Nucleophilic addition to complexes of (Ph2P)2C=CH2 as a route to functionalized, redox-active ruthenium (II) - Diphosphine complexes

Article abstract of DOI:10.1021/ic970799o

The ligand (Ph₂P)₂C=CH₂ (dppen) reacts with [RuCl₂(PPh₃)₃] in CH₂Cl₂ to give trans-[RuCl₂(dppen)₂], 1. Complex 1 has reversible Ru(II)/Ru(III) electrochemistry, and on treatment with NOBF₄, 1 affords [RuCl₂(dppen)₂]BF₄, 1⁺. Refluxing solutions of 1 in chlorobenzene affords cis-[RuCl₂(dppen)₂], 2. With excess RNH₂ in chlorobenzene or toluene, 1 reacts to give functionalized diphosphine complexes of general formula trans-[RuCl₂({Ph₂P}₂CHCH₂-NHR) ₂] (R: PhCH₂, 3a; [CH₂]₃NH₂, 3b; n-octyl, 3c; R-α-CH(Me)Ph, 3d; [CH₂]₃Si(OEt)₃, 3e) characterized principally by a small upfield shift in their ³¹P{¹H} NMR spectra compared to that of 1 and by their distinctive ¹H NMR spectra. Complex 3b·4MeOH crystallizes in the space group P1? with a = 11.448(6) A?, b = 13.10(1) A?, c = 11.178(7) A?, α = 93.16(6)°, β= 99.51(5)°, γ = 92.19(5)°, V = 1648(2) A?³, and D_calcd = 1.250 g cm^-3 for Z = 1. Complex 3e reacts with the surface hydroxyl groups of indium-doped tin oxide (ITO) electrodes, to give monolayers of anchored, redox-active Ru(II)-diphosphine complexes, characterized by cyclic voltammetry. The modified electrodes are stable to repetitive cycling over the Ru(II)/Ru(III) redox wave in acetonitrile -tetraethylammonium tetrafluoroborate and in aqueous buffer (pH 8). With anodized Pt electrodes, however, 3e reacts to form multilayers. Reaction of 1 with secondary amines is more sluggish than reaction with primary amines, and only adducts with pyrrolidine (3f) and morpholine (3g; impure) were isolated. With LiC≡C(CH₂)₃-CH₃, 1 reacts to give trans-[Ru(C≡CR)₂({Ph₂P}₂CHCH ₂C≡CR)₂] (R = n-butyl, 5); acetylide nucleophiles displace the chloride ligands as well as adding to the dppen double bonds of 1. Other carbanions fail to react with 1. The cis complex 2 reacts with a limited range of primary amines, to afford cis-[RuCl₂({Ph₂P}₂CHCH₂NHR) ₂] (R: n-hexyl, 4a; [CH₂]₃Si[OEt]³, 4b). However, 4b was too insoluble for electrode derivatization. The half-sandwich complex [CpRuCl(dppen)] (6; Cp = η⁵-C₅H₅) also reacts with RNH₂ to give [CpRuCl({PPh₂}₂CHCH₂NHR)] (R: -[CH₂]₃NH₂, 7a; -CH₂Ph, 7b).

Full text of DOI:10.1021/ic970799o

6188
Inorg. Chem. 1997, 36, 6188-6196
Nucleophilic Addition to Complexes of (Ph2P)2CdCH2 as a Route to Functionalized,
Redox-Active Ruthenium(II)-Diphosphine Complexes
Jim V. Barkley, Simon J. Higgins,* Mark K. McCart, and Thomas J. Pounds
Department of Chemistry, The Donnan Laboratories, University of Liverpool, P.O. Box 147,
Liverpool L69 3BX, United Kingdom
ReceiVed June 26, 1997X
The ligand (Ph2P)2CdCH2 (dppen) reacts with [RuCl2(PPh3)3] in CH2Cl2 to give trans-[RuCl2(dppen)2], 1. Complex
1 has reversible Ru(II)/Ru(III) electrochemistry, and on treatment with NOBF4, 1 affords [RuCl2(dppen)2]BF4,
1+. Refluxing solutions of 1 in chlorobenzene affords cis-[RuCl2(dppen)2], 2. With excess RNH2 in chlorobenzene
or toluene, 1 reacts to give functionalized diphosphine complexes of general formula trans-[RuCl2({Ph2P}2CHCH2-
NHR)2] (R: PhCH2, 3a; [CH2]3NH2, 3b; n-octyl, 3c; R-R-CH(Me)Ph, 3d; [CH2]3Si(OEt)3, 3e) characterized
principally by a small upfield shift in their 31P{1H} NMR spectra compared to that of 1 and by their distinctive
1H NMR spectra. Complex 3b‚4MeOH crystallizes in the space group P1h with a ) 11.448(6) Å, b ) 13.10(1)
Å, c ) 11.178(7) Å, R ) 93.16(6)°, â ) 99.51(5)°, γ ) 92.19(5)°, V ) 1648(2) Å3, and Dcalcd ) 1.250 g cm-3
for Z ) 1. Complex 3e reacts with the surface hydroxyl groups of indium-doped tin oxide (ITO) electrodes, to
give monolayers of anchored, redox-active Ru(II)-diphosphine complexes, characterized by cyclic voltammetry.
The modified electrodes are stable to repetitive cycling over the Ru(II)/Ru(III) redox wave in acetonitrile-
tetraethylammonium tetrafluoroborate and in aqueous buffer (pH 8). With anodized Pt electrodes, however, 3e
reacts to form multilayers. Reaction of 1 with secondary amines is more sluggish than reaction with primary
amines, and only adducts with pyrrolidine (3f) and morpholine (3g; impure) were isolated. With LiCtC(CH2)3-
CH3, 1 reacts to give trans-[Ru(CtCR)2({Ph2P}2CHCH2CtCR)2] (R ) n-butyl, 5); acetylide nucleophiles displace
the chloride ligands as well as adding to the dppen double bonds of 1. Other carbanions fail to react with 1. The
cis complex 2 reacts with a limited range of primary amines, to afford cis-[RuCl2({Ph2P}2CHCH2NHR)2] (R:
n-hexyl, 4a; [CH2]3Si[OEt]3, 4b). However, 4b was too insoluble for electrode derivatization. The “half-sandwich”
complex [CpRuCl(dppen)] (6; Cp ) η5-C5H5) also reacts with RNH2 to give [CpRuCl({PPh2}2CHCH2NHR)]
(R: -[CH2]3NH2, 7a; -CH2Ph, 7b).
Covalent anchoring of metal complexes to surfaces is a topic
of considerable interest in both catalysis1 and electrochemistry,2-5
although these two fields have developed independently. In
the former field, the chief interest is in attempting to combine
the advantages of homogeneous catalysis (high selectivities, high
activities) and heterogeneous catalysis (ease of separation of
catalyst from products)1 by fixing complexes active in homo-
geneous catalysis to heterogeneous supports; the metal centers
involved are usually platinum metal-phosphine complexes. The
approach employed is either to make a suitable complex of a
ligand such as Ph2P(CH2)3Si(OR)3 and then react this with an
oxide surface or to derivatize the oxide first with the siloxy-
phosphine ligand and subsequently carry out an exchange
reaction with a labile metal complex. A major problem in this
field has been that, under the conditions commonly required
for effective catalysis, ligand exchange reactions lead to leaching
of the metal centers from the oxide-anchored ligands. The use
of multidentate ligands functionalized with anchoring groups
might limit this, but few such ligands have been described,
probably because of synthetic difficulties.6
In electrochemistry, the anchoring of a monolayer of a redox-
active metal complex to an electrode surface offers the pos-
sibility of controlling the reactivity of the electrode more pre-
cisely than can be achieved simply by controlling its potential.
Such “modified electrodes” have been widely studied in
electrocatalytic reactions and in various sensing applications.2,4
Interest was also stimulated by the possibility of solar harvesting
using silicon covered with monolayers of photoactive species
or simply protecting semiconductors from photocorrosion.7,8 The
first deliberate modification of an electrode was achieved using
a Pt(IV) complex of an alkene-functionalized pyridine ligand.9
More recently, the study of long-chain alkanethiols self-
assembled on gold, including some metal complex-terminated
examples, has become a very active area.10-12 The metal centers
employed are usually classical redox-active species such as
metal porphyrins and metal phthalocyanines, [M(bpy)3]2+ (bpy
) 2,2′-bipyridine), ferrocenes, etc. Although metal-phosphine
complexes are important in homogeneous catalysis, their ap-
plication to monolayer-modified electrodes has not so far
attracted much attention, probably since platinum metal com-
plexes of ligands such as Ph2P(CH2)3Si(OR)3 do not usually
have reversible redox processes.
X Abstract published in AdVance ACS Abstracts, December 1, 1997.
(1) Iwasawa, Y. Inorganic Oxide-attached Metal Catalysts. In Tailored
Metal Catalysts; Iwasawa, Y., Ed.; D. Reidel Publishing Co.: Dor-
drecht, The Netherlands, 1986.
(2) Abrun˜a, H. D. Coord. Chem. ReV. 1988, 86, 135.
(3) Miller, J. S. Chemically-Modified Surfaces in Catalysis and Electro-
catalysis; American Chemical Society: Washington, DC, 1982.
(4) Murray, R. W. Molecular Design of Electrode Surfaces; John Wiley:
New York, 1992.
(7) Anderson, S.; Constable, E. C.; Dare-Edwards, M. P.; Goodenough,
J. B.; Hamnett, A.; Seddon, K. R.; Wright, R. D. Nature (London)
1979, 280, 5723.
