated metal moiety in noncoordinating solvent. At first glance the more favorable reduction potentials of the bimetallic species relative to the free allenylidene 1 may readily be attributed to stabilization of the added electron density by the coordinated metal cation. The additional cathodic process observed at ca. −2 V may then be interpreted as the onset of reduction for a terminal bpy ligand of 6 or the “[(η6-pcymene)RuCl]” moiety of 7. However, such an interpretation is complicated in the latter case on comparison to the monocation metallo-alkynol 5, which shows a reduction process at −1.43 V thought to proceed by dissociation of a chloride ion, and single-electron dissociation of chloride from the neutral, doubly reduced allenylidene [7]0 should be favored (although a two-electron heterolytic bond cleavage as proposed for 5 would still be electrostatically disfavored). Similarly, the oxidation potential of the RuII/RuIII couple in tricationic 6 would be expected to show an anodic shift relative to the dicationic metallo-alkynol 2 if the remote metal centers are indeed conjugated through the allenylidene bridge, yet the potential of this RuII/RuIII couple is practically unchanged between 6 and 2.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.organomet.5b00569. NMR spectra of complexes 2−7, CVs of 3 and 4, experimental details of some related compounds, molecular depiction of a weakly diffracting sample of 2, table of crystallographic data for 2−5, crystallographic experimental details (PDF) CIF files for compounds 2−5 (CCDC 989023−989026) (CIF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Present Addresses

† Department of Chemistry, University of British Columbia, Vancouver, BC, V6T 1Z1, Canada. ‡ Centre for Microscopy, Characterization and Analysis, University of Western Australia, Western Australia, 6009, Australia.

Author Contributions

P.A.S. performed all of the experimental work reported in this paper except the crystallographic analysis, which was performed L


Article

Organometallics

(32) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.; Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I. Organometallics 2010, 29, 2176−2179. (33) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910. (34) Armarego, W. L. F.; Chai, C. L. L. Purification of Laboratory Chemicals, 7th ed.; Elsevier: Waltham, MA, USA, 2013. (35) Chaudret, B.; Commenges, G.; Poilblanc, R. J. Chem. Soc., Dalton Trans. 1984, 1635−1639. (36) Anderson, G. K.; Lin, M. Inorg. Synth. 1990, 28, 60−63. (37) Bennett, M. A.; Huang, T. N.; Matheson, T. W.; Smith, A. K. Inorg. Synth. 1982, 21, 74−78. (38) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978, 17, 3334−3341. (39) Anderson, P. A.; Deacon, G. B.; Haarmann, K. H.; Keene, F. R.; Meyer, T. J.; Reitsma, D. A.; Skelton, B. W.; Strouse, G. F.; Thomas, N. C.; Treadway, J. A.; White, A. H. Inorg. Chem. 1995, 34, 6145− 6157. (40) Bruce, M. I. Chem. Rev. 1998, 98, 2797−2858. (41) Constable, E. C. Adv. Inorg. Chem. 1989, 34, 1−63. (42) Housecroft, C. E., Constable, E. C., Dilworth, J. R., Eds. Comprehensive Coordination Chemistry II; Elsevier Pergamon: San Diego, CA, 2004; Vol. 5, pp 555−731. (43) Klein, R. A.; van Belzen, R.; Vrieze, K.; Elsevier, C. J.; Thummel, R. P.; Fraanje, J.; Goubitz, K. Collect. Czech. Chem. Commun. 1997, 62, 238−256. (44) Wehman, P.; Dol, G. C.; Moorman, E. R.; Kamer, P. C. J.; Van Leeuwin, P. W. N. M.; Fraanje, J.; Goubitz, K. Organometallics 1994, 13, 4856−4869. (45) Schager, F.; Bonrath, W.; Porschke, K. R.; Kessler, M.; Kruger, C.; Seevogel, K. Organometallics 1997, 16, 4276−4286. (46) Acetylene Chemistry: Chemistry, Biology and Materials Chemistry; Diedrich, F.; Stang, P. J.; Tykwinski, R. R., Eds.; Wiley VCH: Weinheim, 2004. (47) Maitlis, P. M. Acc. Chem. Res. 1976, 9, 93−99. (48) Zeni, G.; Larock, R. C. Chem. Rev. 2004, 104, 2285−2310. (49) Wang, Y. X.; Jackman, D. C.; Woods, C.; Rillema, D. P. J. Chem. Crystallogr. 1995, 25, 549−553. (50) Wang, Y. X.; Perez, W.; Zheng, G. Y.; Rillema, D. P. Inorg. Chem. 1998, 37, 2051−2059. (51) Henderson, L. J.; Fronczek, F. R.; Cherry, W. R. J. Am. Chem. Soc. 1984, 106, 5876−5879. (52) Spiccia, L.; Deacon, G. B.; Kepert, C. M. Coord. Chem. Rev. 2004, 248, 1329−1341. (53) Clear, J. M.; Kelly, J. M.; O’Connell, C. M.; Vos, J. G.; Cardin, C. J.; Costa, S. R.; Edwards, A. J. J. Chem. Soc., Chem. Commun. 1980, 750−751. (54) Kelly, J. M.; O’Connell, C. M.; Vos, J. G. J. Chem. Soc., Dalton Trans. 1986, 253−258. (55) Schauer, P. A.; Low, P. J. Eur. J. Inorg. Chem. 2012, 2012, 390− 411. (56) Plater, M. J.; Kemp, S.; Lattmann, E. J. Chem. Soc., Perkin Trans. 1 2000, 971−979. (57) Johnson, A. W. The Chemistry of the Acetylenic Compounds. Vol 1: The Acetylenic Alcohols; Edward Arnold & Co.: London, 1946; Vol. 1. (58) Fletcher, N. C.; Nieuwenhuyzen, M.; Rainey, S. J. Chem. Soc., Dalton Trans. 2001, 2641−2648. (59) Grepioni, F.; Cojazzi, G.; Draper, S. M.; Scully, N.; Braga, D. Organometallics 1998, 17, 296−307. (60) Mascal, M. J. Chem. Soc., Perkin Trans. 2 1997, 1999−2001. (61) Bruce, M. I.; Butler, I. R.; Cullen, W. R.; Koutsantonis, G. A.; Snow, M. R.; Tiekink, E. R. T. Aust. J. Chem. 1988, 41, 963−969. (62) Haukka, M.; Kiviaho, J.; Ahlgren, M.; Pakkanen, T. A. Organometallics 1995, 14, 825−833. (63) Stein, E.; Oki, S. Y.; Vichi, E. J. Braz. Chem. Soc. 2000, 11, 252− 256. (64) Selegue, J. P. Organometallics 1982, 1, 217−218. (65) Byrne, L. T.; Koutsantonis, G. A.; Sanford, V.; Selegue, J. P.; Schauer, P. A.; Iyer, R. S. Organometallics 2010, 29, 1199−1209.

