Preparation and characterization of diarylphosphazene and diarylphosphinohydrazide complexes of titanium, tungsten and ruthenium and phosphorylketimido complexes of rhenium

Article abstract of DOI:10.1039/b417068g

Reaction of the proligand Ph₂PN(SiMe₃)₂ (L¹) with WCl₆ gives the oligomeric phosphazene complex [WCl₄(NPPh₂)]_n, 1 and subsequent reaction with PMe₂Ph or NBu₄Cl gives [WCl₄(NPPh ₂)(PMe₂Ph)] (2) or [WCl₅(NPPh ₂)][NBu₄] (3), respectively. DF calculations on [WCl ₅(NPPh₂)][NBu₄] show a W=N double bond (1.756 A) and a P-N bond distance of 1.701 A, which combined with the geometry about the P atom suggests, there is no P-N multiple bonding. Reaction of L¹ with [ReOX₃(PPh₃)₂] in MeCN (X = Cl or Br) gives [ReX₂(NC(CH₃)P(O)Ph₂)(MeCN) (PPh₃)] (X = Cl, 4, X = Br, 5) which contains the new phosphorylketimido ligand. It is bound to the rhenium centre with a virtually linear Re-N-C arrangement (Re-N-C angle = 176.6°. when X = Cl) and there is multiple bonding between Re and N (Re-N = 1.809(7) A when X = Cl). The proligand Ph₂PNHNMe₂ (L²H) reacts with [(C ₅H₅)TiCl₃] to give [(C₅H ₅)TiCl₂(Me₂NNPPh₂)] (6). An X-ray crystal structure of the complex shows the ligand (L²) is bound by both nitrogen atoms. Reaction of the proligands Ph₂PNHNR₂ [R₂ = Me₂ (L²H), -(CH₂CH ₂)₂NCH₃ (L³H), (CH ₂CH₂)₂CH₂ (L⁴H)] with [{RuCl(μ-Cl)(η⁶-p-MeC₆H₄Pr)} ₂] gave [RuCl₂(η⁶-p-MeC₆H ₄Pr)(L)] {L = L²H (7), L³H (8), L⁴H (9)}. The X-ray crystal structures of 7-9 confirmed that the phosphinohydrazine ligand is neutral and bound via the phosphorus only. Reaction of complexes 7-9 with AgBF₄ resulted in chloride ion abstraction and the formation of the cationic species [RuCl(η⁶-p-MeC₆H ₄Pr)(L)]⁺BF₄{(L = L²H (10), L ³H (11), L⁴H (12)}. Finally, reaction of complex 6 with [{RuCl(μ-Cl)(η⁶-p-MeC₆H₄Pr)} ₂] gave the binuclear species [(η⁶-p-MeC ₆H₄Pr)Cl₂Ru(μ²,η³- PNNMe₂)TiCl₂C₅H₅)], 13. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b417068g

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Preparation and characterization of diarylphosphazene and
diarylphosphinohydrazide complexes of titanium, tungsten and
ruthenium and phosphorylketimido complexes of rhenium
Andrew R. Cowley, Jonathan R. Dilworth,* Alison K. Nairn and Alasdair J. Robbie
Chemistry Research Laboratory, University of Oxford, 12 Mansfield Road, Oxford,
UK OX1 3TA. E-mail: jon.dilworth@chem.ox.ac.uk; Fax: 01865-272690; Tel: 01865-285151
Received 9th November 2004, Accepted 15th December 2004
First published as an Advance Article on the web 24th January 2005
Reaction of the proligand Ph2PN(SiMe3)2 (L1) with WCl6 gives the oligomeric phosphazene complex
[WCl4(NPPh2)]n, 1 and subsequent reaction with PMe2Ph or NBu4Cl gives [WCl4(NPPh2)(PMe2Ph)] (2) or
=
[WCl5(NPPh2)][NBu4] (3), respectively. DF calculations on [WCl5(NPPh2)][NBu4] show a W N double bond
(1.756 A) and a P–N bond distance of 1.701 A, which combined with the geometry about the P atom suggests, there
˚
˚
is no P–N multiple bonding. Reaction of L1 with [ReOX3(PPh3)2] in MeCN (X = Cl or Br) gives
[ReX2(NC(CH3)P(O)Ph2)(MeCN)(PPh3)] (X = Cl, 4, X = Br, 5) which contains the new phosphorylketimido ligand.
It is bound to the rhenium centre with a virtually linear Re–N–C arrangement (Re–N–C angle = 176.6◦. when X =
˚
Cl) and there is multiple bonding between Re and N (Re–N = 1.809(7) A when X = Cl). The proligand
Ph2PNHNMe2 (L2H) reacts with [(C5H5)TiCl3] to give [(C5H5)TiCl2(Me2NNPPh2)] (6). An X-ray crystal structure of
the complex shows the ligand (L2) is bound by both nitrogen atoms. Reaction of the proligands Ph2PNHNR2 [R2 =
Me2 (L2H), –(CH2CH2)2NCH3 (L3H), (CH2CH2)2CH2 (L4H)] with [{RuCl(l-Cl)(g -p-MeC6H4 Pr)}2] gave
6
i
[RuCl2(g -p-MeC6H4 Pr)(L)] {L = L2H (7), L3H (8), L4H (9)}. The X-ray crystal structures of 7–9 confirmed that the
6
i
phosphinohydrazine ligand is neutral and bound via the phosphorus only. Reaction of complexes 7–9 with AgBF4
6
i
−
resulted in chloride ion abstraction and the formation of the cationic species [RuCl(g -p-MeC6H4 Pr)(L)]+BF4
{(L = L2H (10), L3H (11), L4H (12)}. Finally, reaction of complex 6 with [{RuCl(l-Cl)(g -p-MeC6H4 Pr)}2] gave the
6
i
6
i
3
binuclear species [(g -p-MeC6H4 Pr)Cl2Ru(l2,g -Ph2PNNMe2)TiCl2(C5H5)], 13.
disproportionate readily in reactions with Ni to give a new
Introduction
bidentate N–P–N ligand [(NPh)2PPh2] together with a bridging
PPh2 ligand. In fact, the chemistry of these systems is dominated
by the tendency for disproportionation and cleavage of the P–N
bond.
The chemistry of metal complexes which contain metal–nitrogen
multiple bonds has been very comprehensively studied.1–3 The
drive to study these complexes comes from a desire to understand
the mechanism by which the nitrogenase enzymes operate, for
which the key intermediates in the conversion of coordinated N2
to ammonia are the diazenido (M N NH), isodiazene (M≡N–
NH2) and the hydrazido(1−) (M–NH–NH2) species.4,5 These
types of metal complex have a rich and diverse chemistry in
their own right; they show a wide range of different binding
modes with a large number of different metals.
Roesky et al. prepared [Li(THF)4][(Ph2P–NPh)4Ln] (Ln = Y,
Yb and Lu) using the same lithiated ligand (Ph2P–NPh− Li+).
=
=
2
These complexes are eight coordinate incorporating four g -P–
N ligands.18 Subsequent work by Ku¨hl et al. showed that Zr
complexes of the same ligand could be prepared with the same
coordination geometry.19,20 They also prepared the Ti complex
[TiCl2{N(PPh2)2}2], by reaction of TiCl4 with Li[N(PPh2)2], in
In stark contrast to the many M–N–N species known, there
are very few examples of complexes containing coordinated N–
P(III) ligands, and in almost all of these examples the ligand is
bound in neutral form to the metal via phosphorus only. Many of
2
which the P–N–P ligand binds in the same g -fashion with the
second phosphorus uncoordinated.
We here report our efforts to prepare the hitherto unknown
phosphorus(III) substituted analogues [e.g. R2P(III)NM] of the
isodiazene ligand systems and to study their chemistry and the
nature of the bonding when coordinated. We have adopted
the diarylphosphazene nomenclature for these complexes to
parallel that for the NNR2 (isodiazine) analogues, although the
calculations and observed chemistry suggest that there is no P–
N multiple bonding and that the formal charge is 2−. We also
report investigations into the synthesis of complexes of ligands
[R2P–NRꢀ]− and the impact of the nature of Rꢀ on the complex
stability.
these examples incorporate ligands of the type NHR1–PR2 .6–10
2
For example, the Mo complex in which R1 = H and R2 = Ph has
the molecular structure [Mo(CO)4(PPh2NH2)2], containing two
H2N–PPh2 ligands bound trans to each other via phosphorus.11
This is in contrast to P(V)–N systems, for which there are
many examples of complexes containing the M–N–P unit.
Roesky et al. have prepared the complex, [TiCl2(py)3(NP(S)Ph2)]
12
=
which incorporates the Ti N–P(S)Ph2 unit, and the complex
[(C5H5)TiCl2(NPPh3)] was prepared by Dilworth et al.13,14 Ti
complexes with this ligand type have seen recent applications as
very effective olefin polymerisation catalysts,15,16 and there are
other examples of Ph3P–N complexes with metals such as V, Mo
and W.
Experimental
General procedures
The chemistry of P(III)-containing analogues of hy-
drazido(1−) complexes (phosphaneazenides) has also been com-
paratively little explored. Work by Fenske et al.17 involved the
use of [Ph2P–NPh]−Li+, which can bind through nitrogen and
phosphorus {e.g. [Pd(Ph2PNPh)(PPh3)]} and can also bridge
two metal centres {e.g. [{M(Ph2PNPh)(Ph2PNHPh)}2], M =
Pd, Pt}. The lithiated ligand was however also found to
All reactions were carried out using standard Schlenk-line
techniques under an atmosphere of dry N2. CH2Cl2 and MeCN
were both dried by distillation from CaH2, Et2O was dried by
distillation from Na/benzophenone. All solvents were degassed
thoroughly with N2 before use. All reagents were purchased
from Aldrich and used as received unless stated other-
wise. Ph2PN(SiMe3)2 was prepared according to the literature
6 8 0
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
©