(8) Wrighton, M. S. Acc. Chem. Res. 1979, 12, 303.
(9) Lane, R. F.; Hubbard, A. T. J. Phys. Chem. 1973, 73, 1401.
(10) Chidsey, C. E. Science 1991, 251, 919.
(5) Murray, R. W. Acc. Chem. Res. 1980, 13, 135.
(6) Fox, M. A.; Chandler, D. A.; Wang, P. Macromolecules 1991, 24,
4626.
(11) Fox, M. A. Acc. Chem. Res. 1992, 25, 569.
(12) Hickman, J. J.; Laibinis, P. E.; Auerbach, D. I.; Zou, C.; Gardner, T.
J.; Whitesides, G. M.; Wrighton, M. S. Langmuir 1992, 8, 357.
S0020-1669(97)00799-4 CCC: $14.00 © 1997 American Chemical Society

Redox-Active Ru(II)-Diphosphine Complexes
Inorganic Chemistry, Vol. 36, No. 27, 1997 6189
Vacuo. Solvents were dried by standard techniques.27 The following
The ligand 1,1-bis(diphenylphosphino)ethene (dppen) under-
goes ready nucleophilic addition when chelated to metal centres
such as M(CO)4 (M ) Cr, Mo, W),13-15 PtMe2 and PtIMe3,16
MX2 (M ) Pd, Pt; X ) Cl, I; M ) Pd, X ) OAc),17-19 and
NiCl2.20 The driving force for the reaction has not been
established with certainty but is probably a combination of relief
of angle strain at the carbon adjacent to the donor atoms and
polarization of the double bond by coordination of the phos-
phines. Indirect evidence for the latter is that the reactivity of
the ligand depends markedly upon the particular metal center
to which it is coordinated. Thus, the least reactive examples,
[M(CO)4(dppen)], react with acetylide anions and, on heating,
with amines and hydrazines (in large excess).13-15 The most
reactive example so far, [Pd(OAc)2(dppen)], reacts readily with
primary alcohols, extremely weak nucleophiles, in CH2Cl2
solution at room temperature.17,19
were used as received: KOtBu, 18-crown-6, CDCl3.
Characterization. Microanalyses were performed in house by Mr.
Steve Apter; the presence of solvent of crystallization inferred from
the analytical results was in all cases confirmed by 1H NMR
spectroscopy. 1H and 31P NMR spectra were recorded on a Bruker
WM 250 MHz spectrometer at 250 and 101 MHz, respectively, in
CDCl3 unless otherwise stated; broad-band proton decoupling and line
1
broadening (8 Hz) were employed for 31P NMR spectra. Some H
NMR spectra were recorded at 400 MHz on a Bruker AMX spectrom-
eter. Chemical shifts are reported in ppm (relative to TMS [1H] or
85% H3PO4 [31P]). Low-resolution fast-atom-bombardment mass
spectra (FAB MS) of the complexes, in 3-nitrobenzyl alcohol matrices,
were recorded with a VG 7070E instrument in positive-ion mode.
Electrochemical experiments were conducted using a Sycopel Scientific
Scanning Ministat potentiostat and Lloyd PL3 X-Y-t recorder. A three-
electrode one-compartment cell, with the reference electrode separated
by a glass frit and a Luggin capillary, was employed. For solution
studies, working electrodes were Pt disks (Sycopel). Throughout this
paper, a CH2Cl2 solution of TBAT was used as electrolyte for solution
studies except where otherwise stated. Modified electrodes were either
indium-doped tin oxide (ITO)-coated glass (a gift from Pilkington Group
Research, Lathom, U.K.) or anodized Pt disks. ITO-coated glass was
cleaned in 0.1% NaOH solution. The slides were dried, briefly (15-
30 s) exposed to “pirhana solution” (30% H2O2/H2SO4; care! explosiVe
when mixed with organic material), rinsed thoroughly with triply-
distilled water, and dried at 60 °C. They were then refluxed in propan-
2-ol and dried before derivatization. The Pt disks were anodized by
potential cycling in aqueous sulfuric acid.28 They were then washed
with triply-distilled water and dried at 60 °C before derivatization. The
counter electrode was a Pt gauze, and the reference electrode was an
Ag wire in 0.01M AgNO3/CH3CN/0.2 M Bu4NBF4 (CH2Cl2) or a
commercial saturated calomel electrode (CH3CN). Because different
reference electrodes and electrolytes were employed for solution and
modified electrode voltammetry, all potentials in this paper are quoted
with respect to the ferrocene/ferrocenium couple, which was routinely
monitored after each set of voltammetric measurements. Its potential
occurred at +0.19 V Vs the Ag/Ag+ electrode (CH2Cl2) and +0.58 V
Vs SCE (CH3CN).
Syntheses. (a) trans-[RuCl2(dppen)2] (1). To [RuCl2(PPh3)3] (4.0
g, 4.18 mmol) in CH2Cl2 (20 cm3) was added dppen (3.37 g, 8.5 mmol),
and the mixture was stirred at room temperature for 3 h. The orange
product was filtered off, washed with CH2Cl2 (2 cm3) and then Et2O
(20 cm3), and dried in Vacuo. Yield: 3.42 g, 85%. A further quantity
could be obtained by adding EtOH to the CH2Cl2 mother liquor. Anal.
Calcd for C52H44Cl2P4Ru‚CH2Cl2: C, 60.64; H, 4.42. Found: C, 60.52;
H, 4.40. FAB MS, m/z: 965 (100) M+; 930 (27) [M - Cl]+; 894 (16)
[M - Cl - HCl]+. IR data (cm-1): 328 (m) (νRu-Cl). Electronic
spectral data, Emax/10-3 cm-1 (ꢀ/L mol-1 cm-1) (CH2Cl2): 24.15 (2240).
31P NMR: δ 14.7 (s). 1H NMR data: δ 7.43-7.08 (40H, Ph), 6.12
(4H, quintet, CdCH2). Ru(II)/Ru(III) E1/2: +0.01 V.
Coordinatively saturated complexes of Ru(II) have the twin
advantages of chemical stability and accessible, reversible Ru-
(III)/Ru(II) redox processes.21,22 We have therefore employed
nucleophilic addition to dppen complexed to Ru(II) to afford
complexes suitable for modification of electrode surfaces with
monolayers, using cyclic voltammetry to characterize the
anchored complexes. Herein we report the syntheses and
characterization of trans-[RuCl2(dppen)2] (1) and cis-[RuCl2-
(dppen)2] (2), their reactivity with primary and secondary
amines, and the derivatization of electrode surfaces with the
amine adduct trans-[RuCl2({Ph2P}2CHCH2NH{CH2}3Si{OEt}3)2]
(3e). We also describe some attempts to extend this chemistry
to the “half-sandwich” complex [CpRuCl(dppen)] (6; Cp ) η5-
C5H5).23 A preliminary account of some aspects of the work
has appeared.24
Experimental Section
General Considerations. Reactions were carried out under nitrogen,
using Schlenk-line techniques for transfer of organolithium reagents.
Reaction solvents were reagent grade and were dried by distillation
from CaH2 (CH2Cl2) or sodium benzophenone ketyl (thf, diethyl ether,
heptane). Perdeuterated solvents for NMR studies were used as
received. The ligand dppen was prepared by the published method25
(tetrahydrofuran was used instead of benzene as solvent for Cl2CdCH2),
as were [RuCl2(PPh3)3]26 and [CpRu(dppen)Cl] (Cp ) η5-C5H5).23
Tetrabutylammonium tetrafluoroborate (TBAT) was prepared from the
bromide (Lancaster Synthesis) and 50% HBF4 (Fluka) in water and
recrystallized twice from acetone-water. It was dried at 60 °C in
(13) Cooper, G. R.; McEwan, D. M.; Shaw, B. L. Inorg. Chim. Acta 1986,
122, 207.
(14) Hassan, F. S. M.; Shaw, B. L.; Thornton-Pett, M. J. Chem. Soc., Dalton
Trans. 1988, 89.
(15) Haupt, H.-J.; Kleineberg, E.; Flo¨rke, U. Z. Anorg. Allg. Chem. 1993,
619, 869.
(16) Hassan, F. S. M.; Higgins, S. J.; Jacobsen, G. B.; Shaw, B. L.;
Thornton-Pett, M. J. Chem. Soc., Dalton Trans. 1988, 3011.
(17) King, G.; Higgins, S. J.; Hopton, A. J. Chem. Soc., Dalton Trans.
1992, 3403.
(b) trans-[RuCl2(dppen)2]BF4 (1+). To 1 (0.60 g, 0.62 mmol) in
CH2Cl2 (40 cm3) at 0 °C was added NOBF4 (0.073 g, 0.62 mmol) with
stirring and with N2 bubbling to displace NO. After 2 h, the solvent
volume was reduced in Vacuo and the product was precipitated with
diethyl ether, filtered off, and dried in Vacuo. Yield: 0.60 g, 92%.
Anal. Calcd for C52H44BCl2F4P4Ru‚CH2Cl2: C, 56.01; H, 4.08.
Found: C, 56.50; H, 4.29. FAB MS, m/z: 965 (100) M+; 930 (20)
[M - Cl]+; 894 (8) [M - Cl - HCl]+. Selected IR data (cm-1): 1058
(s) (BF4-). Electronic spectral data, Emax/10-3 cm-1 (CH2Cl2): 14.16,
20.8 (br, sh), 24.10.
(18) Higgins, S. J.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1989, 1527.
(19) Herring, A. M.; Higgins, S. J.; Jacobsen, G. B.; Shaw, B. L. J. Chem.
Soc., Chem. Commun. 1986, 882.
