Ruthenium bipyridyl compounds with two terminal alkynyl ligands

Article abstract of DOI:10.1039/b411609g

Compounds of the form Ru(X₂bipy)(PPh₃) ₂(-C≡CC₆H₄NO₂-p)₂ (X₂ bipy = 4,4′-X₂-2,2′-bipyridine, X = Me 3a, Br 3b, I 3c) have been synthesised from the mono-alkynyl precursors Ru(X ₂bipy)(PPh₃)₂(-C≡CC₆H ₄NO₂-p)Cl (X = Me 2a, Br 2b, I 2c); the former are the first ruthenium bis-alkynyl compounds that also contain a bipyridyl ligand. Spectroelectrochemical investigation of 3a shows that the metal is readily oxidised to form the ruthenium(III) compound 3a⁺, and will also undergo a single-electron reduction at each nitro group to form 3a^2-. ESR and UV/visible spectra of these redox congeners are presented. We also report the synthesis of [Ru(Me₂bipy)(PPh₃) ₂-(-C≡CBu¹)(N≡N)][PF₆] 4 during the attempted synthesis of Ru(Me₂bipy)(PPh₃) ₂(-C≡CBu¹)₂, and report its X-ray crystal structure and IR spectrum. X-Ray crystal structures of 3b and 3c (as two different solvates) are presented, and the nature of the inlermolecular interactions seen therein is discussed. Z-Scan measurements on Ru(Me ₂bipy)(PPh₃)₂-(-C≡CR)Cl (R = C ₆H₄NO₂-p 2a, Bu^t, Ph, C ₆H₄Me) are also reported, and show that Ru(Me ₂bipy)(PPh₃)₂(-C≡CR)-Cl (R = C ₆H₄NO₂-p 2a, Ph) exhibit moderate third-order non-linearities.

Full text of DOI:10.1039/b411609g

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Ruthenium bipyridyl compounds with two terminal alkynyl ligands
Christopher J. Adams,*a Lucy E. Bowen,a Mark G. Humphrey,b Joseph P. L. Morrall,b
Marek Samocc and Lesley J. Yellowleesd
a School of Chemistry, University of Bristol, Bristol, UK, BS8 1TS. E-mail: chcja@bris.ac.uk
b Department of Chemistry, Australian National University, Canberra, Australia, ACT 0200.
E-mail: Mark.Humphrey@anu.edu.au
c Laser Physics Centre, Research School of Physical Sciences and Engineering, Australian
National University, Canberra, Australia, ACT 0200
d School of Chemistry, Joseph Black Building, West Mains Road, Edinburgh, UK, EH9 3JJ.
E-mail: l.j.yellowlees@ed.ac.uk
Received 28th July 2004, Accepted 6th October 2004
First published as an Advance Article on the web 9th November 2004
Compounds of the form Ru(X2bipy)(PPh3)2(–C≡CC6H4NO2-p)2 (X2bipy = 4,4ꢀ-X2-2,2ꢀ-bipyridine, X = Me 3a, Br
3b, I 3c) have been synthesised from the mono-alkynyl precursors Ru(X2bipy)(PPh3)2(–C≡CC6H4NO2-p)Cl (X = Me
2a, Br 2b, I 2c); the former are the first ruthenium bis-alkynyl compounds that also contain a bipyridyl ligand.
Spectroelectrochemical investigation of 3a shows that the metal is readily oxidised to form the ruthenium(III)
compound 3a+, and will also undergo a single-electron reduction at each nitro group to form 3a2−. ESR and
UV/visible spectra of these redox congeners are presented. We also report the synthesis of [Ru(Me2bipy)(PPh3)2-
(–C≡CBut)(N≡N)][PF6] 4 during the attempted synthesis of Ru(Me2bipy)(PPh3)2(–C≡CBut)2, and report its X-ray
crystal structure and IR spectrum. X-Ray crystal structures of 3b and 3c (as two different solvates) are presented, and
the nature of the intermolecular interactions seen therein is discussed. Z-Scan measurements on Ru(Me2bipy)(PPh3)2-
(–C≡CR)Cl (R = C6H4NO2-p 2a, But, Ph, C6H4Me) are also reported, and show that Ru(Me2bipy)(PPh3)2(–C≡CR)-
Cl (R = C6H4NO2-p 2a, Ph) exhibit moderate third-order non-linearities.
Ru(Me2bipy)(PPh3)2(–C≡CR)2 from the mono(alkynyl) com-
pounds mentioned above. This has been attempted in the same
way as before, via intermediate vinylidene compounds, although
a one-pot procedure has been developed so that these vinylidenes
do not need to be isolated. Success has been achieved with R =
C6H4NO2-p, but not with R = But or C6H4Me-p.
Introduction
We have recently published the first results from our investi-
gation into the synthesis and properties of ruthenium–alkynyl
compounds that also contain a bipyridyl ligand,1 stimulated by
an analogy between them and a class of platinum compounds
that have received extensive investigation over the past four years.
Such bipyridyl–platinum–bis(alkynyl) compounds have been
shown to luminesce strongly from a 3MLCT excited state in both
frozen and fluid solution, and have also recently demonstrated
vapoluminescence and interesting interactions with solvent
molecules in the crystal lattice.2 The same arrangement of
bipyridyl and alkynyl ligands can be envisaged in the equatorial
positions of an octahedral ruthenium centre, with secondary
ligands occupying the axial positions. Such d6 octahedral
ruthenium compounds could display the same optoelectronic
properties as the d8 square-planar platinum species, as well as
additional properties such as redox-switchable behaviour and
tunability through the alteration of the axial ligands.
Thus far, we have reported on compounds of the gen-
eral formula Ru(Me2bipy)(PPh3)2(–C≡CR)Cl (Me2bipy = 4,4ꢀ-
dimethyl-2,2ꢀ-bipyridine; R = But, Ph, C6H4Me); that is, mono-
alkynyl species. These compounds are synthesised from the
corresponding dichloride precursor via an intermediate cationic
vinylidene compound, which is formed when one of the chloride
ligands is replaced with a terminal alkyne, and which may then
be deprotonated to yield the alkynyl complex. Although the
spectra of these mono-alkynyl compounds possess the same
MLCT bands as the aforementioned platinum compounds they
do not appear to show the same emissive behaviour, which
could be because of the presence of the chloride ligands, which
dissociate to some extent in solution and may quench any
emission. The compounds do however show redox-switchable
optical properties, changing colour as the lowest energy band in
the UV/visible spectrum changes from MLCT to LMCT upon
oxidation from Ru(II) to Ru(III).1
Experimental
General information
All new compounds except 4 are air stable in the solid
state and reasonably air stable in solution, but standard
inert-atmosphere techniques were used throughout. All sol-
vents were purified using an Anhydrous Engineering Grubbs-
type solvent system.3 The starting materials Ru(PPh3)3Cl2,
4
Ru(Me2bipy)(PPh3)2Cl2 1a,5 Ru(Me2bipy)(PPh3)2(–C≡CR)Cl
(R
=
But, Ph, p-C6H4Me),1 4,4ꢀ-dibromo-2,2ꢀ-bipyridine
(Br2bipy),6 4,4ꢀ-dinitro-2,2ꢀ-bipyridine N,Nꢀ-dioxide7 and p-
nitrophenylacetylene8 were prepared by literature methods, and
all other chemicals were used as purchased. IR and UV/visible
spectra were recorded in dichloromethane solution on a Perkin-
Elmer 1600 series FTIR spectrometer and a Perkin-Elmer k-19
spectrophotometer respectively. Cyclic voltammetry was carried
out under an atmosphere of nitrogen using the standard three-
electrode configuration, with platinum working and counter
electrodes, an SCE reference electrode, dichloromethane as
solvent, 0.1 M [Bu4N][PF6] as electrolyte and FeCp2 or FeCp*2
as internal calibrant, and a substrate concentration of approxi-
mately 1 mM. All potentials are reported vs. the SCE reference
electrode, against which the FeCp2/[FeCp2]+ couple comes at
0.46 V and FeCp*2/[FeCp*2]+ is −0.02 V.9 X-Band ESR spectra
of 2a+ and 3a+ were recorded at the University of Bristol on a
Bruker ESP300E spectrometer in 2 : 1 thf–CH2Cl2 at 110 K, and
that of 3a2− at the University of Edinburgh on a Bruker ER200D
spectrometer. NMR spectra were recorded in CDCl3 (unless
otherwise stated) on a JEOL ECP300 spectrometer, at 300 MHz
(1H) and 121 MHz (31P), and referenced to external TMS and
The current work reports the results of our investigations
of the synthesis of bis(alkynyl) compounds of the form
4 1 3 0
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8
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 4
©

View Article Online
external 85% H3PO4, respectively. UV/visible spectroelectro-
chemical measurements were performed in CH2Cl2 at 243 K
using a locally constructed OTTLE (optically transparent thin-
layer electrode) cell in a Perkin-Elmer k-19 spectrophotometer,
as described previously.10 Microanalyses were carried out by the
staff of the Microanalytical Service of the School of Chemistry
at the University of Bristol.