View Article Online
procedure,21,22 but was used without distillation as this caused
extensive decomposition. Ph2PNHNMe2 was prepared accord-
ing to the literature method.23,24 (C5H5)TiCl3 was prepared from
(C5H5)2TiCl2 and TiCl4 by the literature procedure.25,26
a colour change to dark grey–purple was observed. The volume
was then reduced to ca. 2 cm3 and then Et2O (10 cm3) was added
to precipitate an oily blue–purple solid. The filtrate was removed
and then another portion of Et2O was added (10 cm3) and the
mixture was left to stir for 30 min forming a fine blue–purple
powder which was collected by filtration, washed with Et2O
(5 cm3) and then dried under vacuum (0.378 g, 0.47 mmol, 36%
yield). Elemental analysis for C28H37Cl5N2PW, found (calc.%):
C 41.4 (41.9), H 5.2 (5.8), N 3.2 (3.5). 1H NMR (CD2Cl2): d 7.4–
8.0 (weak m), 3.2 (br s, 2H, –CH2(CH2)2CH3), 1.65 (br s, 2H,
CH2CH2CH2CH3), 1.35 (br s, 2H, (CH2)2CH2CH3), 0.9 (br s,
3H, (CH2)3CH3). IR/cm−1 (KBr disk): 3385 (m, br), 3058 (w),
1
1
All 1H, 13C{ H} and 31P{ H} NMR spectra were recorded on a
1
Varian Mercury VX 300 spectrometer (1H at 300 MHz, 13C{ H}
1
at 75.5 MHz and 31P{ H} at 121.5 MHz) or a Varian Unit
1
500 MHz spectrometer (1H at 499.9 MHz, 13C{ H} at 125.7 MHz
and 31P at 202.4 MHz). Chemical shifts were referenced against
the internal solvent. Mass spectra were recorded on a Micromass
LCT ToF Spectrometer. IR spectra were recorded on a Perkin-
Elmer 1710 spectrometer as KBr disks. Elemental analyses were
performed by the microanalysis laboratory of the Inorganic
Chemistry Laboratory, University of Oxford.
2960 (s), 2873 (s), 1615 (w), 1590 (w), 1481 (s), 1438 (s, vP Ph),
–
1381 (m), 1264 (m, br), 1127 (s, vW P), 1090 (m), 1027 (w), 996
–N–
(w), 882 (w), 749 (m), 728 (s), 691 (s), 552 (m), 518 (m).
Preparations
=
Preparation of [ReCl2(MeCN)(PPh3)2(N C(CH3)P(O)Ph2)], 4.
Preparation of [WCl4(NPPh2)], 1. Ph2PN(SiMe3)2 (0.508 g
1.47 mmol) was dissolved in Et2O (15 cm3) to give a pale yellow
solution. WCl6 (0.518 g, 1.47 mmol) was dissolved in Et2O
(15 cm3) to give and intense orange–brown coloured solution.
The Ph2PN(SiMe3)2 solution was added to the WCl6 solution
and a colour change to intense dark blue was observed over a
few minutes. The solvent was removed under vacuum to leave a
sticky dark blue solid which was re-dissolved in CH2Cl2 (10 cm3).
The solution was filtered and the volume was then reduced to ca.
3 cm3 and Et2O (15 cm3) added to precipitate a sticky dark blue
solid. The filtrate was removed and Et2O (10 cm3) was added,
after stirring for 30 min a fine dark blue powdery solid formed
which was collected by filtration, washed with Et2O (2 × 5 cm3)
and then dried under vacuum (0.360 g, 0.691 mmol, 47% yield).
Elemental analysis for C12H10Cl4NPW, found (calc.%): C 27.2
(27.4), H 2.2 (1.9), N 2.3 (2.7). 1H NMR (CD2Cl2): d 7.2–7.8 (m,
Method A. [ReOCl3(PPh3)2] (0.998 g, 1.20 mmol) was placed
in a Schlenk flask and MeCN (10 cm3) was added to give a
lime green suspension. Ph2PN(SiMe3)2 (0.469 g, 1.36 mmol) was
dissolved in MeCN (10 cm3) to give a pale yellow solution.
The Ph2PN(SiMe3)2 solution was added to the ReOCl3(PPh3)2
suspension and then the mixture was heated under reflux for 2 h.
The dark green–brown solution was filtered hot into a clean flask
and then allowed to cool overnight. A green crystalline material
(4) formed which was collected by filtration and washed with
MeCN (2 × 5 cm3) and then Et2O (2 × 5 cm3) and then dried
under vacuum (0.147 g, 0.14 mmol, 9% yield).
Method B. [ReOCl3(PPh3)2] (0.998 g, 1.20 mmol) was placed
in a Schlenk flask along with Ph2PN(SiMe3)2 (0.413 g,
1.20 mmol) and PPh3 (0.692 g 2.64 mmol). MeCN (15 cm3)
was added to give a lime green suspension and the mixture was
then heated under reflux for 2 h. The solution was filtered hot
to collect a green crystalline material, 4, which was washed with
MeCN (2 × 5 cm3) and dried under vacuum (0.432 g, 0.41 mmol,
34% yield). The filtrate was left to cool and after 24 h a mixture of
green and orange crystals had formed in the solution (0.169 g).
The green crystals were identified as the desired product (4)
and the orange crystals as [Re(MeCN)Cl3(PPh3)2]. Analysis
1
10H, Ph). 31P{ H} NMR (CD2Cl2): d 68 (s, NPPh2). IR/cm−1
(KBr disk): 3405 (m, br), 2963 (m), 1731 (w), 1636 (w), 1590 (w),
1540 (w), 1439 (s, vP Ph), 1292 (s), 1131 (s, vW P), 1096 (s), 1023
–
– –
N
(s), 801 (s), 752 (w), 730 (m), 691 (m), 520 (m).
Preparation of [WCl4(NPPh2)(PMe2Ph)], 2. Ph2PN(SiMe3)2
(0.474 g, 1.37 mmol) was dissolved in Et2O (10 cm3) to give a
pale yellow solution. WCl6 (0.498 g, 1.26 mmol) was dissolved
in Et2O (10 cm3) to give an intense dark orange–brown solution.
The Ph2PN(SiMe3)2 solution was added to the WCl6 and a colour
change to intense dark blue was observed over a few minutes.
The solvent was removed under vacuum to leave a sticky dark
blue solid which was re-dissolved in CH2Cl2 (10 cm3). A solution
of PMe2Ph (0.18 cm3, 1.25 mmol) in CH2Cl2 (5 cm3) was added
to the reaction mixture and a colour change to intense red–
brown was observed. The solution was filtered and then the
volume was reduced to ca. 5 cm3, Et2O (20 cm3) was added and
a dark red–brown solid was formed. The solid was collected by
filtration, washed with Et2O (2 × 5 cm3) and then dried under
vacuum (0.331 g, 0.47 mmol, 37% yield). Elemental analysis for
C20H21Cl4NP2W, found (calc.%): C 36.4 (36.2), H 2.9 (3.2), N
1.6 (2.1). 1H NMR (CD2Cl2): d 7.2–7.8 (m, 15H, Ph), 5.2 (s, 6H,
1
of 4: H NMR (CD2Cl2): d 6.80–7.70 (m, 40H, P(C6H5)3 and
−P(O)(C6H5)2), 2.85 (d, 3JH = 9 Hz, 3H, N C(CH3)P), 2.20
=
–
P
(s, 3H, MeCN). 31P{ H} NMR (CD2Cl2): d 6 (s, P(O)Ph2), −11
1
(s, PPh3). MS (ES): m/z 1024 [M − MeCN]. IR/cm−1 (KBr disk):
3414 (m, br), 3053 (m), 2918 (w), 1541 (m, v N), 1483 (m), 1434
=
C
(s), 1358 (w), 1189 (m), 1114 (m), 1093 (m), 999 (m), 771 (w), 748
(m), 721 (m), 696 (s), 585 (m), 560 (m), 541 (m), 519 (s), 500 (m).
=
Preparation of [ReBr2(MeCN)(PPh3)2(N C(CH3)P(O)Ph2)],
5. [ReOBr3(PPh3)2] (0.501 g, 0.52 mmol) was placed in a
Schlenk flask along with Ph2PN(SiMe3)2 (0.184 g, 0.53 mmol)
and PPh3 (0.301 g, 1.15 mmol). MeCN (15 cm3) was added
to give a dark yellow suspension and the mixture was then
heated under reflux for 2 h. The solution was filtered hot to
collect a green crystalline material, 5, which was washed with
MeCN (2 × 5 cm3) and then dried under vacuum (0.311 g,
0.27 mmol, 52% yield). The filtrate was left to cool and after
24 h a mixture of green and orange/brown crystals formed from
the solution (0.106 g). The green crystals were identified as the
product described previously (5), and the orange–brown crystals
1
Me). 31P{ H} NMR (CD2Cl2): d 37 (s, NPPh2), 28.6 (s, PMe2Ph).
IR/cm−1 (KBr disk): 3385 (s, br), 3060 (m), 2987 (m), 1730 (w),
1654 (w), 1636 (w), 1590 (w), 1541 (w), 1436 (s, vP Ph), 1416 (m),
–
1282 (m), 1125 (s, vW P), 943 (s), 908 (s), 743 (s), 728 (m), 693
–N–
1
(s), 549 (m), 483 (m).
as [Re(MeCN)Br3(PPh3)2]. Analysis of 5: H NMR (CDCl3):
d 6.86–7.85 (m, 40H, P(C6H5)3 and −P(O)(C6H5)2), 3.10 (d,
Preparation of [WCl5(NPPh2)][NBu4], 3. Ph2PN(SiMe3)2
(0.448 g, 1.30 mmol) was dissolved in Et2O (10 cm3) to give
a pale yellow solution. WCl6 (0.511 g, 1.29 mmol) was dissolved
in Et2O (10 cm3) to give an intense dark orange–brown solution.
The Ph2PN(SiMe3)2 solution was added to the WCl6 solution
and a colour change to intense dark blue was observed over a few
minutes. The solvent was removed under vacuum to leave a sticky
dark blue solid which was re-dissolved in CH2Cl2 (10 cm3). A
solution of NBu4Cl (0.354 g, 1.27 mmol, vacuum dried at 120 ◦C
for 48 h) in CH2Cl2 (5 cm3) was added to the reaction mixture and
1
3JH P = 10 Hz, 3H, N C(CH3)P), 2.28 (s, 3H, MeCN). 31P{ H}
=
=
=
NMR (CD2Cl2): d 4 (s, N C(CH3)P(O)Ph2), −17 (s, PPh3).
MS (FAB+): m/z 1033 [M − Br]+, 1074 [M − MeCN − Br)]+.
IR/cm−1 (KBr disk): 3052 (w), 1542 (m, v N), 1482 (m), 1434
=
C
(s), 1400 (m), 1361 (w), 1184 (w), 1114 (w), 1092 (m), 998 (w), 770
(w), 746 (m), 721 (w), 695 (s), 583 (w), 558 (w), 540 (w), 520 (s).
Analysis of [Re(MeCN)Br3(PPh3)2]. MS (FAB+): m/z 991
[M]+, 950 [M − MeCN]+, 871 [M − Br − MeCN)]+. IR/cm−1
(KBr disk): 3433 (br, w), 3056 (w), 1482 (m), 1434 (s), 1187 (w),
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3
6 8 1