(20) Brunner, H.; Limmer, S. J. Organomet. Chem. 1991, 417, 173.
(21) Sullivan, B. P.; Meyer, T. J. Inorg. Chem. 1982, 21, 1037.
(22) Champness, N. R.; Levason, W.; Pletcher, D.; Webster, M. J. Chem.
Soc., Dalton Trans. 1992, 3243.
(23) Brown, M. P.; Burns, D.; Das, R.; Dolby, P. A.; Harding, M. M.;
Jones, R. W.; Robinson, E. J.; Smith, A. K. J. Chem. Soc., Dalton
Trans. 1991, 351.
(24) Higgins, S. J.; McCart, M. K.; McElhinney, M.; Nugent, D. C.; Pounds,
T. J. J. Chem. Soc., Chem. Commun. 1995, 2129.
(25) Colquhoun, I. J.; McFarlane, W. J. Chem. Soc., Dalton Trans. 1982,
1915.
(26) Hallman, P. S.; Stevenson, T. A.; Wilkinson, G. Inorg. Synth. 1970,
12, 237.
(c) Isomerization of 1 to cis-[RuCl2(dppen)2] (2). 1 (3.24 g, 3.35
mmol) in chlorobenzene (150 cm3) was heated to reflux for 16 h under
nitrogen; samples were withdrawn periodically and the 31P{1H} NMR
spectra recorded. Once no further reaction occurred, solvent was
removed in Vacuo. The product was extracted with MeOH (200 cm3),
the extract was filtered, the filtrate was evaporated to dryness, and the
(27) Perrin, D. D.; Armarego, W. L. F. The Purification of Laboratory
Compounds, 3rd ed.; Pergamon: Oxford, U.K., 1988.
(28) Casado, C. M.; Mora´n, M.; Losada, J.; Cuadrado, I. Inorg. Chem.
1995, 34, 1668.

6190 Inorganic Chemistry, Vol. 36, No. 27, 1997
Barkley et al.
residue was dried in Vacuo. The complex was stored in the dark.
Yield: 1.04 g, 32%. Anal. Calcd for C52H44Cl2P4Ru‚H2O: C, 63.55;
H, 4.72. Found: C, 63.33; H, 4.60. FAB MS, m/z: 965 (100) M+;
930 (27) [M - Cl]+; 894 (16) [M - Cl - HCl]+. Selected IR data
(cm-1): 3400 (br) (H2O); 300 (w) and 270 (w) (νRu-Cl). Electronic
spectral data, Emax/10-3 cm-1 (CH2Cl2): 22.73 (sh), 27.17. 31P NMR:
added n-hexylamine (0.5 cm3, 9.2 mmol) with stirring. After 16 h,
solvent was removed in Vacuo, and the yellow oil remaining was
triturated with hexanes. The product was filtered off and dried in Vacuo.
4b. To a mixture of dry toluene (20 cm3) and 2 (0.296 g, 0.307
mmol) in a foil-covered flask under N2 was added (3-aminopropyl)-
triethoxysilane (1.6 cm3, 6.7 mmol) with stirring. After 24 h, the solvent
was evaporated off and the yellow oil remaining was triturated with
hexanes. The product was filtered off and dried in Vacuo.
1
δ 13.84 (s, P-trans-P), -3.57 (s, P-trans-Cl). Selected H NMR
data: δ 8.25 (m, 4H, aromatic H), 7.5-6.7 (m, 28H, aromatic H), 6.52
(dt, 8H, aromatic H), 6.02 (4H, complex m, CdCH2). Ru(II)/Ru(III)
E1/2 (CH2Cl2/0.2 M TBAT): +0.41 V.
4a. Yield: 0.23 g, 63%. 31P NMR: δ 12.41, 27.27 (two t, AA′XX′;
1
|JAX + JAX′| ) 59 Hz). Selected H NMR data: δ 7.99 (m, br, Ph),
(d) Primary Amine Adducts of 1: 3a-e. 1 (0.2 g) was suspended
in toluene (10 cm3), and the appropriate RNH2 (1.0 cm3) was added.
The mixture was stirred for 16 h. The orange suspension dissolved to
give a yellow solution. The toluene was removed under reduced
pressure, to ca. 5 cm3, and the product was left to crystallize at room
temperature for 48 h (R ) -CH2Ph (3a)). Alternatively, the product
was precipitated with methanol (R ) -(CH2)3NH2 (3b), n-octyl (3c),
R-R-CH(Me)Ph (3d)) or dry diethyl ether (R ) -(CH2)3Si(OEt)3 (3e)).
The product was filtered off, washed with a little diethyl ether or
methanol, and dried in Vacuo.
7.84 (m, Ph), 7.58 (m, Ph), 7.29 (m, Ph), 7.11 (m, Ph), 6.71 (m, Ph)
(total: 40H), 5.32 (2H, br, P2CHCH2), 2.30 (6H, br, overlapping m’s,
P2CHCH2 and NH), 1.60 (4H, br m, -NHCH2C4H8Me), 1.25 (m, 16H,
-NHCH2C4H8Me), 0.81 (m, 6H, -NHCH2C4H8Me).
4b. Yield: 0.28 g, 65%. FAB MS, m/z: 1406 (85) M+; 1371 (100)
[M - Cl]+; 1335 (12) [M - Cl - HCl]+. 31P NMR (obtained on crude
product; see Results and Discussion): δ 12.30, 27.23 (two t, AA′XX′;
|JAX + JAX′| ) 58 Hz). 1H NMR data (obtained on crude product; see
Results and Discussion): δ 8.08 (br m, Ph), 7.89 (br m, Ph), 7.61 (br
m, Ph), 7.28 (m, Ph), 6.75 (br m, Ph), 5.38 (br m, P2CHCH2), 3.78 (m,
JHH 5.8 Hz, -Si{OCH2CH3}3), 2.67 (t, JHH 7.1 Hz, -NHCH2CH2-),
2.20 (complex m, P2CHCH2), 1.53 (m, -NHCH2CH2CH2-), 1.22 (m,
-Si{OCH2CH3}3), 0.62 (m, -CH2Si-).
3a. Yield: 78%. Anal. Calcd for C66H62Cl2N2P4Ru‚0.5C7H8: C,
68.23; H, 5.39; N , 2.29. Found: C, 68.20; H, 5.46; N, 2.05. FAB
MS, m/z: 1178 M+; 1143 [M - Cl]+; 1107 [M - Cl - HCl]+. 31P
NMR: δ 12.95 (s). Selected 1H NMR data: δ 5.49 (2H, m, P2CHCH2),
3.62 (4H, s, NHCH2Ph), 3.11 (2H, m, P2CHCH2, JHH 6.4 Hz, apparent
JPH 5.85 Hz), 1.53 (br s, NH). Ru(II)/Ru(III) E1/2: -0.02 V.
3b. Yield: 52%. Anal. Calcd for C58H64Cl2N4P4Ru‚2H2O: C,
60.63; H, 5.97; N, 4.88. Found: C, 61.02; H, 5.77; N, 4.82. FAB
MS, m/z: 1112 M+; 1077 [M - Cl]+; 1042 [M - Cl - HCl]+.
Electronic spectral data, Emax/10-3 cm-1 (ꢀ/L mol-1 cm-1) (CH2Cl2):
20.58 (110), 23.92 sh (210), 31.85 sh (2370). 31P NMR: δ 10.70 (s).
(g) Acetylide Adduct 5. 1 (0.3 g, 0.31 mmol) was dissolved/
suspended in THF (10 cm3). A solution of the acetylide (prepared from
hexyne [0.229 g, 2.78 mmol] and n-BuLi [1.17 mL of 1.6 M; 1.87
mmol]) in THF (2 cm3) was added Via cannula at -78 °C, and the
mixture was then allowed to warm to room temperature. The color
lightened, and the suspended complex dissolved. Solvents were
removed under reduced pressure, and the resulting light brown oil
was triturated with MeOH at 0 °C. The tan solid 5 was filtered
off and dried in Vacuo. Yield: 0.15 g, 42%. Anal. Calcd for
C72H78Cl2N2P4Ru‚0.25CH2Cl2: C, 72.93; H, 7.06. Found: C, 72.08;
H, 6.47. FAB MS, m/z: 1221 (7) M+; 1139 (67) [M - HCtCR]+;
1057 (100) [M - 2HCtCR]+; 975 (15) [M - 3HCtCR]+. 31P
1
Selected H NMR data: δ 5.38 (2H, m, P2CHCH2) 2.96 (4H, m, P2-
CHCH2), 2.55 (4H, t, JHH 6.0 Hz, -NHCH2CH2), 2.43 (4H, t, JHH 6.0
Hz, -CH2CH2NH2), 1.41 (4H, qnt, -CH2CH2CH2-), 1.25 (ca. 10H,
br s, NH, H2O). Ru(II)/Ru(III) E1/2: 0.00 V.
1
NMR: δ 8.01 (s). Selected H NMR data: δ 7.90, 7.62, 7.22, 7.06
3c. Yield: 56%. Anal. Calcd for C68H82Cl2N2P4Ru‚2MeOH: C,
65.30; H, 7.05; N, 2.18. Found: C, 65.01; H, 6.90; N, 2.08. FAB
MS, m/z: 1222 (100) M+; 1187 (31) [M - Cl]+; 1147 (11) [M - Cl
(m’s, Ph), 5.50 (m, apparent JPH 6.0 Hz, JHH 4.9 Hz, P2CHCH2), 3.49
(d, JHH 4.9 Hz, P2CHCH2), 2.25, 2.02 (br m’s, CtCCH2R), 1.52
(overlapping m’s, CtCCH2CH2CH2CH3), 1.28 (overlapping m’s,
CtCCH2CH2CH2CH3), 0.81 and 0.73 (overlapping t’s, JHH 6.9 Hz,
-CH2CH3).
