Investigation of low-valent ruthenium aquo complexes Citation Haney, Susan Elaine (1992) Investigation of low-valent ruthenium aquo complexes. Master's thesis, California Institute of Technology. doi:10.7907/bd27-j183. https://resolver.caltech.edu/CaltechTHESIS:10032011-090148072 Abstract The work presented in this thesis investigates the synthesis of low oxidation state coordination complexes of ruthenium incorporating the anionic ligand, [CpCo(P(O)(OMe)_2)_3]-,(L_(OMe)). Several routes to RuL_(OMe)(H_2O)_3^+ are outlined. These routes produced two major products, possibly related by acid-base chemistry. A new dimeric ruthenium aquo species with the proposed structure [Ru(III)L_(OMe)(H_2O)(µ-OH)_2OTs_2 was isolated. This complex does not provide a straightforward route to the target monomeric oxo compounds, but could conceivably provide a entry into synthesis of ruthenium dimers containing arnines and other ligands. Finally, a new route to [Ru(IV)(H_20)(µ-O)]_2OTf_2 was discovered which is more convenient and is higher yielding than the previously published procedure. Item Type: Thesis (Master's thesis) Subject Keywords: Chemistry Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Bercaw, John E. Thesis Committee: Unknown, Unknown Defense Date: 1 January 1992 Record Number: CaltechTHESIS:10032011-090148072 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:10032011-090148072 DOI: 10.7907/bd27-j183 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 6699 Collection: CaltechTHESIS Deposited By: Tony Diaz Deposited On: 03 Oct 2011 21:21 Last Modified: 16 Apr 2021 23:02 Thesis Files PDF - Final Version See Usage Policy. 4MB Repository Staff Only: item control page

INVESTIGATION OF LOW-VALENT RUTHENIUM AQUO COMPLEXES Thesis by Susan Elaine Haney In Partial Fulfillment of the Requirements for the Degree of Master of Science California Institute of Technology Pasadena. Califonria 1992 ii Acknowledgements All my sincerest thanks I give to Dr. John Bercaw, who is an unfailingly rigorous scientist and a generous, understanding human being. Thanks go to the Bercaw group for help and solace when I was confronted with Ruthenium Road Tar. Thanks, Stew, for being my best friend . iii Abstract The work presented in this thesis investigates the synthesis of low oxidation state coordination complexes of ruthenium incorporating the anionic ligand, [CpCo(P(O)(OMehh]-, (LOMe). Several routes to RuLoMe(lhO)3+ are outlined. These routes produced two major products, possibly related by acid-base chemistry. A new dimeric ruthenium aquo species with the proposed structure [Ru(III)LoMe(H20)(~-OH)hOTs2 was isolated. This complex does not provide a straightforward route to the target monomeric oxo compounds, but could conceivably provide a entry into synthesis of ruthenium dimers containing arnines and other ligands. Finally, a new route to [Ru(IV)(H20)(~-O)hOTf2 was discovered which is more convenient and is higher yielding than the previously published procedure. IV Table of Contents Section Acknowledgements ii Abstract iii Abbreviations Used in the Text v Introduction 1 Results and Discussion 7 Experimental 17 Supplementary NMR Spectra 21 I Introduction Oxidation of organic compounds by metal oxo compounds is one of the most fundamental processes in organic chemistry, and has been known for over two hundred years! In this light perhaps it is surprising that this topic is still fertile ground for research. However, there is still a need for more active and selective oxidants. A great deal of research in the past decade has been focused on achieving the selective functionalization of unactivated carbon- hydrogen bonds. As well as being of general chemical interest, the oxidation of alkanes would have obvious practical significance. Of particular interest is the oxidation of methane, which is abundant but unwieldy, to methanol, for fuel. Previous research in this area by the Bercaw group centered on oxo complexes of early transition metals such as Ta and W. However, these metals proved too oxophilic to serve as oxidation catalysts. For this reason, investigations of oxo complexes of the later transitions metals, which form weaker metal oxygen bonds, were undertaken. Ruthenium, in particular was chosen, because its high