View Article Online
1
5H, C5H5), 3.10 (s, 6H, N(CH3)2). 13C{ H} NMR (CDCl3): d
1090 (s), 1025 (w), 998 (w), 745 (s), 696 (s), 517 (s), 497 (m),
458 (w).
131.7 (d, 2JC P = 20 Hz, o-Ph), 129.7 (s, p-Ph), 128.9 (d, 3JC
=
–
–P
3
6 Hz, m-Ph), 118.1 (s, C5H5), 50.4 (d, JC = 10 Hz, NCH3).
–P
Preparation of 2-diphenylphosphino-1,1-dimethylhydrazine
(L2H). Ph2PNHNMe2 was prepared according to the
literature,23,24 producing a free flowing white solid (3.95 g,
31P{ H} NMR (CDCl3): d 55.8 (s). Selected IR/cm−1 (KBr disk):
1
2961 (s, br, vC H), 1434 (s, vP Ph), 742 (s, vP N).
–
–
–
1
Preparation of [RuCl2(g6-p-MeC6H4 Pr)(L2H)], 7. To a clear
i
16.2 mmol, 90% yield). H NMR (CD2Cl2): d 7.5–7.6 (m, 4H,
o-Ph), 7.3–7.5 (m, 6H, m- and p-Ph), 3.31 (d, 2JH P = 12 Hz, 1H,
colourless solution of L2H (0.5 g, 2.05 mmol) in THF (15 cm3),
–
1
NH), 2.46 (s, 6H, N(CH3)2). 13C{ H} NMR (CD2Cl2): d 132.2
6
i
[{RuCl(l-Cl)(g -p-MeC6H4 Pr)}2] (0.612 g, 1.0 mmol) was
added, producing a dark red suspension. The mixture was
heated under reflux for 1 h and then allowed to cool to room
temperature. The volume was reduced by half and the solution
cooled to 0 ◦C producing a red precipitate which was isolated by
filtration and washed with cold THF (5 cm3) and pentane (2 ×
5 cm3). It was then recrystallized from CH2Cl2–pentane, and
dried under vacuum (0.85 g, 1.5 mmol, 77% yield). Elemental
analysis for C24H31Cl2N2PRu, found (calc.%): C 52.8 (52.4), H
(d, 2JC P = 20 Hz, o-C), 129.0 (s, p-C), 128.4 (d, 3JC P = 5.2 Hz,
–
–
m-C), 51.1 (s, CH3). 31P{ H} NMR (CD2Cl2): d 38.6 (s). MS
1
(ES+): m/z 245 [L + H]+.
Preparation of 1-diphenylphosphinoamino-4-methylpiperazine
(L3H). To
a clear colourless solution of 1-amino-4-
methylpiperazine (1.72 g, 1.8 cm3, 15 mmol) and triethylamine
(1.5 g, 2.1 cm3, 15 mmol) in Et2O (15 cm3), chlorodiphenylphos-
phine (3.3 g, 2.7 cm3, 15 mmol) was added dropwise over 10 min
whilst cooling in an ice bath, producing a white precipitate im-
mediately. The mixture was slowly warmed to room temperature
and stirred for a further 1 h. The solid produced was removed by
filtration and washed with Et2O (2 × 15 cm3). The Et2O solutions
were combined, evaporated to dryness and dried under vacuum
producing a clear, colourless oil in quantitative yield. Elemental
analysis for C17H22N3P, found (calc.%): C 68.2 (68.2), H 7.9
1
5.7 (5.7), N 4.8 (5.1), Cl 12.8 (12.9). H NMR (CDCl3): d 7.92
3
(m, 4H, o-Ph), 7.3–7.5 (m, 6H, m- and p-Ph), 5.19 (d, JH
=
–H
5 Hz, 2H, CHC(Me)), 5.01 (d, JH = 5 Hz, 2H, CHC(iPr)),
3
–H
3.95 (d, 2JH P = 30 Hz, 1H, NH), 2.46 (septet, 3JH H = 7 Hz, 1H,
–
–
RCHMe2), 1.95 (s, 6H, N(CH3)2), 1.86 (s, 3H, CHC(CH3)), 0.78
(d, JH = 7 Hz, 6H, RCH(CH3)2). 13C{ H} NMR (CDCl3):
3
1
–H
1
2
d 134.4 (d, JC = 45 Hz, i-Ph), 133.8 (d, JC = 11 Hz, o-
–
P
–
P
4
3
1
Ph), 130.7 (d, JC = 2 Hz, p-Ph), 127.7 (d, JC = 10 Hz,
–P
–P
(7.4), N 13.7 (14.0). H NMR (CDCl3): d 7.45–7.6 (m, 4H, o-
m-Ph), 108.0 (s, C(iPr)), 94.8 (s, C(Me)), 90.6 (d, 2JC = 5 Hz,
Ph), 7.3–7.45 (m, 6H, m- and p-Ph), 3.35 (d, 2JH P = 10 Hz, 1H,
–P
–
CHCMe), 86.5 (d, JC = 6 Hz, CHC(iPr)), 49.5 (s, NCH3),
2
–P
NH), 2.3–3.0 (m, br, 8H, N(C2H4)2NCH3), 2.27 (s, 3H, NCH3).
1
30.0 (s, RCHMe2), 21.5 (s, RCH(CH3)2), 17.4 (s, CHC(CH3)).
13C{ H} NMR (CDCl3): d 131.9 (d, 2JC P = 20 Hz, o-Ph), 128.7
–
31P{ H} NMR (CDCl3): d 65.7 (s). MS (ES+): m/z 551.5 [M +
1
(s, p-Ph), 128.2 (d, 3JC = 7 Hz, m-Ph), 59.1 (d, 3JC = 4 Hz,
–P
–P
1
H]+. Selected IR/cm−1 (KBr disk): 3275 (m, sh, vN H), 1435 (s,
NHNCH2), 55.0 (s, CH2NCH3), 45.7 (s, NCH3). 31P{ H} NMR
(CDCl3): d 39.2 (s). Selected IR/cm−1 (pure oil): 3196 (w, br,
vN H), 3052 (m, vC Haromatic), 1585 (w, vN H), 1434 (s, vP Ph), 954
–
vP Ph), 747 (s, vP N).
–
–
Preparation of [RuCl2(g6-p-MeC6H4 Pr)(L3H)], 8. To a clear
i
–
–
–
–
(w, vN N), 743 (s, vP N).
colourless solution of L3H (0.40 g, 1.3 mmol) in THF (15 cm3),
–
–
6
i
Preparation of 1-diphenylphosphinoaminopiperidine (L4H).
To a clear colourless solution of 1-aminopiperidine (1.4 g,
1.5 cm3, 14 mmol) and triethylamine (1.5 g, 2.1 cm3, 15 mmol)
in Et2O (15 cm3), chlorodiphenylphosphine (3.1 g, 2.5 cm3,
14 mmol) was added dropwise over 10 min at −78 ◦C, producing
a white precipitate immediately. The mixture was then warmed
to room temperature over the period of 30 min and stirred for
a further 1 h. The solid produced was removed by filtration
and washed with Et2O (3 × 10 cm3). The Et2O solutions were
combined, evaporated to dryness and dried under vacuum at
80 ◦C producing a clear colourless oil in quantitative yield.
Elemental analysis for C17H21N2P, found (calc.%): C 72.0 (71.8),
[{RuCl(l-Cl)(g -p-MeC6H4 Pr)}2] (0.413 g, 0.65 mmol) was
added, producing adarkorange–brownsuspension. The mixture
was stirred at room temperature for 18 h, during which time a
precipitate formed. The solid was isolated by filtration, washed
with THF (2 × 5 cm3) and dried under vacuum, yielding a
red–brown microcrystalline solid (0.67 g, 1.1 mmol, 85% yield).
Elemental analysis for C27H36Cl2N3PRu, found (calc.%): C 53.7
(53.6), H 6.0 (6.0), N 6.9 (6.9). 1H NMR (CDCl3): d 7.85–
8.0 (m, 4H, o-Ph), 7.35–7.45 (m, 6H, m- and p-Ph), 5.22 (d,
3
3JH = 6 Hz, 2H, CHC(Me)), 5.03 (d, JH = 6 Hz, 2H,
–
H
–
H
CHC(iPr)), 4.10 (d, 2JH = 30 Hz, 1H, NH), 2.44 (septet,
–P
3JH H = 7 Hz, 1H, RCHMe2), 2.22 (m, 4H, N(CH2CH2)2NCH3),
–
1
1.99 (m, 7H, N(CH2CH2)2NCH3 and N(CH2CH2)2NCH3), 1.86
H 7.4 (7.4), N 9.5 (9.8). H NMR (CDCl3): d 7.35–7.45 (m,
3
4H, o-Ph), 7.2–7.3 (m, 6H, m- and p-Ph), 3.25 (d, 2JH
=
(s, 3H, CHC(CH3)), 0.78 (d, JH = 7 Hz, 6H, RCH(CH3)2).
–H
–
P
1
13C{ H} NMR (CDCl3): d 134.6 (d, 1JC = 56 Hz, i-Ph), 133.9
–
P
8 Hz, 1H, NH), 2.52 (m, 4H, N(CH2CH2)2CH2), 1.49 (m, 4H,
2
4
N(CH2CH2)2CH2), 1.23 (m, 2H, N(CH2CH2)2CH2). 13C{ H}
1
(d, JC = 11 Hz, o-Ph), 130.7 (d, JC = 2 Hz, p-Ph), 127.6
–
–
P
–P
3
(d, JC = 10 Hz, m-Ph), 108.0 (s, C(iPr)), 94.8 (s, C(Me)),
NMR (CDCl3): d 131.8 (d, 2JC = 20 Hz, o-Ph), 128.6
P
–P
90.5 (d, 2JC = 5 Hz, CHC(Me)), 86.6 (d, 2JC = 6 Hz,
(s, p-Ph), 128.1 (d, 3JC = 6 Hz, m-Ph), 60.9 (d, 3JC
=
–P
–P
–
P
–
P
CHC(iPr)), 57.1 (d, 3JC P = 3 Hz, N(CH2CH2)2NCH3), 54.8 and
–
4 Hz, N(CH2CH2)2CH2), 26.0 (s, N(CH2CH2)2CH2), 23.4 (s,
1
N(CH2CH2)2CH2). 31P{ H} NMR (CDCl3): d 35.6 (s). Selected
45.6 (s, N(CH2CH2)2NCH3 and N(CH2CH2)2NCH3), 29.9 (s,
RCHMe2), 21.4 (s, RCH(CH3)2), 17.4 (s, CHC(CH3)). 31P{ H}
1
IR/cm−1 (pure oil): 3224 (w, br, vN H), 3052 (m, vC Haromatic), 1586
–
–
NMR (CDCl3): d 67.9 (s). MS (ES+): m/z 605.6 [M + H]+.
(w, vN H), 1434 (s, vP Ph), 966 (w, vN N), 742 (s, vP N).
–
–
–
–
Selected IR/cm−1 (KBr disk): 3236 (m, sh, vN H), 1437 (s, vP Ph),
–
–
Preparation of [(C5H5)TiCl2(Ph2PNNMe2)], 6. To a clear
colourless solution of L2H (0.5 g, 2.05 mmol) in Et2O (15 cm3),
[(C5H5)TiCl3] (0.225 g, 1.03 mmol) was added, producing a
bright yellow solution and white precipitate immediately. The
mixture was stirred at room temperature for 2 h and then
cooled to −78 ◦C and stirred for a further 2 h. The white
precipitate was then removed by filtration at −18 ◦C, and
the bright yellow Et2O solution left to stand, producing large
bright orange crystals suitable for X-ray analysis. The remaining
Et2O solution was evaporated to dryness producing a bright
yellow–orange crystalline solid (0.35 g, 0.82 mmol, 80% yield).
Elemental analysis for C19H21Cl2N2PTi, found (calc.%): C 53.7
(53.4), H 5.3 (5.0), N 6.4 (6.6), Cl 16.4 (16.6). 1H NMR (CDCl3):
d 7.4–7.5 (m, 6H, m- and p-Ph), 7.3–7.4 (m, 4H, o-Ph), 6.19 (s,
743 (s, vP N).
–
Preparation of [RuCl2(g6-p-MeC6H4 Pr)(L4H)], 9. To a clear
i
colourless solution of L4H (0.5 g, 1.75 mmol) in THF (20 cm3),
6
i
[{RuCl(l-Cl)(g -p-MeC6H4 Pr)}2] (0.538 g, 0.88 mmol) was
added, producing adarkorange–brownsuspension. The mixture
was stirred at room temperature for 2 h, during which time a
precipitate formed. The volume of the solution was reduced
to ca. 5 cm3, and pentane (30 cm3) was added to complete
precipitation. The solid was isolated by filtration, washed with
pentane (2 × 5 cm3) and dried under vacuum, producing an
orange–red solid (0.97 g, 1.6 mmol, 91% yield). Elemental
analysis for C27H35Cl2N2PRu·CH2Cl2, found (calc.%): C 50.0
(49.8), H 5.8 (5.5), N 4.2 (4.2). 1H NMR (CDCl3): d 7.91
6 8 2
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3

View Article Online
Preparation of [RuCl(g6-p-MeC6H4 Pr)(L4H)]+[BF4]−, 12.
To a dark red–brown solution of 9 (0.20 g, 0.34 mmol) in
toluene (15 cm3), a benzene solution (15 cm3) of AgBF4 (0.058 g,
0.30 mmol) was added in the dark over the period of 5 min,
producing an orange precipitate immediately. The mixture was
stirred at room temperature for 15 min and then left to stand
for ca. 1 h. The solid was isolated by filtration, washed with
toluene (2 × 10 cm3), dissolved in CH2Cl2 (15 cm3) and left to
stand for a further 10 min. This suspension was then filtered
through Celite and the volume reduced under vacuum to ca.
2 cm3. Pentane was added to precipitate the product as a bright
orange microcrystalline solid, which was isolated by filtration
and dried under vacuum, (0.155 g, 0.24 mmol, 80% yield).
Elemental analysis for C27H35BClF4N2PRu, found (calc.%): C
50.6 (50.5), H 5.5 (5.5), N 4.3 (4.4). 1H NMR (CD2Cl2): d 7.5–
7.65 (m, 2H, m-Ph), 7.4–7.5 (m, 5H, o-, m- and p-Ph), 7.25–7.4
(m, 3H, o- and p-Ph), 7.01 (d, 2JH = 3 Hz, 1H, NH), 6.01 (d,
i
(m, 4H, o-Ph), 7.35–7.45 (m, 6H, m- and p-Ph), 5.20 (d,
3
3JH = 6 Hz, 2H, CHC(Me)), 5.00 (d, JH = 6 Hz, 2H,
–
H
–
H
CHC(iPr)), 4.00 (d, 2JH = 31 Hz, 1H, NH), 2.43 (septet,
–P
3JH = 7 Hz, 1H, RCHMe2), 2.13 (m, 4H, N(CH2CH2)2CH2),
–H
1.86 (s, 3H, CHC(CH3)), 1.06 (m, 6H, N(CH2CH2)2CH2 and
3
N(CH2CH2)2CH2), 0.79 (d, JH = 7 Hz, 6H, RCH(CH3)2).
–H
1
13C{ H} NMR (CDCl3): d 134.8 (d, 1JC = 56 Hz, i-Ph), 134.0
–
P
2
4
(d, JC = 10 Hz, o-Ph), 130.6 (d, JC = 2 Hz, p-Ph), 127.5
–
–
–
P
–P
(d, 3JC P = 10 Hz, m-Ph), 107.9 (s, C(iPr)), 94.8 (s, C(Me)), 90.4
(d, 2JC P = 5 Hz, CHC(Me)), 86.5 (d, 2JC P = 6 Hz, CHC(iPr)),
–
3
58.9 (d, JC = 3 Hz, N(CH2CH2)2CH2), 29.9 (s, RCHMe2),
–P
25.9 (s, N(CH2CH2)2CH2), 23.3 (s, N(CH2CH2)2CH2), 21.5 (s,
RCH(CH3)2), 17.4 (s, CHC(CH3)). 31P{ H} NMR (CDCl3): d
1
66.4 (s). MS (ES+): m/z 591 [M + H]+, 555 [M − Cl]+. Selected
IR/cm−1 (KBr disk): 3299 (m, sh, vN H), 1433 (s, vP Ph), 746 (s,
–
–
vP N).
–
–
P
3JH = 6 Hz, 1H, CHC(iPr)), 5.53 (pseudo-quartet, JH
=
3
–
H
–
H
Preparation of [RuCl(g6-p-MeC6H4 Pr)(L2H)]+[BF4]−, 10.
To a dark red–brown solution of 7 (0.25 g, 0.45 mmol) in
toluene (15 cm3), AgBF4 (0.092 g, 0.47 mmol) was added in
the dark, producing an orange precipitate almost immediately.
The mixture was stirred at room temperature for 30 min and
the solid was then isolated by filtration in the dark. The orange
solid was dissolved in CH2Cl2 (10 cm3) and the solution filtered
through Celite, the volume reduced to ca. 3 cm3 and Et2O
(20 cm3) added to precipitate a microcrystalline yellow–orange
solid (0.24 g, 0.40 mmol, 89% yield). Elemental analysis for
C24H31BClF4N2PRu, found (calc.%): C 47.5 (47.9), H 5.6 (5.2),
N 4.8 (4.7). 1H NMR (CD2Cl2): d 7.5–7.8 (m, 10H, Ph), 7.31 (d,
2JH = 3 Hz, 1H, NH), 6.06 (d, 3JH = 6 Hz, 1H, CHC(iPr)),
i
6 Hz and 3JH = 3 Hz, 1H, CHC(Me)), 5.32 (d, 3JH = 6 Hz,
–
P
–
H
1H, CHC(Me)), 4.96 (d, 3JH = 6 Hz, 1H, CHC(iPr)), 3.4–3.6
–
H
(m, 2H, N(CH2CH2)2CH2), 3.1–3.2 (m, 1H, N(CH2CH2)2CH2),
2.76 (m, 1H, N(CH2CH2)2CH2), 2.70 (septet, 3JH H = 7 Hz, 1H,
–
RCHMe2), 1.9–2.1 (m, 1H, N(CH2CH2)2CH2), 1.4–1.6 (m, 3H,
N(CH2CH2)2CH2), 1.34 (s, 3H, CHC(CH3)), 1.2–1.3 (m, 2H,
N(CH2CH2)2CH2), 1.20 (d, 3JH = 7 Hz, 3H, RCHCH3), 1.14
–
H
3
1
(d, JH = 7 Hz, 3H, RCHCH3). 13C{ H} NMR (CD2Cl2): d
–
H
1
4
132.9 (d, JC = 54 Hz, i-Ph), 132.1 (d, JC = 2 Hz, p-Ph),
–P
–
–
P
2
4
131.8 (d, JC = 12 Hz, o-Ph), 131.5 (d, JC = 2 Hz, p-Ph),
–P
P
3
3
130.8 (d, JC = 12 Hz, m-PH), 129.8 (d, JC = 11 Hz, m-
–P
–
P
2
2
Ph), 128.5 (d, JC = 12 Hz, o-Ph), 121.0 (d, JC = 6 Hz,
–P
–P
–
P
–
H
C(iPr)), 99.4 (s, C(Me)), 88.2 (s, CHC(Me)), 85.9 (d, JC
=
2
–
P
5.70 (d, 3JH = 6 Hz, 1H, CHC(Me)), 5.50 (d, 3JH = 6 Hz,
–H
–
H
9 Hz, CHC(iPr)), 83.5 (s, CHC(Me)), 80.0 (s, CHC(iPr)), 71.8
1H, CHC(Me)), 5.42 (d, JH = 6 Hz, 1H, CHC(iPr)), 3.60
3
–
H
3
3
(d, JC = 2 Hz, N(CH2CH2)2CH2), 62.2 (d, JC = 5 Hz,
–
P
–P
3
and 2.86 (s, 6H, N(CH3)2), 2.82 (septet, JH = 7 Hz, 1H,
–
H
3
N(CH2CH2)2CH2), 31.0 (s, RCHMe2), 24.1, 23.1 and 22.6 (s,
N(CH2CH2)2CH2 and N(CH2CH2)2CH2), 22.7 (s, RCH(CH3)2),
RCHMe2), 1.62 (s, 3H, CHC(CH3)), 1.31 (d, JH = 7 Hz,
–H
3H RCHCH3), 1.27 (d, 3JH = 7 Hz, 3H RCHCH3). 13C{ H}
1
–
H
20.7 (s, RCH(CH3)2), 16.9 (s, C(CH3)). 31P{ H} NMR (CD2Cl2):
1
NMR (CD2Cl2): d 129–132 (m, Ph), 119.1 (s, C(iPr), 99.2 (s,
CHC(Me)), 88.4 (s, CHC(Me)), 86.8 (s, CHC(iPr)), 84.1 (s,
CHC(iPr)), 83.0 (s, CHC(Me)), 64.7 and 55.8 (s, N(CH3)2), 31.4
(s, RCHMe2), 22.7 and 21.3 (s, RCH(CH3)2), 17.3 (s, C(CH3)).
d 45.9. MS (ES+): m/z 555 [M]+. Selected IR/cm−1 (KBr disk):
3300 (m, sh, vN H), 1436 (s, vP Ph), 745 (s, vP N).
–
–
–
Preparation of [(g6-p-MeC6H4 Pr)Cl2Ru(l2,g3-PPh2NNMe2)
Ti(C5H5)Cl2], 13. To a bright yellow–orange solution of 6
i
1
31P{ H} NMR (CD2Cl2): d 43.7 (s). MS (ES+): m/z 515.02 [M]+.
Selected IR/cm−1 (KBr disk): 3279 (m, sh, vN H), 1437 (s, sh,
(0.20 g, 0.47 mmol) in THF (10 cm3), [{RuCl(l-Cl)(g -p-
6
–
i
vP Ph), 748 (m, sh, vP N).
MeC6H4 Pr)}2] (0.14 g, 0.228 mmol) was added and the mixture
–
–
stirred at room temperature for 90 min during which time
the solution became dark orange–brown. The volume was
then reduced to ca. 3 cm3 and Et2O (45 cm3) added to cause
precipitation. The dark orange solid produced was isolated
by filtration, washed with Et2O (2 × 5 cm3) and dried under
vacuum (0.20 g, 0.27 mmol, 61% yield). Elemental analysis
for C29H35Cl4N2PRuTi, found (calc.%): C 47.4 (47.5), H 4.9
Preparation of [RuCl(g6-p-MeC6H4 Pr)(LL3H)]+[BF4]−, 11.
To a dark red–brown solution of 8 (0.27 g, 0.45 mmol) in
toluene (15 cm3), a benzene solution (15 cm3) of AgBF4 (0.092 g,
0.47 mmol) was added in the dark over the period of 5 min,
producing an orange precipitate immediately. The mixture was
stirred at room temperature for 30 min and then left to stand for
a further 1 h. The solid was isolated by filtration, washed with
warm toluene (2 × 15 cm3), dissolved in CH2Cl2 (10 cm3) and
left to stand for a further 5 min. This suspension was filtered
through Celite and the volume reduced under vacuum to ca.
2 cm3. Pentane was added to precipitate the product as an ochre
coloured solid, which was isolated by filtration and dried under
vacuum (0.27 g, 0.41 mmol, 93% yield). Elemental analysis for
C27H36BClF4N3PRu, found (calc.%): C 49.1 (49.4), H 5.6 (5.5),
N 6.2 (6.4), Cl 5.0 (5.4). 1H NMR (CD2Cl2): d 7.3–7.6 (m, 10H,
Ph), 7.09 (d, 2JH P = 3 Hz, 1H, NH), 5.96 (d, 3JH H = 6 Hz, 1H,
i
1
(4.8), N 3.8 (3.8), Cl 19.3 (19.3). H NMR (CD2Cl2): d 8.19
(m, 4H, o-Ph), 7.48 (m, 6H, m- and p-Ph), 6.51 (s, 5H, C5H5),
5.06 (d, JH = 6 Hz, 2H, CHC(iPr)), 4.72 (d, JH = 6 Hz,
3
3
–
H
–H
3
2H, CHC(Me)), 2.81 (s, 6H, N(CH3)2), 2.51 (septet, JH
=
–H
7 Hz, 1H, RCHMe2), 1.63 (s, 3H, CHC(CH3)) 1.12 (d, 3JH
=
–
H
7 Hz, 6H, RCH(CH3)2). 13C{ H} NMR (CD2Cl2): d 134.3 (d,
1
2JC = 12 Hz, o-Ph), 131.5 (d, JC = 2 Hz, p-Ph), 128.1 (d,
4
–P
–
P
3JC = 10 Hz, m-Ph), 120.7 (s, C5H5), 112.1 (s, C(iPr)), 98.1
–P
(s, C(Me)), 89.0 (d, 2JC = 4 Hz, CHC(Me)), 88.0 (d, 2JC
=
–P
–
P
6 Hz, CHC(iPr)), 54.1 (s, NCH3), 30.2 (s, RCHMe2), 22.0 (s,
–
–
1
RCH(CH3)2), 17.5 (s, CHC(CH3)). 31P{ H} NMR (CD2Cl2): d
6
i
6
i
g -p-MeC6H4 Pr), 5.45 (d, 3JH H = 7 Hz, 1H, g -p-MeC6H4 Pr),
–
86.3 (s). Selected IR/cm−1 (KBr disk): 3051 (m, vC Haromatic), 2962
–
3
6
i
3
5.37 (d, JH = 6 Hz, 1H, g -p-MeC6H4 Pr), 5.07 (d, JH
=
–
H
6
–
H
(m, vC Haliphatic), 1436 (s, vP Ph), 739 (s, vP N).
–
–
–
i
7 Hz, 1H, g -p-MeC6H4 Pr), 3.65 (m, 1H, N(C2H4)2NCH3),
3.17 (m, 1H, N(C2H4)2NCH3), 2.85 (m, 1H, N(C2H4)2NCH3),
Molecular modelling
3
2.76 (septet, JH = 7 Hz, 1H, RCHMe2), 2.2–2.6 (m, 5H,
–
H
N(C2H4)2NCH3), 2.25 (s, 3H, NCH3), 1.42 (s, 3H, CHC(CH3)),
All calculations were carried out using the Gaussian 98W suite
of programs.27 Becke’s three parameter hybrid functional using
the correlation functional of Lee, Yang and Parr (B3LYP)28
was used in conjunction with the standard 6-31g(d)29–34 basis
set for all non-metal atoms, and the LanL2DZ35–37 basis set
3
3
1.22 (d, JH = 7 Hz, 3H, RCHCH3), 1.18 (d, JH = 7 Hz,
–
H
–H
3H, RCHCH3). 31P{ H} NMR (CD2Cl2): d 46.9 (s). MS (ES+):
1
m/z 570 [M]+. Selected IR/cm−1 (KBr disk): 3303 (m, br, vN H),
–
1437 (s, sh, vP Ph), 751 (m, sh, vP N).
–
–
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3
6 8 3