by B.W.S. All authors contributed to the writing of the manuscript. All authors approve of the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the University of Western Australia for partly funding this project. P.A.S. was the holder of an Australian Postgraduate Award.

■

REFERENCES

(1) Bosnich, B. Inorg. Chem. 1999, 38, 2554−2562. (2) Sala, X.; Maji, S.; Bofill, R.; García-Antón, J.; Escriche, L.; Llobet, A. Acc. Chem. Res. 2014, 47, 504−516. (3) Thomas, C. M. Comments Inorg. Chem. 2011, 32, 14−38. (4) Astruc, D. Acc. Chem. Res. 1997, 30, 383−391. (5) Braunstein, P. New J. Chem. 1994, 18, 51−60. (6) Low, P. J. Dalton Trans. 2005, 2821−2824. (7) Balzani, V.; Juris, A.; Venturi, M.; Campagna, S.; Serroni, S. Chem. Rev. 1996, 96, 759−833. (8) Liu, Y.; Li, Y.; Schanze, K. S. J. Photochem. Photobiol., C 2002, 3, 1−23. (9) Arifin, K.; Majlan, E. H.; Daud, W. R. W.; Kassim, M. B. Int. J. Hydrogen Energy 2012, 37, 3066−3087. (10) Frischmann, P. D.; Mahata, K.; Würthner, F. Chem. Soc. Rev. 2013, 42, 1847. (11) Inagaki, A.; Akita, M. Coord. Chem. Rev. 2010, 254, 1220−1239. (12) Kärkäs, M. D.; Johnston, E. V.; Verho, O.; Åkermark, B. Acc. Chem. Res. 2014, 47, 100−111. (13) Kaim, W.; Sarkar, B. Coord. Chem. Rev. 2007, 251, 584−594. (14) Low, P. J. Coord. Chem. Rev. 2013, 257, 1507−1532. (15) Paul, F.; Lapinte, C. Coord. Chem. Rev. 1998, 178, 431−509. (16) D’Alessandro, D. M.; Keene, F. R. Chem. Rev. 2006, 106, 2270− 2298. (17) Kaes, C.; Katz, A.; Hosseini, M. W. Chem. Rev. 2000, 100, 3553−3590. (18) Cifuentes, M. P.; Humphrey, M. G.; Koutsantonis, G. A.; Lengkeek, N. A.; Petrie, S.; Sanford, V.; Skelton, B. W.; Stranger, R.; White, A. H. Organometallics 2008, 27, 1716−1726. (19) Koutsantonis, G. A.; Jenkins, G. I.; Szczepaniak, B.; Skelton, B. W.; Tan, C.; White, A. H. Organometallics 2009, 28, 2195−2205. (20) Scheerer, S.; Rotthowe, N.; Abdel-Rahman, O. S.; He, X.; Rigaut, S.; Kvapilová, H.; Záliš, S.; Winter, R. F. Inorg. Chem. 2015, 54, 3387−3402. (21) Di Piazza, E.; Norel, L.; Costuas, K.; Bourdolle, A.; Maury, O.; Rigaut, S. J. Am. Chem. Soc. 2011, 133, 6174−6176. (22) Lang, H.; Packheiser, R. Collect. Czech. Chem. Commun. 2007, 72, 435−452. (23) Norel, L.; Bernot, K.; Feng, M.; Roisnel, T.; Caneschi, A.; Sessoli, R.; Rigaut, S. Chem. Commun. 2012, 48, 3948. (24) Packheiser, R.; Ecorchard, P.; Rüffer, T.; Lang, H. Organometallics 2008, 27, 3534−3546. (25) Packheiser, R.; Ecorchard, P.; Walfort, B.; Lang, H. J. Organomet. Chem. 2008, 693, 933−946. (26) Packheiser, R.; Lang, H. Inorg. Chem. Commun. 2007, 10, 580− 582. (27) Packheiser, R.; Ecorchard, P.; Rueffer, T.; Lang, H. Chem. - Eur. J. 2008, 14, 4948−4960. (28) Norel, L.; Di Piazza, E.; Feng, M.; Vacher, A.; He, X.; Roisnel, T.; Maury, O.; Rigaut, S. Organometallics 2014, 33, 4824−4835. (29) Cadierno, V.; Gimeno, J. Chem. Rev. 2009, 109, 3512−3560. (30) Pelerin, O.; Olivier, C.; Roisnel, T.; Touchard, D.; Rigaut, S. J. Organomet. Chem. 2008, 693, 2153−2158. (31) Wong, C.-Y.; Tong, G. S. M.; Che, C.-M.; Zhu, N. Angew. Chem., Int. Ed. 2006, 45, 2694−2698. M