oxidation states are easily accessible. These high oxidation states (Ru(VI- Vm) help stabilize tenninal oxo groups, while imparting considerable oxidation potential to the complex. A premiere example is Ru04, an extremely active oxidant that can cleave carbon-carbon double bonds and will even oxidize alkanes.! Unfortunately, it is also unstable, photosensitive, volatile, and prone to reaction with conunon solvents. It has been found that by addition of suitable ancillary ligand, the ruthenyl group, Ru<h, can be stabilized while retaining much of its fonnidable oxidation potential. The most important examples in the literature have been compounds of the type Ru(VI)i4<h, which are generally active oxidants. These d 2 Ru(VI) or Os(VI) dioxo species preferentially exhibit a trans disposition of tenninal oxo groups. For example, [Ru<hC41 2- , [Ru<hpY41 2+, and Ru<h(pyh(OAch exist only in the trans configuration. 2a< Only one complex, Os<h(bpyh has been synthesized in both the cis and the trans forms; 3 the cis form was found to be unstable and isomerized to the trans form in refluxing acetonitrile. Various molecular orbital arguments 2 have been advanced to explain the tendency of Ru(Vn to adopt the trans configuration. In simple terms, in the trans case the highest occupied molecular orbital is non-bonding with respect to the the oxygen pi orbitals, whereas in the cis case, all the d orbitals are slightly destabilized by the oxo groups. Since the cis oxo derivatives are expected in general to be thennodynamically less stable than the corresponding trans derivatives, it is believed that the cis dioxo species will be more reactive. For this reason, we decided to synthesize a cis dioxo species, with the goal of fmding a well defined and stable system which could be studied easily. Some stable RuCVn cis dioxo complexes have been reported in the literature. In general these fall into two classes. The first contains two molecules of the type cis-Ru(hL2 where L= 6, 6' dichloro-2,2' bipyridine or 6, 6' dimethyl-2,2' bipyridine. The complexes are kinetically trapped in the cis form, since the bulky substituents at the ortho positions of the bipyridine present a high barrier to isomerization to the trans compound. 4 The dichloro-substituted complex is reponed to be a very active oxidant of alkenes and even alkanes; no data is available about its selectivity of reaction with alcohols. The other type of cis oxo complex that has been documented is one in which the oxo groups are trans to another oxygen-containing ligand. One example is RuCI2(T]2CH3C(h)(Oh,5 which is reported to be an efficient two electron oxidant of alcohols, alkyl halides, and phosphines. The work reported herein uses a different approach. It is proposed that the cis conformation of the oxo groups in a RuCVn compound be enforced by using a facially coordinating tridentate ancillary ligand. Cyc\opentadienyl (Cp) and related compounds have been used extensively in such a manner with early transition metals, but the Cp* ligand has been shown to be susceptible to oxidation. 6 Thus, for use with ruthenium, our attention turned to the formally analogous [CpCo(P(ORhO)31-, [LoRl- (OR=OCH3, OCH2CH3), shown in Figure 1. 3 Figure 1. ~, I OR ___ Co OR .,p.......- ~ " ,. OR'-II OR,p .. OR P"OR o II o II 0 This ligand has been shown to be stable at least to one electron oxidation, since it can form a stable complex with Ce(IV).7 In addition, its electronic properties are quite favorable. It is a hard oxygen-donor ligand, electronically comparable to F- or OH-,s which should help to stabilize high oxidation state complexes. Finally, the solubility of the ligand can be altered greatly by choice of OR substituent. The work described herein uses the loMe ligand exclusively, because it is water soluble. It was hoped that this property would transfer to complexes incorporating the ligand, which would facilitate their electrochemical analysis. In designing a synthetic route to terminal oxo complexes of ruthenium containing the loMe ligand, one encounters a major problem-- the tendency of oxo and hydroxo complexes of ruthenium to form dirners, oligorners and polymers. High oxidation states [Ru(VI-VIII)] stabilize hydroxo and oxo groups by providing strong pi