(h) Addition of Primary Amines to [CpRu(dppen)Cl] (6). To
[CpRu(dppen)Cl] (6) (0.1 g, 0.17 mmol) in toluene (15 cm3) was added
H2N(CH2)3NH2 (7a; 1 cm3) or PhCH2NH2 (7b; 0.043 g, 0.4 mmol).
The mixture was allowed to stand for 16 h. Solvent was removed in
Vacuo, and the mixture was triturated with MeOH (7a) or Et2O (7b).
The product was filtered off and dried in Vacuo. Yields: 0.018 g, 22%
(7a); 0.017 g, 20% (7b).
1
- HCl]+. 31P NMR: δ 12.70 (s). Selected H NMR data: δ 5.45
(2H, m, P2CHCH2), 3.02 (4H, m, P2CHCH2), 2.40 (4H, t, JHH 6.7 Hz,
NHCH2C7H15), 1.40-1.19 (24H, overlapping m’s, NHCH2C6H12CH3),
0.86 (6H, t, JHH 6.6 Hz, NHC7H14CH3). Ru(II)/Ru(III) E1/2: -0.01 V.
3d. Yield: 57%. Anal. Calcd for C68H66Cl2N2P4Ru‚H2O: C, 66.66;
H, 5.60; N, 2.29. Found: C, 66.46; H, 5.59; N, 2.15. FAB MS, m/z:
1206 (100) M+; 1171 [M - Cl]+; 1135 [M - Cl - HCl]+. 31P NMR:
δ 9.14 (m). Selected 1H NMR data: δ 5.45 (2H, m, P2CHCH2), 3.57
(2H, quartet, JHH 6.48 Hz, NHCH(CH3)Ph), 2.97 (4H, br m, P2CHCH2),
1.54 (br s, NH and H2O), 1.14 (6H, d, NHCH(CH3)Ph).
7a. Anal. Calcd for C34H37ClN2P2Ru: C, 60.75; H, 5.56; N, 4.17.
3e. Yield: 65%. Anal. Calcd for C70H90Cl2O6N2P4RuSi2: C, 59.78;
H, 6.45; N, 1.99. Found: C, 58.20; H, 6.40; N, 2.15. FAB MS, m/z:
1409 (100) M+; 1374 (46) [M - Cl]+. 31P NMR: δ 10.89 (s). Selected
1H NMR data: δ 5.40 (m, P2CHCH2), 3.70 (q, JHH 6.9 Hz, Si(OCH2-
CH3)3), 3.00 (br m, P2CHCH2), 2.40 (t, -NHCH2CH2CH2Si), 1.60 (br,
NH), 1.41 (br m, -NHCH2CH2CH2Si), 1.28 (t, Si(OCH2CH3)3), 0.47
(m, -NHCH2CH2CH2Si). Ru(II)/Ru(III) E1/2: +0.03 V.
Found: C, 58.41; H, 5.63; N, 4.62. FAB MS, m/z: 673 (50) M+; 637
1
(20) [M - Cl]+. 31P NMR: δ 33.2 (s). Selected H NMR data 7.82
(m, Ph) and 7.40 (m, Ph), 4.89 (s, C5H5), 4.83 (complex m, P2CHCH2),
2.90 (t, JHH 6.3 Hz, HNCH2(CH2)2NH2), 2.70 (complex m, P2CHCH2),
2.55 (t, JHH 6.3 Hz, HN(CH2)2CH2NH2), 1.50 (qnt, HNCH2CH2CH2-
NH2), 1.40 (br s, NH and OH).
7b. Anal. Calcd for C38H36ClNP2Ru: C, 63.10; H, 5.30; N, 1.94.
Found: C, 63.20; H, 5.56; N, 2.21. FAB MS, m/z: 705 (100) M+;
670 (30) [M - Cl]+; 598 (10) [M-PhCH2NH2]. 31P NMR: δ 33.0
(e) Secondary Amine Adducts of 1: 3f,g. 1 (0.2 g) was dissolved/
suspended in toluene (10 cm3), and the amine (3 cm3) was added. The
mixture was stirred for 16 h (pyrrolidine, 3f) or refluxed gently for 30
min (morpholine, 3g). Solvent was removed under reduced pressure,
and the product was precipitated by addition of methanol. It was filtered
off, washed with methanol, and dried in Vacuo.
1
(s). Selected H NMR data: δ 7.87 (m, Ph) and 7.37 (m, Ph), 4.84
(6H, s and overlapping m, C5H5 and P2CHCH2), 3.58 (2H, s, PhCH2-
NH), 2.65 (2H, dt, JHH 7.2 Hz, JPH 11.7 Hz, P2CHCH2), 1.56 (3H, br
s, NH and OH).
Derivatization of Electrodes with 3e. The appropriate electrodes
(ITO-coated glass or anodized Pt disks) were soaked in a solution of
3e (10 mM) in chlorobenzene for 24-48 h under nitrogen. They were
then rinsed thoroughly with CH2Cl2, dried, and stored under dust-free
conditions prior to cyclic voltammetry experiments.
3f. Yield: 69%. Anal. Calcd for C60H62Cl2N2P4Ru: C, 65.10; H,
5.65; N, 2.53. Found: C, 64.60; H, 5.69; N, 2.41. FAB MS, m/z:
1107 (100) M+; 1072 [M - Cl]+; 1036 [M - Cl - HCl]+. 31P NMR:
δ 11.85 (s). Selected 1H NMR data: δ 5.49 (2H, m, P2CHCH2), 2.84
(4H, m, P2CHCH2), 2.44 (8H, br s, pyrrolidine ring H), 1.80 (8H, br s,
pyrrolidine ring H).
3g. Yield: 79%. FAB MS, m/z: 1139 M+; 1104 [M - Cl]+, 1068
Structure Determination. Data were collected using a Rigaku
AFC6S diffractometer. Cell dimensions and intensities were determined
using ω/2θ scans. Structures were solved by direct methods,29,30 and
1
[M - Cl - HCl]+. 31P NMR: δ 12.24 (s). Selected H NMR data:
δ 5.52 (2H, m, P2CHCH2).
(f) Primary Amine Adducts of 2. 4a. To a mixture of chloroben-
zene (20 cm3) and 2 (0.271 g, 0.279 mmol) in a foil-covered flask was
(29) Gilmore, C. J. J. Appl. Crystallogr. 1984, 17, 42.

Redox-Active Ru(II)-Diphosphine Complexes
Table 1. Crystallographic Data for 3b‚4MeOH
Inorganic Chemistry, Vol. 36, No. 27, 1997 6191
Results and Discussion
empirical formula C62H80Cl2N4O4P4Ru
a ) 11.448(6) Å
b ) 13.10(1) Å
c ) 11.178(7) Å
R ) 93.16(6)°
fw 1241.21
Synthesis and Oxidation of trans-[RuCl2(dppen)2] (1).
Treatment of [RuCl2(PPh3)3] in CH2Cl2 with 2 mol equiv of
dppen gave orange crystals of trans-[RuCl2(dppen)2]‚CH2Cl2
(1) in over 85% yield. Complex 1 is sparingly soluble in toluene
and is more soluble in chlorinated solvents. It is insoluble in
acetone, acetonitrile, dmf, diethyl ether, and ethanol. It was
reported previously as a serendipitous product of the reaction
of [Ru2(O2CCH3)4Cl] with dppen in the presence of LiCl, and
its crystal structure was determined.35 Oxidation of 1 in CH2-
Cl2 with NOBF4 gave dark green [RuCl2(dppen)2]BF4 (1+). This
had satisfactory microanalyses, an FAB mass spectrum similar
to that of 1, and an IR spectrum similar to that of 1 except for
space group P1h (No. 2)
T ) 25 °C
λ ) 0.710 69 Å
F
calcd ) 1.250 g cm-3
â ) 99.51(5)°
µ(Mo KR) ) 4.53 cm-1
Ra ) 0.063
γ ) 92.29(5)°
V ) 1648(2) Å3
Z ) 1
Rwb ) 0.070
2
a R ) ∑||Fo| - |Fc||/∑|Fo|. b Rw ) [∑w(|Fo| - |Fc|)2/∑wFo ]1/2
.
Table 2. Significant Bond Lengths (Å) and Angles (deg) for 3b
Bond Distances
-
an additional very strong band at 1055 cm-1 due to the BF4
Ru(1)-Cl(1)
Ru(1)-P(1)
Ru(1)-P(2)
P(1)-C(5)
P(1)-C(10)
P(1)-C(25)
2.432(4)
2.349(4)
2.365(4)
1.82(1)
1.82(2)
1.85(1)
P(2)-C(13)
P(2)-C(19)
P(2)-C(25)
C(25)-C(26)
C(26)-N(1)
N(1)-C(27)
1.83(1)
anion. The electronic spectrum of 1+ in CH2Cl2 is very similar
to that of trans-[RuCl2(dppm)2]BF4.22 This, and the IR spec-
trum, suggests that the trans geometry is retained on oxidation.
Complex 1+ reverts to 1, slowly in the solid state but much
more rapidly in solution, making reliable extinction coefficients
impossible to determine.
