Coordination studies of 1,2-bis(diphenylphosphino)ethane with di-μ-hydroxo dinuclear complexes of tungsten(IV) and molybdenum(IV) J. Serb. Chem. Soc. 81 (1) 47–55 (2016) UDC 546.774+546.784+547.636’416:544.433.21 JSCS–4826 Original Scientific paper 47 Coordination studies of 1,2-bis(diphenylphosphino)ethane with di-μ-hydroxo dinuclear complexes of tungsten(IV) and molybdenum(IV) MAKOTO MINATO1*, TAKASHI ITO1 and JIAN-GUO REN2 1Department of Materials Chemistry, Graduate School of Engineering, Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku, Yokohama, 240-8501, Japan and 2Department of Chemistry, Shanxi University, Taiyuan, 030006, China (Received 1 May, revised 28 July, accepted 29 July 2015) Abstract: New trifluoroethoxido phosphine complexes [Cp2M(η 1-dppe)- (CF3CH2O)] + and [Cp2(CF3CH2O)M(μ-dppe)MCp2(CF3CH2O)] 2+ (M = Mo or W, Cp = η-C5H5 and dppe = Ph2PCH2CH2PPh2) were prepared by reaction of the cationic di-μ-hydroxido dinuclear complex of molybdenocene or tungs- tenocene [Cp2M(μ-OH)2MCp2] 2+ with dppe. From the 1H- and 31P-NMR data, the configurations of the products could be assigned. Furthermore, X-ray crystallography was used to definitively identify one of the products [Cp2(CF3CH2O)Mo(μ-dppe)MoCp2(CF3CH2O)] 2+, which crystallizes in the space group P21/c(#14) with a = 12.230(5) Å, b = 11.149(5) Å, c = 28.966(7) Å, β = 101.07(3)°, V = 3876(2) Å3 and Z = 2. It was ascertained that the amount of dppe added to the reaction mixture could influence the product distribution. A mechanism involving initial replacement of the hydroxido ligand by the alkoxido group followed by nucleophilic attack of the phosphine is proposed based on the reaction profile. Keywords: molybdenocene; tungstenocene; dinuclear complexes; dppe. INTRODUCTION Previously, it was shown that cationic di-μ-hydroxo dinuclear complexes of molybdenocene and tungstenocene [Cp2M(μ-OH)2MCp2]2+ (Cp = η-C5H5; M = = Mo (1a) or W (1b)) could be conveniently prepared by reactions of Cp2MH2 and Cp2M(OTs)2 (OTs = p-CH3C6H4SO3) in aqueous acetone.1 These novel dinuclear complexes have attracted much attention due to their ability to catalyze intra- and inter-molecular H/D exchange reactions,2 and reduction of ketones3 and nitriles.4 * Corresponding author. E-mail: minato@ynu.ac.jp  The authors are pleased to dedicate this paper to Professor Rastko D. Vukicevic in the year of his 65th birthday with continued best wishes for his retirement. doi: 10.2298/JSC150501066M 48 MINATO, ITO and REN The hydroxo groups in complexes 1 were sufficiently labile to undergo dis- placement by a wide variety of substrates, affording molybdenocene and tungs- tenocene derivatives and hence, they are useful as precursors of these types of compounds.1,5,6 It was found that reactions between complexes 1 and monoden- tate tertiary phosphines always proceeded with the concomitant incorporation of co-existing alcohols to yield novel alkoxido phosphine complexes [Cp2M(PR′3)- (RO)]+(OTs–) (2, R′ = Et, Bun, and Ph; R = Me, Et, Pri, CF3CH2 and Ph) (Scheme 1).1c It was proven that these reactions occurred spontaneously under mild conditions (20–50 °C). Furthermore, on dissolving in benzene containing a small quantity of water, the resulting complexes 2 readily and quantitatively rev- erted to the original complexes 1 with liberation of the phosphine ligands, which suggests the reversibility of the reactions. Scheme 1. The reactions between complexes 1 and monodentate tertiary phosphines. The syntheses of complexes 2 showed the following intriguing trends. In methanol, ethanol, or 2-propanol, only basic phosphines, such as