bonding. However, pi bonding in lower oxidation states [Ru (II-V)] results in population of antibonding molecular orbitals. For this reason the low oxidation state complexes have weak metal - oxygen pi bonding, and the oxygen atoms have more basic character, which leads to the formation of oxo- or hydroxo-bridged polymeric species. The tendency of lower oxidation state ruthenium to dimerize is demonstrated by early attempts in our group to synthesize oxo complexes from a mixture of RU04 and NaLoR in acid solution.9 Under these conditions, RU04 is believed to be reduced to Ru(lV).IO This reaction yielded [Ru(lv)LoR(~-O)(OH)h (Figure 2) as the major Ru containing species. 4 Figure 2. This dimer has shown promise as a two electron oxidant of alcohols, and its reactions are a current topic of research in the Bercaw group. It has an extremely stable RU(I!-OhRu core, and up to this time no method has been found to convert it to a monomeric species. Therefore, to synthesize a monomeric ruthenium oxo complex, a route must be found that does not allow dimers to form in the process. One obvious strategy would be to avoid the lower oxidation states altogether by using high oxidation state Ru(Vl) and Ru(VII) complexes as starting materials. Unfortunately, this does not seem to be a viable option, since most readily available high oxidation state complexes feature a trans disposition of the oxo groups. These complexes have never been reported to isomerize to the corresponding cis compounds. Attempts in our lab to add LoR to [Ru(Vl)03(OHh1 2- , [Ru(VII)041' or [Os(Oh(OH)41 2. proved fruitless . For that reason we were forced to adopt a different synthetic strategy. The guiding principle was to begin with low oxidation state ruthenium materials that could be substituted with a variety of sterically bulky "blocking" groups. These groups would prevent dirnerization until the ruthenium center was brought to a high oxidation state which should be less likely to dimerize. The blocking groups could then be altered or removed 5 from the molecule if necessary. Figure 3 outlines several general synthetic routes based upon this strategy. Figure 3. M=II,m 1-= labile ligand B= blocking group. [0) The work presented in this thesis describes the first step of the project, which is to find low oxidation state starting materials which provide a general route to high oxidation state ruthenium oxo complexes that contain facially coordinating tridentate ligands. 6 References 1.. Nugent, W.A; Mayer, J.M. Metal-Ligand Multiple Bonds; J. Wiley and Sons: New York, 1988, pp. 264-268. 2. (a)Perrier, S.; Kechi, J.K. lnorg . Chern. 1988,27, 4165-73.(b)EI-Hendawy, AM.; Griffith, W.P.; Taha, F.I.; Moussa, M.N. J . Chern. Soc. Dalton Trans. 1989,901-906. (c)Kechi, J.k.; Lau, T.C. J. Chern. Soc. Chern. Commun. 1987, 798-799. 3. Dobson, J.C.; Takeuchi, K.J.; Pipes, D.W.; Geselowitz, D.A.; Meyer, T.J. /norg . Chern, 1986, 25, 2357-65. 4. Che, C-M.; Leung, W-H. J. Chern. Soc. Chern. Cornrnun. 1987, 1376-1377. 5. Griffith, W.P.; Jolliffe, J. M.; Ley, S.Y., Williams, D.J. J. Chern. Soc . Chern Cornrnun . 1990, 1219-1221. 6. Parkin, G.; Marsh, R.E.; Schaefer, W.P., Bercaw, I.E. /norg. Chern. 1988,27, 3262-3264. 7. KHiui, W.; Eberspach, W.; Schwarz, R. J. Organornet. Chern. 1983,252,347. 8. Klaui, W. Angew. Chern. Int. Ed. Engl. 1990,29, 627-637. 9. Power, I.M.; Evertz, K.; Henling, L.; Marsh, R.; Schaefer, W.P.; Labinger, I .A ; Bercaw, I .E. lnorg. Chern., 1990,29, 5058-5065. 10. Rard, I .A . Chern. Rev. 1985,85, 1-39. 7 Results and Discussion In accordance with the synthetic scheme outlined in the introduction, we wished to find a starting material incorporating [CpCo(P(O)(0CH3hh)]-, [LoMe] which was monomeric, well-defined, substitutionally labile, and easily prepared in large quantities. Unfortunately, the most commonly available starting materials are RuCI3·H20, an amorphous solid which probably contains Ru(IV) and oxo ligands, 1 and RU04, which is rather indiscriminate in its reactivity. In our search for alternative starting materials, one compound which seemed promising was [RuLoMe(~)]+PF6-, a crystalline, wellcharacterized complex previously synthesized by Kiaui, et al. 2 This compound was interesting in light of studies done on the analogous