Pyramidal Stability of Chiral-at-Metal Half-Sandwich 16-Electron

3514

Organometallics 2008, 27, 3514–3525

Pyramidal Stability of Chiral-at-Metal Half-Sandwich 16-Electron Fragments [CpRu(P-P′)]† Henri Brunner,*,‡ Manfred Muschiol,‡ Takashi Tsuno,*,§ Takemoto Takahashi,§ and Manfred Zabel‡,# Institut fu¨r Anorganische Chemie, UniVersita¨t Regensburg, 93040 Regensburg, Germany, and Department of Applied Molecular Chemistry, College of Industrial Technology, Nihon UniVersity, Narashino, Chiba 275-8575, Japan ReceiVed February 19, 2008

The chiral-at-metal diastereomers (RRu,RC)- and (SRu,RC)-[CpRu(P-P′)Hal] (P-P′ ) (R)-Prophos and (R,R)-Norphos, Hal ) Cl, Br, and I) were synthesized, separated, and characterized by X-ray crystallography. Halide exchange and epimerization reactions were studied in 9:1 and 1:1 chloroform/ methanol mixtures proceeding at room temperature or slightly above according to first-order. The ratedetermining step in the Hal exchange reactions was the dissociation of the Ru-Hal bond, forming the pyramidal 16-electron intermediates (RRu,RC)- and (SRu,RC)-[CpRu(P-P′)]+, which maintain the metal configuration. These intermediates can invert their metal configuration or react with nucleophiles with retention of the metal configuration. The measured competition ratios showed that the inversion of the intermediates was slow compared to quenching with nucleophiles, indicating a high pyramidal stability of the 16-electron fragments (RRu,RC)- and (SRu,RC)-[CpRu(P-P′)]+ toward inversion in agreement with a basilica-type energy profile. Stereochemically this implies that substitution reactions in (RRu,RC)- and (SRu,RC)-[CpRu(P-P′)Hal] occur with predominant retention of configuration, however, accompanied by a well-defined share of inversion, a point overlooked when (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] were extensively used as starting materials in the synthesis of a variety of organometallic derivatives. The rates of the approach to the epimerization equilibrium were much smaller than those of the Hal exchange reactions, because in the basilica-type energy profile the intermediates (RRu,RC)-/(SRu,RC)[CpRu(P-P′)]+, formed in the cleavage of the Ru-Cl bond, have to cross another barrier of appreciable height for inversion. Increasing the methanol content of the solvent increased the rates of the Hal exchange and epimerization reactions. Obviously, the pyramidality of the fragments [CpRu(P-P′)]+ is enforced by the small P-Ru-P angles (83° in the Prophos derivatives and 86° in the Norphos derivatives of (RRu,RC)- and (SRu,RC)-[CpRu(P-P′)Hal]). Due to these small angles, intermediates (RRu,RC)- and (SRu,RC)[CpRu(P-P′)]+ resist planarization and thus inversion of the metal configuration. Introduction Dissociation of a ligand from an 18-electron complex leaves an unsaturated 16-electron species, which may be a stable compound or an intermediate ready for subsequent addition reactions. Does the 16-electron species maintain its structure with a vacant site in place of the dissociated ligand, or does it rearrange simultaneously to its formation or subsequently? Are the 16-electron species [CpRu(P-P′)]+, obtained on dissociation of Cl- from the three-legged piano stool complex [CpRu(PP′)Cl], planar or pyramidal (Scheme 1)? The alternative is relevant in particular for chiral-at-metal compounds1–3 such as [CpRu(P-P′)Cl], because an intermediate [(η-CnHn)MLL′] retains chirality as long as it is pyramidal, whereas it loses † Dedicated to Prof. Dr. Dr. h. c. mult. W. A. Herrmann, President of TU Munich, on the occasion of his 60th birthday. * To whom correspondence should be addressed. (H.B.) Fax: +49-9419434439. E-mail: [email protected]. (T.T.) Fax: +81 47 474 2579. E-mail: [email protected]. ‡ Universita¨t Regensburg. § Nihon University. # X-ray structure analyses. (1) Brunner, H. AdV. Organomet. Chem. 1980, 18, 223–278. (2) Consiglio, G.; Morandini, F. Chem. ReV. 1987, 87, 761–778. (3) Brunner, H. Angew. Chem. 1999, 111, 1248–1263; Angew. Chem., Int. Ed. 1999, 38, 1194-1208.

Scheme 1. Dissociation of Cl- from the Three-Legged Piano Stool Complex [CpRu(P-P′)Cl]: Is the Resulting 16-Electron Intermediate [CpRu(P-P′)]+ Pyramidal or Planar?

chirality when it is planar. Chiral-at-metal compounds are especially suitable to investigate such problems. In the present paper we describe our studies concerning [CpRu(P-P′)Hal] compounds and compare the new results with the [CpMn(NO)(PPh3)X] system, for which pyramidal intermediates have been established.

10.1021/om800148r CCC: $40.75  2008 American Chemical Society Publication on Web 06/11/2008

Half-Sandwich 16-Electron Fragments [CpRu(P-P′)]

Organometallics, Vol. 27, No. 14, 2008 3515

Scheme 2. Diastereomers 1a-c and 1a′-c′ of [CpRu(Prophos)Hal] and 2a-c and 2a′-c′ of [CpRu(Norphos)Hal] with Their Respective Configurationsa

a The priority sequence of the ligands for [CpRu(Prophos)Cl] is Cp > Cl > PCHMe > PCH2, whereas for the corresponding bromo and iodo compounds it is Br(I) > Cp > PCHMe > PCH2, which leads to different configurational symbols for the same relative configurations. The priority sequences for the Norphos derivatives are Cp > Cl > Pendo > Pexo and Br(I) > Cp > Pendo > Pexo (subrule 3 of the CIP system8–10). The two configurational symbols of (R,R)-Norphos were reduced to one (R) in formulas such as (RRu,RC) and (SRu,RC).

The parent [CpRu(P-P′)Hal] type compound is [CpRu(PPh3)2Cl], the structure of which is known.4 Chiral analogues are the diastereomers (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl], (R)-Prophos ) (R)-1,2-bis(diphenylphosphanyl)propane, which have been separated5,6 and used as starting materials for the preparation of optically active organometallic [CpRu(Prophos)X] compounds with retention of configuration.2 These reactions have been carried out in methanol at room temperature and in boiling methanol, which, as we will show in this paper, is crucial, because (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] (Scheme 2) epimerize in methanol solution (see below). Likewise the diastereomers of (RRu,RC)- and (SRu,RC)-[CpRu(Norphos)Cl] as well as (RRu,RC)- and (SRu,RC)-[CpRu(Norphos)I] (Scheme 2), (R,R)-Norphos ) (R,R)-2,3-bis(diphenylphosphanyl)bicyclo[2.2.1]hept-5-ene, have been separated.7 The configurational lability at the Ru center has been noticed, but detailed investigations have not been carried out. Scheme 3 with Fast and Scheme 4 with Slow Pyramidal Inversion. The chiral-at-metal half-sandwich compounds [CpMn(NO)(PPh3)COOR′] (R′ ) CH3 and L-C10H19)11–14 and [CpMn(NO)(PPh3)C(O)R′] (R′ ) CH3, C6H5, and p-C6H4R”)15–19 (4) Bruce, M. I.; Wong, F. S.; Skelton, B. W.; White, A. H. J. Chem. Soc., Dalton Trans. 1981, 1398–1405. (5) Consiglio, G.; Morandini, F.; Bangerter, F. Inorg. Chem. 1982, 21, 455–457. (6) Morandini, F.; Consiglio, G.; Straub, B.; Ciani, G.; Sironi, A. J. Chem. Soc., Dalton Trans. 1983, 2293–2298. (7) Nishiyama, H.; Brunner, H.; Jones, P. G. J. Organomet. Chem. 1991, 405, 247–255. (8) Brunner, H.; Rahman, A. F. M. M.; Bernal, I. Inorg. Chim. Acta 1984, 83, L93–L96. (9) Brunner, H.; Grau, I.; Zabel, M. Organometallics 2004, 23, 3788– 3799. (10) Cahn, R. S.; Ingold, C.; Prelog, V. Angew. Chem. 1966, 78, 413– 447; Angew. Chem., Int. Ed. Engl. 1966, 5, 385-415. (11) Brunner, H. Angew. Chem. 1969, 81, 395–396; Angew. Chem., Int. Ed. Engl. 1969, 8, 382-383. (12) Brunner, H.; Schindler, H.-D. Chem. Ber. 1971, 104, 2467–2474. (13) Brunner, H.; Schindler, H.-D. J. Organomet. Chem. 1971, 24, C7– C10. (14) Brunner, H.; Schindler, H.-D. Z. Naturforsch 1971, 26 b, 1220– 1225. (15) Brunner, H.; Aclasis, J.; Langer, M.; Steger, W. Angew. Chem. 1974, 86, 864–865; Angew. Chem., Int. Ed. Engl. 1974, 13, 810-811.