triethylphos- phine or tributylphosphine, reacted with 1 and no substitution of the hydroxo bridging groups by triphenylphosphine were observed in these solvents. How- ever, in moderately acidic alcohols, such as trifluoroethanol, or in the presence of phenol, the less reactive triphenylphosphine reacted smoothly to afford 2 in good yields. Therefore, the outcome of the reactions appears to be dependent on the nucleophilicity of the tertiary phosphines and the acidity of the co-existing alco- hols. In addition, of particular interest was the fact that no compounds resulting from the incorporation of two phosphine ligands were formed; the labile alkoxido ligand bound to the central metals of 2 was not displaced by a second phosphine ligand even in the presence of excess tertiary phosphine. These results are some- what puzzling and questions remain regarding a reasonable reaction mechanism. As a natural extension of this study, the study of the reactions of the dinuc- lear complexes 1 with a chelating ligand was of interest. The focus of the present paper was the reactions between 1 and dppe (where dppe represents 1,2-bis- (diphenylphosphino)ethane), since Green et al. reported that the reactions of analogous halogenido complexes [Cp2MI2] (M = Mo or W) with dppe resulted in formation of complexes [Cp2M(dppe)]2+ in which the dppe ligand coordinates to the metals in a bidentate mode.7 Thus, considering the chelate effect of the dppe ligand, the formation of a similar bidentate-type complex could be anticipated in THE NEW ALKOXIDO DPPE COMPLEXES 49 the present case. The results of such experiments are reported herein. In addition, another purpose of this study was to propose a reasonable reaction mechanism pertaining to all reactions between 1 and tertiary phosphines. RESULTS AND DISCUSSION The reactions of 1 with dppe were run in CF3CH2OH/C6H6 at room tem- perature. The preliminary results demonstrated that the reaction afforded two organometallic species that were assigned as mononuclear complexes 3 con- taining a monodentate dppe ligand and bridged dimetallic complexes 4 (Scheme 2). The reaction of dppe did not proceed at all in methanol or in ethanol, which is similar to the reaction of triphenylphosphine. It was further ascertained that the amount of dppe added to the reaction mixtures could influence the product dis- tribution. It was then decided to investigate the reaction conditions in order to gain a better understanding of the factors controlling the selectivity of the pro- ducts. The results are summarized in Table I. Scheme 2. The reactions between complexes 1 and dppe to produce mononuclear complexes 3 and bridged dimetallic complexes 4. TABLE I. Products obtained from the reactions of 1 with dppe in the presence of CF3CH2OH under several conditions Compound Complex / mmol dppe, mmol CF3CH2OH/C6H6, mL/mL Yield, % 3a 1a / 0.294 1.300 1.5/5.0 81 3b 1b / 0.135 0.880 1.5/5.0 82 4a 1a / 0.341 0.339 1.5/1.5 50 4b 1b / 0.159 0.158 2.0/2.0 43 As shown in Table I, good yields of complexes 3 could be obtained if a large excess of dppe was added to 1 in solution at room temperature. On the other hand, the reactions of 1 with dppe in a mole ratio of 1:1 gave 4 as the major pro- ducts. It is worth emphasizing that complexes containing a chelating dppe ligand were not observed in the reactions of complexes 1. Thus, dppe does not yield the expected chelated complexes. Complexes 3 are soluble in methanol, ethanol, trifluoroethanol and acetone, while 4 are soluble in the foregoing alcohols and essentially insoluble in acetone. It was found that complexes 3 and 4 were stable to air