cyclopentadienyl compound, [RuCp(~)]PF6.3 It was shown that the cyclopentadienyl complex could be photolyzed cleanly to replace the benzene with a coordinating solvent such as acetonitrile. In hopes that the LoMe complex would display analogous chemistry, an investigation of its photolytic behavior was undertaken. Synthesis of [RuLOMe(CH3CN}J]PFfj The first photolysis of [RuLoMe(~)]+PF6- was done in acetonitrile, as was reported for the Cp compound. This reaction is quite slow, requiring 46 hr. to go to 95% completion under the reaction conditions. The reaction was extremely clean, leading to RuLoMe(CH3CNhPF6, with no decomposition of starting material or product observed by NMR spectroscopy. Unfortunately, the tris acetonitrile complex was exceedingly inert to substitution. No substitution of the acetonitrile ligands by water was apparent in aqueous solution by lH NMR spectroscopy. In addition, oxidations attempted with H2~, terrbutyl hydroperoxide, or 3-chloroperbenzoic acid yielded either no reaction, or ill-defined products. This lack of reactivity is consistent with the exceptional stability of related acetonitrile complexes of Ru(ll). For example, the reported rate constant for acetonitrile exchange onto Ru(CH3CN)62+ is 8.9 ± 2 xlO-ll sol. 4 For this reason, [RuLoMe(CH3CNh]+ was abandoned as a direct precursor to RuCVn oxo complexes. 8 Instead, our attention was turned to the synthesis of ruthenium aquo species containing loMe. A complex such as [RuLoMe(H20)31+ was expected to be more labile to substitution. s In addition, it was possible that the aquo ligands could be oxidized directly to oxo groups, thus simplifying the synthesis of our target compounds. Three routes to Ru(Il)LoMe(H20)3+ were investigated, the results of which are summarized in Fig. 1. Figure 1 [RuloMe(C6 Hs))CI 1 2 I) 3.70 (virt. q., OCH3l I) 3.62 (virt. q., OCH 3l 5.29 (5, C sHsl 5.22 (5, C sHs) + minor Au products The first reaction attempted was the photolysis of RuL(Q;If(;)Cl in unbuffered D20. This was done at the NMR tube scale under Ar. As was the case in acetonitrile, the photolysis was quite slow, at best going to about 40% completion after 17 hr as calculated from the ratios of ligand resonances in the lH NMR spectrum. The major species in solution, 1, exhibited lH NMR (500 MHz) resonances at /) 3.70(vin. q, OCH3) and 5.29 (s, CSHS). In addition, a minor species at /) 3.59 (vin. q, OCH3) and 5.45 (s, CsHs) was present which could not be assigned. A smaller amount of a species later identified as [RuIIILoMe(I1-0H)(H20)12 (3) was also present. Because the reaction products seemed confusing at the time, another route was investigated. 9 The second route was the reaction of [Ru(lhO)6]OTs2 with NaLoMe in D20 at the NMR tube scale under argon. When the two were mixed in 0.1 M p-toluenesulfonic acid, no reaction was observed by IH NMR spectroscopy after 3 hr. However, when the reaction was done in unbuffered D20, a reaction was observed, with the fonnation of a small amount of precipitate. The major species observed by IH NMR spectroscopy was 1, the same complex produced by the photolysis of RuLoMe(~)CI. A small amount of 2 was fonned, with resonances at /)5.22 (s, 5H, C5H5) and 3.62 (vin. q, 18H, OCH3). Minor resonances which could be assigned to (3) were also present, which might have originated from Ru(I1I)(H20)63+ present in the starting material. A consistent picture was beginning to fonn, which became more complete after examination of the third route, which utilized [RuLoMe(CH3CN»]+ as a starting material. Despite its discouraging inertness to substitution by water, it was thought that the acetonitrile ligands might be displaced by a stronger nucleophile than water, such as hydroxide ion. Indeed, when [RuLoMe(CH3CN»]+ was dissolved in 0.1 or 1.0 MOD' in D20 under Ar, complex 2 was seen in the 500 MHz IH NMR spectrum, with peaks at /) 5.22 (s, 5H, C5H5) and 3.62 (virt. q, 18H, OCR3). A representative spectrum of the reaction in progress appears in Fig. 2. During the reaction the bound acetonitrile resonance at /) 2.63 was replaced by several resonances between 1.9 and 2.2 ppm that are characteristic of the products of base-catalyzed hydrolysis of acetonitrile. During the course of the reaction two additional peaks were seen in the bound acetonitrile region. These are tentatively assigned to intennediates in the stepwise replacement of the three acetonitrile ligands by hydroxide, where one or two acetonitrile ligands have been replaced. A slight shift in the methoxy peak of the LoMe ligand was also observed. It could not be detennined from the NMR data whether the mechanism of reaction was straightforward displacement of the acetonitrile with OH', followed by hydrolysis of free acetonitrile, or rather hydrolysis of the coordinated acetonitrile followed by substitution with water. !