are configurationally stable in the solid state. In solution, however, they are configurationally labile. They change the metal configuration in first-order reactions approaching the racemization or in the case of L-menthyl the epimerization equilibria. In nonpolar solvents the half-lives τ1/2 are in the range of minutes to hours at ambient temperatures.11,12,15–17 The concentration and solvent dependence of the racemization and epimerization have been investigated including Hammett correlations and the influence of added triphenylphosphine.12,14–17 These studies as well as exchange reactions with deuterated triphenylphosphine14 and substitution reactions with other phosphines, which occurred with partial retention of configuration,15,18,19 are in accord with Scheme 3. Reviews summarizing these mechanistic studies are available.1,20–22 The rate-determining step in the racemization of (R)[CpMn(NO)(PPh3)C(O)R] in Scheme 3 is the cleavage of the manganese-triphenylphosphine bond, giving free triphenylphosphine and a pyramidal intermediate (R)-[CpMn(NO)C(O)R] with an empty coordination site. Activation energies are on the order of 105-125 kJ/mol.12,16,17 Activation entropies are strongly positive, in agreement with an increase in the number of species. The unsaturated intermediate (R)-[CpMn(NO)C(O)R] is high in energy and cannot be detected by spectroscopy. It can react with the dissociated or added triphenylphosphine, with deuterated triphenylphosphine, or with other added ligands. All these reactions occur with retention of the metal configuration. However, the pyramidal intermediate (R)-[CpMn(NO)C(O)R] can also invert to its mirror image (S)-[CpMn(NO)C(O)R], from which the opposite configurational side is accessible. The barriers to the right (k3) and to the left (k2) from the pyramidal intermediate (R)-[CpMn(NO)C(O)R] are assumed to have about (16) Brunner, H.; Langer, M. J. Organomet. Chem. 1975, 87, 223–240. (17) Brunner, H.; Aclasis, J. A. J. Organomet. Chem. 1976, 104, 347– 362. (18) Brunner, H.; Steger, W. J. Organomet. Chem. 1976, 120, 239– 256. (19) Brunner, H.; Steger, W. Bull. Soc. Chim. Belg. 1976, 85, 883–891. (20) Brunner, H. Angew. Chem. 1971, 83, 274–285; Angew. Chem., Int. Ed. Engl. 1971, 10, 249-260. (21) Brunner, H. J. Organomet. Chem. 1975, 94, 189–194. (22) Brunner, H. Top. Curr. Chem. 1975, 56, 67–90.

3516 Organometallics, Vol. 27, No. 14, 2008

Brunner et al.

Scheme 3. Energy Diagram for the Racemization of the Compounds [CpMn(NO)(PPh3)C(O)R]

the same height. Due to the air-sensitivity and the instability (formation of paramagnetic impurities), detailed investigations of the competition ratio k3/k2 have not been possible in the CpMn system. Omitting the crossed-out part, the energy profile of Scheme 3 resembles the cross-section of a hall church in which the central nave and the side naves have the same height, predominantly built in the late gothic period in Europe. A crucial point is the middle part of Scheme 3. Is the planar species [CpMn(NO)C(O)R] an intermediate or a transition state in the interconversion of the pyramids (R)- and (S)-[CpMn(NO)C(O)R]? Theoretical studies showed that 16-electron fragments [(η-CnHn)M(CO)2] are pyramidal, the planar species being transition states, whereas 16-electron fragments [(ηCnHn)M(NH3)2] adopt planar configurations.23,24 In accord with these older calculations newer calculations predict a planar configuration for the 16-electron fragment [CpRu(NH3)2]+.25 However, ref 24 assigns a planar configuration to the 16-electron fragment [CpFe(PH3)2]+, whereas according to ref 25 the homologous [CpRu(PH3)2]+ should have a pyramidal configuration. In Cp*Ru chemistry many 16-electron complexes have been prepared. In contrast, the number of characterized 16electron CpRu complexes, such as [CpRu(TMEDA)]X,26 is rather limited. Complexes [Cp*Ru(P-P)]X have been isolated; however, complexes [CpRu(P-P)]X are unknown. Planarity was assigned to species such as [CpRu(P-P)]+ on the basis of spectroscopy.27,28 However, the problem has not been investigated with the help of chirality, which allows additional and unequivocal decisions, not accessible with other methods. Using stereochemistry we will show in the present paper that 16electron species [CpRu(P-P′)]+ are pyramidal, the planar fragments being high-energy transition states as indicated in Scheme 4. (23) Hofmann, P. Angew. Chem. 1977, 89, 551–553; Angew. Chem., Int. Ed. Engl. 1977, 16, 536-537. (24) Ward, T. R.; Schafer, O.; Daul, D.; Hofmann, P. Organometallics 1997, 16, 3207–3215. (25) Aneetha, H.; Jimeńez-Tenorio, M.; Puerta, M. C.; Valerga, P.; Sapunov, V. N.; Schmid, R.; Kirchner, K.; Mereiter, K. Organometallics 2002, 21, 5334–5346. (26) Gemel, C.; Huffman, J. C.; Caulton, K. G.; Mauthner, K.; Mereiter, K.; Kirchner, K. J. Organomet. Chem. 2000, 593-594, 342–353. (27) Joslin, F. L.; Johnson, M. P.; Mague, J. T.; Roundhill, D. M. Organometallics 1991, 10, 2781–2794. (28) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A. J. Appl. Crystallogr. 1993, 26, 343–350.

The main difference between Scheme 3 and Scheme 4 is the competition ratio k3/k2 of the pyramidal intermediates for the pyramidal inversion involving the change of the metal configuration versus the back reaction with Hal- (or other nucleophiles). For the Mn compounds in Scheme 3 the competition ratio k3/k2 is about 1. That means that substitution reactions with added nucleophiles inevitably will be accompanied by extensive racemization. For the Ru compounds in Scheme 4, however, the competition ratio k3/k2 is much smaller than 1, as k3 is much smaller than k2. As a consequence the back reaction of the unsaturated pyramidal intermediates with Hal- (or other nucleophiles) is fast compared to the change of the metal configuration. Of course k2 depends on the concentration of the nucleophile used. For high nucleophile concentrations pseudofirst-order conditions prevail. Actually, we did not investigate the [CpRu(P-P′)Hal] system with enantiomers (RRu)/(SRu) but with diastereomers (RRu,RC)/ (SRu,RC) (ligand P-P′ ) (R)-Prophos and (R,R)-Norphos, Hal ) Cl, Br, I), in which the diastereomers (RRu,RC) and (SRu,RC) give rise to different signals in the NMR. Using diastereomers the energy profile becomes unsymmetrical, right and left halves being different (Scheme 4). For enantiomers the energy profile of the CpRu system in Scheme 4 would resemble the crosssection of a basilica in which the central part is much higher than the side parts, found in buildings and churches since the late Roman times.

Materials and Measurements 1 H NMR and 31P NMR spectra: Bruker Avance-400. Mass spectra: Finnigan MAT 710A. Elemental analyses: Heraeus Elementar Vario EL III. X-ray structure analyses: Oxford Diffraction Gemini (Cu KR radiation, graphite monochromator), 123 K, SIR9728 and SHELXS-97.29 The diastereomer mixtures (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl], (RRu,RC)- and (SRu,RC)-[CpRu(Norphos)Cl], and (RRu,RC)and (SRu,RC)-[CpRu(Norphos)I] were prepared and separated as described in refs 6 and 7. (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Br]. (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] (80 mg, 0.12 mmol) were dissolved in 10 mL of methanol. NaBr (750 mg, 6.3 mmol) was added, and

(29) Sheldrick, G. M. SHELXL-97, Program for Crystal Structure Refinement; University of Go¨ttingen: Germany, 1997.



Scheme 4. Energy Diagram for the Epimerization of the Compounds [CpRu(P-P′)Hal]

the reaction mixture was heated to 50 °C for 20 h. Yield: quantitative, orange solid. (RRu,RC)-[CpRu(Prophos)Br]. 1H NMR (CDCl3): δ 7.98-7.92 (m, 2H, ArH), 7.60-7.53 (m, 2H, ArH), 7.48-7.20 (m, 16H, ArH), 4.46 (s, 5H, CpH), 3.19-3.05 (m, 1H, CH), 2.78-2.60 (m, 1H, CH), 1.96-1.86 (m, 1H, CH), 1.12 (dd, 3JH-H ) 6.7 Hz, 3JP-H ) 10.2 Hz, 3H, Me). 31P{1H} NMR (CDCl3): δ 86.32 (d, 2JP-P ) 35.8 Hz, 1P), 61.84 (d, 2JP-P ) 35.8 Hz, 1P). EI MS (70 eV): m/z 660.1 [M+]. C32H31BrP2Ru (658.0). (SRu,RC)-[CpRu(Prophos)Br]. 1H NMR (CDCl3): δ 7.89-7.95 (m, 2H, ArH), 7.47-7.21 (m, 18H, ArH), 4.37 (s, 5H, CpH), 3.02-2.59 (m, 3H, CH-CH2), 1.01 (dd, 3JH-H ) 7.1 Hz, 3JP-H ) 12.8 Hz, 3H, Me). 31P{1H} NMR (CDCl3): δ 81.98 (d, 2JP-P ) 30.5 Hz, 1P), 73.72 (d, 2JP-P ) 30.5 Hz, 1P). (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)I]. Synthesis is in analogy to the corresponding bromo complex. Yield: quantitative, orange solid. (RRu,RC)-[CpRu(Prophos)I]. 1H NMR (CDCl3): δ 8.02-7.97 (m, 2H, ArH), 7.61-7.57 (m, 2H, ArH), 7.49-7.24 (m, 16H, ArH), 4.61 (s, 5H, CpH), 3.46-3.34 (m, 1H, CH), 2.94-2.69 (m, 1H, CH), 2.03-1.93 (m, 1H, CH), 1.27 (dd, 3JH-H ) 6.8 Hz, 3JP-H ) 10.9 Hz, 3H, Me). 31P{1H} NMR (CDCl3): δ 86.03 (d, 2JP-P ) 33.6 Hz, 1P), 62.17 (d, 2JP-P ) 33.6 Hz, 1P). EI MS (70 eV): m/z 705.9 [M+]. Anal. Calcd for C32H31IP2Ru (705.0): C, 54.18; H, 4.85. Found: C, 54.51; H, 4.40. (SRu,RC)-[CpRu(Prophos)I]. 1H NMR (CDCl3): δ 8.07-8.02 (m, 2H, ArH), 7.94-787 (m, 2H, ArH), 7.52-7.22 (m, 14H, ArH), 7.15-7.09 (m, 2H, ArH), 4.56 (s, 5H, CpH), 3.13-2.90 (m, 1H, CH), 2.94-2.69 (m, 1H, CH), 2.03-1.93 (m, 1H, CH), 1.23 (dd, 3 JH-H ) 7.0 Hz, 3JP-H ) 13.6 Hz, 3H, Me). 31P{1H} NMR (CDCl3): δ 84.41 (d, 2JP-P ) 28.4 Hz, 1P), 72.04 (d, 2JP-P ) 28.4 Hz, 1P).