View Article Online
for W of complex 3 and 6-311(g)d38–42 for complex 6. The
structures were optimised and the presence of a minimum on
the potential energy surface was confirmed by a vibrational
frequency calculation. The basis sets were chosen in order to
provide a trade off between accuracy and computational power
required.
Crystallography
X-Ray crystal structures were determined by mounting a
single crystal encased in perfluoropolyether oil on a glass fibre
and then cooling rapidly to 150 K in a stream of cold N2
using an Oxford Cryosystems CRYOSTREAM unit. Diffrac-
tion data were measured using an Enraf-Nonius KappaCCD
diffractometer (graphite-monochromated Mo-Karadiation, k =
˚
0.71073 A). Intensity data were processed using the DENZO-
SMN package.43 The structures were solved using the direct-
methods program SIR92,44 which located all non-hydrogen
atoms of the complexes. Subsequent full-matrix least-squares
refinement was carried out using the CRYSTALS program
suite.45 Hydrogen atoms were positioned geometrically after
each cycle of refinement. A three-term Chebychev polynomial
weighting scheme was applied. Details of the crystal data and
refinement can be found in Tables 1 and 2.
CCDC reference numbers 255284–255294.
See http://www.rsc.org/suppdata/dt/b4/b417068g/ for cry-
stallographic data in CIF or other electronic format.
Results and discussion
Tungsten phosphazene complexes
WCl6 reacted rapidly with Ph2PN(SiMe3)2 in Et2O to give a
dark blue species that analysed as WCl4(NPPh2) 1, the reaction
proceeding with the elimination of two molecules of ClSiMe3.
1
Complex 1 showed a single peak at 68 ppm in the 31P{ H}
NMR spectrum, which is consistent with the P(III)-containing
structure proposed. (cf. the free ligand at 50.2 ppm).46 It seems
probable that the complex exists as a Cl-bridged dimer to achieve
octahedral coordination of the tungsten centres (Fig. 1(A)), as
in the analogous isodiazene species, [{WCl4(NNPh2)}2].47 The
Ph2PN complex 1 proved to be very oxygen sensitive in both
solution and in the solid state. The IR spectrum showed a
strong peak at 1439 cm−1 which is attributed to a vP Ph stretching
–
frequency.24 There is a second strong peak at 1131 cm−1 which
is attributed to a stretching frequency associated with the W–
N–P unit.46 The general similarity of the IR spectra to that
observed for complexes 2 and 3 (discussed below) suggests that
the complex is not a polymer linked together via the Ph2P groups
(Fig. 1(B)), although in the absence of a structure this possibility
cannot be eliminated entirely.
Fig. 1 Proposed structure of [WCl4(NPPh2)] (A) and possible poly-
meric structure of [WCl4(NPPh2)] complex (B).
It was hoped that if an additional ligand could be introduced
into the coordination sphere the stability of the complexes could
be increased. The reaction of 1 with PMe2Ph gave a dark red
complex 2 (Fig. 2) which had an elemental analysis consistent
1
with the formulation [WCl4(NPPh2)(PMe2Ph)]. The H NMR
(CD2Cl2) of this complex showed the expected protons in the
aromatic region as a multiplet between 7.2 and 7.8 ppm. The
31P NMR (CD2Cl2) showed two singlets, one at 37 ppm which
6 8 4
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3

View Article Online
6
i
Table 2 Crystallographic data for the X-ray crystal structures of 8, 9, 10, 12 and [RuCl2(g -p-MeC6H4 Pr)(PPh2NHNHMe2)]+Cl−
6
i
[RuCl2(g -p-MeC6H4 Pr)
Compound
8a,b,c
9a
10a
12a
(L2H2)]+Cl−·THFa
Chemical formula
Formula weight
Crystal system
Space group
C28H38Cl4N3PRu
690.48
C28H37Cl4N2PRu
675.47
Monoclinic
P21/n
10.0630(2)
14.6827(3)
20.0092(5)
91.8167(9)
2954.91(11)
4
C24H31BClF4N2PRu
601.82
Orthorhombic
Pna21
14.7058(5)
16.8368(5)
10.5355(4)
90
2608.57(15)
4
C28H37BCl3F4N2PRu
726.82
Monoclinic
C2/c
29.8867(4)
10.3937(2)
23.4925(4)
121.4595(7)
6224.88(19)
8
C28H40Cl3N2OPRu
659.04
Monoclinic
P21/c
9.5016(2)
20.4988(3)
15.5874(2)
95.2880(7)
3023.06(9)
4
Orthorhombic
P212121
7.4308(2)
13.7935(2)
29.5131(4)
90
3025.00(10)
4
0.948
˚
a/A
˚
b/A
˚
c/A
b/◦
3
˚
V/A
Z
l/mm−1
0.967
0.809
0.859
0.860
Reflections measured
Rint
Goodness of fit
R
wR [I > 3r(I)]
23307
0.046
1.0803
0.0398
22074
0.054
1.1084
0.0308
18147
0.054
1.1040
0.0343
31308
0.055
1.1179
0.0359
34647
0.034
1.0780
0.0282
0.0438
0.0340
0.0367
0.0409
0.0321
a The NH hydrogen atom was located in a difference Fourier map and its coordinates and isotropic thermal parameter subsequently refined. Other
hydrogen atoms were positioned geometrically after each cycle of refinement. b The refined thermal parameters of the C and Cl atoms of the
solvent were extremely large and highly anisotropic, indicative of disorder. This was modelled as disorder over two nearly-overlapping orientations.
Coordinates, isotropic thermal parameters and site occupancies were refined for the◦disordered C and Cl atoms. Geometric restraints were applied:
the C–Cl bond lengths were restrained to 1.78(2) A and the Cl–C–Cl angles to 112(2) . c Refinement of the Flack enantiopole parameter gave a value
˚
of 0.05(4), indicating that the crystal consisted predominantly of a single enantiomer73 despite the lack of inherent chirality of the complex.
Unfortunately no X-ray suitable crystals of the tungsten
complexes could be obtained. Thus in order to gain some insight
into the bonding in these complexes Density Functional (DF)
calculations were carried out on 3 using the Gaussian 98W suite
of programs.27 Fig. 3 shows representations of the HOMO and
LUMO orbitals obtained from the calculations while Fig. 4
shows the 2nd and 3rd highest occupied orbitals.
Fig. 2 Reaction scheme for the formation of the tungsten complexes.
is attributed to the NPPh2 ligand and the second at 29 ppm
which corresponds to the PMe2Ph ligand. P–P coupling is not
1
observed in the 31P{ H} NMR spectrum. The IR spectrum of the
complex showed two distinctive bands analogous to those seen
for 1, slightly shifted to lower frequencies at 1436 and 1125 cm−1.
Despite repeated attempts in a variety of media, no mass spectra
could be obtained using ES, MALDI or FAB techniques.
Reaction of 1 with [NBu4]Cl in dry dichloromethane (Fig. 2)
resulted in the formation of a dark purple complex, 3, with
Fig. 3 DFT model of 3 (HOMO on the left and LUMO on the right).
an elemental analysis consistent with [WCl5(NPPh2)][NBu4]. As
observed for the other tungsten complexes, two bands were
observed in the IR at 1438 cm−1 and 1127 cm−1. Despite
1
repeated attempts, no 31P{ H} NMR signals could be observed
and all 1H NMR spectra showed very broad peaks. This
may be due to a paramagnetic tungsten(V) impurity such
as [WCl5(NHPPh2)][NBu4] which arises from reduction and
1
protonation of 1. Attempts to obtain 13C{ H} NMR and to
carry out further reactions on these complexes was thwarted by
their instability in solution.
Reaction of WCl6 with more than one equivalent of the
Ph2PN(SiMe3)2 ligand only resulted in the formation of 1 with
no evidence for a bis complex analogous to those known for the
isodiazene system with tungsten (e.g. [WCl2(NNR2)2L2] where
R = Ph and L = 1,2-dimethoxyethane).48
t
Several variants of the R2PN(SiMe3)2 proligand (R = Bu,
Fig. 4 DFT model of 3 (2nd highest occupied orbital on the left and
iPr) were prepared. The complexation reactions showed the same
3rd highest occupied orbital on the right).
dark blue colouration on initial reaction with WCl6, but rapid de-
˚
The calculations suggested a W–N bond distance of 1.756 A
cay to dark brown and then purple/blue solutions then occurred,
and no meaningful analysis was obtained. P(V) analogues of
the type Ph2P(X)N(SiMe3)2 (X = O or S) were also prepared
but again reactions with the tungsten starting materials gave a
mixture of products which could not be separated or analysed.
˚
and a N–P bond length of 1.701 A which are typical lengths for
49,46
=
a W N double bond and a P–N single bond, respectively.
The Cl–W–N–P unit was predicted to be virtually linear with
an N–W–Cl bond angle of 179.83◦, and a P–N–W angle of
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3
6 8 5