in the solid state. Com- 50 MINATO, ITO and REN plexes 3 and 4 were characterized by standard methods, in particular 31P{1H}- and 1H-NMR spectroscopies, as well as by X-ray structural determination of 4a. In addition, the combustion analyses for 3b and 4a were consistent with their spectroscopic properties (see Experimental).8 Selected NMR data are collected in Table II. TABLE II. Selected 1H-NMR (270 MHz, CD3OD, 293 K) and 31P-NMR data (202 MHz, CD3OD, 293 K, J in Hz) for complexes 3 and 4 (δ / ppm) Compound 1H-NMR 31P-NMR 3a 7.2–7.6 (20H, m, Ar-H), 5.46 (10H, d, JPH = 1.83, Cp), 3.20 (2H, q, JFH = 9.56, CF3CH2O), 2.6–2.9 (2H, br, MoPCH2CH2), 1.6–1.9 (2H, br, MoPCH2CH2) 29.9 (s), –12.0 (br) 3b 7.2–7.6 (20H, m, Ar-H), 5.42 (10H, d, JPH = 1.22, Cp), 3.45 (2H, q, JFH = 9.56, CF3CH2O), 2.6–2.8 (2H, br, WPCH2CH2), 1.7–1.9 (2H, br, WPCH2CH2) Not measured 4a 7.2–7.6 (20H, m, Ar-H), 5.38 (20H, d, JPH = 1.83, Cp), 3.20 (4H, q, JFH = 9.56, CF3CH2O), 2.2–2.4 (4H, br, MoPCH2) 28.0 (s) 4b 7.2–7.6 (20H, m, Ar-H), 5.35 (20H, d, JPH = 1.22, Cp), 3.40 (4H, q, JFH = 9.56, CF3CH2O), 2.1–2.3 (4H, br, WPCH2) Not measured The 1H-NMR spectra of complexes 3 are similar, showing five different resonances besides the resonances due to the TsO protons, and resemble those of 2 in a previous paper.1c As shown in the Table II, resonances for the cyclopen- tadienyl ring protons of 3a and 3b appear at around δ 5.5 ppm as a doublet coupled to phosphorus; this represents significant shielding of these protons com- pared with the chemical shifts in the parent dinuclear complexes 1 (δ = 6.0 ppm) but are compatible with the observed chemical shifts of 2. As expected, the spectra of 3 show two separate signals for the CH2CH2 fragment of the dppe ligand at δ around 2.7 and 1.8 ppm. The CF3CH2O units in 3a and 3b show a quartet at δ 3.20 (3a, JFH = 9.56 Hz) and 3.45 ppm (3b, JFH = 9.56 Hz), res- pectively. Assignment of the monodentate coordination mode of the dppe ligand was based on the observation of two discrete equal intensity resonances at δ around 29 and –12 ppm in the 31P{1H}-NMR spectrum of 3a. The downfield resonance is assignable to a metal-bonded phosphorus atom, while the high-field resonance can readily be assigned to an uncoordinated phosphorus atom since this chemical shift is very similar to that of free dppe. In the 1H-NMR spectrum of complex 4a, the Cp protons occur at δ = 5.4 ppm as a doublet with a P–H coupling constant of 1.83 Hz, while the CF3CH2 protons appear as a quartet with an F–H coupling constant of 9.56 Hz at δ = 3.2 ppm. Unlike complex 3a, the spectrum of 4a shows only one multiplet for the THE NEW ALKOXIDO DPPE COMPLEXES 51 CH2CH2 fragment at around δ = 2.3 ppm. Furthermore, the 31P{1H}-NMR spectrum of 4a contains only one resonance at around δ 28 ppm, which was assigned to the phosphorus atom on molybdenum. Evidently, these results indi- cate that 4a contains the Mo(μ-dppe)Mo group. The spectrum of complex 4b is almost identical to that of 4a, supporting a structure analogous to that of 4a. Complex 4a was fully characterized by X-ray crystal structure determin- ation. Dark red crystals suitable for the X-ray analysis were obtained by recrys- tallization from CF3CH2OH/Et2O. The more important bond lengths and bond angles are given in Table III. A summary of the crystallographic data is given in Table IV (see Experimental). As was anticipated from the NMR consideration, the analysis indicated that the molecule has a symmetric structure in the solid state with the formulation [Cp2(CF3CH2O)Mo(μ-dppe)Mo(OCH2CF3)Cp2]- (OTs)2(CF3CH2OH)2 in which two molecules of trifluoroethanol