Tuning of charge-transfer absorption and molecular quadratic non-linear optical properties in ruthenium(II) ammine complexes

Article abstract of DOI:10.1039/a905652a

The ligands N-methyl-2,7-diazapyrenium (Medap⁺), N-(2-pyrimidyl)-4,4′-bipyridinium (PymQ⁺), N-methyl-4-[trans-2-(4-pyridyl)ethenyl]pyridinium (Mebpe⁺) and N-phenyl-4-[trans-2-(4-pyridyl)ethenyl]pyridinium (Phbpe⁺

Full text of DOI:10.1039/a905652a

View Article Online / Journal Homepage / Table of Contents for this issue
Tuning of charge-transfer absorption and molecular quadratic
non-linear optical properties in ruthenium(II) ammine complexes†
Benjamin J. Coe,*a James A. Harris,a Inge Asselberghs,b André Persoons,b,c John C. Jeffery,d
Leigh H. Rees,d Thomas Gelbriche and Michael B. Hursthousee
a Department of Chemistry, University of Manchester, Oxford Road, Manchester,
UK M13 9PL. E-mail: b.coe@man.ac.uk
b Laboratory of Chemical and Biological Dynamics, Center for Research on Molecular
Electronics and Photonics, University of Leuven, Celestijnenlaan 200D, B-3001 Leuven,
Belgium
c Optical Sciences Center, University of Arizona, Tucson, Arizona, AZ 85721, USA
d Department of Chemistry, University of Bristol, Cantock’s Close, Bristol, UK BS8 1TS
e EPSRC X-ray Crystallography Service, Department of Chemistry, University of
Southampton, Highfield, Southampton, UK SO17 1BJ
Received 13th July 1999, Accepted 1st September 1999
The ligands N-methyl-2,7-diazapyrenium (Medapϩ), N-(2-pyrimidyl)-4,4Ј-bipyridinium (PymQϩ), N-methyl-4-[trans-
2-(4-pyridyl)ethenyl]pyridinium (Mebpeϩ) and N-phenyl-4-[trans-2-(4-pyridyl)ethenyl]pyridinium (Phbpeϩ) have
been used to prepare a series of complex salts trans-[RuII(NH3)4(LD)(LA)][PF6]3 [LD = NH3 and LA = Medapϩ 1,
PymQϩ 2, Mebpeϩ 3 or Phbpeϩ 4; LD = pyridine (py) and LA = Medapϩ 8, PymQϩ 9, Mebpeϩ 10 or Phbpeϩ 11;
LD = 1-methylimidazole (mim) and LA = Medapϩ 12, PymQϩ 13, Mebpeϩ 14 or Phbpeϩ 15]. The salt trans-[RuII(NH3)4-
(py)(4,4Ј-bpy)][PF6]2 (4,4Ј-bpy = 4,4Ј-bipyridine) 16 has also been prepared. The dipolar complexes in 1–4 and 8–15
exhibit intense dπ(RuII)→π*(LA) metal-to-ligand charge-transfer (MLCT) absorptions in the region 560–700 nm. For
a given LA, the MLCT energy decreases as the donor strength of LD increases, in the order py < NH3 < mim. Within
the pairs of Medapϩ/PymQϩ complexes, the energy of the Ru-based HOMO is constant and the MLCT energy
decreases by ca. 0.3 eV as the acceptor strength of LA increases on going from Medapϩ to PymQϩ. The complexes of
Mebpeϩ or Phbpeϩ also have similar HOMO energies which are lower than those of their Medapϩ/PymQϩ counter-
parts due to the increased basicity of LA. Replacement of Mebpeϩ by Phbpeϩ decreases the MLCT energy by ca. 0.1
eV due to the greater electron-withdrawing ability of Phbpeϩ. Single-crystal structures of 8ؒ4MeCN and 16ؒ2MeCN
have been determined. Molecular first hyperpolarizabilities β of 1–4 and 8–15, obtained from hyper-Rayleigh
scattering measurements at 1064 nm, are in the range (579–1068) × 10Ϫ30 esu. Static hyperpolarizabilities β0 derived
by using the two-level model are also very large, with 13 having the largest at 336 × 10Ϫ30 esu. In general, β0 increases
as the absorption energy decreases, in keeping with the two-level model.
