T.S. Venkatakrishnan et al. / Journal of Organometallic Chemistry 691 (2006) 224–228
225
3
2. Experimental
N2C3HMe2-3,5), 1.33 (d, J(H,H) = 7.0 Hz, CH3, CHMe-
Ph). 31P{1H} NMR (CDCl3, ppm) for 3a and 3b: 3a (major
2
2.1. General
diastereomer) 333.4 (d, J(P,P) = 43.6 Hz, PPh), 76.0 (d,
PPh2); 3b (minor diastereomer) 333.9 (d, 2J(P,P) =
All reactions and manipulations were carried out under
an atmosphere of dry nitrogen using standard Schlenk and
vacuum-line techniques. The solvents were purified by
standard procedures and distilled under nitrogen prior to
use. The NMR spectra (1H, 31P{1H}) were recorded in
CDCl3 at 298 K using Bruker Avance-400 MHz spectrom-
eters. IR spectra were recorded using a Bruker FT-IR spec-
trometer as a thin film on a KBr disk. Elemental analyses
were carried out using a Perkin–Elmer 2400 CHN analyser.
Melting points were recorded in a Buchi B-540 melting
point apparatus and were uncorrected. The ligands
(SCRP)-1 [7] and 2 [8] were prepared by previously reported
procedures. Me3NO (Aldrich) and Ru3(CO)12 (Strem
Chemicals) were used as received.
43.6 Hz, PPh), 72.7 (d, PPh2).
2.4. Ru3(CO)5(lsb-CO)2{l3-N,N0-g1:g1:g1-N2C3HMe23,5}-
{l-P,P0-Ph2PN(R)PPh}] (R = CHMe2) (4):
Yield: 0.009 g (12%). Anal. Calc. for C33.5H32N3O8-
P2Ru3Cl: C, 40.0; H, 3.2; N, 4.2; Found: C, 39.2; H, 2.6;
N, 4.3%. M.p 186–188 ꢁC (d). IR (neat, mCO cmꢀ1):
2042(s), 1990(vs), 1976(w, sh), 1957(w, br), 1929(m),
1887(m, br), 1793(s, br, lsb-CO). 1H NMR (400 MHz,
CDCl3, ppm): 5.28 (s, CH, N2C3HMe2-3,5), 3.30 (m, CH,
CHMe2), 1.63, 1.43 (s, CH3, N2C3HMe2-3,5), 0.95, 0.80
(d, 3J(H,H) = 7.0 Hz, CH3, CHMe2). 31P{1H} NMR
2
(CDCl3, ppm): 331.3 (d, J(P,P) = 43.0 Hz, PPh), 69.0 (d,
PPh2).
2.2. Ru3(CO)5(lsb-CO)2{l3-N,N0-g1:g1:g1-N2C3HMe23,5}-
{l-P,P0-Ph2PN(R)PPh}] (3, 4)
2.5. X-ray crystallography
To a solution of Ru3(CO)12 (0.050 g, 7.82 · 10ꢀ5 mol) in
THF (6 cm3) were added a few drops of Bruce catalyst [9]
using a syringe until the solution darkened. Immediately
the ligand (0.040 g of 1 or 0.035 g of 2, 7.82 · 10ꢀ5 mol)
was added. The resulting solution was heated under reflux
for 2 h. The 31P{1H} NMR spectrum of the reaction mix-
ture revealed the presence of several products. Solvent
was evaporated from the reaction mixture in vacuo to ob-
tain a dark colored residue. The residue was dissolved in
dichloromethane (2 cm3) and subjected to chromato-
graphic separation by thin layer chromatography over sil-
ica gel using CH2Cl2/petrol (1:1 v/v) as eluant. The red
orange band at Rf = 0.70 was isolated and the product ex-
tracted into dichloromethane. Evaporation of the solvent
afforded a dark red residue which was crystallized from
dichloromethane–petrol to yield the compounds, [Ru3-
(CO)5(lsb-CO)2{l3-N,N0-g1:g1:g1-N2C3HMe2-3,5}{l-P,P0-
Ph2PN(R)PPh}] [R = (S)-*CHMePh (3) or CHMe2 (4)] as
dark red orange solids. Efforts to isolate the other prod-
uct(s) from the reaction mixture were unsuccessful.
The crystal was mounted on a glass fibre and the inten-
sity data was obtained at room temperature from a Bruker
SMART APEX CCD diffractometer equipped with fine fo-
cus 1.75 kW sealed tube Mo Ka X-ray source with increas-
ing x (width of 0.3ꢁ/frame) at a scan speed of 5 s/frame.
The SMART [10a] software was used for cell-refinement
and data acquisition and the SAINT [10b] software was used
for data reduction. Lorentzian and polarization corrections
were made on the intensity data. An absorption correction
was made on the intensity data using the SADABS [10c] pro-
gram. The structure was solved using SHELXTL [10d] and the
WINGX graphical user interface [11]. Least-square refine-
ments were performed by the full-matrix method with
SHELXL-97 [12]. All non-hydrogen atoms were refined aniso-
tropically and hydrogen atoms were refined isotropically
and were fixed at idealised positions. The carbon atom
C(21) attached to C(19) was disordered over two sites
and was modelled with 50:50 occupancy. Hence, the hydro-
gen atom bound to C(19) was not fixed. The lattice held
water and the solvent dichloromethane molecules were re-
fined separately during final refinement cycles. The carbon
and chlorine atoms of the solvent dichloromethane were re-
fined isotropically. The hydrogen atoms of the lattice held
dichloromethane and water were not located.
2.3. Ru3(CO)5(lsb-CO)2{l3-N,N0-g1:g1:g1-N2C3HMe23,5}-
{l-P,P0-Ph2PN(R)PPh}] (R = (S)-*CHMePh)-
{(SCRP)-(3a) and (SCSP)-(3b)}
Yield: 0.008 g (10%). Anal. Calc. for C38H31N3O7P2Ru3:
C, 45.3; H, 3.1; N, 4.2; Found: C, 44.7; H, 3.1; N, 4.0%.
M.p. 181–183 ꢁC (d). IR (neat, mCO cmꢀ1): 2040(w),
2025(m), 1991(vs), 1976(w), 1953(sh), 1930(w), 1892(w,
br), 1791(s, br, lsb-CO). 1H NMR (400 MHz, CDCl3,
ppm) for 3a and 3b: 3a (major diastereomer) 5.27 (s, CH,
N2C3HMe2-3,5), 3.90 (m, CH, CHMePh), 1.59, 1.46 (s,
CH3, N2C3HMe2-3,5), 1.44 (d, 3J(H,H) = 7.0 Hz, CH3,
CHMePh); 3b (minor diastereomer) 5.23 (s, CH,
N2C3HMe2-3,5), 4.20 (m, CH, CHMePh), 1.38 (s, CH3,
3. Results and discussion
In contrast to the reactivity of diphosphazanes reported
earlier [4], the unsymmetrical pyrazolyl substituted diphos-
phazanes Ph2PN(R)PPh(N2C3HMe2-3,5) [R = (S)-*CHMe-
Ph (SCRP)-(1), CHMe2 (2)] display a different type of
reactivity towards Ru3(CO)12. The reaction of Ru3(CO)12
with an equimolar quantity of the diphosphazane 1 or 2
in the presence of benzophenone-ketyl radical in boiling
THF results in the formation of several products from