deep purple crystals of the desired product as a dichloromethane
solvate. These were isolated by filtration, washed with hexane
1
and dried to give 0.112 g (0.102 mmol, 74%) of product. H
NMR d: 6.52 (2H, dd, 3JHH = 6.4 Hz, 4JHH = 2.2 Hz, H5); 7.04–
7.20 (18H, m, Ph); 7.46–7.54 (12H, m, Ph); 7.63 (2H, d, 4JHH
=
2.2 Hz, H3); 8.58 (2H, d, 3JHH = 6.4 Hz, H6). 31P NMR d: 22.0.
Anal. Calc. for C46H36N2P2Cl2Br2Ru·CH2Cl2: C 51.53; H 3.50;
N 2.66%. Found: C 51.71; H 3.54; N 2.65%. kmax/nm: 532 (e =
3167 M−1 cm−1), 368 (7421), 306 (17065), 272 (34040). E◦
=
ꢀ
Syntheses
0.58 V.
4,4ꢀ-Diamino-2,2ꢀ-bipyridine. This was synthesised by the re-
ported procedure,11 but isolated differently. In a typical example,
3.83 g (0.013 mol) of 4,4ꢀ-dinitro-2,2ꢀ-bipyridine N,Nꢀ-dioxide
and 1 g of 10% palladium on carbon were suspended in 100 cm3
of degassed ethanol. A solution of hydrazine monohydrate
(4 cm3) in ethanol (20 cm3) was added dropwise to this over
the course of an hour, and the solution was then refluxed for
8 h. After cooling it was filtered through Celite (which was then
washed with more ethanol), and the combined ethanol solutions
were reduced in volume to approximately 100 cm3, using a rotary
evaporator with a hot water-bath. At this stage the solution
should still be clear; any yellow precipitate can be redissolved by
addition of a little more ethanol and further heating. 50 cm3 of
distilled water was then added, and the solution again reduced
to about 100 cm3. At this point, slow addition of a further
100 cm3 of distilled water caused the product to form as a
white crystalline solid; the solution was refrigerated overnight to
ensure complete crystallisation. Following isolation by filtration,
washing with distilled water, and drying, the product was
Ru(I2bipy)(PPh3)2Cl2 1c. 0.074 g (0.18 mmol) of 4,4ꢀ-diiodo-
2,2ꢀ-bipyridine and 0.172 g (0.13 mmol) of RuCl2(PPh3)3 were
stirred for 45 min in 10 cm3 of CH2Cl2 to give a deep purple
solution. To this was slowly added 20 cm3 of hexane, and the
solution was refrigerated overnight to yield deep purple crystals
of the desired product. These were isolated by filtration, washed
with hexane and dried to give 0.094 g (0.085 mmol, 47%) of
1
3
4
product. H NMR d: 6.68 (2H, dd, JHH = 6.2 Hz, JHH = 1.8
Hz, H5), 7.03–7.20 (18H, m, Ph), 7.42–7.54 (12H, m, Ph), 7.76
4
3
(2H, d, JHH = 1.8 Hz, H3), 8.39 (2H, d, JHH = 6.2 Hz, H6).
31P NMR d: 22.0. Anal. Calc. for C46H36N2P2Cl2I2Ru: C 50.02;
H 3.29; N 2.54%. Found: C 50.17; H 3.29; N 2.50%. kmax/nm):
531 (e = 4174 M−1 cm−1), 371 (9996), 311 (21732), 272 (43413).
E◦ = 0.57 V.
ꢀ
Ru(Me2bipy)(PPh3)2(–C≡CC6H4NO2-p)Cl 2a. 0.304 g of
Ru(Me2bipy)(PPh3)2Cl2 1a (0.34 mmol), 0.070 g of p-nitro-
phenylacetylene (0.48 mmol) and 0.121 g (0.34 mmol) of TlPF6
were stirred for 2 h in 10 cm3 of CH2Cl2. To the resulting orange
solution with a white precipitate was added 0.3 g (2.1 mmol)
of potassium carbonate, and the solution was allowed to stir
overnight. The resulting purple solution containing a white
precipitate was filtered, and the residual solid washed with a
further 2 × 5 cm3 portions of CH2Cl2. 100 cm3 of diethyl
ether was then added to the combined solutions, and overnight
refrigeration yielded 2a as a purple–brown solid with a green
1
isolated as a white (lit11: yellow) solid (1.76 g, 69%). H NMR
(d6-DMSO) d: 6.01 (4H, s, NH2), 6.41 (2H, dd, 3JHH = 5.5 Hz,
4JHH = 2.3 Hz, H5), 7.51 (2H, d, 3JHH = 2.3 Hz, H3), 8.00 (2H, d,
4JHH = 5.5 Hz, H6). Anal. Calc. for C10H10N4: C 64.50; H 5.41;
N 30.09%. Found: C 64.40; H 5.58; N 30.29%.
4,4ꢀ-Diiodo-2,2ꢀ-bipyridine (I2bipy). This compound was pre-
pared from 4,4ꢀ-diamino-2,2ꢀ-bipyridine in a manner analogous
to the preparation of 4-iodopyridine from 4-aminopyridine.12
Thus, 4,4ꢀ-diamino-2,2ꢀ-bipyridine (0.9 g, 4.5 mmol) was sus-
pended in 48% aqueous HBF4 (25 cm3) and cooled to −10 ◦C.
To the resulting slurry was added, portionwise over the course of
about 1 h, powdered sodium nitrite (1.3 g, 18.8 mmol), at such a
rate that no nitric oxide evolution was detected. After a further
15 min stirring, the precipitate of diazonium salt was isolated
by suction filtration, but not allowed to dry out. This was then
added, portionwise over the course of about 30 min, to a cooled
(−10 ◦C) solution of potassium iodide (5.1 g, 30.7 mmol) in an
acetone–water (40 : 60, 100 cm3) solution. The reaction was then
stirred for 30 min, being allowed to warm to room temperature,
giving a yellow–brown suspension. This was neutralised by
adding aqueous sodium carbonate, and then a little aqueous
sodium thiosulfate solution was added to decolorise the solution
phase. Extraction with dichloromethane (4 × 50 cm3) gave a
yellow solution, which was dried over magnesium sulfate and
treated with activated carbon. Following filtration through an
alumina pad (2 cm diameter × 2 cm depth), the solvent was
removed under vacuum to give the crude product. This was
redissolved in hot ethanol, treated again with activated carbon,
and filtered hot through Celite. Upon cooling and storage at
−10 ◦C the compound formed as a white crystalline solid, which
was isolated by filtration and dried (0.66 g, 1.61 mmol, 33%).1H
NMR d: 7.63 (2H, dd, 3JHH = 5.4 Hz, 4JHH = 1.9 Hz, H5), 8.23
1
iridescence (0.238 g, 0.24 mmol, 70%). H NMR d: 2.20, 2.34
(each 3H, s, Me); 5.91 (1H, d, 3JHH = 4.7 Hz, H5); 6.64 (1H, d,
3
3JHH = 4.7 Hz, H5); 6.80 (2H, d, JHH = 8.8 Hz, C6H4); 7.00–
7.17 (18H, m, Ph); 7.37 (1H, s, H3); 7.52–7.61 (13H, m, Ph and
3
3
H3); 7.96 (2H, d, JHH = 8.8 Hz, C6H4); 8.02 (1H, d, JHH
=
5.8 Hz, H6); 8.87 (1H, d, 3JHH = 5.8 Hz, H6). 31P NMR d: 30.5.
Anal. Calc. for C56H46N3P2ClRuO2: C 67.84; H 4.68; N 4.24%.