(RRu,RC)- and (SRu,RC)-[CpRu(Norphos)Br]. Synthesis is in analogy to the corresponding iodo complex in ref 7. Yield: quantitative, orange solid. EI MS (70 eV): m/z 710.1 [M+, 100%]. Anal. Calcd for C36H33BrP2Ru (708.6): C, 61.04; H, 4.66. Found: C, 60.44; H, 4.92. For the Hal exchange and epimerization reactions highly enriched samples of (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Hal] and (RRu, RC)- and (SRu,RC)-[CpRu(Norphos)Hal] were used. Dissolution of the samples (for the Hal exchange reactions in the presence of the halide salts) sometimes was reluctant, taking 10-15 min to make the first measurement in the NMR spectrometer adjusted to the respective temperature. During this time interval halide exchange and epimerization had already started for some samples. Therefore, the diastereomer ratios of the first measurement under controlled conditions were used as “starting ratios” in tables, figures, and calculations. In all the kinetic measurements the integrals of the Cp or Me signals in the 1H NMR spectra were used, which proved to be more sensitive than the 31P signals in the 31P NMR spectra. The problem of line broadening and signal shifting will be discussed in the section “Cl/Br Exchange in [CpRu(Prophos)Cl] in 1:1 CDCl3/CD3OD”.

Configurations The Prophos ligand was used in its (R)-configuration.30 The configuration of the less soluble diastereomer of [CpRu(Prophos)Cl] having the low-field Cp signal in the NMR had been determined as (SRu,RC).6 It was the diastereomer that dominated (30) Fryzuk, M. D.; Bosnich, B. J. Am. Chem. Soc. 1977, 99, 6262– 6267.


Brunner et al.

Table 1. Crystallographic Data for [CpRu(P-P′)Hal] Complexes 1b, 1c, 2b′, 2c′, and 2c

empirical formula fw cryst syst space group a (Å) b (Å) c (Å) R (deg) β (deg) γ (deg) V (Å3) Z Fcalcd (Mg/m3) abs coeff (mm-1) abs correct. transmn max./min. F(000) cryst size (mm) θ range (deg) rflns/unique Rint no. of data/params goodness of fit F2 R1/wR2 (I > 2σ(I)) R1/wR2 (all data) abs struct param largest diff peak and hole (e Å-3) CCDC no.

(RRu,RC)[CpRu(Prophos)Br]

(RRu,RC)[CpRu(Prophos)I]

(RRu,RC)[CpRu(Norphos)Br]

(RRu,RC)[CpRu(Norphos)I]

(SRu,RC)[CpRu(Norphos)I]

C32H31BrP2Ru 658.48 monoclinic P21 9.65602(10) 14.95694(10) 10.48788(10) 90 112.4786(12) 90 1399.62(2) 2 1.562 7.424 semiempirical 1.17771/0.73228 664 0.128 × 0.106 × 0.088 4.56-51.38 17986/2986 0.0328 2986/326 1.081 0.0168/0.0425 0.0173/0.0426 -0.029(6) 0.269/-0.295

C32H31IP2Ru 705.48 monoclinic P21 8.4821(15) 15.0654(8) 11.0948(8) 90 91.851(9) 90 1417.0(3) 2 1.653 1.775 semiempirical 1.07185/0.93327 700 0.260 × 0.249 × 0.229 2.98-29.00 22392/6459 0.0283 6459/326 0.970 0.0244/0.0476 0.0303/0.0482 -0.037(11) 1.159/-0.418

C36H33BrP2Ru 708.53 orthorhombic P212121 12.45942(10) 15.52669(14) 15.69572(14) 90 90 90 3036.39(5) 4 1.550 6.892 semiempirical 1.14583/0.63065 1432 0.229 × 0.213 × 0.147 4.00-63.11 43053/4873 0.0289 4873/361 1.046 0.0203/0.0526 0.0208/0.0528 -0.028(6) 0.606/-0.320

C36H33IP2Ru,CH2Cl2 840.46 monoclinic I2 16.3029(3) 11.3714(3) 20.1949(4) 90 112.352(2) 90 3462.57(14) 4 1.612 1.417 semiempirical 1.07793/0.91776 1672 0.170 × 0.130 × 0.040 3.08-29.00 28060/8136 0.0452 8136/403 0.971 0.0316/0.0655 0.0427/0.0677 0.014(15) 0.856/-0.665

C36H33IP2Ru 755.53 orthorhombic P212121 12.5156(3) 15.5113(4) 15.8381(5) 90 90 90 3074.70(15) 4 1.632 1.642 semiempirical 1.03461/0.94835 1504 0.180 × 0.060 × 0.030 3.26-29.16 16329/6624 0.0487 6624/361 0.934 0.0385/0.0498 0.0692/0.0543 -0.006(16) 0.890/-0.715

659547

659548

659549

659550

659551

the equilibrium.6 The same relative configuration was assigned to the less soluble diastereomers (RRu,RC)-[CpRu(Prophos)Br] and (RRu,RC)-[CpRu(Prophos)I], which were the major diastereomers in the (RRu,RC)/(SRu,RC) equilibrium having the lowfield Cp signals. The Norphos ligand was used in its (R,R)-configuration.31,32 The configuration of the major diastereomer of [CpRu(Norphos)I] in the equilibrium had been determined as (RRu,RC).7 It had the low-field olefin NMR signals at 5.57 and 6.42 ppm. The same relative configuration was assigned to the major diastereomers (SRu,RC)-[CpRu(Norphos)Cl] and (RRu,RC)-[CpRu-

(Norphos)Br] in the (RRu,RC)/(SRu,RC) equilibria having the same NMR properties in the olefin region. For the X-ray structure determinations (RRu,RC)-[CpRu(Prophos)Br] and (RRu,RC)-[CpRu(Prophos)I] were crystallized from methanol. A sample of (RRu,RC)- and (SRu,RC)-[CpRu(Norphos)I] of composition 33:67 was crystallized from a 1:3 mixture of CH2Cl2/CH3OH to give a crystalline fraction of (RRu,RC)-/ (SRu,RC)-[CpRu(Norphos)I] (27:73) that contained crystals of both diastereomers. The molecular structures of different diastereomers of [CpRu(Prophos)Br], [CpRu(Prophos)I], [CpRu(Norphos)Br], and

Figure 1. Molecular structures of (RRu,RC)-[CpRu(Prophos)Br], 1b, and (RRu,RC)-[CpRu(Prophos)I], 1c. Table 2. Bond Lengths [Å] and Angles [deg] in [CpRu(P-P′)Hal] Complexes 1b, 1c, 2b′, 2c′, and 2c

Centr-Ru Ru-P Ru-P′ Ru-Hal Centr-Ru-P Centr-Ru-P′ Centr-Ru-Hal P-Ru-P′ P-Ru-Hal P′-Ru-Hal

(RRu,RC)[CpRu(Prophos)Br]

(RRu,RC)[CpRu(Prophos)I]

(RRu,RC)[CpRu(Norphos)Br]

(RRu,RC)[CpRu(Norphos)I]

(SRu,RC)[CpRu(Norphos)I]

1.8387(3) 2.2755(11) 2.2762(9) 2.5654(4) 129.119(26) 130.981(27) 121.981(12) 83.080(35) 84.997(25) 93.007(25)

1.8518(2) 2.2768(8) 2.2859(8) 2.7320(4) 129.162(23) 131.010(21) 119.147(10) 83.448(28) 87.575(21) 93.918(21)

1.8405(2) 2.2989(8) 2.2985(8) 2.5767(3) 130.647(21) 126.616(21) 123.957(10) 86.105(28) 84.629(21) 91.579(21)

1.8387(3) 2.2957(13) 2.2974(14) 2.7312(4) 130.957(36) 126.654(36) 122.245(16) 86.139(47) 85.690(33) 92.475(34)

1.8427(4) 2.3022(12) 2.2963(10) 2.7185(4) 126.867(33) 130.023(31) 120.930(17) 85.107(37) 94.719(29) 87.516(30)



Figure 2. Molecular structures of (RRu,RC)-[CpRu(Norphos)I], 2c′, and (SRu,RC)-[CpRu(Norphos)I], 2c.

[CpRu(Norphos)I] have been determined (Table 1). The major diastereomers 1b of [CpRu(Prophos)Br] and 1c of [CpRu(Prophos)I] (Figure 1) have (RRu,RC)-configuration and closely related structures. In both of them the methyl groups of the Prophos ligand are oriented away from the Ru-Hal bond. The bond lengths Ru-PCHMe and Ru-PCH2 and the angles Centr-Ru-PCHMe and Centr-Ru-PCH2 as well as PCHMeRu-Hal and PCH2-Ru-Hal obey the same trends (Table 2). The same tendencies in bond lengths and angles are observed in the major (RRu,RC)-diastereomers 2b′ of [CpRu(Norphos)Br] and 2c′ of [CpRu(Norphos)I] (Figure 2). 2c′ contains one molecule CH2Cl2 per formula unit. The minor (SRu,RC)diastereomer 2c of [CpRu(Norphos)I], on the other hand, is distinctly different (Figure 2 and Table 2).