View Article Online
171.77◦. The compound has a plane of symmetry running
through P–N–W and the Ph2PN ligand shows evidence of a
strong trans influence as the W–Cl bond trans to the ligand
◦
˚
Table 3 Selected bond lengths (A) and angles ( ) for 4
Re(1)–Cl(1)
Re(1)–Cl(2)
Re(1)–N(1)
Re(1)–N(2)
Re(1)–P(2)
Re(1)–P(3)
2.4637(19) N(1)–C(1)
2.4123(18) C(1)–C(2)
1.27(1)
1.552(11)
1.801(8)
1.489(6)
1.80(1)
˚
has a calculated bond length 3.5% longer (2.486 A) than the
1.809(7)
2.049(6)
2.434(2)
C(1)–P(1)
P(1)–O(1)
P(1)–C(3)
˚
average W–Cl bond length in the equatorial plane (2.402 A).
The HOMO and LUMO orbitals have no contribution from the
phosphazene ligand and involve non-bonding electron pairs on
the halogen ligands. The second highest occupied orbital showed
the P(III) lone pair and an anti-bonding W–Cl orbital trans to the
phosphazene ligand. The third highest MO clearly shows the p-
bonding component of the W–N multiple bond, consistent with
2.4284(19) P(1)–C(9)
1.806(9)
Cl(1)–Re(1)–Cl(2)
Cl(1)–Re(1)–N(1)
Cl(2)–Re(1)–N(1)
Cl(1)–Re(1)–N(2)
Cl(2)–Re(1)–N(2)
N(1)–Re(1)–N(2)
Cl(1)–Re(1)–P(2)
90.88(7)
170.3(2)
98.6(2)
N(1)–Re(1)–P(2)
Cl(1)–Re(1)–P(3)
N(1)–Re(1)–P(3)
91.8(2)
91.06(7)
91.0(2)
177.13(7)
176.6(6)
120.1(7)
123.1(6)
79.01(18) P(2)–Re(1)–P(3)
169.80(18) Re(1)–N(1)–C(1)
˚
the short W–N distance of 1.756 A. There was also evidence for
P–C antibonding orbitals on the P(III) atom, but these were not
extended towards the nitrogen and there was no evidence for any
P–N multiple bonding. This may well account for the observed
instability of these complexes.
91.5(3)
N(1)–C(1)–C(2)
C(2)–C(1)–P(1)
86.08(6)
=
[Re(OH)(N CMe2)(dppe)2][HSO4], for which Re–N = 1.901(5)
and Re–N–C = 178.9(5)◦).52 The C(1)–N(1) bond length is also
Rhenium phosphorylketimido complexes
˚
consistent with a double bond at 1.27(1) A. The carbon C(1)
is sp2 hybridised with an N(1)–C(1)–P(1) angle of 116.8◦ and
Reaction of the Ph2PN(SiMe3)2 proligand with [ReOCl3(PPh3)2]
in MeCN under reflux resulted in the formation of a dark
green crystalline material 4 in 10% yield on cooling (method
A). Addition of two equivalents of PPh3 to the reaction mixture
improved the yield of the green crystalline material to around
40%. Further material could be obtained from the filtrate
but was contaminated with the other orange product of this
reaction which is identified from its X-ray crystal structure as
the known complex [ReCl3(MeCN)(PPh3)2].50,51 The analogous
reaction of the proligand with [ReOBr3(PPh3)2] gave 5 and
[ReBr3(MeCN)(PPh3)2].
an N(1)–C(1)–C(2) angle of 120.9◦. These bond distances and
angles are similar to those seen in other ketimido complexes
{e.g. [(C5H5)(Bu 2C N)TiMe2]}53,54 and the distances and angles
t
=
around the phosphorus are typical for a phosphine oxide.55 The
bromo analogue 5 was found to be essentially isostructural and
its details are not discussed here.
Electrospray mass spectrometry of 4 in CH2Cl2 showed an
ion of 100% relative abundance at m/z 1024, which corresponds
to the species formed by protonation and loss of MeCN. In
the case of the bromide complex, 5, the FAB+ mass spectrum
showed a peak at m/z 1033 (100%) and one at 1074 (40%)
which corresponds to the cations formed by the loss of both
The X-ray crystal structure of 4
Br− and MeCN or only Br−, respectively. The 31P{ H} NMR
1
The crystals deposited directly from the reaction mixture proved
suitable for an X-ray structure determination (details in Table 1).
An ORTEP representation of the structure appears in Fig. 5
and selected bond lengths and angles appear in Table 3. The
geometry about the rhenium centre is pseudo-octahedral with
spectrum (in CD2Cl2) of 4 shows two singlets at −11 and
+6 ppm. The resonance at −11 ppm is due to the two equivalent
PPh3 ligands (cf. 31P NMR for [Re(NPh)Cl3(PPh3)2] at −19 ppm
in CDCl3)56,57 and the resonance at 6 ppm is due to the
P(V) in the NC(CH3)P(O)Ph2 ligand [cf. Ph2PN(SiMe3)2 at
=
previously unknown phosphorylketimido ligand –N C(CH3)–
50.2 ppm].22 The 31P{ H} NMR spectrum of 5 shows the same
1
P(O)Ph2) bound trans to one chloride and the two triphenylphos-
phine ligands bound trans to each other. The Re–Cl distances are
virtually identical and suggest that the ketimide ligand exerts a
similar trans influence to that of MeCN. The length of the Re(1)–
peaks slightly shifted to −17 and +4 ppm assigned as before.
The IR spectra of complexes 4 and 5, showed bands due to
v
C
at 1541 cm−1 and 1542 cm−1, respectively.58 It was not
=
N
possible to obtain elemental analysis of either complex 4 or
5 due to the co-crystallization with [ReCl3(MeCN)(PPh3)2] or
[ReBr3(MeCN)(PPh3)2], respectively.
˚
N(1) bond at 1.809(7) A, combined with the Re(1)–N(1)–C(1)
bond angle of 176.6◦ is consistent with a double bond (cf . trans-
A proposed mechanism for the formation of complexes 4
and 5 is shown in Fig. 6. Acetonitrile appears essential as
Fig. 5 ORTEP view (40% probability ellipsoids) of the X-ray crystal
structure of 4 with hydrogen atoms omitted for clarity.
Fig. 6 Proposed mechanism for the formation of the phosphoryke-
timide ligand.
6 8 6
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3

View Article Online
the use of other solvents (THF or toluene) led to brown
solutions, and no clean products could be isolated. It is known
that [ReOCl3(PPh3)2] reacts irreversibly with acetonitrile and
triphenylphosphine to give [ReCl3(MeCN)(PPh3)2] and this
appears to be in competition with the formation of 4 since the
Re(III) acetonitrile complex does not react with Ph2PN(SiMe3)2
to give complex 4. The chloride trans to the oxo group is more
labile due to the higher trans effect of the oxo group and will be
readily replaced through elimination of trimethylchlorosilane to
gives species A.
Reduction of the rhenium–oxo group by triphenylphosphine
and coordination of MeCN to the coordinatively unsaturated
Re(III) species formed produces B. The acetonitrile ligand
in [ReCl3(MeCN)(PPh3)2] is known to be susceptible to nu-
cleophilic attack, and so it is possible that intramolecular
nucleophilic attack by P at the nitrile carbon can then occur
as shown in species B, giving C.59,60 However, this requires that
the N(SiMe3)PPh2 ligand be cis to the nitrile ligand, otherwise
intramolecular attack is not possible. Finally, hydrolysis of the
negatively charged phosphinamine group gives species D.
Attempts to verify the proposed mechanism by a variety
of spectroscopic techniques was not successful, furthermore,
the reaction of one of the possible hydrolysis products of the
ligand, Ph2P(O)H, with [ReCl3(MeCN)(PPh3)2] does not yield
4. However, the mechanism is consistent with the observation of
two products and the known chemistry of these systems.
hexane. The 1H NMR spectra (in CD2Cl2 or CDCl3) of L2H, L3H
and L4H showed a doublet at 3.31 (2JH P = 12 Hz), 3.35 (2JH
=
–
–
P
10 Hz) and 3.25 ppm (2JH = 8 Hz), respectively, attributed to
–P
NH. The 31P{ H} NMR spectra of the three ligands showed a
1
single resonance at 38.6, 39.2 and 35.6 ppm.
The 31C{ H} NMR spectra were unambiguously assigned
1
using 1H–13C HMQC and 1H–1H COSY experiments and
showed doublets in the aromatic region due to 13C–31P coupling,
with coupling constants similar to those previously reported
for Ph3P. 62 The N(CH3)2 carbon resonances are also split due
to 13C–31P coupling, with a coupling constant of about 4 Hz
as expected for a three-bond aliphatic coupling.63 The ipso-
carbon resonances were not observed. It was not possible to
obtain a mass spectrum of either L3H or L4H, however the ES+
MS for L2H showed [L + H]+ as the major peak and suitable
elemental analyses were obtained for all three. The IR spectrum
of L2H was published by Sisler et al. showing strong peaks
at 1432 cm−1 (vP Ph), 1583 cm−1 (vN H) and 740 cm−1 (vN P).
–
–
–
Unsurprisingly, the new ligands L3H and L4H have very similar
IR spectra showing the same characteristic peaks.
Recrystallisation of L2H from hot hexane produced crystals
suitable for X-ray analysis (Fig. 9, Tables 1 and 4). The NH group
projects towards the terminal N of a neighbouring molecule,
ꢀ
˚
but the N · · · N distance (N(1) · · · N(2) = 3.396(2) A, symmetry
operator 1/2 − x, −1/2 + y, z) is greater than that generally
associated with N · · · N hydrogen bonds. The coordination
geometry around N(1) is considerably flatter than is normal
for a secondary amine (P(1)–N(1)–N(2) = 115.50(10)◦), which,
Phosphinohydrazine and phosphinohydrazido ligands
˚
together with the short N–P bond length (1.6815(13) A),
All attempts to prepare early and middle transition metal
complexes of phosphazenido (R2P–NRꢀ, R = Ph, Rꢀ = Ph, Me,
cyclohexyl) ligands were unsuccessful in our hands. Reaction of
R2PNHR with appropriate metal halides in the presence of a
range of bases at various temperatures and in several different
solvents gave intractable mixtures. Complexes of [Ph2P–N–
PPh2] with Ti(IV) and Zr(IV) are known19,20 and we investigated
the effect of replacing one PPh2 with an NR2 group on the
coordination geometry of the complexes, with the possibility of
P–N, or N–N donation and the formation of three- or four-
membered chelate rings (Fig. 7).
suggests that there is an interaction between the N lone pair
and the P atom giving some multiple bond character.
Fig. 7 Possible phosphinohydrazido ligand binding modes.
Ph2PNHNMe2 was first produced by Sisler et al. in 1966,
and subsequently used as a neutral phosphorus donor with
several metals.23,24 The synthesis involves reaction of PPh2Cl
with two equivalents of NH2NMe2 producing the desired ligand
PPh2NHNMe2, L2H (Fig. 8) and the HCl salt of the hydrazine,
(the use of Et3N as a base yields only (PPh2)2NNMe2 and the HCl
salt of the hydrazine, presumably due to the lower solubility in
Et2O of the hydrazine hydrochloride compared to [Et3NH]+Cl−).
In contrast the new ligands L3H and L4H (Fig. 8) were both
prepared through reaction of one equivalent of each of the
respective hydrazine, Ph2PCl and Et3N in Et2O. On removal
of the Et3NH+Cl− by filtration and evaporation of the solution
to dryness, the two ligands were obtained pure as colourless
oils in quantitative yield. They are all soluble and stable in dry,
degassed toluene, benzene, CH2Cl2, CHCl3, Et2O, pentane and
Fig. 9 ORTEP view (40% probability ellipsoids) of the X-ray crystal
structure of L2H with geometrically positioned hydrogen atoms omitted
for clarity.
Phosphorus(III) hydrazine and hydrazide complexes
In the limited number of complexes published containing the
ligand L2H,64–66 none have contained it as an anionic ligand.
However, reaction of two equivalents of L2H with [(C5H5)TiCl3]
in Et2O produced [Ph2PNHNHMe2]+Cl− and the relatively air-
and moisture-stable complex [(C5H5)TiCl2(Ph2PNNMe2)] 6 in
high yield. Complete separation of the HCl salt of the ligand and
the complex proved difficult as they have very similar solubility
in a wide range of solvents, however, filtering the Et2O solution
at −18 ◦C proved successful.
1
The 31P{ H} NMR spectrum of 6 (CDCl3) showed a single
resonance at 55.8 ppm. In the 1H NMR spectrum, the NH peak
of the free ligand (observed at 3.31 ppm) is missing and the peak
at 2.46 ppm, assigned to N(CH3)2 has shifted to 3.10 ppm. The
C5H5 resonance has also shifted, from 7.0 ppm67 in [(C5H5)TiCl3]
to 6.19 in the complex 6. As in the free ligand, the resonances
1
corresponding to the ortho- and meta-carbons in the 31C{ H}
Fig. 8 Phosphinohydrazine ligands.
NMR spectrum of the complex were split due coupling to 31P
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3
6 8 7