Found: C 67.66; H 4.90; N 4.32%. FT-IR m(C≡C)/cm−1: 2043s,
2014 sh. kmax/nm: 515 (e = 22552 M−1 cm−1), 332sh (11920),
298 (26923), 277 (36485). E◦ = 0.39, −1.23 V.
ꢀ
Ru(Br2bipy)(PPh3)2(–C≡CC6H4NO2-p)Cl 2b. 0.185 g of
Ru(Br2bipy)(PPh3)2Cl2·CH2Cl2 1b·CH2Cl2 (0.17 mmol), 0.064 g
of TlPF6 (0.18 mmol) and 0.040 g of p-nitrophenylacetylene
(0.27 mmol) were stirred for 2 h in 10 cm3 of CH2Cl2 to give
an orange–brown solution. To this was added 0.180 g of K2CO3
(1.30 mmol), and the reaction stirred for a further 20 h to give
a red–purple solution. This was filtered through a filter-paper
tipped cannula, and the residual solid washed with a further
5 cm3 of CH2Cl2. 35 cm3 of hexane were added to the combined
solutions, which were then stored at 4 ◦C for 48 h, after which
time filtration allowed the isolation of the desired product as
a microcrystalline brown solid, which was washed with hexane
and dried under vacuum to give 0.141 g of product (0.13 mmol,
75%). 1H NMR d: 6.27 (1H, dd, 3JHH = 6.2 Hz, 4JHH = 2.2 Hz,
3
3
H5); 6.74 (2H, d, JHH = 8.8 Hz, C6H4); 6.97 (1H, dd, JHH
=
3
4
(2H, d, JHH = 5.4 Hz, H6), 8.73 (2H, d, JHH = 1.9 Hz, H3).
Anal. Calc. for C10H6N2I2: C 29.44; H 1.48; N 6.87%. Found: C
29.73; H 1.52; N 6.88%.
6.2 Hz, 4JHH = 1.9 Hz, H5); 7.02–7.18 (18H, m, Ph); 7.48–7.58
(12H, m, Ph); 7.65 (1H, d, 4JHH = 2.2 Hz, H3); 7.80 (1H, d, 4JHH
=
3
1.9 Hz, H3); 7.96 (2H, d, JHH = 8.8 Hz, C6H4); 8.06 (1H, d,
3JHH = 6.6 Hz, H6); 8.88 (1H, d, 3JHH = 6.0 Hz, H6). 31P NMR d:
28.5. Anal. Calc. for C54H40Br2ClN3O2P2Ru: C, 57.85; H 3.60; N
3.75%. Found: C, 57.53; H 3.63; N 3.79%. FT–IR m(C≡C)/cm−1:
2048s, 2011sh. kmax/nm: 518 (e = 20442 M−1 cm−1), 362 (8219),
Ru(Br2bipy)(PPh3)2Cl2 1b. 0.043 g (0.14 mmol) of 4,4ꢀ-
dibromo-2,2ꢀ-bipyridine and 0.125
g
(0.13 mmol) of
RuCl2(PPh3)3 were stirred for 90 min in 10 cm3 of CH2Cl2 to
give a deep purple solution. To this was slowly added 30 cm3
of hexane, and the solution was refrigerated overnight to yield
306 (25126), 276 (38234). E◦ = 0.56 V.
ꢀ
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8
4 1 3 1

View Article Online
Ru(I2bipy)(PPh3)2(–C≡CC6H4NO2-p)Cl 2c. 0.107
g
of
Ru(I2bipy)(PPh3)2(–C≡CC6H4NO2-p)2 3c. 0.072
g
of
Ru(I2bipy)(PPh3)2Cl2 1c (0.097 mmol), 0.034 g of TlPF6
(0.097 mmol) and 0.028 g of p-nitrophenylacetylene (0.19 mmol)
were stirred for 4 h in 10 cm3 of CH2Cl2 to give an orange–brown
solution. To this was added 0.100 g of K2CO3 (0.72 mmol),
and the reaction stirred for a further 20 h to give a red–purple
solution. This was filtered through a filter-paper tipped cannula,
and the residual solid washed with a further 5 cm3 of CH2Cl2.
40 cm3 of hexane were added to the combined solutions, which
were then stored at 4 ◦C for 24 h, after which time filtration
allowed the isolation of the desired product as a microcrystalline
brown solid, which was washed with hexane and dried under
vacuum to give 0.062 g of product (0.051 mmol, 52%). 1H NMR
d: 6.42 (dd, JHH = 6.2 Hz, 1.8 Hz, 1H, H5); 6.83 (d, JHH = 8.8
Hz, 2H, C6H4); 7.04–7.21 (19H, m, Ph, H5); 7.57–7.50 (12H, m,
Ru(I2bipy)(PPh3)2Cl(–C≡CC6H4NO2-p) 2c (0.06 mmol), 0.021 g
of TlPF6 (0.06 mmol) and 0.017 g of p-nitrophenylacetylene
(0.12 mmol) were stirred for 2 days in 20 cm3 of CH2Cl2 to
give an orange–brown solution. To this was added 0.075 g of
K2CO3 (0.54 mmol), and the reaction stirred for a further 2 d
to give a purple solution. This was filtered through Celite and
evaporated to dryness, and then extracted into toluene. This
toluene solution was filtered through Celite, and then reduced
in volume on the rotary evaporator until solid material was
observed around the side of the flask. This was redissolved by
swirling, and the solution was stored at −10 ◦C for two weeks,
during which time the product was deposited as deep-purple
crystals. The solution was decanted from these, and they were
then washed with hexane and dried under vacuum to give
0.027 g (0.02 mmol, 29% yield {based on 2.5 toluene solvent
ꢀ
Ph); 7.79 (d, JHH = 1.8 Hz, 1H, H3 ); 7.91 (d, JHH = 6.2 Hz,
1
1H, H6); 7.98 (m, 3H, H3, C6H4); 8.71 (d, JHH = 5.9 Hz, 1H,
H6). 31P NMR d: 28.43. Anal. Calc. for C54H40I2ClN3O2P2Ru: C,
53.37; H 3.32; N 3.46%. Found: C, 53.70; H, 3.13; N, 3.34%.
molecules per molecule of 3c}) of product. H NMR d: 6.78
3
3
(4H, d, JHH = 8.8 Hz, C6H4); 6.92 (2H, dd, JHH = 6.1 Hz,
4JHH = 1.5 Hz, H5); 7.26–7.08 (12H, m, Ph); 7.61–7.58 (18H,
FT-IR m(C≡C)/cm−1: 2048s, 2012sh. kmax/nm: 526 (e = 20
m, Ph); 7.94 (4H, d, 3JHH = 8.8 Hz, C6H4); 7.98 (2H, d, 4JHH
=
392 M−1 cm−1), 369 (9197), 310 (26596), 275 (38722). E◦
=
1.8 Hz, H3); 8.20 (2H, d, 3JHH = 6.1 Hz, H6). 31P NMR d: 34.68.
Anal. Calc. for C62H44I2N4O4P2Ru: C, 56.16; H 3.34; N 4.23%.
