Silica supported Pd/Ba (and Pd/Sr, see Fig.) have been prepared using {(DMF)xBa(or Sr)(CN)4}∞ as precursor. The bimetallics delivered activities that were up to two orders of magnitude greater than that of Pd/SiO2, a response that we attribute to a surface Ba/Pd (and Sr/Pd) synergy that enhances Pd dispersion (and H2 chemisorption capacity) allied to a more effective C–Cl activation for hydrogen attack.

A series of alkaline earth-transition metal heterobimetallic complexes {(DMF)(x)M(mu-CN)(y)M*(CN)(4-y)}(infinity) [M = Ba: M* = Ni 1, M* = Pd 2, M* = Pt 3; M = Sr: M* = Ni 4, M* = Pd 5, M* = Pt 6; x = 3 or 4; y = 3 or 4] was prepared. Single-crystal X-ray diffraction analysis revealed that {(DMF)(4)Ba(mu-CN)(3)Ni(CN)}(infinity) 1, {(DMF)(4)Sr(mu-CN)(3)Ni(CN)}(infinity) 4, {(DMF)(4)Sr(mu-CN)(3)Pd(CN)}(infinity) 5, and {(DMF)(4)Sr(mu-CN)(3)Pt(CN)}(infinity) 6 form isostructural one-dimensional "ladder" arrays through isocyanide linkages (M-NC-M*), while {(DMF)(3)Ba(mu-CN)(4)Pd}(infinity) 2, and {(DMF)(3)Ba(mu-CN)(4)Pt}(infinity) 3 are two-dimensional puckered sheet like arrays. These bimetallic complexes can serve as effective precursors for the synthesis of supported bimetallic catalysts. It has been established that Ba-Pd/SiO(2) and Sr-Pd/SiO(2) catalysts, prepared from 2 and 5 loaded onto a silica support, delivered specific reaction rates in the hydrodechlorination of mono- and dichloro benzene that were over an order of magnitude greater than that achieved with conventional Pd/SiO(2).

The gas phase hydrodechlorination (HDC) of chlorobenzene (CB), 1,2-dichlorobenzene (1,2-DCB) and 1,3-dichlorobenzene (1,3-DCB) has been investigated over Pd/SiO2 and a series of Ln-Pd/SiO2 prepared from the organometallic precursor{(DMF)10Ln2[Pd(CN)4]3}∞ where Ln=La, Ce, Sm, Eu, Gd and Yb; Pd loading=5%, w/w. Under identical reaction conditions, the following overall sequence of increasing initial fractional dechlorination has been established: Pd/SiO2<Yb-Pd/SiO2≈Sm-Pd/SiO2<Gd-Pd/SiO2<La-Pd/SiO2≈Ce-Pd/SiO2<Eu-Pd/SiO2; reaction over Ln/SiO2 resulted in a negligible conversion. HDC activity declined with time-on-stream but the Ln-Pd/SiO2 catalysts maintained a significantly higher fractional HDC than Pd/SiO2; loss of activity is attributed to deleterious HCl/surface interactions. The pre- and post- reaction catalyst samples have been characterized in terms of BET area, TPR, TEM, H2chemisorption/TPD, XRD and XPS analyses. When compared with Pd/SiO2, Pd is present in the Ln-Pd/SiO2 samples as much smaller particles, while the lanthanide component is finely dispersed over the surface, i.e. Ln is in intimate contact with Pd. The promotional effect of Ln in Ln-Pd/SiO2 is attributed to a surface Pd/Ln synergism resulting in an enhancement of surface reactive hydrogen and a more effective CCl bond activation for hydrogenolytic attack. HDC performance is discussed in terms of surface composition, Pd particle size, Ln electronic structure and H2 uptake/release dynamics.

An ionic dinuclear triple-hydrogen-bridged 9-BBN hydroborate zirconium complex, [K(Et2O)4][{(μ-H)2BC8H14}3Zr(μ-H)3Zr{(μ-H)2BC8H14}3] (1) was formed from the reaction of Zr{(μ-H)2BC8H14}4 with KH and aniline in diethyl ether. Its molecular structure was determined. 9-BBN hydroborate ligands are coordinated to Zr atoms via bridging H atoms, with two Zr atoms bridged to each other by three hydrogens.

Systematic synthetic procedures produced several different structural types of extended lanthanide–transition metal (group 10) complexes with cyanide bridges. Of these, one-dimensional ladder arrays containing a Yb–Pd combination have been converted to bimetallic heterogeneous catalysts on an oxide (SiO2) surface that is more effective than supported Pd alone. Two lanthanide–Cu(I) complexes have been prepared. One type, an inclusion complex consists of lanthanide(III) cations encapsulated in the pockets of a three-dimensional anionic array that contains Cu(I)–CN–Cu(I) bridges. The second type, an extended layer complex, consists of joined five-membered rings in a “tile-like” pattern with Ln–CN–Cu and Cu–CN–Cu bridges.