with {RuII(η5-LЈ)L2}ϩ [LЈ = cyclopentadienyl or indenyl; L2 =
Introduction
mono- or bi-dentate phosphine ligand(s)] or trans-{RuIICl-
Molecular materials possessing non-linear optical (NLO)
(dppm)2}ϩ [dppm = bis(diphenylphosphino)methane] electron
properties are of great interest for applications in optoelectronic
donor groups.5–8 We are investigating the quadratic NLO
and photonic devices of the 21st century.1 Organotransition
behaviour of ruthenium ammines which combine synthetic
metal complexes form an important sub-class of NLO
accessibility with especially well understood redox and spectro-
materials which have recently attracted increasing attention.2
scopic properties. Initial studies have shown that {RuII-
Such compounds are promising because polarizable d electrons
can contribute to enhanced NLO responses, and redox activity
provides extensive opportunities for modulation of NLO
properties. The creation of materials for quadratic (second
order) NLO applications begins with the molecular engineering
of chromophores possessing large first hyperpolarizabilities,
(NH3)5}2ϩ or trans-{RuII(NH3)4L}2ϩ (L = 4-dimethylamino-
pyridine or 1-methylimidazole) complexes of N-methyl/aryl-
4,4Ј-bipyridinium ligands possess very large β values.9 In such
dipolar molecules the d6 ruthenium() centre acts as a powerful
π donor whilst the pyridinium ring is an acceptor. These com-
plexes hence exhibit intense, low energy metal-to-ligand charge-
β. Detailed structure–activity relationships for β of purely
transfer (MLCT) absorptions, and a good correlation exists
organic compounds are well established,1 but comparable
between β0 (the static first hyperpolarizability) and 1/E2 where E
understanding for organotransition metal complexes is poorly
is the MLCT energy.9c This is in accord with the two-level
developed.2 Systematic investigations into the quadratic NLO
model which is often used to relate charge-transfer absorption
activity of metal complexes are hence timely.
The quadratic NLO properties of ruthenium complexes have
and NLO properties in classical dipolar organic chromo-
phores.10 Furthermore, the NLO responses of the {RuII-
proven an especially popular research topic, and large β
(NH3)5}2ϩ complexes can be reversibly and very effectively
responses have been reported.3–8 Most of these studies have
involved organometallic σ-acetylide or allenylidene complexes
switched via RuIII/II redox.11
We have begun to develop a versatile family of complexes,
the MLCT absorptions and quadratic NLO responses of which
are readily controlled. The subject of this report is the further
tuning of these properties in ruthenium() ammine complexes,
† Supplementary data available: rotatable 3-D crystal structure diagram
in CHIME format. See http://www.rsc.org/suppdata/dt/1999/3617/
J. Chem. Soc., Dalton Trans., 1999, 3617–3625
This journal is © The Royal Society of Chemistry 1999
3617

View Article Online
principally by changing the structure of the acceptor-
substituted ligand. Previous work with NLO-active transition
metal complexes has often borrowed “molecular engineering”
concepts from purely organic molecules,2 for example by simply
replacing an amino donor with an electron-rich metal centre.
We are seeking to develop novel approaches towards modifying
the properties of MLCT chromophores which have no existing
counterparts in simple organics. The results of these studies are
of direct relevance to the design of other transition metal-based
NLO compounds and also related organic chromophores.
ethanol (30 cm3), aniline (5.6 cm3, 61 mmol) was added and the
resulting solution heated at reflux for 3 h. The solution was
cooled to room temperature and reduced to half volume in
vacuo. On addition of water a green material precipitated which
was filtered off, washed with water and discarded. The filtrate
was reduced to dryness and dissolved in ethanol. The addition
of diethyl ether afforded a golden-brown precipitate which was
filtered off, washed with diethyl ether and dried: yield 2.24 g
(39%). δH(D2O) 8.85 (2 H, d, J 6.6, C5H4N), 7.99 (2 H, d, J 6.9
Hz, C5H4N), 7.67 (5 H, s, Ph) and 2.71 (3 H, s, Me) (Found:
C, 63.30; H, 6.06; N, 6.23. Calc. for C12H12ClNؒ1.25H2O:
C, 63.16; H, 6.40; N, 6.14%).