Found: C, 56.36; H 3.22; N 4.13% (It was necessary to grind
and thoroughly dry the sample in order to remove toluene and
obtain satisfactory analysis). FT-IR m(C≡C)/cm−1: 2060m,
2040s, 2011w. kmax/nm: 525 (e = 34656 M−1 cm−1), 312 (30285),
ꢀ
0.57 V.
Ru(Me2bipy)(PPh3)2(–C≡CC6H4NO2-p)2 3a. 0.150
g
of
Ru(Me2bipy)(PPh3)2(Cl)(–C≡CC6H4NO2-p) 2a (0.15 mmol),
0.053 g of TlPF6 (0.15 mmol) and 0.033 g of p-nitrophenyl-
acetylene (0.22 mmol) were stirred for 48 h in 20 cm3 of CH2Cl2
to give an orange–brown solution. To this was added 0.150 g of
K2CO3 (1.08 mmol), and the reaction stirred for a further 20 h
to give a purple solution. This was filtered through Celite, and
evaporated to dryness. The resulting purple solid was extracted
into 20 cm3 of warm toluene, filtered through Celite, and then
allowed to cool before being stored at −10 ◦C for 48 h. Decanting
the liquid from the resulting mixture allowed the isolation of the
desired product 3a as dark crystals, which were washed with
hexane and dried under vacuum to give 0.070 g of product
271 (41447). E◦ = 0.54 V.
ꢀ
[Ru(Me2bipy)(PPh3)2(–C≡CBut)(N≡N)][PF6] 4.Ru(Me2bipy)
(PPh3)2(–C≡CBut)Cl (0.053 g, 0.057 mmol) and TlPF6 (0.020 g,
0.057 mmol) were stirred in 10 cm3 of nitrogen-saturated thf for
2 h to give an orange solution. This was filtered with a filter-paper
tipped cannula, and 10 cm3 of hexane were added. Overnight
refrigeration caused the formation of the desired product as red
crystals, which were isolated by filtration, washed with hexane,
and dried under vacuum, to give 4 as an orange–brown solid
(0.038 g, 0.036 mmol, 63%). 1H NMR d: 1.17 (9H, s, But); 2.41
1
(0.063 mmol, 42%). H NMR d: 2.31 (6H, s, Me); 6.39 (2H,
3
3
3
d, JHH = 5.7 Hz, H5); 6.74 (4H, d, JHH = 8.4 Hz, C6H4);
(6H, s, Me); 6.31 (1H, d, JHH = 5.8 Hz, H5); 6.38 (1H, d,
7.00–7.16 (18H, m, Ph); 7.54 (2H, s, H3); 7.56–7.63 (12H, m,
3JHH = 4.8 Hz, H5); 7.10–7.20 (12H, m, Ph); 7.20–7.30 (6H,
3
3
Ph); 7.92 (4H, d, JHH = 8.4 Hz, C6H4); 8.35 (2H, d, JHH
=
m, Ph); 7.35–7.45 (12H, m, Ph); 8.13 (1H, s, H3); 8.17 (1H, s,
5.7 Hz, H6). 31P NMR d: 36.4. Anal. Calc. for C64H50N4P2RuO4:
C 69.75; H 4.57; N 5.08%. Found: C 69.54; H 4.05; N 5.22%.
FT-IR m(C≡C)/cm−1: 2010m, 2035s, 2056m. kmax/nm: 507 (e =
H3); 8.34 (1H, d, JHH = 5.8 Hz, H6). 31P NMR d: 30.3. Anal.
3
Calc. for C54H51N4P3RuF6: C, 60.96; H, 4.83; N, 5.27%. Found:
C, 60.95; H, 4.94; N, 5.05%. FT-IR/cm−1: 2149s (m{N≡N}),
2101w (m{C≡C}).
36896 M−1 cm−1), 330 (19856), 274 (39986). E◦ = 0.42 V,
ꢀ
−1.20 V.
Crystal structure determinations
Ru(Br2bipy)(PPh3)2(–C≡CC6H4NO2-p)2 3b. 0.086
g of
Ru(Br2bipy)(PPh3)2Cl(–C≡CC6H4NO2-p) 2b (0.08 mmol),
0.027 g of TlPF6 (0.08 mmol) and 0.022 g of p-nitrophenyl-
acetylene (0.15 mmol) were stirred for 2 days in 20 cm3 of CH2Cl2
to give an orange–brown solution. To this was added 0.086 g of
K2CO3 (0.62 mmol), and the reaction stirred for a further 20 h
to give a red–purple solution. This was filtered through Celite
and evaporated to dryness, and then extracted into toluene. This
toluene solution was filtered through Celite, and then reduced
in volume on the rotary evaporator until solid material was
observed around the side of the flask. This was redissolved by
swirling, and the solution was stored at −10 ◦C for four days
during which time the product was deposited as rectangular
crystals. The solution was decanted from these, and they were
then washed with hexane and dried under vacuum to give the
desired product 3b as brown crystals (0.044 g, 0.04 mmol, 47%
yield). 1H NMR d: 6.72–6.78 (6H, m, C6H4 and H5); 7.05–7.12
(12H, m, Ph); 7.55–7.22 (6H, m, Ph); 7.58–7.65 (12H, m, Ph);
Data collection was carried out on a Bruker SMART diffrac-
tometer with the crystal mounted in a nitrogen stream at
−100 ◦C or on a Bruker APEX diffractometer with the crystal
mounted in a nitrogen stream at −173 ◦C. Structures were
solved by direct methods and refined by full-matrix least-
squares techniques against F2 using the programs SHELXS-
9713 and SHELXL-97.14 The asymmetric unit of the structure
of 3b·3C7H8 contains two half molecules of toluene solvent that
are badly disordered about inversion centres, which have been
refined isotropically without the addition of hydrogen atoms.
There are also two fully ordered toluene molecules that have
been refined anisotropically with hydrogen atoms. The crystal
structure of 3c·2.5C7H8 has Zꢀ = 2, so the asymmetric unit
contains two molecules of 3c and five of toluene. In one of the
PPh3 ligands of one of these two molecules of 3c two of the
phenyl rings are each disordered equally over two sites.
4
3
7.82 (2H, d, JHH = 2.2 Hz, H3); 7.94 (4H, d, JHH = 9.1 Hz,
Crystal data. 3b·3C7H8C83H60Br2N4O4P2Ru: M = 1500.18,
3
C6H4); 8.38 (2H, d, JHH = 5.9 Hz, H6). 31P NMR d: 34.7.
triclinic, space group P1 (no. 2), a = 11.939(3), b = 13.588(2), c =
¯
◦
˚
Anal. Calc. for C62H44Br2N4O4P2Ru: C, 60.45; H 3.60; N 4.55%.
Found: C, 60.71; H 3.61; N 4.69%. FT-IR m(C≡C)/cm−1: 2060m,
2040s, 2010w. kmax/nm: 523 (e = 36255 M−1 cm−1), 306 (29993),
24.678(4) A, a = 82.868(17), b = 84.101(16), c = 64.082(15) ,
3
˚
˚
U = 3567.6(12) A , T = 173(2) K, k = 0.71073 A, Z = 2, l(Mo-
Ka) = 1.439 mm−1, 44608 reflections measured, 16145 unique
(Rint = 0.0607), R1 [I > 2r(I)] = 0.0503, R1 [all data] = 0.1084.
273 (43749). E◦ = 0.55 V.
ꢀ
4 1 3 2
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8

View Article Online
3c·2.5C7H8: C79.5H64I2N4O4P2Ru, M = 1556.16, triclinic,
be synthesised, the reaction of the mono-alkynyl com-
pounds Ru(Me2bipy)(PPh3)2(–C≡CR)Cl (R = But, C6H4Me-
p, C6H4NO2-p 2a) with further terminal alkyne in the pres-
ence of TlPF6 and K2CO3 was investigated. We have re-
ported the preparation of the first two of these mono-
alkynyl starting materials previously, and that of the third,
Ru(Me2bipy)(PPh3)2(–C≡CC6H4NO2-p)Cl 2a from the dichlo-
ride Ru(Me2bipy)(PPh3)2Cl2 1a is given herein. Although this
preparation is analogous to those previously published, re-
placement of a chloride ligand with an alkyne (as a vinylidene
ligand) and subsequent deprotonation, the synthesis has been
modified to allow a one-pot procedure without isolation of the
intermediate vinylidene compound (Scheme 1).