The gas-phase hydrodechlorination (HDC) of chlorobenzene (CB), 1,2-dichlorobenzene (1,2-DCB), and 1,3-dichlorobenzene (1,3-DCB) was investigated over a 5% w/w Pd/SiO2 and a series of SiO2-supported Yb-Pd (5% w/w Pd and Yb/Pd = 2/3 mol/mol) catalysts. The Pd/SiO2 catalyst was prepared by Pd(C2H3O2)(2) impregnation, supported Yb synthesized by contacting SiO2 with Yb powder in liquid NH3 and the bimetallic catalysts prepared by two stepwise routes: Pd(C2H3O2)(2) impregnation of Yb/SiO2 (Pd-Yb/SiO2-step) or Pd impregnation preceding Yb introduction (Yb-Pd/SiO2-step) and a single-step simultaneous introduction of Pd and Yb from a {(DMF)(10)Yb-2[Pd(CN)(4)](3)}(infinity) precursor (Yb/Pd/SiO2-sim). Under identical reaction conditions the following specific initial CB HDC rate sequence was established: Pd-Yb/SiO2-step (0.21 mol(Cl) h(-1) m(-2)) Pd/SiO2 (0.24 mol(Cl) h(-1) m(-2)) < Yb/Pd/SiO2-sim (0.41 mol(Cl) h(-1) m(-2)) approximate to Yb-Pd/SiO2)-step (0.46 mol(Cl) h(-1) m(-2)): reaction over Yb/SiO2 resulted in a negligible conversion. Yb acts as a CB HDC promoter through a Surface synergism with Pd: the extent of this promotion depends on the nature of the catalyst precursor and the sequence of metal(s) introduction to the support. The promotional effect of Yb extends to DCB HDC, where Yb-Pd/SiO2-step outperforms Yb/Pd/SiO2-sim (with a specific HDC rate 25 times greater than that delivered by Pd/SiO2) and to catalytic hydrogenation (benzene -> cyclohexane). The prereaction and postreaction catalyst samples were characterized in terms of BET area, TPR, TEM-EDX, H-2 chemisorption/TPD, XRD. and XPS measurements. The role of Yb as a promoter is discussed in terms of electron donation and impact on Pd dispersion but is attributed to the action of YbH2, which serves as an additional source of surface reactive hydrogen: Yb activation of the C-Cl bond(s) for hydrogenolytic attack is also considered. HDC activity decreased with time-on-stream. an effect that we link to deleterious HCl/catalyst interactions that modify surface composition. leading to a disruption in H-2 uptake/release: XPS and TEM-EDX were used to characterize the residual surface Cl post-HDC. (c) 2006 Elsevier Inc. All rights reserved.

The complex Zr(9-BBN)(4) [9-BBN = (mu-H)(2)BC8H14] has been synthesized via the reaction of K(9-BBN) with ZrCl4 in diethyl ether. The structure of the title compound has been determined by X-ray and neutron single-crystal diffraction techniques. Each 9-BBN ligand is coordinated to the Zr atom via two B-H-Zr bridges, and these metalligand bonding interactions are further augmented by three prominent C-H---Zr agostic interactions. Average molecular parameters derived from the neutron analysis: Zr-H = 2.051(8) angstrom, B-H = 1.286(7) angstrom, Zr---B = 2.409(6) angstrom, Zr-H-B = 87.7(4)degrees, H-Zr-H = 58.9(3)degrees. The Zr---H distances corresponding to the three C-H... Zr agostic interactions are 2.424(7), 2.663(8), and 2.551(7) angstrom. The fourth potential C-H---Zr interaction has a Zr---H distance [3.146(7) angstrom] that is too long to be considered in the agostic range. Single-crystal X-ray diffraction data were collected on an Enraf-Nonius Kappa CCD diffraction system, and neutron diffraction data were collected on the quasi-Laue diffractometer VIVALDI at the Institut Laue-Langevin; the final agreement factor for the neutron analysis is 6.52% for 2557 reflections with l > 2 sigma(l).

9-BBN hydroborate complexes Ti{(mu-H)(2)BC8H14}(3)(THF)(2) (1), Ti{(mu-H)(2)BC8H14}(3)(OEt2) (2), and [K(OEt2)(4)]-[Ti{(mu-H)(2)BC8H14}(4)] (4) were formed from the reaction of TiCl4 with K[H2BC8H14] in diethyl ether or THF. Ti{(mu-H)(2)BC8H14}(3)(PhNH2) (3) was isolated from the reaction of 2 with aniline in diethyl ether. In the formation of these complexes, Ti(IV) is reduced to Ti(III). The coordinated diethyl ether in 2 can be displaced by the stronger bases THF and aniline, to form 1 and 3, respectively. All of the compounds were characterized by single-crystal X-ray diffraction analysis. In complex 1, which contains two coordinated THF ligands, the titanium possesses a 17 electron configuration and there is no evidence for agostic interaction. Complexes 2 and 3 contain only one coordinated ether or aniline ligand, and the titanium possesses a 15 electron configuration. In these compounds, a C-H hydrogen on an a carbon on the BC8H14 unit of a 9-BBN hydroborate ligand forms an agostic interaction with the titanium. Criteria for assessing the existence of agostic interactions are discussed. As the potassium salt, the anion of complex 4 is more stable than the complexes 1-3. Organometallic anions of the type [ML4](-) for titanium(III) are rare.