Halide Exchange in [CpRu(Prophos)Cl] and [CpRu(Norphos)Cl] Cl/I Exchange in [CpRu(Prophos)Cl] in 9:1 CDCl3/ CH3OH. The kinetics of the Cl/I exchange in (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] was measured with an excess of [Bu4N]I in the solvent mixture CDCl3/CH3OH (9:1, v/v) at 300 K. Figure 3 shows the reaction of a sample of 85:15 (RRu,RC)-/ (SRu,RC)-[CpRu(Prophos)Cl] with a 14-fold excess of [Bu4N]I. The concentration of (RRu,RC)-[CpRu(Prophos)Cl] decreased with time, whereas the concentrations of (SRu,RC)-[CpRu(Prophos)I] and (RRu,RC)-[CpRu(Prophos)I] increased. Surprisingly, the concentration of (SRu,RC)-[CpRu(Prophos)Cl] did not change appreciably, which means that the (SRu,RC)-diastereomer reacted much more slowly with [Bu4N]I than the (RRu,RC)-diastereomer. The first-order rate constant for the disappearance of (RRu,RC)[CpRu(Prophos)Cl] at 300 K was k ) 7.2 × 10-3 min-1. As the concentration of (SRu,RC)-[CpRu(Prophos)Cl] stayed almost constant, the two substitution products (SRu,RC)-[CpRu(Prophos)I] and (RRu,RC)-[CpRu(Prophos)I] originated from the disappearing (RRu,RC)-[CpRu(Prophos)Cl]. It mainly transformed to (SRu,RC)-[CpRu(Prophos)I]. (RRu,RC)-[CpRu(Prophos)Cl] and (SRu,RC)-[CpRu(Prophos)I] have the same relative configuration, because the change of the configurational symbol is only due to a change in the ligand priority sequence. Thus, the Cl/I exchange in (RRu,RC)-[CpRu(Prophos)Cl] occurred predominantly with retention of configuration at the metal atom. Definitely, however, a small amount of (RRu,RC)-[CpRu(Prophos)Cl] was converted into (RRu,RC)-[CpRu(Prophos)I] with inversion of configuration. These experimental facts can be explained by assuming an energy profile as in Scheme 4. The Cl/I exchange in (RRu,RC)(31) Brunner, H.; Pieronczyk, W. Angew. Chem. 1979, 91, 655–656; Angew. Chem., Int. Ed. Engl. 1979, 18, 620-621. (32) Brunner, H.; Pieronczyk, W.; Scho¨nhammer, B.; Streng, K.; Bernal, I.; Korp, J. Chem. Ber. 1981, 114, 1137–1149.

Figure 3. Cl/I exchange reaction in 85:15 (RRu,RC)-/(SRu,RC)[CpRu(Prophos)Cl] (2/9; 18.4 mmol L-1) with [Bu4N]I (0.25 mol L-1) in CDCl3/CH3OH (9:1, v/v) at 300 K. The products are (SRu,RC)-[CpRu(Prophos)I] (0) and (RRu,RC)-[CpRu(Prophos)I] (O).

[CpRu(Prophos)Cl] starts on the left side with the cleavage of the Ru-Cl bond. This is the rate-determining step, which requires a high activation energy. An unsaturated intermediate (RRu,RC)-[CpRu(Prophos)]+ with pyramidal geometry is formed that has kept its metal configuration. This intermediate can react in a fast reaction with iodide to give the substitution product (SRu,RC)-[CpRu(Prophos)I] with retention of the metal configuration. It can also react with CH3OH, present in the solvent mixture, to give the species (RRu,RC)-[CpRu(Prophos)CH3OH]+ as a temporary parking lot. However, the intermediate (RRu,RC)[CpRu(Prophos)]+, which cannot be observed experimentally, can also invert its configuration, crossing the middle of Scheme 4, to form its diastereomer (SRu,RC)-[CpRu(Prophos)]+, which subsequently is quenched to (RRu,RC)-[CpRu(Prophos)I]. As the iodo complexes (SRu,RC)- and (RRu,RC)-[CpRu(Prophos)I] are more stable than the chloro starting material (RRu,RC)-[CpRu(Prophos)Cl], particularly in the presence of an excess of iodide, the chloro complex will be quantitatively converted into the iodo products. The crucial point in the Cl/I exchange in (RRu,RC)-[CpRu(Prophos)Cl] is the choice of the intermediate (RRu,RC)[CpRu(Prophos)]+ to invert its configuration (k3 path) or to react to the iodo complex (SRu,RC)-[CpRu(Prophos)I] with retention of configuration (k2 path). This choice is best treated with the competition ratio k3/k2. The competition ratio k3/k2 can be


Brunner et al.

Table 3. Kinetics of the Disappearance of (RRu,RC)-[CpRu(Prophos)Cl] and (SRu,RC)-[CpRu(Prophos)Cl] in the Cl/Hal Exchange Reactions with [Bu4N]I (upper part) and [Bu4N]Br (lower part) in CDCl3/CH3OH (9:1, v/v) and Activation Parameters reaction

added salt

temp (K)

k1 or k1′ (min-1) -3

τ1/2 (min)

(RRu,RC)-[CpRu(Prophos)Cl] f (SRu,RC)-[CpRu(Prophos)I]

[Bu4N]I

activation enthalpy ∆Hq(300 K) ) 69 kJ mol-1 activation entropy ∆Sq(300 K) ) -90 J mol-1 K-1 Gibbs free energy ∆Gq(300 K) ) 96 kJ mol-1 (SRu,RC)-[CpRu(Prophos)Cl] f (RRu,RC)-[CpRu(Prophos)I]

300 308 313 323

7.2 × 10 1.5 × 10-2 2.7 × 10-2 5.4 × 10-2

96 46 26 13

[Bu4N]I

activation enthalpy ∆Hq(300 K) ) 88 kJ mol-1 activation entropy ∆Sq(300 K) ) -47 J mol-1 K-1 Gibbs free energy ∆Gq(300 K) ) 102 kJ mol-1 (RRu,RC)-[CpRu(Prophos)Cl] f (SRu,RC)-[CpRu(Prophos)Br]

300 308 313 323

6.2 × 10-4 1.3 × 10-3 3.1 × 10-3 7.7 × 10-3

1120 530 220 90

[Bu4N]Br

activation enthalpy ∆Hq(300 K) ) 78 kJ mol-1 activation entropy ∆Sq(300 K) ) -59 J mol-1 K-1 Gibbs free energy ∆Gq(300 K) ) 96 kJ mol-1 (SRu,RC)-[CpRu(Prophos)Cl] f (RRu,RC)-[CpRu(Prophos)Br]

300 308 313 323

6.2 × 10-3 1.8 × 10-2 3.9 × 10-2 6.0 × 10-2

110 38 18 12

[Bu4N]Br

300 308 313 323

7.8 × 10-4 1.7 × 10-3 3.1 × 10-3 9.0 × 10-3

890 400 220 77

activation enthalpy ∆Hq(300 K) ) 84 kJ mol-1 activation entropy ∆Sq(300 K) ) -60 J mol-1 K-1 Gibbs free energy ∆Gq(300 K) ) 102 kJ mol-1

determined from the product ratio (RRu,RC)-[CpRu(Prophos)I]/ (SRu,RC)-[CpRu(Prophos)I], which is constant throughout the reaction. We neglected the first three measurements of Figure 3, because the concentration of the new products was still too low, and calculated the average of all the other measurement points. The competition ratio k3/k2 ) 0.08 shows that the inversion of the intermediate (k3 path) is slow compared to the reaction of the intermediate with excess iodide (k2 path), indicating a basilica-type energy profile as in Scheme 4. The rate constants of the Cl/I exchange in (RRu,RC)-[CpRu(Prophos)Cl] with [Bu4N]I increased with rising temperature (Table 3, upper part). A large excess of [Bu4N]I was used in all these reactions to guarantee pseudo-first-order conditions. In this temperature variation the competition ratio k3/k2 showed considerable scattering due to the inaccuracy of the determination of the concentration of (RRu,RC)-[CpRu(Prophos)I], which is formed only in small amounts. In these measurements too a reluctance of (SRu,RC)-[CpRu(Prophos)Cl] to undergo Cl/I exchange was noticed. To measure the Cl/I exchange in the (SRu,RC)-diastereomer, a sample of 97:3 (SRu,RC)-/(RRu,RC)-[CpRu(Prophos)Cl] was treated with an excess of [Bu4N]I in CDCl3/CH3OH (9:1, v/v) at 300-323 K. Table 3 shows that the rate constants for the disappearance of (SRu,RC)-[CpRu(Prophos)Cl] in the Cl/I exchange are about 10 times slower than for (RRu,RC)-[CpRu(Prophos)Cl]. The reason is that (SRu,RC)-[CpRu(Prophos)Cl], the right side species in Scheme 4, is the thermodynamically more stable diastereomer. It dominates the equilibrium (SRu,RC)-[CpRu(Prophos)Cl]/ (RRu,RC)-[CpRu(Prophos)Cl] 85:15 (see below), which means that it is more stable (and less reactive) than (RRu,RC)[CpRu(Prophos)Cl]. This reflects a higher activation energy for k1′ and k3′ compared to k1 and k3 (see Scheme 4). We did not determine the competition ratios k3′/k2′ for the Cl/I exchange in (SRu,RC)-[CpRu(Prophos)Cl], because samples even highly enriched in (SRu,RC)-[CpRu(Prophos)Cl] contained some (RRu,RC)-[CpRu(Prophos)Cl], which in a fast reaction gave (SRu,RC)- and (RRu,RC)-[CpRu(Prophos)I] (see above), before