The formation of the bis-alkynyl compound Ru(Me2bipy)-
(PPh3)2(–C≡CC6H4NO2-p)2 3a from 2a may indeed be achieved
in the same manner as that of 2a from 1a (Scheme 1),
although the first half of the reaction proceeds more slowly;
only after reaction for two days (cf. 2 h) is complete replacement
of chloride by vinylidene achieved. 2a and 3a display more
absorption bands (two and three respectively) in the C≡C
region of the IR spectrum than there are carbon–carbon
triple bonds, because of Fermi coupling15 – the same is also
true of Ru(Me2bipy)(PPh3)2(–C≡CC6H4Me-p)Cl.1 In contrast,
the attempted synthesis of Ru(Me2bipy)(PPh3)2(–C≡CC6H4Me-
p)2 from Ru(Me2bipy)(PPh3)2(–C≡CC6H4Me-p)Cl by the same
method was not possible. It appears to be the formation
of the intermediate mixed vinylidene–alkynyl complex that is
unsuccessful; although a white precipitate of TlCl is observed in
the reaction vessel, the resulting solution contains a multitude
of 31P NMR signals.
¯
space group P1 (no. 2), a = 15.158(3), b = 20.783(4), c =
◦
˚
21.935(4) A, a = 90.29(3), b = 92.81(3), c = 100.88(3) , U =
3
˚
˚
6777(2) A , T = 100(2) K, k = 0.71073 A, Z = 4, l(Mo-Ka) =
1.244 mm−1, 68419 reflections measured, 30886 unique (Rint
0.0299), R1 [I > 2r(I)] = 0.0587, R1 [all data] = 0.0782.
=
3c·CH2Cl2: C63H46Cl2I2N4O4P2Ru, M = 1410.75, monoclinic,
space group P21/c (no. 14), a = 18.979(2), b = 13.4126(18),
◦
3
˚
˚
c = 22.890(3) A, b = 101.9882(8)) , U = 5699.6(12) A , T =
−1
˚
100(2) K, k = 0.71073 A, Z = 4, l(Mo-Ka) = 1.561 mm , 40
426 reflections measured, 13089 unique (Rint = 0.0656), R1 [I >
2r(I)] = 0.0426, R1 [all data] = 0.0731.
4·CH2Cl2: C55H53Cl2F6N4P3Ru, M = 1148.89, monoclinic,
space group P21/c (no. 14), a = 16.692(6), b = 11.874(4), c =
◦
3
˚
˚
26.902(8) A, b = 97.69(3) , U = 5284(3) A , T = 173(2) K, k =
−1
˚
0.71073 A, Z = 4, l(Mo-Ka) = 0.551 mm , 33710 reflections
measured, 12121 unique (Rint = 0.0866), R1 [I > 2r(I)] = 0.0663,
R1 [all data] = 0.1372.
CCDC reference numbers 246092–246095.
See http://www.rsc.org/suppdata/dt/b4/b411609g/ for cry-
stallographic data in CIF or other electronic format.
Z-Scan studies
Z-Scans were recorded for solutions of the samples in
dichloromethane with concentrations in the range 0.5–4% w/w.
Most experiments were performed at 800 nm using an amplified
Ti-sapphire laser system delivering lJ range pulses of ca. 150 fs
duration at a repetition rate of 30 Hz. Samples for Z-scan were
placed in 1 mm path glass cells and the scans were recorded
using an f = 275 mm lens providing a focal plane spot size of
about w0 = 42 lm. The light intensity was adjusted to keep the
nonlinear phase shift in the solvent cell and cells with solutions
in the Dφ0 = 0.5–1 rd range (roughly on the order of 100 GW
cm−2). The phase shifts and the imaginary (absorptive) parts
of the nonlinearity were obtained by numerical fitting of both
closed aperture and open aperture Z-scan traces and the absolute
values were determined using a comparison with Z-scans on a
1 mm thick silica plate for which the nonlinear refractive index
n2 = 3×10−16 cm2 W−1 was assumed.
Preparation of Ru(Me2bipy)(PPh3)2(–C≡CBut)2 by the same
method was also unsuccessful, although in this case 31P NMR re-
veals a relatively clean reaction. Isolation of the product and sub-
sequent characterisation reveals it to be [Ru(Me2bipy)(PPh3)2-
(–C≡CBut)(N≡N)][PF6] 4, in which the chloride ligand of the
starting material has been replaced not with a molecule of tert-
butylacetylene but with one of dinitrogen (Scheme 2). The reason
for this is presumably steric; inspection of the crystal structure
of Ru(Me2bipy)(PPh3)2(–C≡CBut)Cl1 reveals that even after
the chloride ligand has been eliminated there is little space in
In addition, Ru(Me2bipy)(PPh3)2(–C≡CPh)Cl was also inves-
tigated at 670 nm. This wavelength was obtained from another
amplified femtosecond system comprising a Clark MXR CPA-
2001 regenerative amplifier and a Light Conversion TOPAS
optical parametric amplifier. This system was operated at the
repetition rate of 250 Hz and the beam parameters were similar
to those for 800 nm measurements (w0 = 50 lm was used).
2
the coordination sphere of the metal for the g -coordination
of another alkyne molecule that precedes isomerisation to a
vinylidene group. The IR spectrum of 4 contains two bands in
the triple bond region, a strong one at 2146 cm−1 and a weak one
at 2101 cm−1. These may be assigned by synthesising 4 under an
atmosphere of 15N2, which leads to the appearance of a strong
new absorption at 2073 cm−1. This is the correct position to be the
counterpart of the absorption at 2146 cm−1, which is therefore
m(N≡N), and so the absorption at 2101 cm−1 is m(C≡C). This is
consistent with the observation that N≡N absorptions tend to
be strong.16
Results and discussion
Synthesis of new alkynyl compounds
In order to discover whether bis-alkynyl compounds of
the general formula Ru(Me2bipy)(PPh3)2(–C≡CR)2 could
A crystal suitable for X-ray diffraction was grown from
a dichloromethane–hexane solution of 4, and the cation of
Scheme 1 Synthesis of compounds 2 and 3.
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8
4 1 3 3

View Article Online
˚
tetraazacyclohexadecane) {1.005(10) A}. The fact that the N≡N
distance does not normally vary much from the free gas has
been interpreted by invoking contributions from two opposing
effects of approximately equal magnitude – donation from a r*
lone-pair on the dinitrogen to the metal, which causes the
N≡N distance to decrease by removing electron density from
an antibonding orbital, and back donation from the metal into
the p* orbitals of the dinitrogen, which causes the N≡N distance
to increase.21 The shortness of the N≡N distance in 4 implies that
this back-bonding contribution is very small, although why this
should be the case is not apparent; though [RuTp(PEt3)2(N2)]+,
[Ru(16-TMC)(N2)Cl]+ and 4+ are all cationic, octahedral, d6
ruthenium(II) fragments, the same is also true of other fragments
which do not show the same shortening of the N≡N bond.
Scheme 2 Synthesis of 4+.
the resulting structure is shown in Fig. 1. Selected bond
lengths and angles are presented in Table 1, and show that the
geometry of the ruthenium centre is much more similar to that
of the (neutral) ruthenium(II) complex Ru(Me2bipy)(PPh3)2(–
C≡CBut)Cl than to that of the (cationic) ruthenium(III)
compound [Ru(Me2bipy)(PPh3)2(–C≡CBut)Cl][PF6].1 For ex-
ample, the distances Ru(1)–C(13), C(13)–C(14) and Ru(1)–
N(1) of 2.032(5), 1.179(7) and 2.124(4), respectively, in 4 are
well matched by those of 2.053(5), 1.174(6) and 2.120(4) in
Ru(Me2bipy)(PPh3)2(–C≡CBut)Cl,1 which seems to imply that
it is the oxidation state of the metal rather than the charge of
the complex that is more important in determining these bond
lengths.