are included as a crystallization solvent. An ORTEP drawing of the cation of 4a is shown in Fig. 1. The TsO molecules were found to be disordered and were omitted for clarity. From this drawing, it is clear that the two molybdenum centers are held together by a dppe ligand. The coordination sphere around the metal center is completed by bridging dppe ligand, CF3CH2O group, and two cyclopentadienyl rings, which are arranged in the form of a distorted tetrahedron. The cyclopentadienyl rings are bound to molybdenum in an η5 fashion and each of the ring carbon atoms are coplanar. Thus, complex 4a has geometry typical of bent metallocene. TABLE III. Selected bond distances and angles for 4a Bond Distance, Å Bond Distance, Å Mo1–P1 2.541(4) Mo1–O1 2.070(10) Mo1–C1 2.29(2) Mo1–C2 2.33(2) Mo1–C3 2.37(3) Mo1–C4 2.33(2) Mo1–C5 2.31(3) Mo1–C6 2.31(2) Mo1–C7 2.27(2) Mo1–C8 2.35(2) Mo1–C9 2.37(2) Mo1–C10 2.34(2) O1–C11 1.45(2) C11–C12 1.37(4) Bond Angle,  Bond Angle,  P1–Mo1–O1 75.5(3) Mo1–P1–C13 113.0(5) Mo1–P1–C14 111.1(5) Mo1–P1–C20 116.9(5) Mo1–O1–C11 119.4(10) P1–Mo1–C2 132.2(6) The structure can be compared with that of the related alkoxido phosphine complex [Cp2Mo(PBun3)(CF3CH2O)]+ (2a).1c The Mo–P bond distance (2.541(4) Å) is remarkably similar to that exhibited by 2a (2.540(4) Å). On the other hand, the Mo–O bond distance of 2.070(10) Å is significantly longer than that found in 2a (2.019(8) Å). Furthermore, the O–Mo–P angle of 75.5(3)° is slightly greater than the corresponding angle of 74.4(3)° found in 2a. The most 52 MINATO, ITO and REN interesting feature of the structure is the observation of the long O–C bond dis- tance of 1.45(2) Å, which is considerably longer than the ca. 1.39 Å found in 2a. TABLE IV. Summary of the crystal structure data for 4a•2CF3CH2OH Formula C68H68F12P2O10Mo2S2 M / g mol-1 1591.21 Crystal system monoclinic Space group P21/c (#14) a / Å 12.230(5) b / Å 11.149(5) c / Å 28.966(7) β / ° 101.07(3) V / Å3 3876(2) Z 2 ρc / g cm -3 1.363 μ / cm-1 4.97 R / wR 0.102/0.143 GOF 2.28 T / °C 25 Fig. 1. ORTEP drawing of the cation of complex 4a with thermal ellipsoid plots (40 % probability). The hydrogen atoms have been omitted for clarity. A mechanism accounting for the reaction pathways is proposed in Scheme 3 based on literature precedents and the reaction profile. In addition, certain gen- eral observations pertained to all reactions between 1 and the tertiary phosphines are included in this study. As mentioned in the Introduction, each step in Scheme 3 is likely to be reversible. It is well known that most 18-electron complexes undergo ligand substitution reactions via dissociative pathways.9 Hence, it is conceivable that the first step in the sequence leading to formation of 2–4 (D) is THE NEW ALKOXIDO DPPE COMPLEXES 53 dissociation of 1 into the monomeric 16-electron complexes A. In this case, it is possible that it is the resonance limiting form A′, which may be formally viewed as a protonated oxo-complex.10 Taking into account the fact that the reaction between 1 and a tertiary phosphine is very susceptible to the co-existing alcohol as noted in the Introduction, it seems likely that the next step consists of nucleophilic attack of an alcohol on the metal center. Then proton transfer to the hydroxy group occurs to give alkoxido complex B; this process is quite similar to the reported hydrolysis of 1.2 Subsequent elimination of water from B produces an unsaturated species C. This dehydration step seems to be a facile process since π-donation by an alkoxido ligand is well established toward early transition metals11 and so the canonical