Experimental
N-Phenyl-4-[trans-2-(4-pyridyl)ethenyl]pyridinium chloride,
[Phbpe؉]Cl. A mixture of [Phpicϩ]Clؒ1.25H2O (3.08 g, 13.5
mmol), pyridine-4-carbaldehyde (1.5 cm3, 15.8 mmol) and
piperidine (0.4 cm3) in ethanol (3 cm3) was heated at reflux
for 15 min. The resulting solution was cooled in an ice-bath,
causing formation of a solid mass. A black liquid was decanted
off and the residue dissolved in ethanol and precipitated by
addition of diethyl ether. The pink solid was filtered off, washed
with diethyl ether and dried: yield 1.04 g (23%). δH(D2O) 8.94
(2 H, d, J 6.8, C5H4N), 8.48 (2 H, d, J 5.9, C5H4N), 8.19 (2 H, d,
J 6.7, C5H4N), 7.71 (1 H, d, J 16.4, CH), 7.69 (5 H, s, Ph), 7.59
(2 H, d, J 6.2, C5H4N) and 7.51 (1 H, d, J 16.4 Hz, CH) (Found:
C, 64.19; H, 5.38; N, 8.06. Calc. for C18H15ClN2ؒ2.25H2O: C,
64.48; H, 5.86; N, 8.35%).
Materials and procedures
The compound RuCl3ؒ2H2O was supplied by Johnson Matthey
plc. The salts [RuII(NH3)5(H2O)][PF6]2,12 trans-[RuIICl(NH3)4-
(SO2)]Cl,12 trans-[RuIII(SO4)(NH3)4(py)]Cl,12 N-methyl-2,7-di-
azapyrenium chloride ([Medapϩ]Cl)13 and N-methyl-4-[trans-
2-(4-pyridyl)ethenyl]pyridinium iodide ([Mebpeϩ]I)14 were pre-
pared according to published procedures. The latter two salts
were metathesized to [Medapϩ]PF6 and [Mebpeϩ]PF6, respec-
tively, by precipitation from water–aqueous NH4PF6. 2,7-
Diazapyrene was prepared by using a combination of published
procedures,15,16 with one simple modification: in the synthesis
of the intermediate 1,3,6,8-tetrahydro-2,7-dimethyldiazapyrene
we found that extraction into boiling methanol affords a higher
yield of a purer product compared with the published THF
extraction method.16 All other reagents were obtained com-
mercially and used as supplied. Products were dried overnight
at room temperature in a vacuum desiccator (CaSO4) prior to
characterization.
N-Phenyl-4-[trans-2-(4-pyridyl)ethenyl]pyridinium hexafluoro-
phosphate, [Phbpe؉]PF6. The salt [Phbpeϩ]Clؒ2.25H2O (200 mg,
0.596 mmol) was dissolved in the minimum volume of water
and aqueous NH4PF6 added dropwise. The white precipitate
was filtered off, washed with water and dried: yield 210 mg
(87%). δH(CD3COCD3) 9.31 (2 H, d, J 6.8, C5H4N), 8.71 (2 H,
d, J 6.1, C5H4N), 8.57 (2 H, d, J 6.8, C5H4N), 8.16 (1 H, d,
J 16.4, CH), 8.00–7.95 (2 H, m, Ph), 7.94 (1 H, d, J 16.4, CH),
7.82–7.78 (3 H, m, Ph) and 7.72 (2 H, d, J 6.0 Hz, C5H4N)
(Found: C, 53.56; H, 3.77; N, 6.72. Calc. for C18H15F6N2P:
C, 53.48; H, 3.74; N, 6.93%).
Physical measurements
Proton NMR spectra were recorded on a Varian Gemini 200
spectrometer and all shifts are referenced to SiMe4. The fine
splitting of pyridyl or phenyl ring AAЈBBЈ patterns is ignored
and the signals are reported as simple doublets, with J values
referring to the two most intense peaks. Solvents of crystalliz-
ation were detected as singlets at δ 2.86 (water) or 2.09 (acetone)
in acetone-d6 solutions. Elemental analyses were performed
by the Microanalytical Laboratory, University of Manchester
and UV/VIS spectra were obtained by using a Varian Cary 1E
spectrophotometer.