Spectroscopic behaviour
We have previously shown that in the mono-alkynyl compounds
Ru(Me2bipy)(PPh3)2(–C≡CR)Cl (R = But, Ph, p-C6H4Me), the
HOMO is a metal t2g-derived orbital and the LUMO is a p*
orbital on the bipyridine. The lowest energy band seen in their
UV/visible spectra is thus metal to bipyridine MLCT in nature,
appearing at around 490–500 nm with an extinction coefficient
in the order of 103 M−1 cm−1.1 The spectra of compounds 2 and
3 also have a lowest energy band in approximately the same
position, but this is much broader and an order of magnitude
stronger (e ≈ 104 M−1 cm−1), indicating that it has a different
origin (below).
The HOMOs of compounds 1a, 2a and 3a are metal based,
as before. These compounds all show RuII to RuIII oxidations at
relatively low (<0.6 V vs. SCE) potentials, and the ESR spectra of
samples of the representative compounds 2a+ and 3a+ (generated
in situ by oxidation of the neutral compound with ferrocenium
hexafluorophosphate) show the rhombic pattern† typical of
22
similar compounds.1
Oxidising these compounds creates a
vacancy in the dxy orbital of the metal,1 causing their colour to
change as an alkynyl-to-metal LMCT band appears. This has
been investigated using UV/visible spectroelectrochemistry for
2a and 3a, and the results for 3a are shown in Fig. 2. Upon
generation of 3a+ the broad band observed between 400 and
700 nm in the spectrum of 3a disappears and is replaced by
several other features, notably a band of medium strength at
around 400 nm, and two LMCT features at 740 (e ≈ 6532 M−1
cm−1) and 798 nm (e ≈ 6662). Analogous bands are also present
in the spectrum of electrochemically generated 2a+, at 611 (e ≈
2101) and 678 nm (e ≈ 3395).
Fig. 1 ORTEP (50% probability level) of 4+. Hydrogen atoms have
been removed for clarity.
˚
The N≡N bond distance of 1.011(6) A seen in the structure of
4 is among the shortest observed in ruthenium–dinitrogen com-
pounds. A search of the Cambridge Structural Database17 shows
that it contains 13 other mono-ruthenium compounds with a r-
bonded, non-bridging, dinitrogen ligand. Within experimental
˚
error, 11 of these possess N≡N distances of around 1.08–1.11 A,
comparable to the distance observed in free N2 (1.097 A).18 How-
˚
ever, there are two complexes that show short N≡N distances
19
˚
of around 1.01 A like 4, which are [RuTp(PEt3)2(N2)][BPh4]
˚
(Tp = hydrotris(pyrazolyl)borate) {1.01(2) A} and [Ru(16-
TMC)(N2)Cl][PF6]20 (16-TMC = 1,5,9,13-tetramethyl-1,5,9,13-
Table 1 Selected bond lengths and angles in the structure of 4+
Fig. 2 The UV/visible spectra of compound 3a before and after
oxidation and reduction.
Ru(1)–N(3)
Ru(1)–C(13)
Ru(1)–N(2)
Ru(1)–N(1)
1.994(5) Ru(1)–P(1)
2.032(5) Ru(1)–P(2)
2.084(4) C(13)–C(14)
2.124(4) N(3)–N(4)
2.3769(17)
2.4010(17)
1.179(7)
1.011(6)
N(3)–Ru(1)–C(13)
P(1)–Ru(1)–P(2)
94.79(19) C(14)–C(13)–Ru(1) 177.0(5)
176.29(5)
† For 2a+, g1 = 2.292, g2 = 2.260 and g3 = 1.891. For 3a+, g1 = 2.358,
g2 = 2.272 and g3 = 1.860.
4 1 3 4
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8

View Article Online
The UV/visible spectrum of electrochemically reduced 3a is
also shown in Fig. 2, which again shows the loss of the broad
band observed between 400 and 700 nm in the spectrum of 3a
and the production of several new features. Particularly strong
are two bands at 339 and 415 nm, each with e ≈ 40000; the
former of these is characteristic of reduced para-substituted
nitrobenzenes.24
Given the nature of the HOMOs and LUMOs of 2 and 3, the
broad, intense band visible in their UV/visible spectra between
about 400 and 700 nm is thus either metal-to-nitro charge
transfer, or more probably (given the distance between metal
and nitro group) a nitrophenylalkynyl localised p–p* band.25
This latter hypothesis is supported by the fact that upon adding
a second such ligand (i.e. on moving from 2 to 3), the strength of
the absorption increases by a factor of approximately 2. Judged
by the similarity of their UV/visible spectra this is true of all six
of the nitro-alkynyl compounds reported herein, although it is
worth noting that compounds 2b, 2c, 3b and 3c (vide infra) show
only irreversible cathodic processes in the cyclic voltammogram.
Nonlinear optical studies
Because a large number of ruthenium compounds containing
nitro-substituted arylalkynyl ligands show non-linear optical
activity,26 compounds Ru(Me2bipy)(PPh3)2(–C≡CR)Cl (R =
C6H4NO2 2a, But, Ph, C6H4Me) were investigated for non-linear
optical response at 800 nm. Cubic optical non-linearities were
determined by Z-scans, typical traces for the solvent and two
concentrations of 2a, together with theoretical curves, being
shown in Fig. 4. The deviations of Z-scans from that predicted
by theory that are seen in the wing regions are due to some
non-Gaussian character of the beam which, however, does not
influence the accuracy of the determination of the nonlinear
parameters.
Fig. 3 Simulated (top) and observed (bottom) ESR spectra of 3a2−
.
Unlike the mono-alkynyl complexes Ru(Me2bipy)(PPh3)2-
(–C≡CR)Cl (R = But, Ph, p-C6H4Me), the LUMO of com-
pounds 2a and 3a is not bipyridyl based. Cyclic voltammetry
reveals that they each have a reversible reduction wave at around
−1.20 V vs. SCE that is not displayed by the aforementioned
mono-alkynyl compounds, and this is due not to the reduction
of a bipyridyl p* based LUMO, but of a p* LUMO (or two
degenerate LUMOs in the case of 3a) based on the nitrophenyl
group(s) of the alkynyl ligand. This is confirmed by the ESR
spectrum displayed by an electrochemically reduced sample of
3a, which shows coupling to one nitrogen atom and to two pairs
of protons (Fig. 3 and Table 2), unequivocally confirming the ni-
trophenyl groups as the location of the reduction, since reduced
bipyridyl ligands show a five line pattern due to coupling to two
nitrogen nuclei. Upon reduction, 2a and 3a therefore behave sim-
ilarly to the platinum bis-alkynyl compound Pt(tBu2bipy)(–C≡
CC6H4NO2)2 that we have reported on previously.23 This too has
two degenerate nitrophenyl-based LUMOs that are populated
upon reduction, and also displays a diagnostic ESR pattern.
Fig. 4 Closed aperture Z-scans for pure CH2Cl2 and 1.5 and 3%
solutions of 2a. Wavelength = 800 nm, pulse duration ca. 150 fs.
Table 3 lists the real (c real) and imaginary (c imag) parts of
the cubic hyperpolarizability obtained from these studies. It
should be stressed that, primarily because of the limited range
of concentrations that were available, the results for two of the
compounds (R = p-C6H4Me, But) have error brackets that are
too high to be of use. For the remaining two compounds (R =
p-C6H4NO2 2a, Ph), the estimates of the real part of c are of
the order of 10−34 esu, which indicates moderate nonlinearities.
The imaginary parts of the nonlinearities are negligible (i.e. two-
photon absorption at 800 nm is quite weak), except for 2a, which
has a relatively pronounced absorptive nonlinearity at 800 nm,
Table 2 Parameters used for simulating the ESR spectrum of 3a2−
shown in Fig. 3. A Lorentzian linewidth of 0.45 G and a g value of
2.0155 were used
Nucleus
No. of nuclei
a/G
14N
1
2
2
1
9.856
3.278
1.052
0.53
1H
1H
99, 101Ru
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8
4 1 3 5

View Article Online
Table 3 Cubic nonlinear optical data for Ru(Me2bipy)(PPh3)2(–
C≡CR)Cl. Results of Z-scan measurements at 800 nm in CH2Cl2 (except
* at 670 nm)
R
c real/10−36 esu
c imag/10−36 esu
p-C6H4NO2 2a
−200 30
110 200
100 30
−65
4
3
2
p-C6H4Me
2
7
Ph
Ph
But
−1070 300*
13 8*
10
280 200
3
Fig. 5 A cruciform donor (D)–acceptor (A) system can lead to a polar
two-dimensional sheet of molecules.
its negative sign indicating that the effect is of the induced
transmission type rather than the two-photon absorption type
as for the other compounds. Indeed, 2a shows some one-photon
absorption at 800 nm and absorption saturation effects are seen.