form C′ would make a greater contribution to the hybrid. Inevitably, the final step in the mechanism is nucleophilic attack of a tertiary phosphine on the central metal to afford the product D. Scheme 3. Possible mechanism for the formation of the complexes 2–4. EXPERIMENTAL General procedures All manipulations were performed under an inert nitrogen or argon atmosphere using standard Schlenk techniques. Commercially available reagent grade chemicals (Wako Chemical) were used as such without any further purification. Solvents were purified according to standard procedures. All NMR spectra were recorded on a JEOL JNMEX-270 spectrometer or a JEOL JNMGX-500 spectrometer. 31P{1H}-NMR peak positions were referenced to external H3PO4. The di-μ-hydroxido dinuclear complexes [Cp2M(μ-OH)2MCp2] 2+ (Cp = η-C5H5; M = Mo (1a) or W (1b)) were prepared by literature procedures. 1c Reaction of 1 with excess dppe A solution containing 1a (0.244 g, 0.294 mmol) and dppe (0.500 g, 1.260 mmol) in CF3CH2OH/C6H6 (1.5 mL/5 mL) was stirred at room temperature for 140 h. During this time, the solution changed from green to red. From the resulting solution, the solvent was evapor- ated to dryness under reduced pressure. The residue was washed successively with hexane and diethyl ether and then extracted with acetone. The extract was reduced to dryness and the residue was washed with hexane and diethyl ether to yield 3a (0.430 g, 81 %) as an orange– 54 MINATO, ITO and REN –red powder. This procedure was also applicable to the synthesis of the tungsten analogue 3b (yield = 82 %). 3b: Anal. Calcd. for C45H43F3O4SP2W: C, 55.00; H, 4.41 %. Found: C, 54.28; H, 4.53 %. Reaction of 1 with 1 equiv. of dppe A solution containing 1a (0.289 g, 0.341 mmol) and dppe (0.135 g, 0.339 mmol) in CF3CH2OH/C6H6 (1.5 mL/1.5 mL) was stirred at room temperature for 15 h. During this time, the solution changed from green to red. From the resulting solution, the solvent was evapor- ated to dryness under reduced pressure. The residue was washed successively with hexane, diethyl ether, and acetone, and then extracted with ethanol. The extract was reduced to dryness and the residue was washed with hexane and diethyl ether to yield 4a (0.237 g, 50%) as an orange–red powder. Purification of the product by recrystallization from CF3CH2OH/ /Et2O afforded dark red crystals in the form of flat plates. This procedure was also applicable to the synthesis of the tungsten analogue 4b (yield = 43 %). 4a: Anal. Calcd. for C64H62F6O8S2P2Mo2: C, 55.26; H, 4.49 %. Found: C, 55.03; H, 4.53 %. X-Ray crystallographic study of 4a A crystal suitable for X-ray crystallography was grown in CF3CH2OH–Et2O. The thus obtained dark red crystal was mounted on a glass fiber. Measurement was made on a Rigaku AFC5R diffractometer using Mo Kα radiation (λ = 0.71068 Å) for data collection. The unit- -cell parameter was determined by least squares fitting of 25 reflections with a range 21.39 < < 2θ < 25.95°. The parameters used during the collection of diffraction data are given in Table IV. The structure was solved and refined using Fourier techniques. The non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included but not refined. CONCLUSIONS New trifluoroethoxido phosphine complexes [Cp2M(η1-dppe)(CF3CH2O)]+ (3) and [Cp2(CF3CH2O)M(μ-dppe)MCp2(CF3CH2O)]2+ (4) were prepared by reactions of the cationic di-μ-hydroxido dinuclear complex of molybdenocene and tungstenocene [Cp2M(μ-OH)2MCp2]2+ (1) with dppe. The products of the reactions were identified using 1H- and 31P-NMR spectroscopy. In addition, X-ray structural data on 4a clearly established