~ , 1 ) U r 2. Cs1I.1 [Rul{)Me<CH3CN lJ)PF.. cs1I.1 , , < , n - ~-,.---.---,------y-T- )' \.. l ) -j rf l' r) [Ru4>Me(CH.lCN}J)PF6 2. OCtla [Ru4>Me(CH3CN}J)pF6 • OCJjl rI Fig. 2. [RULOMe(CH)CN}J)PF6 + 1M KOH/D10, 1 day under Ar. o 11 It is believed that both 1 and 2 are ruthenium aquo or hydroxo species, since the same products are formed from widely varying starting materials. Because 2 formed at high pH, and 1 was found at more neutral conditions, it seems likely that the two are related by acid-base reactions. This is not an unreasonable supposition, since pKas for related Ru(II) tris aquo species range from 7.9 for [Ru(0iH6)(H20)3l2+ 6to 12.3 for [Ru(tpm)(H20bl (tpm = tris(l-pyrazolyl)methane).7 Structures for 1 and 2 are suggested in Fig. 3. + Fig. 3. Possible structures for 1(above) and 2 (below) 12 All structures are consistent with the available NMR data if fast proton exchange between the aquo and hydroxo ligands is occurring. In the presence of air, 2 is converted to a new product with a green color, which displays a splitting pattern in the NMR spectrum that indicates C3v symmetry about the ligand During every attempt at isolation of a clean product, whether by precipitation, extraction, or neutralization, this product decomposed to NaloMe and a black intractable solid (presumably Ru(OHh or Ru(h:nH20). Because of this behavior, it seems likely that the green product is aCTUally polymeric Ru(OHh. This result is not really a surprising one, since ruthenium(Im and ruthenium(lV) aquo and oxo complexes have been shown to polymerize under basic conditions. l Keeping this in mind, it would be interesting to treat RuLoMe(CH3CNh+ with strong acid. If indeed its decomposition in base relies upon hydrolysis of coordinated acetonitrile, hydrolysis of [RuLoMe(CH3CNh]PF6 in ~ solution in air may produce a stable Ru(lln tris aquo species. Reactions of (RuLOMeCI2hO An investigation was begun of the possibilities of using (RuLoMeC12hO (R=Me, Et) as a starting material for synthesis of Ru(VI) oxo complexes. This compound was synthesized and structurally characterized by Kaspar Evertz. Its advantage as a starting material is its ease of synthesis; NaloMe is simply stirred with commercial RuCI3 in ethanol in air. Its disadvantage is obvious: the compound is already an oxo-bridged dimer, which was to be avoided. In hopes of obtaining a monomeric complex, the somewhat counter-inTUitive step of reducjne: (RuLoMeCI2hO was taken. Meyer, et al., set some precedent for this type of reaction in their synthesis of [(tpm)Ru(H20h]2+ by zinc reduction of [(tpm)RuIII(~-O)(~­ D2P(O)(OH)h RuIII(tpm)] (tpm = tris(l-pyrawlyl)methane).1 Reduction of (RuLoMeC12)20 by a large excess of Zn/Hg amalgam in O.IM ptoluenesulfonic acid under Ar resulted in a purple solution. After oxidation of the solution 13 by air, a yellow crystalline product, 3, slowly forms in about 30% yield, assuming an approximate molecular weight of 750 glmol per Ru center. Unfortunately, crystals suitable for x·ray structure determination were not forthcoming, but based upon the following chemical and spectroscopic evidence, the structure shown in Fig. 4 is proposed. Figure 4. Proposed structure of 3 2+ The complex is of the type Rul<JMe(AhB, and contains a mirror plane, as evidenced by the IH NMR spectrum of the product, which displayed a quartet:quartet:triplet (I: I: I) pattern in D20 for the OCH3 peaks of the ligand. OTs· resonances were also present and integrated as lOTs· per ligand resonance, indicating that each ruthenium center is singly charged. Addition of silver nitrate to the sample resulted in a color change, but no precipitation of AgCl, suggesting that the Cl· in the starting