k3/k2 0.08 0.06 0.04 0.11

0.04 0.05 0.05 0.04

(SRu,RC)-[CpRu(Prophos)Cl] was slowly converted to (RRu,RC)and (SRu,RC)-[CpRu(Prophos)I], from which the ratios k3′/k2′ would have to be calculated. The activation parameters for the Cl/I exchange reactions in (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] with [Bu4N]I are given in Table 3 (upper part). The reactions start with the cleavage of the Ru-Cl bond. Negative values were found for the entropy of activation, whereas for reactions of the Mn compounds, described above, in which the initial step is the cleavage of the Mn-PPh3 bond, activation entropies had been positive. Both reaction types are dissociations in which the number of particles increases. However, in the Mn system a neutral complex breaks up into two neutral fragments, leading to a positive entropy of activation. Differently, in the Ru system a neutral complex forms two ions, which are strongly solvated by the polar solvent methanol. This increases the order reflected by a negative entropy of activation. Cl/Br Exchange in [CpRu(Prophos)Cl] in 9:1 CDCl3/ CH3OH. The kinetics of the Cl/Br exchange in (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] was measured with an excess of [Bu4N]Br under similar conditions to the Cl/I exchange in the solvent mixture CDCl3/CH3OH (9:1, v/v). As in the Cl/I exchange the concentration of (RRu,RC)-[CpRu(Prophos)Cl] decreased with time, whereas the concentrations of (SRu,RC)[CpRu(Prophos)Br] and (RRu,RC)-[CpRu(Prophos)Br] increased. Similar to the Cl/I system, the concentration of (SRu,RC)[CpRu(Prophos)Cl] did not change appreciably during the Cl/ Br exchange reaction. The first-order rate constant for the disappearance of (RRu,RC)-[CpRu(Prophos)Cl] in the Cl/Br exchange at 300 K was k ) 6.2 × 10-3 min-1 compared to k ) 7.2 × 10-3 min-1 in the Cl/I exchange. Thus, within the limits of error, the rate constants for the Cl/Br and the Cl/I exchange are the same. This also holds for the temperature dependence of the rate constants (Table 3, lower part). The reason for this is obvious from Scheme 4. For both processes, the Cl/Br and the Cl/I exchange, the rate-determining step is the cleavage of the Ru-Cl bond in (RRu,RC)-[CpRu(Prophos)Cl],



Table 4. Kinetics of the Disappearance of (RRu,RC)-[CpRu(Prophos)Cl] and (SRu,RC)-[CpRu(Prophos)Cl] in the Cl/Hal Exchange Reaction with [PyCH2Ph]I (upper part) and [PyCH2Ph]Br (lower part) in CDCl3/CD3OD (1:1, v/v) and Activation Parameters. For the Cl/Br substitution of (SRu,RC)-[CpRu(Prophos)Cl] many 1H NMR spectra were measured in the presence of a small amount of Cp2Co (see text) temp (K)

k1 or k1′ (min-1)

τ1/2 (min)

k3/k2


[PyCH2Ph]I


283 288 293 300

8.1 × 10-3 1.5 × 10-2 3.1 × 10-2 6.3 × 10-2

86 45 23 11

0.08 0.15 0.12 0.09

[PyCH2Ph]I

activation enthalpy ∆Hq(300 K) ) 90 kJ mol-1 activation entropy ∆Sq(300 K) ) -22 J mol-1 K-1 Gibbs free energy ∆Gq(300 K) ) 96 kJ mol-1 (RRu,RC)-[CpRu(Prophos)Cl] f (SRu,RC)-[CpRu(Prophos)Br]

300 308 313 323

8.8 × 10-3 1.5 × 10-2 3.3 × 10-2 1.1 × 10-1

82 46 21 6

[PyCH2Ph]Br

283a 288a 293a 293b 300a

1.1 × 10-2 2.0 × 10-2 3.0 × 10-2 3.0 × 10-2 5.9 × 10-2

61 35 23 23 12

300b 305b 318b 323b

6.6 × 10-3 1.4 × 10-2 4.9 × 10-2 9.0 × 10-2

110 48 17 8

reaction

activation enthalpy ∆Hq(300 K) ) 86 kJ mol-1 activation entropy ∆Sq(300 K) ) -12 J mol-1 K-1 Gibbs free energy ∆Gq(300 K) ) 90 kJ mol-1 (SRu,RC)-[CpRu(Prophos)Cl] f (RRu,RC)-[CpRu(Prophos)Br]

added salt

[PyCH2Ph]Br

activation enthalpy ∆Hq(300 K) ) 85 kJ mol-1 activation entropy ∆Sq(300 K) ) -36 J mol-1 K-1 Gibbs free energy ∆Gq(300 K) ) 96 kJ mol-1 a

0.09 n.d. 0.11 0.13

In the absence of Cp2Co. b In the presence of Cp2Co.

in which the unsaturated intermediate (RRu,RC)-[CpRu(Prophos)]+ is formed. Quenching of the intermediate with excess bromide or excess iodide is a fast reaction, which does not affect the rate-determining step. The competition ratios k3/k2 in the Cl/Br exchange reactions were similar to those of the Cl/I systems (Table 3, lower part). Cl/I Exchange in [CpRu(Prophos)Cl] in 1:1 CDCl3/ CD3OD. The kinetics of the Cl/I exchange in (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] was measured with an excess of benzylpyridinium iodide ([PyCH2Ph]I) in the solvent mixture CDCl3/CD3OD (1:1, v/v) at different temperatures. With respect to the I- salt we changed to [PyCH2Ph]I because of overlap of the methyl signals of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] and [Bu4N]I in the 1H NMR spectrum, and with respect to the methanol admixture we changed from CH3OH to CD3OD because of problems with the internal lock signals. Table 4 shows that the rates in 1:1 CDCl3/CD3OD are much faster than the rates in 9:1 CDCl3/CH3OH. Suitable comparisons are possible for the measurements at 300 K in Tables 3 and 4. Going from 9:1 to 1:1 chloroform/methanol, the rate of the Cl/I substitution increased by a factor of about 10 for both (RRu,RC)[CpRu(Prophos)Cl] and (SRu,RC)-[CpRu(Prophos)Cl]. Similar to 9:1 CDCl3/CH3OH the Cl/I exchange in 1:1 CDCl3/CD3OD was 7 times faster for (RRu,RC)-[CpRu(Prophos)Cl] than for its more stable diastereomer (SRu,RC)-[CpRu(Prophos)Cl]. Cl/Br Exchange in [CpRu(Prophos)Cl] in 1:1 CDCl3/ CD3OD. Contrary to the Cl/I exchange in 1:1 CDCl3/CD3OD, the Cl/Br exchange in the system (SRu,RC)- and (RRu,RC)[CpRu(Prophos)Cl]/[PyCH2Ph]Br in 1:1 CDCl3/CD3OD suffered from extensive line broadening of the NMR spectra. The reason for this was that traces of air-oxygen introduced during sample preparation oxidized a small amount of [CpRu(Prophos)Cl] to the paramagnetic radical cation [CpRu(Prophos)Cl]+•, leading to the observed broadening and shifting of the NMR signals.

Concomitantly with the line broadening, the quartets of the methyl signal of the Prophos ligand in the 1H NMR spectra became doublets, because the coupling to phosphorus was lost. Addition of a small amount of the strongly reducing agent Cp2Co removed the radical cation [CpRu(Prophos)Cl]+•, allowing for good 1H NMR spectra including a regaining of the phosphorus coupling to the H atoms in the Prophos ligand.33–35 Measurements in the presence and absence of Cp2Co gave the same rate constants (Table 2), although in the presence of Cp2Co the signals of (SRu,RC)- and (RRu,RC)-[CpRu(Prophos)D]35 appeared in low intensity. As in CDCl3/CH3OH (9:1) the rate constants within the limits of error were identical for the Cl/Br and the Cl/I exchange in CDCl3/CD3OD (1:1) (Table 4). Thus, the rate-determining steps k1 and k1′ are the same for both systems. The competition ratios were a little larger in CDCl3/CD3OD (1:1) than in CDCl3/ CH3OH (9:1), indicating a slight adjustment of the central and peripheral heights of the basilica-type energy profile. The broadening of the NMR signals due to the formation of the paramagnetic radical cation [CpRu(Prophos)Cl]+•, observed in the Cl/Br exchange reactions with [PyCH2Ph]Br, had not been a problem in the Cl/I substitution with [PyCH2Ph]I. Interestingly, on addition of [PyCH2Ph]I the signals of the system [CpRu(Prophos)Cl]/[PyCH2Ph]Br sharpened. Therefore, it must be assumed that iodide reduces the radical cation [CpRu(Prophos)Cl]+•, whereas bromide does not. Nevertheless, addition of Cp2Co is the best choice to avoid line broadening. (33) Disley, S. P. M.; Grime, R. W.; McInnes, E. J. L.; Spencer, D. M.; Swainston, N.; Whiteley, M. W. J. Organomet. Chem. 1998, 566, 151– 158. (34) Brunner, H.; Klankermayer, J.; Zabel, M. Eur. J. Inorg. Chem. 2002, 2494–2501. (35) Morandini, F.; Consiglio, G.; Lucchini, V. Organometallics 1985, 4, 1202–1208.