1.85(1)
1.86(1)
1.49(2)
1.47(2)
1.45(2)
Bond Angles
Cl(1)-Ru(1)-P(1)
Cl(1)-Ru(1)-P(2)
P(1)-Ru(1)-P(2)
Ru(1)-P(1)-C(5)
Ru(1)-P(1)-C(10)
Ru(1)-P(1)-C(25)
C(5)-P(1)-C(10)
C(5)-P(1)-C(25)
Ru(1)-P(2)-C(13)
95.5(1)
98.9(1)
71.5(1)
123.1(5)
123.7(5)
94.8(5)
100.2(7)
107.0(6)
118.4(5)
Ru(1)-P(2)-C(19)
Ru(1)-P(2)-C(25)
C(13)-P(2)-C(19)
C(13)-P(2)-C(25)
C(19)-P(2)-C(25)
P(1)-C(25)-P(2)
P(1)-C(25)-C(26)
P(2)-C(25)-C(26)
N(1)-C(26)-C(25)
127.9(5)
94.1(5)
100.3(6)
109.4(7)
105.0(6)
95.8(6)
120(1)
Synthesis of Amine Adducts of 1: 3a-g. We wished to
examine the reactivity of 1 toward different nucleophiles in order
to assess its utility in preparing metal-diphosphine complexes
suitable for oxide surface derivatization. First, we explored
reactions of 1 with primary amines. When 1 was suspended in
toluene, or partially dissolved in chlorobenzene, and treated with
a large (g50-fold) excess of primary amine at room temperature,
the color paled from orange to yellow and complete dissolution
occurred over a period ranging from minutes to 16 h. The
yellow solids isolated after workup were the hoped-for adducts
trans-[RuCl2({Ph2P}2CHCH2NHR)2] (R ) PhCH2, 3a; R )
(CH2)3NH2, 3b; R ) n-octyl, 3c; R ) R-R-CH(Me)Ph, 3d; R
) (CH2)3Si(OEt)3, 3e). The adducts 3a-e were soluble in CH2-
Cl2, CHCl3, and chlorobenzene (a particularly useful solvent
because it does not react with amines), sparingly soluble in
aromatic solvents, but insoluble in polar aprotic solvents such
as CH3CN. The more polar 3b was sparingly soluble in EtOH.
Secondary amines did not add so readily to 1. Pyrrolidine
added at room temperature under essentially the same conditions
as for the primary amines, to give trans-[RuCl2({Ph2P}2CHCH2-
NC4H8)2], 3f. However, the addition of morpholine to 1a to
give 3g required refluxing in toluene for 30 min. The adduct
was not analytically satisfactory, although it displayed the
expected M+, [M - Cl]+, and [M - Cl - HCl]+ clusters in its
FAB mass spectrum and had a singlet in its 31P{1H} NMR
124(1)
112(1)
successive difference Fourier syntheses were used to locate all non-
hydrogen atoms, followed by final refinements using full-matrix least-
squares procedures. The TEXSAN crystallographic software package,31
running on a microVAX computer, was used for computation of the
structure, and ORTEP32 was used to plot the molecular diagrams.
Scattering factors and corrections for anomalous dispersion33 were taken
from standard references.34 Experimental parameters and crystal data
are given in Table 1, and selected bond lengths and angles are listed in
Table 2. Full listings of H atom coordinates, bond distances and angles,
and anisotropic thermal parameters have been deposited with the
Cambridge Crystallographic Data Centre, University Chemical Labora-
tory, Lensfield Road, Cambridge CB2 1EW, U.K.
3b‚4MeOH. Orange crystals of 3b suitable for X-ray diffraction
were obtained on layering a solution in toluene and excess 1,3-
diaminopropane with MeOH. A crystal (0.10 × 0.10 × 0.25 mm) was
mounted on a fiber and used for measurements of precise cell constants
and collection of intensity data. Of 4298 unique reflections scanned
in the range 2 < θ < 25°, 1913 with F > 3.0σ(F) were used for
structure solution and refinement. During data collection, three standard
reflections, measured after every 150 reflections, declined by 1.01%.
A linear correction factor was therefore applied to the data. Diffraction
intensities were corrected for Lorentz and polarization effects.
1
spectrum (below). The H NMR spectrum of 3g showed the
presence of free morpholine in the sample, which could not be
removed by recrystallization or vacuum-drying.
With diethylamine in chlorobenzene at room temperature, no
reaction with 1 occurred, and on prolonged reflux, only a 20:
80 mixture of trans- and cis-[RuCl2(dppen)2] was obtained; no
addition to the dppen double bonds occurred. The same result
was obtained with the less volatile amine HN(CH2CH2OCH3)2.
It is notable that secondary amines also underwent more sluggish
addition to [M(CO)4(dppen)] (M ) Cr, Mo, W) than did primary
amines.36 It is possible that steric factors are important here,
since the bulky primary amine (CH3)3CCH2NH2 also failed to
add to 1.
Non-hydrogen atoms were refined anisotropically. Hydrogen atoms
were placed in idealized calculated positions and allowed to ride on
their corresponding atom. Refinement converged at R ) 0.063 (Rw )
0.070). The final electron density difference synthesis showed no peaks
>0.51 or <-0.42 e Å-3. The two MeOH solvate molecules in each
asymmetric unit were distributed over three sites, and refinement was
completed with 66.67% occupancy for each site.
(30) Beurskens, P. T. DIRDIF: Direct Methods for Difference Structuressan
automatic procedure for phase extension and refinement of difference
structure factors; Crystallography Laboratory: Toernooiveld, The
Netherlands, 1984.
(31) TEXSAN-TEXRAY Structure Analysis Package; Molecular Structure
Corp.: The Woodlands, TX, 1985.
We then tried adding secondary amines to 1+ in the hope
that the more polarizing Ru(III) center would increase the
reactivity of coordinated dppen toward nucleophiles; it is known
that the Pt(IV) complex [PtIMe3(dppen)] is significantly more
(32) Johnson, C. K. ORTEPII; Oak Ridge National Laboratory: Oak Ridge,
TN, 1976.
(33) Ibers, J. A.; Hamilton, W. C. Acta Crystallogr. 1964, 17, 781.
(34) Cromer, D. T.; Waber, J. T. International Tables for X-Ray Crystal-
lography; The Kynoch Press: Birmingham, U.K., 1974; Vol. IV.
(35) Cotton, F. A.; Diebold, M. P.; Matusz, M. Polyhedron 1987, 6, 1131.
(36) Cooper, G. R.; Hassan, F.; Shaw, B. L.; Thornton-Pett, M. J. Chem.
Soc., Chem. Commun. 1985, 614.

6192 Inorganic Chemistry, Vol. 36, No. 27, 1997
Barkley et al.
reactive than [PtMe2(dppen)].16 However, treatment of 1+ with
primary or secondary amines gave exclusively Ru(II) products
with no apparent difference in reactivity.
The robust six-coordinate Ru(II) complexes 3a-g are po-
tentially suitable for further functionalization. In preliminary
experiments, we have found that 3b reacts with p-toluenesulfo-
nyl chloride in CH2Cl2 to afford the corresponding insoluble
tetrasulfonamide derivative (characterized by microanalyses,
infrared spectroscopy, and FAB MS only). With salicylalde-
hyde, the primary amine groups of 3b react to afford the
corresponding bis(imine), which is also too insoluble for NMR
spectroscopy. In principle, this opens the way to incorporation
of 3b in polymers (Via condensation reactions) and supramo-
lecular multimetallic redox centers (utilizing the salicylideamine
groups as metal receptors), although such studies may be
hampered by the limited solubility of the derivatives. There is
also much current interest in polymeric siloxanes incorporating
redox-active complexes,28 and 3e would be ideal for incorpora-
tion in such materials.
Figure 1. 1H NMR spectrum (0-6 ppm region) of 3b (CDCl3, 400
MHz). Peaks (from low field): P2CHCH2 (m), P2CHCH2 (m), CH2-
NHCH2 (t), CH2NH2 (t), CH2CH2CH2 (quintet), NH/NH2 (br s).
Spectroscopic Characterization of Complexes 1 and 3a-
g. Fast atom bombardment mass spectrometry (FAB MS) of
complexes 1 and 3a-g revealed molecular ions as the base
peaks, together with [M - Cl]+ and, usually, [M - Cl - HCl]+
clusters (see Experimental Section). The 31P{1H} NMR spec-
trum of 1 showed a singlet, at δ ) +14.7 ppm, consistent with
trans geometry. The 31P{1H} NMR spectra of 3a-g also
showed singlets (except 3d), shifted slightly upfield from the
value for 1. The 31P{1H} NMR spectrum of 3d showed a triplet
pattern centered at δ ) 9.14 ppm, with a separation of 35 Hz
between the outer lines; this is because the two phosphorus
nuclei in each ligand are rendered slightly inequivalent by the
chiral amine substituent.
1
The H NMR spectrum of 1 showed aromatic resonances at
7.5-7.1 ppm and a singlet due to the CH2Cl2 solvate. The
CdCH2 protons resonate as a pseudoquintet (apparent JPH 12.6
Hz) at δ ) 6.18 ppm. This is a “virtual” quintet37 and arises
as a result of the strong P-trans-P coupling (300-350 Hz)
typically observed for Ru(II)-phosphine complexes.38,39 The
four CdCH2 protons and the four phosphorus nuclei formally
comprise an AA′A′′A′′′XX′X′′X′′′ spin system. On the addition
of nucleophiles to give 3a-g, this resonance is replaced by
multiplets at ca. δ ) 5.4 (1H) and 3.0 ppm (2H) for the P2-
CHCH2 moiety, with both types of proton coupled to each other
and to the phosphorus nuclei (“virtual” coupling), in addition
to resonances characteristic of the amine fragment. The
appearance of these spectra is exemplified by that of 3b, shown
in Figure 1. Like the Pd(II) complexes studied earlier,17 the
adducts crystallize with significant amounts of solvent or water
of crystallization, which cannot readily be removed by drying
Figure 2. Molecular structure of 3b.
possible that this particular stereoisomer is thermodynamically
favored for these adducts for steric reasons. We have since
obtained the crystal structure of an alcohol adduct of 1, trans-
[RuCl2{(Ph2P)2CHCH2OCH2CH2C4H3S-3}2], and this is simi-
larly centrosymmetric.42 It should be noted that 3b crystallized
as 3b‚4MeOH with the MeOH molecules distributed over three
sites per asymmetric unit in the crystal lattice, and refinement
was completed with 66.67% occupancy of each site, which
probably accounts for the rather high Rw. In spite of the
presence of these solvate molecules, no evidence was found
for significant hydrogen bonding between the amine groups in
3b and the solvate molecules.