a bridged dimeric structure. It was ascertained that the amount of dppe added to the reaction mixtures could inf- luence the product distribution. Thus, the reaction of 1 with dppe in a molar ratio of 1:1 gives complexes 4 as the major products, while good yields of complexes 3 were obtained if a large excess of dppe was added to 1. SUPPLEMENTARY MATERIAL Crystallographic data for the structural analysis has been deposited with the Cambridge Crystallographic Data Centre, as CCDC reference number 605332. A copy of this information may be obtained free of charge from The Director, CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K. (fax, +44-1223-336033; e-mail, deposit@ccdc.cam.ac.uk; web, http:// //www.ccdc.cam.ac.uk). Acknowledgments. We are grateful to Dr. Mikio Yamasaki of the Rigaku Corporation for the X-ray structure analysis. We also thank Dr. Masako Tanaka of the Tokyo Institute of Technology for the elemental analysis. THE NEW ALKOXIDO DPPE COMPLEXES 55 И З В О Д ИСПИТИВАЊЕ КООРДИНАЦИЈЕ 1,2-БИС(ДИФЕНИЛФОСФИНО)ЕТАНА СА ДИ--ХИДРОКСИДО ДИНУКЛЕАРНИМ КОМПЛЕКСИМА ВОЛФРАМА(IV) И МОЛИБДЕНА(IV) MAKOTO MINATO1, TAKASHI ITO1 и JIAN-GUO REN2 1 Department of Materials Chemistry, Graduate School of Engineering, Yokohama National University,79-5 Tokiwadai, Hodogaya-ku, Yokohama, 240-8501, Japan и 2 Department of Chemistry, Shanxi University, Taiyuan, 030006, China У реакцији катјонског ди--хидроксидо комплекса молибденоцена или волфра- моцена [Cp2M(μ-OH)2MCp2] 2+ са dppe лигандом синтетисани су нови трифлуороеток- сидо-фосфински комплекси, [Cp2M(η 1-dppe)(CF3CH2O)] + и [Cp2(CF3CH2O)M(μ- -dppe)MCp2(CF3CH2O)] 2+ (M = Mo или W, Cp = η-C5H5 и dppe = Ph2PCH2CH2PPh2). Нађено је да дистрибуција реакционих производа зависи од количине додатог dppe лиганда. Дискутован је механизам испитиване реакције који у првој фази укључује супституцију хидроксидо лиганда са алкоксидо групом, а затим у другој фази долази до нуклеофилног напада фосфинског лиганда. (Примљено 1. маја, ревидирано 28. јула, прихваћено 29. јула 2015) REFERENCES 1. a) J.-G. Ren, H. Tomita, M. Minato, K. Osakada, T. Ito, Chem. Lett. (1994) 637; b) M. Minato, J. Ren, H. Tomita, T. Tokunaga, F. Suzuki, T. Igarashi, T. Ito, J. Organomet. Chem. 473 (1994) 149; c) J.-G. Ren, H. Tomita, M. Minato, K. Osakada, M. Yamasaki, T. Ito, Organometallics 15 (1996) 852 2. a) C. Balzarek, D. R. Tyler, Angew. Chem. Int. Ed. 38 (1999) 2406; b) C. Balzarek, T. J. R. Weakley, L. Y. Kuo, D. R. Tyler, Organometallics 19 (2000) 2927; c) K. L. Breno, D. R. Tyler, Organometallics 20 (2001) 3864 3. a) L. Y. Kuo, T. J. R. Weakley, K. Awana, C. Hsia, Organometallics 20 (2001) 4969; b) L. Y. Kuo, D. M. Finigan, N. N. Tadros, Organometallics 22 (2003) 2422 4. K. L. Breno, M. D. Pluth, D. R. Tyler, Organometallics 22 (2003) 1203 5. J.-G. Ren, Z.-W. Wang, Z.-M. Zhang, D.-S. Liu, M. Minato, T. Ito, Chin. J. Appl. Chem. 17 (2000) 81 6. M. Minato, J.-G. Ren, M. Kasai, K. Munakata, T. Ito, J. Organomet. Chem. 691 (2006) 282 7. T. Aviles, M. L. H. Green, A. R. Dias, C. Romao, J. Chem. Soc., Dalton Trans. (1979) 1367 8. For complexes 3a and 4b, no samples suitable for combustion analyses are available at present. 9. J. P. Collman, L. S. Hegedus, J. R. Norton, R. G. Finke, Principles and Applications of Organotransition Metal Chemistry, University Science Books, Mill Valley, CA, 1987, p. 247 10. a) M. L. H. Green, A. H. Lynch, M. G. Swanwick, J. Chem. Soc., Dalton Trans. (1972) 1445; b) N. D. Silavwe, M. Y. Chiang, D. R. Tyler, Inorg. Chem. 24 (1985) 4219 11. J. C. Huffman, K. G. Moloy, J. A. Marsella, K. G. Caulton, J. Am. Chem. Soc. 102 (1980) 3009.