Brunner et al.

Table 5. Kinetics of the Disappearance of (RRu,RC)-[CpRu(Prophos)Cl] and (SRu,RC)-[CpRu(Prophos)Cl] in the Cl/I Exchange Reactions with [Bu4N]I in CDCl3 reaction

added salt

temp (K)

k1 or k1′ (min-1) -4


[Bu4N]I


300 308 313 323

4.0 × 10 8.5 × 10-4 1.4 × 10-3 3.0 × 10-3

[Bu4N]I

300 308 313 323

9.7 × 10-5 1.6 × 10-4 3.3 × 10-4 6.6 × 10-4

activation enthalpy ∆Hq (300 K) ) 67 kJ mol-1 activation entropy ∆Sq(300 K) ) -133 J mol-1 K-1 Gibbs free energy ∆Gq(300 K) ) 107 kJ mol-1

Cl/I and Cl/Br Exchange in [CpRu(Prophos)Cl] in CD3OD. The kinetics of the Cl/I and the Cl/Br exchange in (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] was measured with an excess of [PyCH2Ph]I and [PyCH2Ph]Br in pure CD3OD at different temperatures. The results were similar to those in the solvent mixture 1:1 CDCl3/CD3OD. The rates in pure CD3OD were only a little higher than the rates given in Table 4 for 1:1 CDCl3/CD3OD (up to a factor of 2). Thus, the rates of the Hal substitution reactions changed on going from 9:1 CDCl3/CH3OH to 1:1 CDCl3/CD3OD by a factor of about 10. However, on going from 1:1 CDCl3/CD3OD to pure CD3OD they increased only marginally. Cl/I Exchange in [CpRu(Prophos)Cl] in CDCl3 and Toluene-d8. As expected the Cl/I exchange in (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] in pure chloroform is much slower than in methanol containing solvents. It is just possible to measure the temperature dependence by NMR spectroscopy without coming too close to the boiling point of CDCl3 (Table 5). Compared to 9:1 CDCl3/CH3OH the rate for (RRu,RC)[CpRu(Prophos)Cl] decreased by a factor of close to 20. For (SRu,RC)-[CpRu(Prophos)Cl] the rate decreased by a factor of about 10. Qualitative measurements showed that the Cl/I exchange in (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] occurred even in toluened8, saturated with [Bu4N]I. However, temperatures as high as 353 K were necessary, indicating the low ability of toluene to stabilize the ions (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)]+ and Clformed in the rate-determining step. As in the chloroform/ methanol mixtures the (RRu,RC)-diastereomer of [CpRu(Prophos)Cl] reacted much faster than the (SRu,RC)-diastereomer. Cl/I Exchange in [CpRu(Norphos)Cl] in CH3OH. Addition of an excess of NaI to a methanol solution of [CpRu(Norphos)Cl] at room temperature caused the NMR signals of [CpRu(Norphos)Cl] to become smaller, whereas the signals of CpRu(Norphos)I] appeared and grew until no more [CpRu(Norphos)Cl] was left. Half-quantitative kinetic measurements were carried out at 273 K. Samples were taken from the methanol solution and evaporated at low temperatures. Then, the NMR spectra were measured in CDCl3 solution. For analysis the wellseparated olefinic signals of the Norphos ligand were chosen, which form four-line patterns. Starting with a diastereomer mixture 42:58 (SRu,RC)-/(RRu,RC)-[CpRu(Norphos)Cl] Hal exchange was relatively slow at 273 K. After 2 h at 273 K conversion was a little more than 50% for (RRu,RC)-[CpRu(Norphos)Cl] and less than 50% for (SRu,RC)-[CpRu(Norphos)Cl]. After 4 h at 273 K conversion was complete to almost 80%. Treatment of the data obtained for the reaction (SRu,RC)[CpRu(Norphos)Cl] f (RRu,RC)-[CpRu(Norphos)I] according

τ1/2 (h) 29 14 8.3 3.9

120 72 35 18

to first-order gave a rate constant k1 ) 1.7 × 10-2 [min-1] and a half-life of τ1/2 ) 41 min at 273 K. Thus, the Cl/I exchange was much faster in the Norphos complexes than in the corresponding Prophos complexes. Room-temperature spectra of the Cl/I exchange reaction in [CpRu(Norphos)Cl] in methanol showed that the half-life of the reaction (SRu,RC)-[CpRu(Norphos)Cl] f (RRu,RC)-[CpRu(Norphos)I] was about 30 min. The half-life of the corresponding reaction of the diastereomer (RRu,RC)-[CpRu(Norphos)Cl] f (SRu,RC)-[CpRu(Norphos)I] was appreciably shorter. The Cl/I exchange reaction occurred with predominant retention of the metal configuration. In fact, neither at 273 K nor at room temperature there was a change in the Ru configuration during the halide substitution reactions, indicating a higher degree of retention in the reactions of the Norphos complexes with respect to the Prophos complexes.

Epimerization of [CpRu(Prophos)Cl], [CpRu(Prophos)I], and [CpRu(Norphos)Cl] Epimerization of [CpRu(Prophos)Cl] in 9:1 CDCl3/ CH3OH. (SRu,RC)- and (RRu,RC)-[CpRu(Prophos)Cl] have extensively been used as starting materials for reactions occurring with retention of configuration in methanol at room temperature and in boiling methanol.2 Thus, they were assumed to be configurationally stable in solution. However, our studies showed that (SRu,RC)- and (RRu,RC)-[CpRu(Prophos)Cl] did epimerize at the metal center in methanol containing solvents. In 9:1 CDCl3/CH3OH (v/v) a sample (SRu,RC)-/(RRu,RC)[CpRu(Prophos)Cl] (14.3:85.7) epimerized at 293 K to the equilibrium composition (SRu,RC)- and (RRu,RC)-[CpRu(Prophos)Cl] (86.0:14.0) with kep ) 3.3 × 10-4 (min-1), corresponding to a half-life τ1/2 ) 35 h for the approach to equilibrium (Table 6). Using the equilibrium constant K ) 6.1 the rate constants kf and kr for the forward reaction (SRu,RC) f (RRu,RC) and the backward reaction (RRu,RC) f (SRu,RC) could be calculated. Table 6 shows the temperature dependence of the rate constants of the epimerization reaction. At 323 K the half-life for approach to equilibrium was down to 1.4/1.5 h. NMR measurements at higher temperatures would come too close to the boiling point of CDCl3. As to be expected, samples enriched in the diastereomer (RRu,RC)-[CpRu(Prophos)Cl] gave the same equilibrium ratios, rate constants, and half-lives as samples enriched in (SRu,RC)-[CpRu(Prophos)Cl] (Table 6). In the investigation of the epimerization of [CpRu(Prophos)Cl] in methanol-containing solvents we did not add Cp2Co, which had been successful in sharpening the NMR spectra of the Hal exchange reactions (see above). The reason was a side



Table 6. Kinetics of the Epimerization of Enriched Samples of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] in CDCl3/CH3OH (9:1, v/v) and Activation Parametersa starting ratio (SRu,RC):(RRu,RC) act. param 14.3:85.7 5.8:94.2 96.3:3.7 50.0:50.0b 11.9:88.1 50.0:50.0b 27.2:62.8 50.0:50.0b 12.6:87.4 95.0:5.0 50.0:50.0b

temp (K) 293 300 300 300 308 308 313 313 323 323 323

equilibrium ratio (SRu,RC):(RRu,RC) 86.0:14.0 85.0:15.0 84.0:16.0 84.6:15.4 85.6:14.4 83.6:16.4 84.4:15.6 82.8:17.2 81.2:18.8 83.4:16.6 81.1:18.9

K 6.1 5.7 5.3 5.5 6.0 5.1 5.4 4.8 4.3 5.0 4.8

kep (min-1) -4

3.3 × 10 5.0 × 10-4 3.8 × 10-4 6.2 × 10-5 1.2 × 10-3 1.7 × 10-4 2.4 × 10-3 3.5 × 10-4 8.1 × 10-3 7.6 × 10-3 3.5 × 10-4

activation enthalpy ∆Hqf ) 86 kJ mol-1 activation entropy ∆S‡f(300 K) ) -58 J mol-1 K-1 Gibbs free energy ∆G‡f(300 K) ) 103 kJ mol-1 a

τ1/2 (h) 35 23 30 190 9.6 68 4.8 33 1.4 1.5 33

kf (min-1)

kr (min-1)

-4

5.4 × 10-5 8.8 × 10-5 7.2 × 10-5 1.2 × 10-5 2.0 × 10-4 3.0 × 10-5 3.2 × 10-4 6.0 × 10-5 1.9 × 10-3 1.5 × 10-3 6.0 × 10-5

2.8 × 10 4.8 × 10-4 3.1 × 10-4 5.2 × 10-5 1.0 × 10-3 1.4 × 10-4 2.1 × 10-3 2.9 × 10-4 6.2 × 10-3 6.1 × 10-3 2.9 × 10-4

∆H‡r(300 K) ) 93 kJ mol-1 ∆S‡r(300 K) ) -47 J mol-1 K-1 ∆G‡r(300 K) ) 107 kJ mol-1

Measurements were performed using the Cp signals of the 1H NMR spectra. b Addition of a 10-fold excess of [Bu4N]Cl. Table 7. Kinetics of the Epimerization of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] in CDCl3/CD3OD (1:1, v/v) and Activation Parameters

starting ratio (SRu,RC):(RRu,RC) act. param

temp (K)

equilibrium ratio (SRu,RC):(RRu,RC)