1
in Vacuo. This is also evident in the H NMR spectra.
Crystal Structure of 3b. Crystals of 3b suitable for X-ray
structure determination were obtained from CH2Cl2/MeOH by
diffusion. The molecular structure is shown in Figure 2;
significant bond lengths and angles are given in Table 2. As
also found for the analogous complexes trans-[RuCl2(dppm)2],40
trans-[RuCl2(Me2C{PPh2}2)2],41 and 1,35 there is a crystallo-
graphically-imposed center of symmetry at Ru for 3b. It is
In 1, the P-C(dCH2)-P angle is 98.6(1)°,35 compared with
118.96(6)° for uncomplexed dppen,43 a difference of 20.4°. In
3b, the P-CH(CH2R)-P angle is 95.8(6)°. Although a free
ligand closely analogous to the diphosphine unit in 3b (for
instance, HMeC(PPh2)2) has yet to be structurally characterized,
it is unlikely that the P-C-P angle in such a ligand would be
(37) Gu¨nther, H. NMR Spectroscopy: Basic Principles, Concepts, and
Applications in Chemistry, 2nd ed.; Wiley: Chichester, U.K., 1995.
(38) Whinnery, L. L.; Yue, H. J.; Marsella, J. A. Inorg. Chem. 1986, 25,
4136.
(39) Lindner, E.; Mo¨ckel, A.; Mayer, H. A.; Ku¨hbauch, H.; Fawzi, R.;
Steimann, M. Inorg. Chem. 1993, 32, 1266.
(40) Chakravarty, A. R.; Cotton, F. A.; Schwotzer, W. Inorg. Chim. Acta
1984, 84, 179.
(41) Barkley, J. V.; Grimshaw, J. C.; Higgins, S. J.; Hoare, P. B.; McCart,
M. K.; Smith, A. K. J. Chem. Soc., Dalton Trans. 1995, 2901.
(42) Higgins, S. J.; McCart, M. K.; Pounds, T. Unpublished work.
(43) Schmidbauer, H.; Herr, R.; Riede, J. Chem. Ber. 1984, 117, 2322.

Redox-Active Ru(II)-Diphosphine Complexes
Inorganic Chemistry, Vol. 36, No. 27, 1997 6193
larger than that found in free dppm, namely 106.16(5)°.44
Therefore the angle strain has been reduced from 20.4° for 1
to, at most, 10.4° for 3b as a consequence of nucleophilic
addition to dppen. The reduction in angle strain is the likely
driving force for the reaction. Both the metal-ligand distances
and the bond lengths and angles in the four-membered chelate
rings in 2b are very similar to those in trans-[RuCl2(dppm)2].40
Reactions of 1 with Carbanions. The reactivity of 1 toward
amine nucleophiles clearly mirrors that of the isoelectronic
[M(CO)4(dppen)] (M ) Cr, Mo, W). The latter are known to
react with carbanions,13 and we therefore investigated the
reaction of 1 with these nucleophiles. On treatment of an orange
suspension of 1 in THF with a 6-fold molar excess of LiCtC-
(CH2)3CH3 in hexanes/Et2O at ca. -20 °C, the mixture paled
to almost colorless, and all the complex dissolved. Removal
of solvent and trituration with cold MeOH afforded a tan solid
5. The 31P{1H} NMR spectrum of 5 showed a singlet at +17.2
ppm, unusually somewhat downfield of the resonance for 1.
Microanalytical and FAB mass spectral data were consistent
with the formula trans-[Ru(CtCR)2({Ph2P}2CHCH2CtCR)2]
(R ) nBu) for 5. In particular, the FAB mass spectrum showed
a weak cluster of peaks due to M+ and stronger peaks
corresponding to progressive loss of HCtCnBu fragments. No
peaks due to loss of chloride ions could be discerned. Moreover,
Figure 3. 31P NMR spectrum of 2 (CDCl3, 24.3 MHz, line width ca.
3 Hz). The starred peak is due to an impurity of 1 in this sample.
dium diethylmalonate in dry thf adds readily to [M(CO)4(dppen)]
(M ) Mo, W).13
Synthesis and Chemistry of cis-[RuCl2(dppen)2] (2) and
Amine Adducts 4a,b. On reflux in chlorobenzene for 16 h,
with changes being monitored by 31P{1H} NMR spectroscopy,
1 gave a 10:90 mixture of 1 and another complex, which still
analyzed satisfactorily for [RuCl2(dppen)2]. The latter is cis-
[RuCl2(dppen)2] (2). That this 10:90 mixture of 1 and 2 is the
thermodynamic product is strongly suggested by the fact that
prolonged (>48 h) reflux in chlorobenzene (bp 132 °C) or 1,2-
dichlorobenzene (bp 180 °C) resulted in no further change. Pure
2 was obtained by extracting the crude product with MeOH (in
which 2 is sparingly soluble) as a pale yellow solid. In CDCl3,
on exposure to laboratory light, 2 reverts completely to 1 over
a period of a few days. The complex was therefore handled
and stored in the dark. Similar thermal conversion of trans-
[RuCl2(dppm)2] (dppm ) Ph2PCH2PPh2) to cis-[RuCl2(dppm)2]
(16 h reflux in 1,2-dichloroethane), and photochemical conver-
sion of cis-[RuCl2(dppm)2] to trans-[RuCl2(dppm)2] in solution,
has been reported.21
1
an analytical test for chloride proved negative. The H NMR
spectrum showed resonances characteristic of the ligand P2-
CHCH2- moiety (unusually, coupling between the P2CHCH2
protons and the phosphorus nuclei was too weak to be observed
in this complex) and distinct resonances for the two (Ru-CtCR
and P2CHCH2CtCR) sets of n-butyl environments; we do not
know which is which. We also obtained an off-white solid on
treatment of 1 with LiCtCPh, but the insolubility of this product
has hindered characterization. The FAB mass spectrum showed
clusters of peaks at M+ and [M - HCtCPh]+, consistent with
the formulation [Ru(CtCPh)2({Ph2P}2CHCH2CtCPh)2], but
also showed weak peaks at higher mass.
There is considerable current interest in metal-acetylide
conjugated polymers, including {[Ru(diphos)2(-CtC-R-
CtC-)]}n (diphos ) dppe, Et2PCH2CH2PEt2), but these are
normally made from the chlorides using organotin acetylides;45
lithium acetylides have not commonly been used to prepare Ru-
(II)-acetylide complexes. Attempts to react 1 with acetylenes
(BunCtCH, propargyl alcohol) and tertiary amine bases [NEt3
or 1,8-bis(dimethylamino)naphthalene] in the presence of CuI
failed to yield acetylide complexes, although similar reactions
have been employed successfully using cis-[RuCl2(dppm)2].46
The more electron-rich complexes trans-[RuCl2(L-L)2] (L-L
) R2PCH2CH2PR2; R ) Me, Et) are reported to react with
terminal acetylenes R′CtCH in alcohol in the presence of
sodium alkoxide to give trans-[Ru(CtCR′)2(L-L)2],47 but 1
did not react under these conditions.
The addition of amines to 2 was more sluggish than the
addition to 1, and steric requirements appear more stringent.
Thus, adducts with (n-hexyl)NH2 (4a) and with (EtO)3Si(CH2)3-
NH2 (4b) were successfully isolated. Interestingly, whereas 4a
was soluble in chlorinated solvents, 4b was almost totally
insoluble, once isolated by hexane precipitation from solution
in chlorobenzene/excess amine during its synthesis, and ac-
1
cordingly the 31P{1H} and H NMR spectra for 4b (Experi-
mental Section) were obtained using CDCl3 solutions of the
crude oil remaining after pumping off chlorobenzene and most
of the excess amine. The 1H NMR spectrum showed resonances
due to 4b as detailed in the Experimental Section, together with
resonances due to residual (EtO)3Si(CH2)3NH2.
NMR Spectroscopic Characterization of 2 and 4a,b. The
31P{1H} NMR spectrum of 2 at 101.2 MHz showed two
apparent singlets of equal intensity at -3.56 and 13.85 ppm,
which we assign to P-trans-Cl and P-trans-P respectively,
since the latter resonance is close to that of 1. A spectrum
recorded at 24.3 MHz (JEOL FX60; ca. 3 Hz line width; Figure
3) revealed these to be, in fact, second-order multiplets,
consistent with the expected AA′XX′ spin system. The program
gNMR (Version 3.6r5, IvorySoft; Apple Macintosh Performa
6320) was used to model this behavior. The peak positions
and relative intensities could be approximately simulated by
employing the following values for the coupling constants of
Interestingly, other nucleophilic carbanions did not react with
1. For example, 1 was recovered in high yield from reactions
with sodium diethylmalonate or sodium acetylacetonate in
chlorobenzene/butanol and from attempts to react 1 with
H2C(CO2Et) in chlorinated solvents in the presence of 1,8-bis-
(dimethylamino)naphthalene. This was unexpected, since so-
(44) Schmidbauer, H.; Reber, G.; Schier, A.; Wagner, F. E.; Muller, G.