material was removed during the reaction. Perhaps the strongest structural evidence comes from cyclic voltametry experiments. The electrochemistry of 38was identical to that of [Ru(N)l<JMe(~· O)(H20)h2+, the so-called (N·N) dimer, which suggests that 3 is a reduced form of the (N·N) dirner and shares its doubly bridged structure. The electrochemical results could be reproduced chemically; addition of oxidants such as H2~ or 3-chloroperbenwic acid to 3, 14 or heating a neutral aqueous solution of the compound in air resulted in fast, quantitative conversion to the (IV-IV) dimer. At this point it may be fruitful to reflect on the reason a dimeric species was formed in this case. Ru(llI) aquo complexes are generally more acidic than their Ru(II) analogs. Typical pKas are between 2 and 3. 4•7. When Ru(lll) complexes are deprotonated, they are prone to formation of polymeric compounds.9 Although the reduction of (RuLoMeCI2hO is begun at pH 1, by the end of the reaction the pH is substantially higher, approximately 3 by pH paper. The higher pH could encourage the slow formation of 3. Possibly, if the solution were acidified after the removal of Zn, the formation of the dimer could be prevented. Alternatively, if higher yields of the dirner were desired the pH could be adjusted accordingly. Although it is dirneric, 3 is synthetically interesting, because it its aquo ligand can be substituted easily by other ligands. This property was first discovered in an NMR tube sample of 3 in methanol. After 24 hr., 3 had been convened partially to a new species with slighdy different chemical shift. This suggested that a OCH3- complex, perhaps [RuLoMe(OCH3)(Il-OH)h. had formed This result led us to investigate whether 3 would react with other nucleophilic ligands. Indeed, when 3 was dissolved in neat pyridine, a new yellow complex was formed, which displayed resonances in the IH NMR spectrum (CD2CI2, 500 MHz) at /) 2.37 (OTs), 2.69 (OCH3), 3.43 (OCH3), 4.11 (OCR3), 4.96 (C5H5) , 6.88 (py), 7.16 (py, OTs), 7.71(OTs), and 8.31(py). If this reaction is general, it could be used to prepare amino analogs of [Ru(IV)(H20)(Il-O))z2+ and perhaps even a mixed oxe-imido complex. Preparation of [RuLOMe(OH2)(Il-O)hOTf2 In the course of our investigations of the reactivity of (RuLoMeCI2hO an improved route to [RuLoMe(OH2)(Il-O))zOTf2, 4, was discovered. This compound had been prepared previously by reaction of Ru04 and NaLoR in dilute sulfuric acid,10 but the yields for the reaction with NaLoMe were quite low (10-25%). It was found that refluxing a 15 mixture of (RuLoMeCl:2hO and AgOTf followed by addition of HOTf produced 4 in approximately 55% yield In addition to the improved yield, the procedure is advantageous in that the starting materials are easier to prepare and handle. The best yield for the procedure is obtained when the mixture is refluxed for 1 hr. and worked up immediately. Longer heating at lower temperatures leads to the production of yellow and brown products instead of the desired dimer. These products were not characterized. Conclusions The work presented in this thesis investigates the synthesis of low oxidation state coordination complexes of ruthenium incorporating the anionic ligand, [CpCo(P(O)(OMehhl-, (LOMe)· Several routes to RuLoMe(H20h+ were outlined. These routes produced two major products, possibly related by acid-base chemistry. All of these routes may lead to a monomeric aquo species, if the pH of the reactions is carefully controlled. A new dimeric ruthenium aquo species with the proposed structure [RuLoMe(H20)(J.L-OH)h was isolated. This complex does not provide a straightforward route to the target monomeric oxo compounds, but could conceivably provide a entry into synthesis of ruthenium dirners containing amines and other ligands. Finally, a new route to [Ru(IV)(H20)(J.L-O)hOTf2 was discovered which is more convenient and is higher yielding than the previously published procedure. 16 References 1. Rard, J.A. Chern. Rev. 1985,85(1), 13-14. 2. Klliui, W.; Otto, H.; Eberspach, W.; Buchholz, E. Chern. Ber. 1982,115, 19221933. 3. Gill, T.P.; Mann, K.R. Organornet. 1982, 1, 485-88. 4. Rapaport, 1.; Helm, L.; Merbach, A. E.; Bernhard, P.; Ludi, A.lnorg. Chern. 1988,27, 873-879. 