K

kep (min-1)

τ1/2 (min)

kf (min-1)

kr (min-1)

63.2:36.8 13.0:87.0 19.7:80.3 27.2:72.8 51.3:48.7

293a 300b 308b 313b 323b

85.3:14.7 87.0:13.0 83.9:16.1 84.2:15.8 86.8:13.2

5.8 6.8 5.2 6.3 6.6

5.6 × 10-3 1.4 × 10-2 4.5 × 10-2 6.8 × 10-2 1.7 × 10-1

120 50 15 10 4.1

4.6 × 10-3 1.2 × 10-2 3.6 × 10-2 5.7 × 10-2 1.4 × 10-1

9.7 × 10-4 2.1 × 10-3 8.6 × 10-3 1.1 × 10-2 2.6 × 10-2

activation enthalpy ∆H‡f(300 K) ) 87 kJ mol-1 activation entropy ∆S‡f(300 K) ) -26 J mol-1 K-1 Gibbs free energy ∆G‡f(300 K) ) 95 kJ mol-1

∆H‡r (300 K) ) 89 kJ mol-1 ∆S‡r (300 K) ) -34 J mol-1 K-1 ∆G‡r (300 K) ) 99 kJ mol-1

a Measurements were performed using the Cp signals of the 1H NMR spectra. Prophos ligand of the 1H NMR spectra.

reaction, which produced the byproduct (RRu,RC)-/(SRu,RC)[CpRu(Prophos)H] and (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)D], respectively. Cp2Co removes traces of air-oxygen according to the equation 4 Cp2Co + O2 f 2 [Cp2Co]2O. In methanol the oxide ion gives OMe- and OH-. Obviously, at higher OMeconcentrations the intermediates (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)]+, formed in the bond cleavage of (RRu,RC)-/(SRu,RC)[CpRu(Prophos)Cl], react with OMe- to give the complexes (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)OMe], which transform to the hydrides (RRu,RC)-/(SRu,RC)- [CpRu(Prophos)H]. The compounds (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)H] had been prepared by Consiglio et al. from (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] and NaOMe.35 In methanol-containing solvents higher OMe- concentrations arise from addition of either Cp2Co/O2 or NaOMe. In both cases in solutions of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] the NMR signals of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)H] appear at room temperature within minutes and rise according to the kinetics of the cleavage of the Ru-Cl bonds. Compared to the rates k1 and k1′ of the Hal exchange reactions, the rates kf and kr for the forward and backward reactions in the epimerization are about 10 times slower, because in the epimerization reaction the intermediates (SRu,RC)- and (RRu,RC)[CpRu(Prophos)]+ have to cross another barrier of appreciable height in the basilica-type energy profile of Scheme 4. The activation parameters for the epimerization of (SRu,RC)and (RRu,RC)-[CpRu(Prophos)Cl] are given in Table 6. The epimerization in 9:1 CDCl3/CH3OH starts with the cleavage of the Ru-Cl bond (Scheme 4) to form two strongly solvated ions. Therefore, the entropy of activation is negative, as discussed in the context of the Cl/I exchange reactions. After the cleavage of the Ru-Cl bond in the epimerization of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] the dissociation prod-

b

Measurements were performed using the methyl signals in the

ucts (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)]+ and Cl- are present in very low concentration, which disfavors the back reactions to (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] according to Scheme 4. Performance of the epimerization in the presence of a soluble chloride salt, however, should favor the back reactions and lead to a substantial decrease of the rate of epimerization. This is observed in measurements with an excess of [Bu4N]Cl (marked with b in Table 6). A 10-fold excess of [Bu4N]Cl reduced the rate kep by a factor of about 10. Epimerization of [CpRu(Prophos)Cl] in 1:1 CDCl3/ CD3OD. The epimerization of (SRu,RC)-/(RRu,RC)-[CpRu(Prophos)Cl] in 1:1 CDCl3/CD3OD was about 10-20 times faster than in 9:1 CDCl3/CH3OH, the equilibrium compositions and activation parameters remaining about the same (Table 7). Thus, at 323 K the half-life of the epimerization was only a few minutes. Solutions of (SRu,RC)- and (RRu,RC)-[CpRu(Prophos)Cl] in pure CD3OD became dark after a short time. Precise measurements of the epimerization were not possible. Epimerization of [CpRu(Prophos)I]. The epimerization kinetics of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)I] was measured in 9:1 CDCl3/CH3OH and 1:1 CDCl3/CD3OD at 323 K using diastereomerically pure (RRu,RC)-[CpRu(Prophos)I]. The data are given in Table 8. With equilibrium ratios of (RRu,RC)-/ (SRu,RC)-[CpRu(Prophos)I] ) 87:13 for both solvent mixtures the half-lives of the approach to equilibrium were τ1/2 ) 710 min in 9:1 CDCl3/CH3OH and τ1/2 ) 95 min in 1:1 CDCl3/ CD3OD. The same trend had been observed in the epimerization of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl]. A comparison of the iodo system (Table 8) with the chloro system (Tables 6 and 7) shows that the configurational stability of [CpRu(Prophos)I] is higher than that of [CpRu(Prophos)Cl] by a factor of about 10.


Brunner et al.

Table 8. Kinetics of the Epimerization of Diastereomerically Pure (RRu,RC)-[CpRu(Prophos)I] in CDCl3/CH3OH (9:1, v/v) and CDCl3/CD3OD (1:1, v/v) solvent mixture CDCl3/CH3OH 9:1, v/v CDCl3/CD3OD 1:1, v/v

temp (K)

equilibrium ratio (RRu,RC):(SRu,RC)

K

kep (min-1)

710 95

323

87:13

6.7

9.7 × 10

323

87:13

6.7

7.3 × 10-3

Epimerization of [CpRu(Norphos)I]. [CpRu(Norphos)I] was only sparingly soluble in methanol. Therefore, the epimerization kinetics was measured in 1:1 CDCl3/CD3OD. Using a sample (RRu,RC)-/(SRu,RC)-[CpRu(Norphos)I] (30:70) kinetics of the approach to the epimerization equilibrium (RRu,RC) ) (SRu,RC) at 323 K was monitored by integrating the olefinic signals of the Norphos ligands in the diastereomers. After 20 h at 323 K the system was equilibrated, the equilibrium ratio being (RRu,RC)-/(SRu,RC)-[CpRu(Norphos)I] ) 58.4:41.6 (K ) 1.40). Analysis of the data according to first-order gave kf ) 3.0 × 10-4 [min-1] and kr ) 2.1 × 10-4 [min-1], equivalent to a half-life of τ1/2 ) 136 min (k ) 5.1 × 10-4 [min-1]) for the approach to equilibrium. This shows a higher configurational stability of the Norphos complexes compared to the Prophos complexes, which had already been observed in the Hal exchange reactions.

Discussion The Mechanism. In solution the Ru-Hal bond in compounds of the type [CpRu(Prophos)Hal] and [CpRu(Norphos)Hal] tends to dissociate according to eq 1, forming the 16-electron intermediate [CpRu(P-P′)]+ and a halide ion. The cleavage of the Ru-Hal bond is the rate-determining step of the energy profile in Scheme 4.

[CpRu(P - P′)Hal] f [CpRu(P - P′)]+ + Hal-

(1)

The dissociation of the Ru-Hal bond occurs in polar media such as methanol and chloroform at room temperature or somewhat above, obvious from a series of halogen exchange and epimerization reactions presented in this paper. Cleavage of the Ru-Hal bond proceeds more readily the better the solvation of the ions formed in the dissociation according to eq 1. Hal Exchange Reactions. The Cl/I exchange in (RRu,RC)[CpRu(Prophos)Cl], the minor diastereomer in the equilibrium (RRu,RC)-[CpRu(Prophos)Cl] a (SRu,RC)-[CpRu(Prophos)Cl] ) 15:85, with excess [Bu4N]I in 9:1 CDCl3/CH3OH to give (SRu,RC)-[CpRu(Prophos)I] had a half-life of 95 min at 27 °C. Interestingly, the Cl/I exchange in (SRu,RC)-[CpRu(Prophos)Cl], the major diastereomer in the equilibrium, was slower by a factor of about 10, in accord with the energy profile in Scheme 4. The kinetics of the Cl/Br exchange was the same as that of the Cl/I exchange, because in both reactions the dissociation of the Ru-Cl bond is the rate-determining step. Quenching of the 16electron intermediate [CpRu(P-P′)]+ with halides is a fast reaction, which does not affect the rate-determining step. The entropies of activation for the Hal exchange reactions were negative, since a neutral complex formed two ions that were strongly solvated, increasing the order of the systems. The Hal exchange reactions occurred with predominant retention of configuration at the metal atom (see below). In 1:1 CDCl3/CD3OD the Hal exchange reactions were faster than in 9:1 CDCl3/CH3OH by a factor of about 10 due to better solvation of ionic species by the higher methanol concentration. In pure CD3OD, however, the rates of Hal substitution increased

τ1/2 (min)

-4

kf (min-1)

kr (min-1)