Inorg. Chim. Acta 1988, 147, 1988.
(45) Atherton, Z.; Faulkner, C. W.; Ingham, S. L.; Kakkar, A. K.; Khan,
M. S.; Lewis, J.; Long, N. J.; Raithby, P. R. J. Organomet. Chem.
1993, 462, 265.
(46) Pirio, N.; Touchard, D.; Dixneuf, P. H.; Fettouli, M.; Ouahab, L.
Angew. Chem., Int. Ed. Engl. 1992, 31, 651.
(47) Field, L. D.; George, A. V.; Hockless, D. C. R.; Purches, G. R.; White,
A. H. J. Chem. Soc., Dalton Trans. 1996, 2011.

6194 Inorganic Chemistry, Vol. 36, No. 27, 1997
Barkley et al.
1
The H NMR spectra of 4a and 4b are more complicated
than those of 3a-e, and the resonances due to the P2CHCH2-
NHR moiety appear as broad singlets; individual coupling to
phosphorus and to neighboring protons were not resolved. The
resonances due to the P2CHCH2NHR methylene, in particular,
are at significantly higher field than those of the corresponding
protons in 3a-e. It is likely that 4a and 4b are mixtures of at
least some of the eight possible diastereomers (Figure 4B), and
this could be responsible for the broad proton signals. We have
not yet been able to isolate X-ray-quality single crystals of 4a.
It did not prove possible to isolate adducts of 2 with secondary
amines, PhCH2NH2 or R-R-H2NCH(Me)Ph, although partial
addition did occur as revealed by 31P{1H} NMR spectroscopy;
two downfield triplets were seen in the 31P{1H} NMR spectra
of mixtures of 2 and excess PhCH2NH2, and four triplets
(partially overlapping) were seen in the 31P{1H} NMR spectra
of mixtures of 2 and excess R-R-PhCH(Me)NH2.
Electronic Spectra of 1, 2, and 3b. The complexes [RuX2-
(diphosphine)2] range in color from pale yellow to orange. The
electronic spectra of trans-[RuX2(diphosphine)2] usually show
two d-d bands in the visible region, a lower energy band
1
1
assigned as A1g f Eg and a higher energy band assigned as
1A1g f 1A2g, assuming approximate D4h symmetry.51 cis-[RuX2-
(diphosphine)2] complexes usually have d-d bands at somewhat
higher energy, which accounts for their generally paler color.21,51
The spectrum of orange 1 in CH2Cl2 shows a single, rather broad
and surprisingly intense (ꢀ ) 2240 L mol-1 cm-1) band at
24 150 cm-1. The electronic spectrum of a representative amine
adduct, 3b, in CH2Cl2, showed two much weaker d-d bands,
Figure 4. (A) Labeling used in the discussion of the spin system for
the 31P NMR spectrum of 2. (B) Possible isomers of amine adducts of
2. Since the point group of these is C2 or C1, each has a nonsuperim-
posable mirror image, giving eight possible diastereoisomers in total.
the AA′XX′ spin system appropriate for the 31P{1H} NMR
spectrum of cis-[RuCl2(dppen)2]) (Figure 4A): JAA′ +405 Hz,
JAX +40 Hz, JAX′ -44 Hz, and JXX′ -30 Hz. The absolute
values of these coupling constants are reasonable for octahedral
Ru(II).38,48 For the signals at 101.2 MHz to appear as singlets,
it is clear that JAX ()JA′X′) must be similar in magnitude, but
opposite in sign, to JAX′ ()JA′X) and JXX′. Whereas P-cis-P
coupling constants are negative if the coupling is exclusively
“through-the-metal”49 (typically ca. -35 Hz for Ru(II)50), in
chelate ligands the observed P-P coupling is the algebraic sum
of “through-the-metal” and “through-the-backbone” contribu-
tions. The “through-the-backbone” contribution is usually
positive for ligands containing a one-carbon bridge and can be
particularly large for dppen (ca. 100 Hz in Cr(0)-, Mo(0)-,
and W(0)-carbonyl complexes).49 The “through-the-backbone”
contribution in this complex is therefore ca. +84 Hz. This
results in a net positive P-cis-P coupling constant when the P
nuclei are connected by a CdCH2 backbone.
at 20 580 and 23 920 cm-1 (ꢀ ) 110 and 210 L mol-1 cm-1
,
1
1
1
1
respectively), assigned to the A1g f Eg and A1g f A2g
transitions; this spectrum is very similar to that of trans-[RuCl2-
(dppm)2].21 The decrease in extinction coefficient on amine
addition is reflected in the colors of the amine adducts, which
are all paler than 1. The electronic spectrum of 2 in CH2Cl2
shows d-d bands at 22 730 (shoulder) and 27 170 cm-1, at
higher energy than those for 1, but reliable extinction coefficients
could not be obtained since solutions of 2 invariably contain
some 1, as shown by 31P{1H} NMR spectroscopy and cyclic
voltammetry (below).
Solution Electrochemistry of Complexes 1-3e. In cyclic
voltammetry experiments in CH2Cl2, 1 and 3a-e showed
reversible Ru(II)/Ru(III) couples at ca. +0.01 V ((0.03 V;
Experimental Section), similar to those found earlier for a series
of trans-[RuCl2(L-L)2] complexes.22 Since such major varia-
tion as altering the donor atom or chelate ring size for the neutral
ligands has little effect on Ru(II)/Ru(III) potentials for this
family of complexes, it is not surprising that the addition of
nucleophiles to 1 has almost no effect on the Ru(II)/Ru(III)
potential. The voltammetry of 3a-e was complicated by the
precipitation of material from solution. Infrared evidence
suggested that quaternization of the amine group(s) by solvent
had occurred. The precipitates were too insoluble for NMR
characterization.
The cyclic voltammogram of 2 was investigated both in CH3-
CN/0.2 M tetraethylammonium tetrafluoroborate (TEAT), in
which it was sparingly soluble, and in CH2Cl2/0.2 M TBAT. In
CH3CN, a broad oxidation process, centered at +0.52 V, an
associated broad reduction wave at +0.37 V, and an oxidation
at +0.90 V were observed. The voltammogram is similar to
that observed earlier for cis-[RuCl2(dppm)2] in a CH3CN/CH2-
Cl2 mixture.21 The redox waves were assigned respectively to
the formation and re-reduction of chlorine from chloride ions
present as a result of solvolysis of cis-[RuCl2(dppm)2] to
1
The CdCH2 resonance in the H NMR spectrum of 2 is, as
expected, a complex symmetrical multiplet, centered at 6.02
ppm, partially overlapping the residual signal due to 1. Interest-
ingly, whereas the phenyl protons in the spectrum of 1 resonate
as closely-spaced overlapping multiplets from 7.5 to 7.1 ppm,
the phenyl protons of 2 resonate as a series of distinct multiplets
between 8.25 and 6.52 ppm, presumably because steric crowding
in the cis complex causes some of these protons to be in unusual
environments.
The 31P{1H} NMR spectra of 4a and 4b showed two triplets,
at ca. 27 and 12 ppm, with |JAX + JAX′| ) 58 Hz. Interestingly,
these are shifted considerably downfield (by ca. 15 ppm) from
the apparent singlets seen for 2, whereas the resonances for
3a-e were shifted upfield slightly from that due to 1. Also
notable is the reversion to the normal AA′XX′ “virtual” triplets
as the ligand backbone changes from P2CdCH2 to P2CHR.
(48) Chaudret, B.; Commenges, G.; Poilblanc, R. J. Chem. Soc., Dalton
Trans. 1984, 1635.
(49) Bookham, J. L.; Fontaine, X. L. R.; Kennedy, J. D.; McFarlane, W.
Inorg. Chem. 1988, 27, 1111.
(50) Baker, M. V.; Field, L. D. Inorg. Chem. 1987, 26, 2010.
(51) Klassen, D. M.; Crosby, G. A. J. Mol. Spectrosc. 1968, 25, 398.

Redox-Active Ru(II)-Diphosphine Complexes
Inorganic Chemistry, Vol. 36, No. 27, 1997 6195
Figure 6. CV of a monolayer of 3e on an anodized Pt electrode,
recorded in CH3CN/0.2 M TEAT.
anchored to ITO in similar fashion.45,53 The E1/2 value, +0.03
V (Vs ferrocene/ferrocenium), corresponds closely to those for
3a-d in CH2Cl2 solution. As an indirect check that the
anchoring did indeed involve reaction of the triethoxysilane
groups in 3e with the surface, we repeated the derivatization
process using 1, trans-[RuCl2(dppm)2], and 3a. No difference
was observed between the voltammetry of these and untreated
electrodes; the terminal triethoxysilyl groups are essential for
successful derivatization.
Figure 5. CV’s of a monolayer of 3e on an ITO electrode, recorded
in CH3CN/0.2 M TEAT. Sweep rates (with increasing peak currents)
were respectively 50, 75, 100, 150, 200, and 250 mV s-1. Peak currents
were linear with scan rate up to ca. 150 mV s-1
.