5. The rate of water exchange onto the similar [Ru(H20)61 2+ is 1.8 ±O.2 xlO- 2 s-l (Ref. 4). 6. Hung, y.; Kung, W-J.; Taube, H. lnorg. Chern. 1981,20, 457-463. 7. Llobet, A.; Hodgson, D. J.; Meyer, T. J.lnorg. Chern. 1990,29, 3760-3766. 8. pH 5.6 CNa4P20]IH3P04 buffer), H plane graphite electrode. 9 Harzion, Z. ; Navon, G. lnorg. Chern. 1980, 19, 2236-2239. Kim, T.D.; McNeese, T.J.; Rheingold, A.L. lnorg. Chern. 1988,27, 2554-2555. 10 Power, J.M.; Evertz, K.; Henling, L.; Marsh, R.; Schaefer, W.P.; Labinger, J. A.; Bercaw, J.E. lnorg. Chern. 1990,29, 5058-5065. 17 Experimental Section All solvents were reagent grade and were used as received unless otherwise noted. RuC13·3H20 (99.9% metal basis) was purchased from Johnson Matthey. CsH5Li, N(CH2CH3)4CN, HPO(OC{;H5h, KPF6, ~PF6, CF3S03H, and AgCF3S03 were purchased from Aldrich Chemical Co. p-CH3YlH4S03·H20 was obtained from Spectrum Chemical. (Ru(~)Chhl and RuloMe(~)PF62 were prepared by literature methods. NaloMe was prepared by the literature method3, and could be recrystallized from CH2Cl2lacetone. (RuloMeCI2hO, previously prepared by Kaspar Evenz, was prepared by stirring equimolar amounts of NaloMe and RuC13·3H20 in absolute ethanol for 24 hr., removing the solvent by rotary evaporation, and washing the resulting brownred solid with diethyl ether and methanol. RuLOMe(CH3CNhPF6- 0.775 g RuLoMe(C(iH6)PF6 (0.001 mol) were dissolved in 225 ml CH3CN. The solution was degassed by bubbling with N2 for 15 minutes, and irradiated in a 100 W Ace-Hanovia immersion lamp apparatus. The reaction was monitored by IH NMR spectroscopy. After 46 hr the reaction was >98% complete and the solvent was removed in vacuo. Yield 0.682 g (83%) yellow-brown powder. IH NMR (500 MHz, D20) 0 S.30 (s, SH, C5H5), 3.72 (virtual quartet, 18H, OCH3), 2.63 (s, 9H, CH3CN). IR (nujol mUll) 2267 (m), 1601(w), 1617(w), 1174(w), 1120(s), 1107(s), 1040(s), lOO6(s), 873(shoulder), 844(s), 777(m), 731(m), 631(shoulder), 601(m),557(m). [Co(CsHS)(P(OC6HShh]- N(C2HS)4+ (NE4LOPh). 44.8 g CO(C5H5h (0.237 mol) was heated with 200 ml (1.04 mol) HPO(OPhh to 120 °C under Ar for 20 hr. The resulting yellow solid was filtered in air and washed with ethanol and n-pentane. The solid was dissolved in CH2Ch and filtered to remove a black residue. Yield of CoCLoPhh after removal of solvent was 24.4 g (18%). 9.35 g (0.055 mol) CoCLoPhh was dissolved in 500 ml CH2Cl2. S.O g N(C2H5)4CN (0.032 mol) were dissolved in 75 ml CH2Cl2 and added to the CoCLoPh>2. The orange mixture was stirred under air for two days. Solid 18 orange Co(CNMN(C2Hs)41 was removed by filttation. Solvent was removed fonn the filttate by rotary evaporation, and the residue was redissolved in acetone, then filtered to remove excess CoCLOPh )2' The filttate was stripped again, then dissolved in toluene. The solution was filtered to remove any remaining N(~HS)4CN, and the solvent was removed by rotary evaporation to reveal a yellow oil, which was taken up in CH2CI 2 and dried over MgS04 (Caution! LoPh fonns MgCLoPhh upon prolonged standing with MgS04!). Removal of the solvent by rotary evaporation yielded a yellow solid soluble in CHCI3, CH2CIz, and toluene. 1H NMR (500 MHz, CDCI3) 1) 1.34 (t, 12H, N(CH2CH3)4), 3.39 (q, 8H, N(C!!zCH3)4), 5.52 (s, 5H, CSHS), 6.65 (m, 18H, ~S), 7.00 (m, 12H, C611s)· [RUIIILOMe(H20)(~-OH)h[CH3C6H4S03h ,J.. 7.13 g p- CH3C6ll4S03·HzO were added to a Schlenk flask and dissolved in 375 m1 distilled H20 to make a 0.1 M solution, which was degassed by sparging with AI for 20 min. 1.5 g (RuLoMeC12hO (1262.3 g/mol, 1.2 mmol) was added to the flask under AI flush. 15 g granular Zn was swirled with 0.75 g HgCI2 in 50 m1 H20 and rinsed with distilled water. The Zn was added to the Schlenck flask under AI flush, and the solution was stirred magnetically until bubbling has ceased (ca. 10 min.). The purple solution was filtered under air to remove Zn, then stirred under AI overnight, during which time yellow needlelike crystals had fonned. These were filtered and washed with a small amount of H20. Yield 0.600 g (%). This material is soluble in water, CH30H, CH2CI2, and TIIF. It is slightly air-sensitive and should be stored under an inert atmosphere. IH NMR (500 MHz, D20) 1) 2.41 (s, 6H, tos), 3.15 (virtual t, 12H, OCH3), 4.07 (virtual q, 12H, OCH3), 