View Article Online
◦
2
6
i
2
+
−
˚
Table 4 Selected bond lengths (A) and angles ( ) for L H, 7, 8, 9, 10, 12 and [RuCl2(g -p-MeC6H4 Pr)(L H2)] Cl
6
i
L2H
7
8
9
10
12
[RuCl2(g -p-MeC6H4 Pr) (L2H2)]+Cl−
Ru(1)–P(1)
Ru(1)–Cl(1)
Ru(1)–Cl(2)
Ru(1)–cymenea
P(1)–N(1)
N(1)–N(2)
N(2)–C(1)
N(2)–C(X)
N(1)–H(1)
N(2)–H(2)
—
—
—
—
2.3238(8)
2.4122(9)
2.4103(9)
1.69
1.673(3)
1.433(4)
1.457(5)
1.466(5)b
0.80(4)
—
2.3222(12)
2.4235(11)
2.4097(11)
1.70
1.691(4)
1.432(5)
1.453(6)
1.479(6)c
0.86(6)
2.3425(8)
2.4167(8)
2.4120(8)
1.70
1.664(3)
1.434(4)
1.467(4)
1.473(4)d
0.86(5)
—
2.3133(9)
2.3918(11)
2.221(4)e
1.72
1.658(4)
1.457(5)
1.480(6)
1.484(6)b
0.89(6)
—
2.3109(9)
2.3901(9)
2.227(3)e
1.73
1.665(3)
1.444(4)
1.505(4)
1.508(5)d
0.81(5)
—
2.3379(5)
2.4215(5)
2.4163(5)
1.71
1.7017(18)
1.443(2)
1.495(3)
1.494(3)a
0.90(3)
1.6815(13)
1.4388(17)
1.455(2)
1.457(2)b
0.88(2)
—
—
0.90(3)
Ru(1)–P(1)–N(1)
Ru(1)–N(2)–N(1)
P(1)–Ru(1)–N(2)
P(1)–N(1)–N(2)
N(2)–N(1)–H(1)
N(1)–N(2)–C(1)
N(1)–N(2)–C(X)
C(1)–N(1)–C(X)
—
—
—
112.96(11)
—
—
116.9(2)
112(3)
109.2(3)
110.3(3)b
110.8(3)b
112.09(15)
—
—
117.0(3)
114(4)
110.4(4)
109.1(3)c
108.9(4)c
111.30(11)
—
—
117.7(2)
117(3)
110.3(2)
108.8(2)d
111.2(2)d
88.24(13)
97.2(2)
66.79(10)
106.4(3)
119(3)
108.3(4)
109.0(3)b
108.4(4)b
88.48(12)
97.67(19)
66.75(8)
106.7(2)
123(4)
111.4(3)
109.8(3)d
108.0(3)d
114.58(7)
—
—
119.60(14)
108.2(18)
113.54(18)
108.18(18)b
111.64(18)b
115.50(10)
114.1(13)
107.98(12)
111.17(12)b
110.90(14)b
a Distance from Ru(1) to the best plane of the C6 ring. b X = 2. c X = 3. d X = 5. e Length of Ru(1)–N(2) bond, not Ru(1)–Cl(2).
(2JC P = 20 and 3JC P = 6 Hz, respectively, which are consistent
–
–
with literature values).62 The N(CH3)2 carbons appeared as a
doublet due to three-bond coupling to 31P, and not due to
the slight asymmetry seen in the crystal structure. This was
1
confirmed by running the 13C{ H} NMR spectrum at both
300 and 500 MHz, with both spectra showing a peak to peak
separation of 10.4 Hz, which is consistent with 31P coupling and
not magnetic inequivalence.
Despite numerous attempts, it was not possible to obtain a
mass spectrum of the complex. However, on standing, the Et2O
solution produced a large number of orange crystals suitable
for X-ray crystallography (Fig. 10). The structure is similar to
the previously reported [(C5H5)TiCl2(NMe2NMe)],68 (Table 5)
however, the two Ti–N bond lengths are slightly less asymmetric,
˚
˚
with Ti–NMe2 at 2.1689(15) A and Ti–NP at 1.9013(14) A [cf.
˚
2.217 and 1.849 A, respectively, for [(C5H5)TiCl2(NMe2NMe)].
Fig. 10 ORTEP view (40% probability ellipsoids) of the X-ray crystal
structure of 6 with hydrogen atoms omitted for clarity.
The N–N bond shows no multiple bonding character (bond
˚
length = 1.4337(19) A, and similarly, the P–N bond length
69
The NH peak, observed at 1583 cm−1 in the IR spectrum of the
free ligand is missing as expected.
Density functional (DF) calculations were carried out on 6 in
order to compare the optimised geometry with the X-ray crystal
structure, and to confirm the accuracy of the model chosen
˚
(1.7319(14) A) is characteristic of a single bond. As expected,
the soft P lone pair does not coordinate to the hard Ti centre,
and the lone pair is directed away from the Ti metal centre. The
IR spectrum of the complex showed bands at 1434 cm−1 (s, vP
)
–
Ph
and 742 cm−1 (s, vP N), similar to the spectrum of the free ligand.
–
◦
68
˚
Table 5 Selected bond lengths (A) and angles ( ) for 6, the DF optimised structure of 6 and (Cp)TiCl2(NMe2NMe).
6
DFT Calc.
Da
[(Cp)TiCl2(NMe2NMe)]
Ti(1)–N(1)
Ti(1)–N(2)
Ti(1)–Cl(1)
Ti(1)–Cl(2)
Ti(1)–(C5H5)d
N(1)–N(2)
P(1)–N(1)
1.9013(14)
2.1689(15)
2.3270(5)
2.3165(5)
2.3728
1.4337(19)
1.7319(14)
1.917
2.224
2.317
2.312
2.400
1.424
1.751
+0.79
+2.54
−0.43
−0.17
+1.15
−0.70
+1.10
1.849b
2.217c
2.329d
2.352
1.412
Ti(1)–N(1)–P(1)
Ti(1)–N(1)–N(2)
Ti(1)–N(2)–N(1)
N(1)–Ti(1)–N(2)
N(2)–N(1)–P(1)
Ti(1)–N(2)–C(1)
Ti(1)–N(2)–C(2)
C(1)–N(1)–C(2)
161.83(9)
79.79(9)
59.63(8)
159.93
82.03
58.63
−1.17
154.90e
84.57
56.10
2.80
−1.68
−3.06
0.06
40.58(5)
39.34
39.33
117.90(11)
120.49(12)
125.60(12)
111.14(15)
117.97
124.06
121.72
110.65
120.53f
122.93
122.93
112.15
2.96
−3.09
−0.44
a D = {(Calc. value − X-ray value)/X-ray value} × 100%. b Ti–NMe bond length. c Ti–NMe2 bond length. d Average bond length. e Ti–N–C, not
Ti–N–P bond angle. f N–N–C, not N–N–P bond angle.
6 8 8
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3

View Article Online
for representing such P–N complexes. As before, the Gaussian
98W27 suite of programs was used. The optimised bond lengths
(Table 5) show that the optimised structure is in good agreement
with that determined by X-ray crystallography. The calculated
structure generally slightly overestimates the bond lengths by
about 0.75%, up to a maximum of 2.5% for the Ti(1)–N(2)
bond. The HOMO orbital is delocalised over the Ti–N(1)–P–
C(ipso) unit, but also has some phosphorus lone pair character,
and the LUMO is an anti-bonding M–Cl orbital. This is in
comparison to [(C5H5)TiCl2(MeNNMe2)], for which the HOMO
is a Ti–(C5H5) bonding orbital (for 6, the second and third
highest occupied orbitals are Ti–(C5H5) bonding in character),
and the LUMO, which is an anti-bonding M–Cl orbital, as
for 6.
Unfortunately, it was not possible to mimic the above reaction
with either L3H or L4H. Reaction of two equivalents of L3H with
1
[(C5H5)TiCl3] gave a 31P{ H} NMR spectrum showing at least
five 31P environments. The main peaks (from integration) were
two doublets at 35.9 and −22.7 ppm (1JP P = 227 Hz) which were
–
=
=
tentatively assigned to Ph2P–P( N–R)Ph2 [cf. Ph2P–P( O)Ph2
at 34 and −23 ppm, 1JP = 220 Hz70 and Ph2P–P( N–
=
–
P
PyH+)Ph2 at 18.9 and −20.3 ppm, 1JP P = 280 Hz],71 showing the
Fig. 11 ORTEP view (40% probability ellipsoids) of the X-ray crystal
structure of 7 with geometrically positioned hydrogen atoms omitted for
clarity.
–
instability of the P–N bond to hydrolysis. Reaction of L4H gave
very similar results with peaks at 35.4 and −23.4 ppm (1JP
=
–P
226 Hz) in the 31P{ H} NMR spectrum. In the presence of Et3N
as base, the reaction produces a crystalline red solid, however,
the high air- and or moisture-sensitivity of this product has
precluded any meaningful analysis. The difference in stability of
the Ti complex of L2H as opposed to that of the large array of
1
Meanwhile the N–N bond lengths of the complexes 6 and 7 and
the free ligand (L2H) are almost exactly the same (1.4337(19),
˚
1.433(4) and 1.4388(7) A, respectively). The Ru(1)–P(1) bond
˚
length of 2.3238(8) A is similar to that of previously reported
other R1 P–NR2 compounds that we have investigated including
compounds containing a Ru–P(III) bond.72 The NH group does
not participate in hydrogen bonding in the solid state, although
this is probably as a result of its high degree of steric crowding.
Attempts to deprotonate the ligand in complex 7 were made
using a variety of bases including KOtBu, proton sponge, Et3N,
2
those of L3H and L4H is put down to the ability of the sterically
small ligand L2H to bind to the Ti centre through both of
its N atoms (Fig. 7). However, that is obviously not possible
for the diarylphosphazene ligands, and is not possible for L3H
and L4H due to the steric requirements of the cyclohexyl ring,
thus reducing the strength of the bonding interactions between
the ligand and the metal, and so increasing the ease of hydrolysis
of the complexes.
1
1
NaNH2, NaOMe, however, the H and 31P{ H} NMR spectra
of the reaction solutions showed no change.
6
The reaction of L3H and L4H with [{RuCl(l-Cl)(g -p-
i
MeC6H4 Pr)}2] produced complexes analogous to 7. The
1
6
i
In contrast to the above reaction of L2H with [(C5H5)TiCl3],
31P{ H} NMR spectra for both [RuCl2(g -p-MeC6H4 Pr)(L3H)]
the reaction of four equivalents of L2H with [{RuCl(l-Cl)(g -
8 and [RuCl2(g -p-MeC6H4 Pr)(L4H)] 9 showed resonances at
6
6
i
p-MeC6H4 Pr)}2] in THF gave a product with L2H bound
ca. 65 ppm and 1H NMR spectra showed NH resonances at ca.
i
1
via only the phosphorus. In fact, reaction of two equivalents
4 ppm as anticipated (see Table 6). The 13C{ H} NMR spectra
of L2H with [{RuCl(l-Cl)(g -p-MeC6H4 Pr)}2] gave the complex
are also analogous to that seen for 7, with 13C–31P coupling again
evident. The ES+ MS of both showed the peak corresponding
to [M]+, the elemental analyses are satisfactory and the IR
spectra (KBr disc) of both have resonances at ∼3300, ∼1435
and ∼745 cm−1, assigned to vN H, vP and vP N, respectively.
6
i
[RuCl2(g -p-MeC6H4 Pr)(L2H)] (7) in good yield (77%). The 1H
6
i
NMR spectrum of the complex showed that the NH proton
1
chemical shift had moved downfield to 3.95 ppm, and the H–
31P coupling constant had increased to 30 Hz (from 12 Hz
–
–
Ph
–
1
in L2H). The 31P{ H} NMR spectrum shows a single resonance
Recrystallisation of 8 and 9 by layering a CH2Cl2 solution
with pentane produced crystals suitable for X-ray analysis. The
crystal structures of both of the complexes 8 and 9 are totally
isostructural with that of 7 and so are not shown here and are not
worthy of extensive discussion (selected bond lengths and angles
can be found in Table 4, and details of the structure refinement in
Table 2). The only significant differences are in the NH hydrogen
bonding. In 8, the NH group appears to be interacting with
one of the Cl atoms of the same molecule. While both the
at 65.7 ppm (cf. 38.6 ppm for the free ligand) consistent with a
Ru–P(III) complex.72 As in the free ligand and the Ti complex, the
1
13C{ H} NMR spectrum shows 13C–31P coupling with J values
6
i
comparable with the literature.62 The aromatic g -p-MeC6H4 Pr
CH carbon atoms also appear as doublets due to coupling to
31P through the metal centre, with 2JC P values of approximately
–
5 Hz, and the analysis of HMQC and HMBC spectra allowed
6
the unambiguous assignment of the two pairs of aromatic g -p-
MeC6H4 Pr CH protons. The ES+ MS of the complex showed
N(1) · · · Cl(1) distance (3.211(4) A) and the H · · · Cl distance
i
˚
almost exclusively the desired compound, [M + H]+ at m/z
(2.52(6) A) are sufficiently short to suggest that there is an
˚
551.5.
intramolecular hydrogen bond, the N–H · · · Cl angle is relatively
acute (138(5)◦) for a hydrogen bond. The crystal structure of 9
shows that the NH group approaches both Cl ligands of the same
molecule relatively closely [N(1) · · · Cl(1) 3.410(3), N(1) · · · Cl(2)
Slow evaporation of a THF solution of 7 under N2 yielded
crystals suitable of X-ray structure determination (Fig. 11,
6
i
Tables 1 and 4). The phenyl ring of the g -p-MeC6H4 Pr ligand is
◦
˚
˚
coordinated in a symmetric manner and the Ru atom lies 1.69 A
3.129(3) A]. However, the N–H · · · Cl angles are both below 120
from the ring centroid. As with the free ligand, the coordination
geometry around N(1) is considerably flatter than normal for a
secondary amine, which together with an even shorter P(1)–N(1)
suggesting no hydrogen bond is present. The crystal structures
of both 8 and 9 show that the geometry of the protonated N
atom (N(1)) is a flattened trigonal pyramid. This, together with
the large P–N–N angle (117.0(3) and 117.7(2)◦, respectively) and
˚
bond length (1.673(3) A) and a P(1)–N(1)–N(2) bond angle of
116.9(2)◦ suggests that there is some P(1)–N(1) multiple bond
character. This is in contrast to the Ti complex which, as already
discussed, has a P–N bond length characteristic of a single bond.
short P–N bond distance (1.691(4) and 1.664(3) A, respectively)
˚
suggests that, as in complex 7, the P–N bonds have partial
multiple-bond character.
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3
6 8 9