The origin of the absorption is not clear, however, it could be a
tail of short-wavelength absorption or it could result from some
oxidation of the compound. The data for these compounds are
similar in magnitude to previously-reported ruthenium acetylide
complexes of comparable size.27
The data for Ru(Me2bipy)(PPh3)2(–C≡C–Ph)Cl at 670 nm
were obtained under conditions of relatively strong one-photon
absorption. From comparison with the 800 nm data for the same
compound, the change of sign of the nonlinearity and increase
of its absolute value may be postulated to be an effect of moving
the wavelength closer to a resonance.
ever, we reasoned that the ruthenium bis-nitrophenylalkynyl
systems reported above could provide a pathway into crystal
engineering using coordination compounds. Making such a
compound with a 4,4ꢀ-dihalobipyridine would create a cruciform
donor–acceptor system in the equatorial plane of the molecule,
which could cause it to form a two-dimensional network in
the solid state (Fig. 5). We were also interested in investigating
whether the attractive intermolecular forces that could cause
relatively small organic molecules to aggregate as described
above would be strong enough to influence the packing of much
larger metal-containing molecules, where such forces would
contribute a much smaller fraction of the total lattice energy.
The synthesis of Ru(Br2bipy)(PPh3)2(–C≡CC6H4NO2)2 3b
was readily achieved from the corresponding dichloride 1b via
the mono-alkynyl compound 2b, using the procedure described
above for the synthesis of 3a from 1a. Crystals suitable for X-
ray diffraction were grown from a refrigerated toluene solution;
the resulting structure of 3b is shown in Fig. 6, and selected
bond lengths and angles are presented in Table 4. Whilst this
structure confirms the general geometry of the compound – cis-
alkynyl and trans-triphenylphosphine ligands–examination of
the packing reveals that there are no interactions between the
bromo and nitro groups.
Crystal engineering
There have been several papers published recently concerning
the crystal structures of organic compounds containing both
nitro and iodo functionalities.28 It has been demonstrated that
the polarisation-assisted attractive interaction between the two
is strong enough to influence the packing of molecules in the
solid state, and in particular the elegant work of Desiraju and
co-workers has shown how pre-defined packing geometries may
be achieved, since molecules tend to adopt an arrangement
that reflects the disposition of the two halves of the syn-
thon. Thus, to take just three examples, the crystal structures
of para-iodonitrobenzene and 4-iodo-4ꢀ-nitrobiphenyl, where
the iodo and nitro groups in each molecule are at 180◦
to each other, contain linear chains of molecules linked by
iodo–nitro interactions,29 whereas that of 4,4ꢀ-diiodo-4ꢀꢀ,4ꢀꢀꢀ-
dinitrotetraphenylmethane, in which two iodo and two nitro
groups are arranged tetrahedrally around a carbon centre,
contains a diamondoid network.30 Although less commonly
used for crystal engineering than iodo-nitro interactions, bromo-
nitro interactions have also been shown to have the capacity for
structure direction.31
Though 4,4ꢀ-diiodo-2,2ꢀ-bipyridine (I2bipy) has not been
reported before, 4-iodopyridine is readily synthesised from
4-aminopyridine via a Sandmeyer reaction.12 This procedure
also works when applied to 4,4ꢀ-diamino-2,2ꢀ-bipyridine, yield-
ing I2bipy as a stable, crystalline white solid. As expected,
this readily coordinates to ruthenium, and the synthesis of
Ru(I2bipy)(PPh3)2(–C≡CC6H4NO2)2 3c was achieved in an
analogous way to 3a and 3b. The first crystals of 3c for X-ray
diffraction were grown by refrigeration of a toluene solution.
The asymmetric unit of the crystal structure (hereafter referred
to as structure A) contains two crystallographically independent
molecules of 3c and five of toluene. Generally the bond lengths
and angles observed within the two independent molecules are
similar to each other and to the corresponding value in the
structure of 3b (Table 4), but a striking exception is the angle
comprised of the two atoms of the carbon–carbon triple bond
All previous work that has used halo–nitro interactions for
crystal engineering has concerned purely organic systems. How-
◦
˚
Table 4 Selected bond lengths (A) and angles ( ) in the structures of compounds 3b and 3c. Structure A of the latter has two independent molecules
in the asymmetric unit, and thus has two values for each measurement
3c
3b
Structure A
Structure B
Ru(1)–N(1)
Ru(1)–N(2)
Ru(1)–C(11)
Ru(1)–C(19)
C(11)–C(12)
C(19)–C(20)
2.114(3)
2.111(3)
1.991(4)
1.984(4)
1.216(5)
1.220(5)
2.104(4)
2.123(4)
1.993(5)
1.985(5)
1.206(7)
1.210(7)
2.137(4)
2.117(4)
1.986(4)
1.999(4)
1.220(6)
1.213(6)
2.100(3)
2.125(3)
1.992(4)
1.986(4)
1.203(5)
1.190(5)
Ru(1)–C(11)–C(12)
C(11)–C(12)–C(13)
Ru(1)–C(19)–C(20)
C(19)–C(20)–C(21)
P(1)–Ru(1)–P(2)
174.8(3)
176.3(4)
176.7(3)
177.1(4)
173.72(4)
177.9(4)
171.8(6)
178.0(5)
170.2(5)
171.71(5)
177.5(4)
168.4(5)
175.1(4)
177.3(5)
171.33(4)
173.9(4)
173.3(4)
177.0(4)
169.7(4)
173.98(4)
4 1 3 6
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8

View Article Online
˚
Table 5 Iodo–oxygen distances (A) seen in structure A of 3c, illustrat-
ing the unsymmetric nature of the iodo–nitro interaction. The sum of
˚
the van der Waals’ radii of iodine and oxygen is 3.50 A, and thus the
shorter iodo–oxygen contacts d1 constitute stabilising interactions. The
angle v is the torsion angle O2–N–O1–I; in other words, the angle of
elevation of the iodo group from the planar nitro group
Molecule
d1
d2
v
1
1
2
2
3.126
2.924
3.107
3.272
4.743
3.929
4.775
4.115
75.09
78.09
80.00
83.14
Fig. 6 ORTEP (50% probability level) of the structure of 3b. Hydrogen
atoms have been removed for clarity.
and the ipso carbon atom of the nitrophenyl ring, varying from
177.3(4)◦ in one molecule to 168.4(5)◦ in the other.
Examination of the packing within structure A of 3c reveals
that it does contain iodo–nitro interactions that orient the
molecules. However, instead of the envisaged two-dimensional
network, two separate one-dimensional chains running in
opposite directions are observed, each comprised of one of
the two different molecules of 3c (Fig. 7). This is due to
the nature of the iodo–nitro interaction seen (Table 5), which
links molecules together in a convergent, rather than divergent,
fashion. Somewhat unusually, the iodo atoms are not in the plane
of the nitro group, which is by far the most common angle of
interaction,28 but are almost perpendicular to it. It is this feature
that allows the convergence of the two interactions between the
donor and acceptor molecules and leads to chains rather than
sheets.
Fig. 8 ORTEP (50% probability level) of 3c from structure B, showing
that the dichloromethane solvent molecule is oriented to allow H(10D)
to interact with the C(11)–C(12) triple bond. Most hydrogen atoms have
been removed for clarity.