[RuCl(CH3CN)(dppm)2]Cl and to the Ru(II)/Ru(III) wave for
the cationic Ru(II) complex. Clearly, an analogous solvolysis
is occurring for cis-[RuCl2(dppen)2]. The only difference is that
whereas the Ru(II)/Ru(III) wave was reversible in CH3CN/CH2-
Cl2 for the dppm complex, it is totally irreversible for
[RuCl(CH3CN)(dppen)2]+ in CH3CN. The rather positive
oxidation potential, compared with that for 1, for example, arises
because cationic Ru(II) complexes are harder to oxidize than
neutral species and because the metal center has lost a π-donor
halide ligand which would stabilize the Ru(III) state.
The 31P{1H} NMR spectrum of a saturated solution of 2 in
CH3CN showed the formation of a single species, with complex
multiplets centered at -0.95, 13.60, and 20.70 ppm, in ap-
proximately a 2:1:1 intensity ratio. This suggests an AA′MX
spin system, consistent with a cis-[RuCl(CH3CN)(dppen)2]+
moiety. Cis stereochemistry had earlier been assumed for the
complex [RuCl(CH3CN)(dppm)2]Cl21 without supporting spec-
troscopic evidence.
In CH2Cl2, a very different voltammogram was observed.
Upon a positive scan from -0.4 V, a weak, reversible wave at
+0.01 V, due to the presence of traces of 1, was seen, together
with a much stronger redox wave centered at +0.41 V, which
we assign to the 2/2+ couple; for cis-[RuCl2(dppm)2], the Ru-
(II)/Ru(III) couple was similarly 0.37 V positive of the potential
for the trans isomer.21 The reversibility of the 2/2+ couple did
not vary appreciably as a function of scan rate (5-200 mV s-1).
Moreover, the small peaks due to the 1/1+ couple did not
increase in intensity on repetitive scanning through the 2/2+
couple. This suggests that cis-[RuCl2(dppen)2]+ is stable on
the voltammetric time scale, unlike cis-[RuCl2(dppm)2]+, which
was shown to undergo isomerization to trans-[RuCl2(dppm)2]+.21
Functionalization of Electrodes with 3e. Clean ITO-coated
glass was soaked in a chlorobenzene solution (10 mM) of 3e.
After thorough washing with dichloromethane and acetone, these
electrodes were examined by cyclic voltammetry in 0.2 M TEAT
in CH3CN, from ca. -0.7 to +0.3 V. A typical example is
shown in Figure 5. Although for an ideal surface-localized
redox process the cathodic and anodic current peaks should
occur at the same potential,9 and indeed this has been observed
for functionalized Pt electrodes,52 there is a peak-to-peak
separation of 40 mV in our experiment. However, this remains
invariant with scan rate up to 250 mV s-1 (Supporting
Information). Moreover, larger peak-to-peak separations have
been observed in the voltammetry of ferrocenes surface-
The unit cell volume of 3b is 1648(2) Å3 (Z ) 1). Since 3b
and 3e are very similar, the area occupied on the surface by a
molecule of 3e can be approximated by the product of two of
the unit cell dimensions of 3b. An upper limit for this is
therefore approximately 150 Å2. A lower limit for the charge
passed on oxidizing a monolayer of 3e in a 1-electron process
is therefore 10.7 µC cm-2; this assumes a smooth electrode
surface and close packing. The area under the anodic wave in
voltammograms of ITO electrodes modified with 3e was
consistently 15 ( 3 µC cm-2
. This is consistent with an
approximate monolayer coverage of the electrode by 3e; the
difference is accounted for by electrode roughness. Our
modified electrodes therefore have much in common with earlier
work using classical complexes such as Ru(II)-bipyridine
derivatives.3-5
The modified electrodes are highly stable to prolonged cycling
in acetonitrile. Although the anodic peak current for the
electrode of Figure 5 declined by 7% over 100 scans between
-0.7 and +0.3 V, it did not then significantly change after
another 1000 scans. Interestingly, although these complexes
are insoluble in water, the Ru(II)/Ru(III) redox process for
electrodes modified with a monolayer of 3e in aqueous buffer
(0.1 M Na2HPO4/0.1 M NaCl) was seen at the same formal
potential as in CH3CN, and the redox peak currents did not
appreciably decline over 100 scans in this medium.
Another approach often used to attach metal complexes to
oxide electrode surfaces has been to derivatize the surface with
an amine-terminated silyl anchor and then to couple the amine
with some functionality (often a carboxylic acid) on the metal
complex.52 We therefore tried modifying ITO-coated glass
slides with (EtO)3Si(CH2)3NH2 and then reacting these with 1
in chlorobenzene/NEt3. Monolayer or sub-monolayer coverage
of the anchored complex was sometimes observed by cyclic
voltammetry, but more often the procedure failed. This is
consistent with the fact that successful syntheses of 3a-e (as
monitored by 31P{1H} NMR spectroscopy in solution) required
>10-fold excess of amine.
Platinum electrodes become coated with a hydrated platinum
oxide layer on anodizing in aqueous H2SO4, and monolayers
have been successfully deposited on such electrodes.28,52
Therefore we attempted to anchor 3e to anodized Pt electrodes
similarly. Figure 6 shows a voltammogram for such an
electrode. The charge under the anodic wave for a freshly-
(52) Abrun˜a, H.; Meyer, T. J.; Murray, R. W. Inorg. Chem. 1979, 11, 3233.
(53) Wilson, R.; Schiffrin, D. J. Analyst (London) 1995, 120, 175.

6196 Inorganic Chemistry, Vol. 36, No. 27, 1997
Barkley et al.
derivatized electrode typically corresponded to ca. 20 mono-
layers. This is too large to be accounted for by surface
roughness52 and suggests that polymerization of the siloxy
groups of 3e occurs during electrode modification, resulting in
a polymer layer on the electrode. Peak currents did not vary
linearly with scan rate for these electrodes, and the peak-to-
peak separation increased with scan rate, probably owing to slow
electron transfer kinetics. Also, the peak currents declined by
20-30% over 100 scans between -0.6 and +0.3 V, presumably
owing to desorption of material from the electrode surface or
to progressive insulation of some of the redox sites from the
electrode surface. It is known that attempts to modify electrodes
using this reaction in the presence of traces of water can lead
to multilayers.54 Moreover, even after oven drying, hydrous
oxides can still retain water. Polymerization will be more likely
with a substrate such as 3e, bearing two pendant siloxy groups,
than with molecules like [({EtO}2MeSi{CH2}3NHCOC5H4)-
FeCp] used by other workers.28
Attempts to prepare ITO electrodes modified using the cis
isomer 4b were frustrated by its insolubility in any solvent
compatible with electrochemistry. We therefore refluxed ITO
electrodes coated with a monolayer of 3e in chlorobenzene for
up to 16 h, in attempts to isomerize the surface-anchored
complex from trans to cis. Subsequent cyclic voltammetry of
these electrodes in CH2Cl2/0.2 M TBAT revealed a small, broad
redox wave at +0.03 V, probably due to residual surface-
anchored 3e, and a totally irreversible oxidation which com-
menced at +0.50 V and which disappeared on repeated
scanning.
toluene, 6 reacted to give the adduct [CpRuCl({Ph2P}2CHCH2-
NHR)] (R ) (CH2)3NH2 (7a) or CH2Ph (7b), respectively);
characterization data are given in the Experimental Section. It
proved difficult to free these very soluble adducts from excess
amine, and yields were low. With H2N(CH2)3Si(OEt)3, addition
occurred (monitored by 31P{1H} NMR) but the adduct could
not be isolated free of excess amine.
Conclusions
We have used the reactivity of coordinated dppen toward
nucleophiles to construct the new ruthenium(II) complexes of
amine-functionalized diphosphine ligands trans-[RuCl2{(Ph2P)2-
CHCH2NRR′}2], including an example suitable for surface-
anchoring to oxide supports, 3e; these could be electrode
surfaces, as demonstrated here, or inert oxides as commonly
used in the “heterogenizing” of homogeneous catalysts. These
complexes also have potential for incorporation into polymers
Via condensation reactions (e.g., 3b) or silica gels Via controlled
hydrolysis (3e). We have demonstrated the anchoring of a
monolayer of 3e to ITO electrodes using cyclic voltammetry
of the surface-localized complex. Future work will probe the
effects of changing the metal center and coligands upon this
chemistry, and further exploration of the reactivity of 1 and 2
with alkynes is in hand.
Acknowledgment. We thank H. Jones, M. McElhinney, D.
Nugent, and S. R. Churchward, undergraduate project students,
for experimental assistance; the staff at the EPSRC National
Mass Spectrometry Service, Swansea, U.K., for FAB mass
spectra; the Royal Society (Grant 18038) and the EPSRC (Grant
GR/J68625; Ph.D. studentships to M.K.M. and T.J.P.; time at
the EPSRC National Mass Spectrometry Service) for funding;
and Johnson-Matthey for a generous loan of RuCl3‚nH2O.
Amine Adducts of [CpRuCl(dppen)] (6). We have exam-
ined the addition of nucleophiles to the “half-sandwich” complex
[CpRuCl(dppen)]23 (6). The anchoring of this more reactive
organometallic center to electrode or oxide surfaces would be
of interest. Moreover, with only a single pendant trialkoxysilyl
group, it was anticipated that an adduct of 6 with (EtO)3Si-
(CH2)3NH2 would lessen the problems of multilayer formation.
With an excess of 1,3-diaminopropane or benzylamine in
Supporting Information Available: A diagram showing the anodic
peak current Vs scan rate for Figure 5 (1 page). An X-ray crystal-
lographic file, in CIF format, for complex 3b‚4MeOH is available on
the Internet only. Ordering and access information is given on any
current masthead page.
(54) Lenhard, J. R.; Murray, R. W. J. Am. Chem. Soc. 1978, 100, 7870.
IC970799O