-4

8.4 × 10

1.3 × 10-4

6.6 × 10-3

9.6 × 10-4

only marginally. As expected, in the less polar solvent CDCl3 the rates of Hal exchange dropped by a factor of 10-20 compared to 9:1 CDCl3/CH3OH. In the Norphos complexes the Cl/I exchange was faster than in the corresponding Prophos complexes. Epimerization. The epimerization of [CpRu(Prophos)Cl], [CpRu(Prophos)I], and [CpRu(Norphos)I] followed first-order kinetics. A sample of (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] (85.7:14.3) approached equilibrium in 9:1 CDCl3/CH3OH at 20 °C with a half-life of 35 h. This half-life is much longer than that observed in the Hal exchange reaction, because in the basilica-type energy profile of Scheme 4 the intermediates (RRu,RC)-/(SRu,RC)-[CpRu(P-P′)]+, formed in the cleavage of the Ru-Cl bond, have to cross another barrier of appreciable height. As expected from Scheme 4, the presence of excess Clled to a marked retardation of the epimerization of (RRu,RC)-/ (SRu,RC)-[CpRu(Prophos)Cl]. In 1:1 CDCl3/CD3OD the epimerization of (RRu,RC)-/(SRu,RC)[CpRu(Prophos)Cl] was 10-20 times faster than in 9:1 CDCl3/ CH3OH. The iodo complex [CpRu(Prophos)I] proved to be configurationally more stable than the chloro complex [CpRu(Prophos)Cl] by a factor of about 10. The Norphos complexes had a higher configurational stability than the Prophos complexes. Configurational Stability. (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] were synthesized in a ratio close to 1:1 by reacting [CpRu(PPh3)2Cl] and (R)-Prophos in boiling benzene under conditions of kinetic control.6 The diastereomers were separated on the basis of solubility differences. It was claimed that the diastereomers did not epimerize in toluene at 80 °C. However, epimerization took place in C6D5Cl at the same temperature.6 The configurational stability of (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] in methanol solution was not studied, although both diastereomers were extensively used as starting materials in the solvent methanol. To the synthesized organometallic products retention of configuration and, implicitly, the same stereochemical purities as the starting materials were assigned by comparison of CD spectra, etc. As our studies showed that (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] change the metal configuration in methanol solution, slowly at room temperature and fast at higher temperatures, loss of stereoselectivity must have occurred in these substitution reactions. In most of these reactions (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] were dissolved at room temperature in methanol and the other reagents were added.2 In these cases some epimerization must have lowered the stereoselectivity depending on how long it took to form the configurationally stable organometallic compound. However, in the reactions carried out in boiling methanol2 significant epimerization must have taken place and consequently extensive loss of stereoselectivity is to be expected. In the cleavage of the Ru-Cl bond in (RRu,RC)- and (SRu,RC)[CpRu(Prophos)Cl] the high-energy 16-electron intermediates (RRu,RC)- and (SRu,RC)-[CpRu(P-P′)]+ are formed, which have maintained the Ru configuration. They can invert their metal configuration (k3 and k3′ paths) or react with nucleophiles with retention of the metal configuration (k2 and k2′ paths). The


measured competition ratios k3/k2 are on the order of 0.1, showing that the inversion of the intermediates is slow compared to the reaction with nucleophiles, in agreement with a basilicatype energy profile. The statement in ref 6 that the diastereomers (RRu,RC)- and (SRu,RC)-[CpRu(Prophos)Cl] “do not show any epimerization when heated in toluene at 80 °C for 96 h” is incorrect. We heated a diastereomer mixture (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] (53:47) in toluene-d8. After 18 h the diastereomer ratio was 58: 42, and after 46 h it was 69:31. That means the system was on its way to the (RRu,RC)-/(SRu,RC)-[CpRu(Prophos)Cl] equilibrium. The 16-Electron Intermediate [CpRu(Prophos)]+ and Related Compounds. The lability of the chloride ligand in [CpRuP(P′)Hal] and [CpRu(P-P)Hal] as well as their Cp* analogues in solvents such as alcohols was discussed in several papers. The chloro ligand in [Cp*Ru(P-P′)Hal] was so labile that a mobile equilibrium between [Cp*Ru(P-P′)Hal] and [Cp*Ru(P-P′)]+/Hal- was established.36 In a 1975 paper it was claimed on the basis of conductometric measurements that [CpRu(PPh3)2Cl] appreciably dissociated in methanol into [CpRu(PPh3)2(CH3OH)]+ and Cl-.37 However, there are no experimental data in the paper. Species [CpRu(PP)]+ were discussed as intermediates, whereas their [Cp*Ru(PP)]+ counterparts could be isolated. The X-ray analysis showed that in [Cp*Ru(PMeiPr2)2]+ the ring centroid, the Ru atom, and the two P atoms lie in a plane, the angle P-Ru-P being 101.43(5)°.25,38 In contrast to [Cp*Ru(P-P)]+ the planar frame in [Cp*Ru(N-N)]+ and even [CpRu(N-N)]+ tolerated small angles N-Ru-N of 80-81° for the hard donor N,N′tetramethylethylenediamine.25,26,39,40 The planar complexes [Cp*Ru(N-N)], N-N ) amidinate ligands, form a strange Ru-C(amidinate) contact.41 The Angle P-Ru-P. With extended Hu¨ckel methodology as well as density functional theory the structure of coordinatively unsaturated, two-legged 16-electron piano stool complexes of the type [(η-CnHn)MLL′] were analyzed.23,24 Whereas for fragments with strongly π-accepting ligands such as [CpMn(CO)2] and [CpFe(CO)2]+ pyramidal structures with inversion barriers of about 10 kcal/mol were calculated, fragments with σ-donor ligands such as [CpMn(PH3)2] and [CpFe(PH3)2]+ were predicted to adopt planar geometries. In the DFT geometry optimization of [CpFe(CO)2]+ the OC-Fe-CO angle in the (36) De los Rios, I.; Jimeńez-Tenorio, M.; Padilla, J.; Puerta, M. C.; Valerga, P. J. Chem. Soc., Dalton Trans. 1996, 377–381. (37) Haines, R. J.; du Preez, A. L. J. Organomet. Chem. 1975, 84, 357– 367. (38) Jimeńez-Tenorio, M.; Mereiter, K.; Puerta, M. C.; Valerga, P. J. Am. Chem. Soc. 2000, 122, 11230–11231. (39) Gemel, C.; Mereiter, K.; Schmid, R.; Kirchner, K. Organometallics 1997, 16, 5601–5603. (40) Gemel, C.; Sapunov, V. N.; Mereiter, K.; Ferencic, M.; Schmid, R.; Kirchner, K. Inorg. Chim. Acta 1999, 286, 114–120. (41) Yamaguchi, Y.; Nagashima, H. Organometallics 2000, 19, 725– 727.


pyramidal species was 94.14°, while in the planar transition state it was 103.02°. B3LYP energy optimization showed that [CpRu(NH3)2]+ was planar with an angle N-Ru-N of 91°, whereas [CpRu(PH3)2]+ was pyramidal with an angle P-Ru-P of 96°.25 In [CpRu(PPh3)2Cl], the parent compound of the chelate complexes studied here, the P-Ru-P angle42 is 103.79(3)°, similar to the angles in the calculations.24 Usually, in fivemembered chelate rings the P-M-P angles are smaller. Thus, the P-Ru-P angle in [CpRu(Prophos)Cl]6 is 82.9°, and in [CpRu(Norphos)I]7 it is 86.1°. The P-Ru-P angles in the Prophos and Norphos complexes, described in this paper, are 83.1-83.4° and 85.1-86.1°, respectively (Table 2). The puckered Ru-Prophos chelate ring, the λ-conformation of which is dictated by the equatorial orientation of the methyl group, is relatively rigid and cannot increase the P-Ru-P angle appreciably. Imposed by the norbornene skeleton, the Ru-Norphos chelate ring, also having λ-conformation, is completely rigid. Both chelate rings will resist a widening of the P-Ru-P angle, which in the planar complex CpRh(PPh3)2 is 98.17°.43 This is a fact that strongly favors a pyramidal structure of the 16electron fragments [CpRu(P-P′)]+. Not only this, it also will increase the inversion barrier, because in the planar transition state even larger P-Ru-P angles are required. This should be analogous to the inversion of the amine nitrogen, which is slowed dramatically by introducing the nitrogen atom into small rings. Thus, aziridines with their 60° angles greatly resist planarization, which needs expanded angles in the transition state. It is to be expected that increasing the P-Ru-P angle, e.g., with the ligand Chairphos44,45 (Ph2P-CHMe-CH2CH2-PPh2), having one CH2 group more than Prophos, will decrease the barrier of the pyramidal inversion. On the contrary, a decrease of the size of the chelate ring, e.g., to a fourmembered P-P′ system, should increase the inversion barrier significantly. Supporting Information Available: 1H NMR and 31P{1H} NMR spectra of 1b, a mixture of 1b/1b′, and a mixture of 1c/1c′. Data for the molecular structure of [CpRh(PPh3)2] in PDF format. CIF files giving the crystallographic data for compounds 1b, 1c, 2b′, 2c′, 2c, and [CpRh(PPh3)2]. This material is available free of charge via the Internet at http://pubs.acs.org. X-ray data, except for [CpRh(PPh3)2], have been deposited at the CCDC, Cambridge U.K. (659547-659551). OM800148R (42) Perkins, G. J.; Bruce, M. I.; Skelton, B. W.; White, A. H. Inorg. Chim. Acta 2006, 359, 2644–2649. (43) Tsuno, T.; Uehara, T.; Brunner, H.; Zabel, M. Unpublished results. (44) Kagan, H. B.; Fiaud, J. C.; Hoornaert, C.; Meyer, D.; Poulin, J. C. Bull. Soc. Chim. Belg. 1979, 88, 923–931. (45) MacNeil, P. A.; Roberts, N. K.; Bosnich, B. J. Am. Chem. Soc. 1981, 103, 2273–2280.

Pyramidal Stability of Chiral-at-Metal Half-Sandwich 16-Electron

Recommend Documents