View Article Online
1
Table 6 31P{ H} and selected 1H NMR data for compounds L2H, L3H, L4H and 6–13
Chemical shift/ppm
Compound
d 31
P
d NH (2JH P/Hz)
–
L2H
L3H
L4H
6
Ph2PNHNMe2
38.6a
39.2b
35.6b
55.8b
65.7b
67.9b
66.4b
43.7a
46.9a
45.9a
86.3a
3.31a (12)
3.35b (10)
3.25b (8)
—
Ph2PNHNCH2CH2N(CH3)CH2CH2
Ph2PNHNCH2CH2CH2CH2CH2
(C5H5)TiCl2(PPh2NNMe2)
6
i
7
[RuCl2(g -p-MeC6H4 Pr)(L2H)]
3.95b (30)
4.10b (30)
4.00b (31)
7.35a (3)
7.05a (3)
7.01a (3)
—
6
i
8
9
10
11
12
13
[RuCl2(g -p-MeC6H4 Pr)(L3H)]
[RuCl2(g -p-MeC6H4 Pr)(L4H)]
6 i
6
i
[RuCl(g -p-MeC6H4 Pr)(L2H)]+BF4−
−
6
i
[RuCl(g -p-MeC6H4 Pr)(L3H)]+BF4−
6
i
[RuCl(g -p-MeC6H4 Pr)(L4H)]+BF4
6
i
3
[(g -p-MeC6H4 Pr)Cl2Ru(l2,g -Ph2PNNMe2)TiCl2(C5H5)]
a Spectrum run in CD2Cl2. b Spectrum run in CDCl3.
Cationic ruthenium phosphinohydrazine complexes
The reaction of 7 with AgBF4 in toluene resulted in Cl− ion
38.9 ppm, 7 at 65.7 ppm, and 10 at 43.7 ppm, see Table 6).
In the 1H NMR spectrum, the two N(CH3)2 proton resonances
are now coincident and appear at the same chemical shift as in
7. It appears that solvation of the complex in MeCN results in
coordination of at least one molecule of MeCN to the metal
centre and so ring opening the four-membered metal chelate.
Recrystallisation of 10 by layering a CH2Cl2 solution with
pentane produced crystals suitable for X-ray crystal structure
analysis (Fig. 12, Tables 2 and 4). As expected, a chelate ring
is formed in order to maintain 18-electron configuration at the
Ru centre, and a four-membered ring is preferred over the two
possible three-membered rings. This is the first example of a
four-membered Ru–P–N–N metal chelate complex (Fig. 12(A)),
which is a neutral analogue of the known ruthenium structure
abstraction from the Ru coordination sphere and the formation
6
i
of a cationic Ru species [RuCl(g -p-MeC6H4 Pr)(L2H)]+BF4− 10.
The ES+ MS of the resulting orange solid showed the only
significant peak to be due to [M]+ as expected. The 31P{ H}
1
NMR spectrum (CD2Cl2) of the compound showed a single
1
peak at 43.7 ppm, and the H NMR (CD2Cl2) also showed
several distinct differences to that of the 7. The spectrum showed
inequivalent methyl protons of the N(CH3)2 and isopropyl
groups. These inequivalencies are due to the formation of a
chelate ring, which produced a chiral complex and so making
the protons diastereotopic. The close approach of the N(CH3)2
6
i
group to one side of the g -p-MeC6H4 Pr ring restricts the
rotation of the isopropyl group, and removes the magnetic
equivalence of the two methyl groups of the iso-propyl group
(as can be seen in the crystal structure, Fig. 12).
1
2
i
of the triazenido ligand [Ru(N(R ) N–N–R )(PPr 3)Cl3].74 The
=
6
i
phenyl ring of the g -p-MeC6H4 Pr ligand is coordinated in a
˚
symmetrical manner with the Ru atom lying 1.72 A from the
best plane of the C6 ring. The sum of the internal angles of the
Ru(1)–P(1)–N(1)–N(2) chelate ring is 358.63◦, indicating that it
is very close to planar in geometry (Fig. 12(B)). Similarly the
sum of the bond angles around N(1) is 351.4◦, which, combined
˚
with the short P(1)–N(1) bond of 1.658(4) A is indicative of
significant multiple bond character in the P–N bond. The P–
˚
Ru bond distance of 2.3133(9) A is very similar to the neutral
compound 7. The NH group forms a bifurcated hydrogen bond
to two of the F atoms of a neighbouring anion (N(1) · · · F(1)
˚
˚
2.954(5) A and N(1) · · · F(2) 3.034(5) A).
The reaction of 8 with AgBF4 in toluene/benzene also results
in Cl− ion abstraction and the formation of a cationic Ru species
which after rapid recrystallisation from CH2Cl2–E2O produced
6
i
−
[RuCl(g -p-MeC6H4 Pr)(L3H)]+BF4 11. The ES+ MS of the
resulting ochre solid showed the expected peak at 570 m/z due to
1
[M]+. The 31P{ H} NMR spectrum (CD2Cl2) of the compound
shows a single peak at d = 46.9 ppm and as for 10 the 1H NMR
showed inequivalent methyl protons of the iso-propyl group
6
and inequivalent aromatic CH’s for the two sides of the g -p-
i
MeC6H4 Pr ring. This is presumably due to the coordination of
a lone pair on one of the nitrogen atoms to the coordinatively
unsaturated Ru centre forming a four- or seven-membered ring
6
i
which locks the conformation of the g -p-MeC6H4 Pr ring. The
complex is, however, unstable in solution in CH2Cl2, with after
24 h, over 75% of the sample decomposing, precluding any
further meaningful analysis of the complex.
Reaction of 9 with AgBF4 in toluene/benzene produced a
6
i
−
much more stable complex [RuCl(g -p-MeC6H4 Pr)(L4H)]+BF4
Fig. 12 ORTEP view (40% probability ellipsoids) of the X-ray crystal
structure (A, with geometrically positioned hydrogen atoms omitted for
clarity) and selected bond lengths and angles for 10 (B).
12. 1H NMR of a bright orange CD2Cl2 solution showed the NH
proton once again just above 7 ppm, and the four aromatic CH
6
i
protons of the g -p-MeC6H4 Pr ring appearing as four distinct
doublets in the region 4.96–6.01 ppm. The aromatic protons
of P(C6H5)2 also appeared inequivalent, showing the different
environments of the two rings as evident in the crystal structure
The NMR spectra in CD3CN shows very different resonances
1
to that in CD2Cl2. The phosphorus resonance in the 31P{ H}
NMR spectrum in CD3CN shifts to 70.8 ppm (cf. L2H at
6 9 0
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3

View Article Online
(see later). In contrast to the apparently conformationally locked
primary coordination sphere, the peaks assigned to the protons
of the –(CH2CH2)2CH2 ring were broad and ill-defined, possibly
spectrum showed a broad singlet at d = 86.3 ppm (cf. 38.6 ppm
1
for the free ligand, 55.8 in 6 and 65.7 in 7). The H NMR
spectrum showed a downfield shift of the C5H5 resonance to
1
indicative of ring fluxionality. The 13C{ H} NMR spectrum was
6.51 ppm from 6.19 ppm in 6 and 7.00 ppm in [(C5H5)TiCl3]. The
6
i
assigned fully following analysis of the 1H–1H COSY and 1H–13C
HMQC and HMBC spectra, showing six different environments
aromatic protons of the g -p-MeC6H4 Pr ligand again appear as
two doublets (as for 7, 8 and 9), this time at 5.06 and 4.72 ppm
for the ortho-, meta- and para-carbon atoms, all coupled to 31
P
with JH = 6 Hz [cf. 5.19 and 5.01 ppm, JH = 5 Hz for
3
3
–H
–H
1
to varying degrees. The phosphorus resonance appeared as a
7]. The 13C{ H} NMR spectrum shows 10 different carbon
environments as expected (the ipso carbon resonances could not
be detected) with the ortho-, meta- and para- P(C6H5)2 carbon
atoms appearing as doublets due to coupling to 31P as before.
On this occasion, 13C–31P coupling is also seen for the aromatic
1
singlet at 45.9 ppm in the 31P{ H} NMR as expected.
Good elemental analysis of the bulk material was also
obtained and the mass spectrum shows only one set of peaks at
m/z 555 with the correct isotope pattern for [M]+, furthermore,
the high-resolution MS (ES+) showed the main peak at 555.1280
([M]+ requires 555.1270).
6
i
CH carbons of the g -p-MeC6H4 Pr ligand. The N(CH3)2 peak
is just resolved from the solvent peak (CD2Cl2) at 54.2 ppm in
1
the 13C{ H} NMR.
It was not possible to obtain a mass spectrum of the
compound despite numerous attempts, due to fragmentation
under the mass spectrum conditions. However, the elemental
analysis obtained was in excellent agreement with the expected
values and the IR spectrum shows peaks at 1436 and 739 cm−1
The X-ray crystal structure of complex 12
On layering a CH2Cl2 solution of 12 with pentane, crystals
suitable for X-ray crystal analysis were produced. The structure
is very similar to that of 10 and so is not shown here (selected
bond lengths and angles can be found in Table 4, and details
of the structure refinement in Table 2). In particular, bond
lengths of the four-membered chelate ring differ by less than 1%.
Furthermore, the sum of the internal angles of the heterocyclic
ring is 359.6◦, indicating that it is essentially planar in geometry.
attributed to vP and vP N, respectively.
–
Ph
–
Attempts at recrystallization of the orange–brown solid
from THF–Et2O diffusion produced crystals suitable for X-
ray analysis. However, the complex decomposes over time
6
in THF producing crystals of the Ru(II) species [RuCl2(g -p-
˚
The short P(1)–N(1) bond of 1.658(4) A combined with the
MeC6H4 Pr)(L2H2)]+Cl− (Fig. 14). The structure is very similar
i
planar nature of the N(1) atom (the sum of the angles around
N(1) is 356.7◦) is indicative of significantmultiple bond character
in the P–N bond. The most significant difference to the structure
of 10 is in the NH bond length (ca. 9% shorter in 12), however
this is presumably due to the fact that the NH in 12 is hydrogen
to that of 7, with the main difference being the P(1)–N(1) bond
˚
length which is slightly longer at 1.7017(18) A as opposed to
◦
˚
1.673(3) A for 7. The sum of the angles around N(1) is 345.5
(cf. 345.9◦ for 7) which is considerably flatter than is normal for a
secondary amine, and so indicating the possibility of some P–N
multiple bonding. The NH groups both participate in hydrogen
bond formation. The secondary NH forms a hydrogen bond to
−
bonded to only one fluorine atom of the BF4 counterion
˚
(N(1) · · · F(1) 2.918(4) A) instead of two in 10.
The crystal did however, contain some disorder. It was clearly
apparent that the thermal parameters of the carbon atoms of
one of the phenyl substituents [C(13), C(14), C(16) and C(17),
fractional site occupancy = 0.514(15)] were unusually large
and this was readily modelled as being due to disorder of this
group over two positions related by a rotation about the P–
C · · · C(para) axis. In addition the refined thermal parameters
of some of the C atoms of the p-cymene ligand were large
and highly anisotropic. This was most satisfactorily modelled
over two orientations of the isopropyl group [C(24), C(25) and
C(26), fractional site occupancy = 0.687(14)], but neglecting
disorder of the phenyl or methyl fragments. The coordinates,
anisotropic thermal parameters and site occupancies of the
disordered carbon atoms were refined. An attempt was made
to solve and refine the structure on the space group Cc, however
the refinement failed to converge, confirming that the original
choice of space group was correct. Attempts to cool the crystal
slowly to determine if a less-symmetric ordered structure was
preferred at low temperature were frustrated by decomposition
of the crystals, presumably as a result of loss of solvent.
−
˚
a free Cl ion (N(1) · · · Cl(3) 3.1039(19) A), whereas the tertiary
NH group forms a bifurcated intramolecular hydrogen bond
to both Cl ligands (N(2) · · · Cl(1) 3.190(2) and N(2) · · · Cl(2)
˚
3.138(2) A).
Bimetallic complex
Fig. 14 ORTEP view (40% probability ellipsoids) of the X-ray crystal
Reaction of the complex 6 with 0.5 equivalents of [{RuCl(l-
6
i
structure of [RuCl2(g -p-MeC6H4 Pr)(PPh2NHNHMe2)]+Cl− with geo-
6
i
6
Cl)(g -p-MeC6H4 Pr)}2] in THF yielded a bimetallic species, [(g -
metrically positioned hydrogen atoms omitted for clarity.
p-MeC6H4 Pr)Cl2Ru(l2,g -PPh2NNMe2)Ti(C5H5)Cl2] 13, with
the two metal centres bridged by the Ph2PNNMe2 ligand
(Fig. 13). This is the first example of a bimetallic Ru, Ti species
i
3
1
bridged by an amino phosphine ligand. The 31P{ H} NMR
Conclusions
We have prepared a number of tungsten complexes incorpo-
rating the previously unknown phosphazene ligands (N–PPh2).
These complexes have unfortunately proved to be unstable
so X-ray structure determinations were not possible. DFT
calculations, however, showed the P–N ligand to be in a linear
arrangement with the nitrogen doubly bonded to the tungsten
centre and confirmed that there was no evidence for any multiple
bonding between P and N. We have shown that while it is
Fig. 13 Structure of bimetallic Ru–Ti complex 13.
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3
6 9 1