4.18 (virtual q, 12H, OCH3), 5.51 (s, lOH, CSHS), 7.38 (d, 4H, tos), 7.70 (d, 4H, tos). (RuLOMe(OH2)Oh(CF3S03h. 0.1000 g (RuLoMeCI2hO (1262.3 g/mol, 7.9xlO-5 mol) and 0.0812 g AgOTf (3.16 xlO-4 mol) were refluxed in 30 m1 H20 for 1 hr, during which time the solution became green and an AgCI precipitate was fonned. The solution was filtered while hot to remove the AgCI (0.0297 g, 66%). To the yellow-green 19 filtrate was added a few drops of CF3S(>:3H. A green solid precipitated, leaving a yellow solution. The precipitate was collected on a fIne scintered glass frit and washed with a small amount of water. Yield: 0.0692 g green solid. The IH NMR spectrum of this material was identical to material prepared according to the published method,4 but contained small amounts of 3 and a paramagnetic impurity which was detected as broad humps in the IH NMR spectrum. H2LOMePF6. NaLoMe was dissolved in 0.10 M H2S04. An excess of ~PF6 was added to precipitate H2LoMePF6 as large golden crystals. (The neutral monoprotonated form of this ligand is a yellow oil). IH NMR (90 MHz, CDCI3) 1) 13.2 (s, 2H), 5.5 (s, 5H, CsHS), 3.8 (virtual quartet, 18 H, OCH3). 20 References 1. Bennett, M.A.; Smith, A.K. J. Chern. Soc . Dalton Trans. 1974, 233. 2. KHtui, W.; Otto, R.; Ebersprach, W.; Buchholz, E.; Chern. Ber. 1982, 115, 1922-1933. 3. KHiul, W.; Lenders, B.; Ressner, B.; Evertz, K. Organomet. 1988, 7, 13571363. 4. Power, J. M.; Evertz, K. E.; Renting, L.; Marsh, R.; Schaefer, W. P.; Labinger, J. A.; Bercaw, J. E. Inorg. Chern. 1990,29, 5058-5065. I 7.'.. CslJ,j PPM CD)()() jv oCt!;) ~ CD)OD oCl:l:J J 10- ; ; ~ ~~ Cl:I3C 1 6'·'4S03" ..'~ •.'.. .,.lLj~o~JTo~--"Dl,.. . ',. CH3Cr.!i1S03" , -'- - '-"..'.'.-~.O~.J;"O"-~-;J CH3C,.l:!4S03 " ~ [RuIJlL O M e(Il -0H )( H20))2 , CD30D, SOO MHz_ ...1.-.....-..0--1- - . .50 1 . 00 - tv - - I" 9.0 • 0' ,--0--'- '" r 8.5 B. O 7 .5 7. 0 .----r-...,........,.. .. . "' ' 6.5 6 .0 5 .5 .··r ·........ ' --r r ' --,--,--r-.,-. ~ I CSI-I.l 5.0 'i' , .4 . 5 ,. PPM 4.0 - ~r-r . \ I / 3.5 0 I OCHJ I I 3 .0 2.5 2. 0 l ·,..- .. - r-------r-r ~r ....- . 20Clll [Ru(lIz0)61 0Ts z + NaLOMc. OzO. t .5 • I' I t .O • I .5 I ~ ," ,' 0.0 - .5 ' ·....- I - -r- T.~~-.~ N N • '- I ....- -r--r--T - -r---,.----. --,- .- 1-. - .. · · -, 1 - · T' 9.0 8 .5 B.O 7.5 . 1< .----,----.,----.,--.." 7.0 I C5~ \ ,II 1m! '" ,., 1",1" •• I I 'UWll /1 .. ~ Ol '.n I h~~ Ii I I~ IRuU/LoMc(IL·OHl(H20 lI2. 0CH) 6.5 6.0 5.5 5.0 4.5 <1.0 PPM '" N 0 '" • 35 3.0 2.5 2.0 1. 5 1.0 ' .5 I 0.0 , ---.---r"""TT"""""' 1 ,--. - .5 ~J ~ ---!--.-.--._ ._._J------_._________._.__.___ '!? J I OCli~ I I I L ( .. I I I, ..,.--.--.........,...,--. ....... "T--o....,..,. ....I-T-.,. ., . . ,. . . ., . . . .......-,-.-. -.-r-r,.-,--.--.,·..,. . . . . +-rr..,--,~ --."""'T..,.....,....,......., -,- ' - l-r--.-" ,......., 1 , .. " " " ' ' ' ~ . ;' ,l.......__._, :'J ~ \ OCHJ (IL·OHl( fI 20 )12 (Rullll()Mc IRuLoMC< Cr, H 6l JCI. C5 U 5· --··-~<---'~r~.r~' -r - ·- - C6 H 6 I RuLoMC<C6116lJCI. Cr,H6. IRuLoM c(C6H6lJCI.OC H). [RuLOMc(C6116)]CI in D20. hv 17 hr. 8 .5 8 .0 7.5 TT· CHCI, I···-"-·--'-~-.-----r--'-- I 7.0 _-,-__ ',. r 1 _____ . ~ .) 0 ... 6 .5 'T' 6.0 '" ' r- T 5 .5 no -' r- CsH, j 'g: '~' ..;.; ,... , '--;--' .~. .' / Irl' - <lC,-J:!,j ~~ 5.0 -TITT- 4.5 PPM 'I' 4.0 ',' '" iii 'rg' '--r "-' 3.5 ~ r- CllzCH, ~ '" ] ,0 2. 5 'I 2.0 TT'- ~--'----r---'-----"-- no :01 1.5 'I' .... ~' N"; • - LA 1~ , I(CSHS)Co (P(O)(OC6Hs))]IIN(CH2CH})4J; (NEl4LOPh·). CDCl}. 500 MHz. 1.0 'I' if ,5 ..A I RuLoMc(CH3CN))IPF6. C51::1,5. i 1. \ U DSS A DSS IRul()Mc(CI:I~CN)3I1'F6 IRul()Mc(CIl3CN)3IPF6.0C1:!1 ~ ~i j .5 DSS 1,-"--,_ • ! 9 .0 8 .5 0 . 0 7. 5 7 . 0 6 . 5 6 . 0 5 . 5 5 . 0 4 . 5 4 . 0 PPM ) .5 3 .0 2.5 2.01.51.0 0 .0 - .5 ." "." , ". , ,,-1~,,. ,~J.~ , , ~,,~~., ~ _ _ _ _----'----!'c'L'--"l t [RuLOMc(CB)CN}J JPF6, D2() containing I)SS, 500 Mllz. 5000 . 000 16384 RS NS 20 32 VI tv .02 - .15IP 0 ,0 9 . 24SP • 35 . 00 0 ,0 0 .0 6ll PO 9000 . 000 6300 12 100 127 . 5 291 IS SR . 286 I 6639 . 46 PPM/ eM DC 1. 000 HZ/CM t 42 . a40 HI F2 tV FI ex NO lB 68 DP 02 DE DR ON Flit TE 5 . 000 1 . 638 " .5 RD AO 0 .0 VO PN 5M2 5000 . DOO HZ / PT . 610 SM TO SF 500.137 SFO 500 . 130 SF02 500 . 130 SF2 500 . 137 S'( 83 . 0 01 8760 . 000 SI 16384 SOL VENT CDC 13 EPes . 900 BBBA