As the iodo–nitro interaction in structure A of 3c is rather
atypical, and because of the presence of several molecules of
toluene, a second set of crystals (as a dichloromethane solvate)
for X-ray diffraction were grown from a different solvent system
(layered dichloromethane–hexane). The metrics of 3c within
structure B are very similar to those within structure A (Table 4),
but unlike structure A, structure B contains no short iodo–nitro
Conclusions
The results obtained for the reactions of the ruthenium mono-
alkynyl complexes with further alkyne in the presence of TlPF6
allow us to draw some conclusions regarding the stability of
the desired compounds. Both p-nitrophenylacetylene and p-
tolylacetylene are of similar size, yet the alkynyl compound
2 of the former readily forms an alkynyl–vinylidene complex
˚
interactions. It does, however, have a 3 A contact (Fig. 8) between
the acidic hydrogen atom of the CH2Cl2 solvent molecule and
the triple bond of the alkynyl ligand, of the type recently noted.2
Fig. 7 Chains of molecules with iodo–nitro interactions in structure A of 3c. Each chain is composed solely of one of the two independent molecules
in the asymmetric unit.
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8
4 1 3 7

View Article Online
(which is then converted in-situ to the bis-alkynyl complex 3)
whilst the latter does not, implying that an electronic factor
can account for the difference. As p-tolylacetylene is a much
better electron donor than p-nitrophenylacetylene, it seems
that electron-withdrawing groups can stabilise this alkynyl–
vinylidene intermediate and allow subsequent synthesis of
bis-alkynyl compounds, a hypothesis that is currently under
investigation in our laboratory. tert-Butylacetylene is both a
better electron donor and much bulkier than the aforementioned
arylalkynes; under the above postulate, the former property
implies that the alkynyl-vinylidene intermediate would not be
stable, but the latter property seems to prevent formation of
this in the first place, and rather than fit another bulky ligand
around the metal a sterically undemanding dinitrogen ligand is
incorporated instead.
4 P. S. Hallman, T. A. Stephenson and G. A. Wilkinson, Inorg. Synth.,
1970, 12, 237.
5 C. J. Adams, J. Chem. Soc., Dalton Trans., 2002, 1545.
6 G. Maerker and F. H. Case, J. Am. Chem. Soc., 1958, 80, 2745.
7 G.-J. ten Brink, I. W. C. E. Arends, M. Hoogenraad, G. Verspui and
R. A. Sheldon, Adv. Synth. Catal., 2003, 345, 497.
8 S. Takahashi, Y. Kuroyama, K. Sonogashira and N. Hagihara,
Synthesis, 1980, 627.
9 N. G. Connelly and W. E. Geiger, Chem. Rev., 1996, 96, 877.
10 S.-M. Lee, R. Kowallick, M. Marcaccio, J. A. McCleverty and M. D.
Ward, J. Chem. Soc., Dalton Trans., 1998, 3443.
11 A. Hilton, T. Renouard, O. Maury, H. L. Bozec, I. Ledoux and J.
Zyss, Chem. Commun., 1999, 2521.
12 C. Coudret, Synth. Commun., 1996, 26, 3543.
13 G. M. Sheldrick, SHELXS-97, Program for Crystal Structure Deter-
mination, University of Go¨ttingen, 1997.
14 G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
Refinement, University of Go¨ttingen, 1997.
In having a low-lying p* orbital the nitrophenylalkynyl
ligand is not typical of the majority of arylalkynyl ligands.
The consequence of this is that the optical properties of the
bis(nitrophenylalkynyl) ruthenium compounds reported herein
are probably not going to be the same as those of similar
compounds with other ligands.
15 R. Denis, L. Toupet, F. Paul and C. Lapinte, Organometallics, 2000,
19, 4240; F. Paul, J.-Y. Mevellec and C. Lapinte, J. Chem. Soc., Dalton
Trans., 2002, 19, 1783.
16 B. Folkesson, Acta Chem. Scand., 1972, 26, 4008.
17 F. H. Allen, Acta Crystallogr., Sect. B, 2002, 58, 380.
18 P. G. Wilkinson and N. B. Houk, J. Chem. Phys., 1956, 24, 528.
19 M. A. J. Tenorio, M. J. Tenorio, M. C. Puerta and P. Valerga, J. Chem.
Soc., Dalton Trans., 1998, 3601.
Our foray into crystal engineering has also produced some
interesting results. Not the least of these is the first synthesis of
4,4ꢀ-diiodo-2,2ꢀ-bipyridine, which the weakness of the carbon–
iodine bond makes a potentially valuable precursor to 4,4ꢀ-
disubstituted bipyridines. However, using it to interact with
nitro-containing alkynyl ligands has been less successful. Al-
though structure A of 3c does contain short iodo-nitro contacts,
structure B does not, and furthermore the non-planarity of
the iodo-nitro interaction in structure A is not typical of that
normally observed when using it for the construction of a
supramolecular motif. This leads us to conclude that the strength
of this interaction is probably not great enough to be structure
determining in the current system, and that its presence in
structure A, although probably attractive in nature, is not the
major influence on the packing geometry.
20 W.-H. Chiu, C.-M. Che and T. C. W. Mak, Polyhedron, 1996, 15,
4421.
21 G. Trimmel, C. Slugovc, P. Wiede, K. Mereiter, V. N. Sapunov, R.
Schmid and K. Kirchner, Inorg. Chem., 1997, 36, 1076.
22 J. Chakravarty and S. Bhattacharya, Polyhedron, 1994, 13, 2671.
23 C. J. Adams, S. L. James, X. Liu, P. R. Raithby and L. J. Yellowlees,
J. Chem. Soc., Dalton Trans., 2000, 63.
24 R. G. Compton, R. A. W. Dryfe and A. C. Fisher, J. Chem. Soc.,
Perkin Trans. 2, 1994, 1581.
25 C. E. Whittle, J. A. Weinstein, M. W. George and K. S. Schanze,
Inorg. Chem., 2001, 40, 4053.
26 S. K. Hurst, M. P. Cifuentes, J. P. L. Morrall, N. T. Lucas, I. R.
Whittall, M. G. Humphrey, I. Asselberghs, A. Persoons, M. Samoc,
B. Luther-Davies and A. C. Willis, Organometallics, 2001, 20, 4664;
C. E. Powell and M. G. Humphrey, Coord. Chem. Rev., 2004, 248,
725.
27 I. R. Whittall, M. G. Humphrey, M. Samoc, J. Swiatkiewicz and
B. Luther-Davies, Organometallics, 1995, 14, 5493; I. R. Whittall,
M. G. Humphrey, M. P. Cifuentes, M. Samoc, B. Luther-Davies, A.
Persoons and S. Houbrechts, Organometallics, 1997, 16, 2631.
28 F. H. Allen, J. P. M. Lommerse, V. J. Hoy, J. A. K. Howard and G. R.
Desiraju, Acta Crystallogr., Sect. B, 1997, 53, 1006.
29 J. A. R. P. Sarma, F. H. Allen, V. J. Hoy, J. A. K. Howard, R.
Thaimattam, K. Biradha and G. R. Desiraju, Chem. Commun., 1997,
101; V. R. Thalladi, B. S. Goud, V. J. Hoy, F. H. Allen, J. A. K. Howard
and G. R. Desiraju, Chem. Commun., 1996, 401.
30 R. Thaimattam, C. V. K. Sharma, A. Clearfield and G. R. Desiraju,
Cryst. Growth Des., 2001, 1, 103.
31 A. D. Bond, D. A. Haynes and J. M. Rawson, Can. J. Chem., 2002,
80, 1507.
Acknowledgements
C. J. A. thanks the University of Bristol for a Junior Fellowship,
and M. G. H. thanks the Australian Research Council for an
ARC Australian Professorial Fellowship.
References
1 C. J. Adams and S. J. A. Pope, Inorg. Chem., 2004, 43, 3492.
2 W. Lu, M. C. W. Chan, N. Zhu, C.-M. Che, Z. He and K.-Y. Wong,
Chem. Eur. J., 2003, 6155.
3 A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen and
F. J. Timmers, Organometallics, 1996, 15, 1518.
4 1 3 8
D a l t o n T r a n s . , 2 0 0 4 , 4 1 3 0 – 4 1 3 8