Article

Organometallics (66) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; vonZelewsky, A. Coord. Chem. Rev. 1988, 84, 85−277. (67) Lytle, F. E.; Hercules, D. M. J. Am. Chem. Soc. 1969, 91, 253− 257. (68) Gorelsky, S. I.; Dodsworth, E. S.; Lever, A.; Vlcek, A. A. Coord. Chem. Rev. 1998, 174, 469−494. (69) Hogan, C. F.; Bond, A. M.; Myland, J. C.; Oldham, K. B. Anal. Chem. 2004, 76, 2256−2260. (70) Canivet, J.; Karmazin-Brelot, L.; Süss-Fink, G. J. Organomet. Chem. 2005, 690, 3202−3211. (71) Jung, I.; Choi, H.; Lee, J. K.; Song, K. H.; Kang, S. O.; Ko, J. Inorg. Chim. Acta 2007, 360, 3518−3524. (72) Lee, D. N.; Kim, H. M.; Kim, E. H.; Soh, B. K.; Kim, C.-H.; Lee, B. M.; Yoon, S. H.; Lee, W.-Y.; Kim, B. H. Synth. Met. 2006, 156, 885− 892. (73) Pogozhev, D. V.; Bezdek, M. J.; Schauer, P. A.; Berlinguette, C. P. Inorg. Chem. 2013, 52, 3001−3006. (74) Dykeman, R. R.; Luska, K. L.; Thibault, M. E.; Jones, M. D.; Schlaf, M.; Khanfar, M.; Taylor, N. J.; Britten, J. F.; Harrington, L. J. Mol. Catal. A: Chem. 2007, 277, 233−251. (75) Frediani, P.; Giannelli, C.; Salvini, A.; Ianelli, S. J. Organomet. Chem. 2003, 667, 197−208. (76) Lalrempuia, R.; Rao Kollipara, M. Polyhedron 2003, 22, 3155− 3160. (77) Zhou, H.-F.; Bian, Z.-Q.; Liu, Z.-W.; Huang, C.-H.; Zhao, Y.-L. Acta Crystallogr., Sect. E: Struct. Rep. Online 2007, 63, M1851−U1672. (78) Kaim, W.; Reinhardt, R.; Sieger, M. Inorg. Chem. 1994, 33, 4453−4459. (79) Lever, A. B. P. Inorganic Electronic Spectroscopy; Elsevier Science Publishers B. V.: Amsterdam, 1984; Vol. 33. (80) Štěpnička, P.; Ludvík, J.; Canivet, J.; Süss-Fink, G. Inorg. Chim. Acta 2006, 359, 2369−2374. (81) Guesmi, S.; Touchard, D.; Dixneuf, P. H. Chem. Commun. 1996, 2773−2774. (82) Rigaut, S.; Perruchon, J.; Guesmi, S.; Fave, C.; Touchard, D.; Dixneuf, P. H. Eur. J. Inorg. Chem. 2005, 2005, 447−460. (83) Uno, M.; Dixneuf, P. H. Angew. Chem., Int. Ed. 1998, 37, 1714− 1717. (84) Weiss, D.; Dixneuf, P. H. Organometallics 2003, 22, 2209−2216. (85) Winter, R. F.; Hartmann, S.; Záliš, S.; Klinkhammer, K. W. Dalton Trans. 2003, 2342−2352.

N


Coordinating Tectons 4: Coordination Chemistry of ... - ACS Publications

Recommend Documents