View Article Online
=
possible to prepare these M N–PPh2 complexes, they are very
18 T. G. Wetzel, S. Dehnen and P. W. Roesky, Angew. Chem., Int. Ed.,
1999, 38, 1086.
19 O. Kuhl, T. Koch, F. B. Somoza, P. C. Junk, E. Hey-Hawkins, D. Plat
and M. S. Eisen, J. Organomet. Chem., 2000, 604, 116.
20 O. Kuhl, S. Blaurock, J. Sieler and E. Hey-Hawkins, Polyhedron,
2001, 20, 111.
susceptible to further reactions resulting in the breaking of the
P–N bond and the formation of new and unusual ligand types.
Reaction of the same proligand, Ph2PN(SiMe3)2, with rhe-
nium resulted in the formation of a new phosphorylketimido
ligand complex arising from nucleophilic attack of phosphorus
at the nitrile carbon of coordinated acetonitrile.
21 O. Schmitz-DuMont and W. Jansen, Angew. Chem., Int. Ed. Engl.,
1968, 7, 382.
22 B. Wrackmeyer, C. Kohler, W. Milius, J. M. Grevy, Z. Garcia-
Hernandez and R. Contreras, Heteroat. Chem., 2002, 13, 667.
23 R. P. Nielsen and H. H. Sisler, J. Chem. Soc., 1962, 3818.
24 H. H. Sisler and R. P. Nielsen, Inorg. Synth., 1966, 8, 74.
25 R. D. Gorsich, J. Am. Chem. Soc., 1958, 80, 4744.
26 C. L. Sloan and W. A. Barber, J. Am. Chem. Soc., 1959, 81, 1364.
27 GAUSSIAN 98 (Revision A.11.2), M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G.
Zakrzewski, J. A. Montgomery, R. E. Stratmann, J. C. Burant,
S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C.
Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B.
Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A.
Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W.
Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres, M. Head-
Gordon, E. S. Replogle and J. A. Pople, Gaussian, Inc., Pittsburgh,
PA, 1998.
Reaction of the ligand Ph2PNHNMe2 with [(C5H5)TiCl3] has
produced a complex containing an ionic N,N-bound phos-
phahydrazido ligand. However, it was not possible to repeat
this with the other phosphinohydrazine ligands. In the reaction
of these same three ligands with Ru, they acted as simple
monodentate phosphines. Formation of new four-membered
Ru–P–N–N metal chelates was possible on removal of Cl− using
6
AgBF4. We also synthesized a Ru–Ti binuclear complex [(g -p-
MeC6H4 Pr)Cl2Ru(l2,g -PPh2NNMe2)Ti(C5H5)Cl2] 13, through
i
3
6
i
reaction of 6 with [{RuCl(l-Cl)(g -p-MeC6H4 Pr)}2].
The coordination of the phosphorus lone pair to the Ru centre
in complexes 7–12 caused a significant (ca. 1.5%) reduction in
P–N bond length for all the complexes, except 8, for which there
˚
was a 0.01 A increase in bond length. This increase may help to
explain the lower stability of the cationic complex 11 compared
to 10 and 12. The general reduction in bond length is presumably
due to an increase in the nitrogen lone pair donation to the
phosphorus atom, caused by a reduction in the electron density
on the phosphorus on coordination to a metal. This postulate
is reinforced by the significantly longer P–N bond length of the
28 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
29 R. Ditchfield, W. J. Hehre and J. A. Pople, J. Chem. Phys., 1971, 54,
724.
30 W. J. Hehre, R. Ditchfield and J. A. Pople, J. Chem. Phys., 1972, 56,
˚
Ti(IV) complex, 6 (1.7319(14) A), due to the reduction in nitrogen
2257.
lone pair donation to the phosphorus atom on coordination of
the nitrogen to the highly Lewis acidic Ti(IV) centre.
31 P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28, 213.
32 P. C. Hariharan and J. A. Pople, Mol. Phys., 1974, 27, 209.
33 M. S. Gordon, Chem. Phys. Lett., 1980, 76, 163.
34 R. C. Binning Jr. and L. A. Curtiss, J. Comput. Chem., 1990, 11,
1206.
35 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270.
36 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299.
37 W. R. Wadt and P. J. Hay, J. Chem. Phys., 1985, 82, 284.
38 A. J. H. Wachters, J. Chem. Phys., 1970, 52, 1033.
39 P. J. Hay, J. Chem. Phys., 1977, 66, 4377.
Acknowledgements
The authors gratefully acknowledge funding from the EPSRC
and would like to thank Johnson Matthey plc for their generous
donation of ruthenium starting material and Hermann Starck
GmBH for their generous donation of rhenium metal.
40 R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys.,
1980, 72, 650.
41 A. D. McLean and G. S. Chandler, J. Chem. Phys., 1980, 72,
References
5639.
1 W. A. Nugent and B. L. Haymore, Coord. Chem. Rev., 1980, 31, 123.
2 J. Chatt and R. L. Richards, J. Organomet. Chem., 1982, 239, 65.
3 J. R. Dilworth, Coord. Chem. Rev., 1996, 154, 163.
4 J. Chatt, J. R. Dilworth and R. L. Richards, Chem. Rev., 1978, 78,
589.
5 V. Ritleng, D. V. Yandulov, W. W. Weare, R. R. Schrock, A. S. Hock
and W. M. Davis, J. Am. Chem. Soc., 2004, 126, 6150.
6 T. Q. Ly, A. M. Z. Slawin and J. D. Woollins, J. Chem. Soc., Dalton
Trans., 1997, 1611.
42 K. Raghavachari and G. W. Trucks, J. Chem. Phys., 1989, 91, 1062.
43 Z. Otwinowski and W. Minor, Processing of X-ray Diffraction Data
Collected in Oscillation Mode, Methods Enzymol., 1997, 276, 307.
44 A. Altomare, G. Cascarano, G. Giacovazzo, A. Guagliardi, M. C.
Burla, G. Polidori and M. Camalli, J. Appl. Crystallogr., 1994, 27,
435.
45 D. J. Watkin, C. K. Prout, J. R. Carruthers, P. W. Betteridge
and R. I. Cooper, CRYSTALS issue 11, Chemical Crystallography
Laboratory, Oxford, UK, 2001.
7 S. M. Aucott, A. M. Z. Slawin and J. D. Woollins, J. Organomet.
Chem., 1999, 582, 83.
46 J. D. Lichtenhan, J. W. Ziller and N. M. Doherty, Inorg. Chem., 1992,
31, 2893.
8 P. A. Bella, O. Crespo, E. J. Fernandez, A. K. Fischer, P. G. Jones, A.
Laguna, J. M. Lopez-de-Luzuriaga and M. Monge, J. Chem. Soc.,
Dalton Trans., 1999, 4009.
9 S. M. Aucott, A. M. Z. Slawin and J. D. Woollins, J. Chem. Soc.,
Dalton Trans., 2001, 2279.
10 P. Braunstein, B. T. Heaton, C. Jacob, L. Manzi and X. Morise,
Dalton Trans., 2003, 1396.
11 F. C. Bradley, E. H. Wong, E. J. Gabe and F. L. Lee, Inorg. Chim.
Acta, 1986, 120, L21.
47 J. R. Dilworth and S. Morton, Transition Met. Chem. (London),
1987, 12, 41.
48 J. R. Dilworth, V. C. Gibson, C. Lu, J. R. Miller, C. Redshaw and Y.
Zheng, J. Chem. Soc., Dalton Trans., 1997, 269.
49 D. E. C. Corbridge, Phosphorous: An outline of its Chemistry,
Biochemistry and Technology, Elsevier Science Ltd., Oxford, 4th edn.,
1990, p. 47.
50 M. G. B. Drew, D. G. Tisley and R. A. Walton, J. Chem. Soc. D,
1970, 600.
12 H. W. Roesky, H. Voelker, M. Witt and M. Noltemeyer, Angew.
Chem., 1990, 102, 712.
13 J. R. Dilworth, H. J. D. L. Meijer and J. H. Teuben, J. Organomet.
Chem., 1978, 159, 47.
51 M. Davis, F. Belanger-Gariepy, D. Zargarian and A. L. Beauchamp,
Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1997, 53, 428.
52 C. M. P. Ferreira, M. F. C. Guedes da Silva, V. Y. Kukushkin,
J. J. R. Frausto da Silva and A. J. L. Pombeiro, J. Chem. Soc., Dalton
Trans., 1998, 325.
53 S. Zhang and W. E. Piers, Organometallics, 2001, 20, 2088.
54 A. R. Dias, M. T. Duarte, A. C. Fernandes, S. Fernandes, M. M.
Marques, A. M. Martins, J. F. d. Silva and S. S. Rodrigues,
J. Organomet. Chem., 2004, 689, 203.
55 A. L. Spek, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1986,
43, 1233.
56 J. Chatt and G. A. Rowe, J. Chem. Soc., 1962, 4019.
57 J. Chatt, J. D. Garforth, N. P. Johnson and G. A. Rowe, J. Chem.
Soc., 1964, 1012.
14 I. A. Latham, J. G. Leigh, G. Huttner and I. Jibril, J. Chem. Soc.,
Dalton Trans., 1986, 337.
15 E. Hollink, J. C. Stewart, P. Wei and D. W. Stephan, Dalton Trans.,
2003, 3968.
16 D. W. Stephan, J. C. Stewart, F. Guerin, S. Courtenay, J. Kickham, E.
Hollink, C. Beddie, A. Hoskin, T. Graham, P. Wei, P. E. V. H. Spence,
W. Xu, L. Koch, X. Gao and D. G. Harrison, Organometallics, 2003,
22, 1937.
17 D. Fenske, B. Maczek and K. Maczek, Z. Anorg. Allg. Chem., 1997,
623, 1113.
6 9 2
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3

View Article Online
58 J. Chatt, R. J. Dosser, F. King and G. J. Leigh, J. Chem. Soc., Dalton
Trans., 1976, 2435.
59 G. Rouschias and G. Wilkinson, J. Chem. Soc. A, 1968, 489.
60 L. R. Falvello, S. Ferna´ndez, R. Navarro and E. P. Urriolabeitia,
Inorg. Chem., 1997, 36, 1136.
67 H. C. Beachell and S. A. Butter, Inorg. Chem., 1965, 4, 1133.
68 I. A. Latham, G. J. Leigh, G. Huttner and I. Jibril, J. Chem. Soc.,
Dalton Trans., 1986, 385.
69 S. M. Aucott, A. M. Z. Slawin and J. D. Woollins, J. Chem. Soc.,
Dalton Trans., 2000, 2559.
61 F. A. Cotton, S. A. Duraj, C. C. Hinckley, M. Matusz and W. J. Roth,
Inorg. Chem., 1984, 23, 3080.
70 T. Q. Ly, A. M. Z. Slawin and J. D. Woollins, Polyhedron, 1999, 18,
1761.
62 O. A. Gansow and B. Y. Kimura, J. Chem. Soc., Chem. Commun.,
71 Z. Fei, N. Biricik, D. Zhao, R. Scopelliti and P. J. Dyson, Inorg.
1970, 1621.
Chem., 2004, 43, 2228.
63 R. J. Cross, T. H. Green and R. Keat, J. Chem. Soc., Dalton Trans.,
1976, 1424.
72 S. M. Aucott, A. M. Z. Slawin and J. D. Woollins, Polyhedron, 2003,
22, 361.
64 G. E. Graves and L. W. Houk, J. Inorg. Nuclear Chem., 1974, 36,
73 H. D. Flack and G. Bernardinelli, J. Appl. Crystallogr., 2000, 33,
232.
1143.
65 G. E. Graves and L. W. Houk, Inorg. Chem., 1976, 15, 7.
66 C. J. Harlan, T. C. Wright, J. L. Atwood and S. G. Bott, Inorg. Chem.,
1991, 30, 1955.
74 A. A. Danopoulos, R. S. Hay-Motherwell, G. Wilkinson, S. M.
Cafferkey, T. K. N. Sweet and M. B. Hursthouse, J. Chem. Soc.,
Dalton Trans., 1997, 3177.
D a l t o n T r a n s . , 2 0 0 5 , 6 8 0 – 6 9 3
6 9 3