N-(2-Pyrimidyl)-4,4Ј-bipyridinium chloride, [PymQ؉]Cl. A
mixture of 4,4Ј-bipyridine (2.06 g, 13.2 mmol) and 2-chloro-
pyrimidine (1.00 g, 8.73 mmol) was heated quickly to ca. 70 ЊC.
Upon solidification, the green material was dissolved in ethanol
(5 cm3) and heated at reflux for 6 h. Addition of diethyl ether
afforded a green precipitate which was filtered off, washed with
diethyl ether and dried: yield 1.30 g (55%). δH(D2O) 10.08 (2 H,
d, J 7.3, C5H4N), 9.10 (2 H, d, J 4.9, C4H3N2), 8.74 (2 H, d,
J 6.1, C5H4N), 8.61 (2 H, d, J 7.3, C5H4N), 7.96 (2 H, d, J 6.3,
C5H4N) and 7.85 (1 H, t, J 5.0 Hz, C4H3N2) (Found: C, 61.94;
H, 4.42; N, 20.39. Calc. for C14H11ClN4: C, 62.11; H, 4.10;
N, 20.70%).
Cyclic voltammetric measurements were carried out by
using an EG&G PAR model 173 potentiostat/galvanostat with
a model 175 universal programmer. A single-compartment
cell was used with the saturated calomel reference electrode
(SCE) separated by a salt bridge from the platinum-bead
working electrode and platinum-wire auxiliary electrode.
Acetonitrile (HPLC grade) was used as received and tetra-n-
butylammonium hexafluorophosphate, twice recrystallized
from ethanol and dried in vacuo, as supporting electrolyte.
Solutions containing ca. 10Ϫ3 mol dmϪ3 analyte (0.1 mol dmϪ3
electrolyte) were deaerated by purging with N2. All E1/2 values
N-(2-Pyrimidyl)-4,4Ј-bipyridinium
hexafluorophosphate,
[PymQ؉]PF6. The salt [PymQϩ]Cl (500 mg, 1.85 mmol) was
dissolved in the minimum volume of water and aqueous
NH4PF6 added dropwise. The white precipitate was filtered
off, washed with water and dried: yield 665 mg (95%).
δH(CD3COCD3) 10.36 (2 H, d, J 7.4, C5H4N), 9.29 (2 H, d,
J 4.8, C4H3N2), 8.96–8.91 (4 H, m, C5H4N), 8.14 (2 H, d, J 6.3,
C5H4N) and 8.04 (1 H, t, J 4.8 Hz, C4H3N2) (Found: C, 44.36;
H, 2.93; N, 14.88. Calc. for C14H11F6N4P: C, 44.22; H, 2.92;
N, 14.73%).
were calculated from (Epa ϩ Epc)/2 at a scan rate of 200 mV sϪ1
.
Syntheses
4-Methyl-N-phenylpyridinium chloride, [Phpic؉]Cl. A solu-
tion of 4-methylpyridine (2.4 cm3, 25 mmol) and chloro-2,4-
dinitrobenzene (5 g, 25 mmol) in ethanol (25 cm3) was heated
at reflux for 2 h. After cooling to room temperature, diethyl
ether was added and the black precipitate filtered off, washed
with diethyl ether and dried. The crude product was purified by
precipitation from boiling ethanol–diethyl ether to afford a grey
solid which was filtered off, washed with diethyl ether and
dried: 3.68 g. δH(D2O) 9.31 (1 H, d, J 2.5 Hz, H3), 8.89–8.83
(3 H, m, H5 ϩ C5H4N), 8.19–8.06 (3 H, m, H6 ϩ C5H4N)
and 2.79 (3 H, s, Me). This material was dissolved in boiling
[RuII(NH3)5(Medap؉)][PF6]3 1. A solution of [RuII(NH3)5-
(H2O)][PF6]2 (250 mg, 0.505 mmol) and [Medapϩ]PF6 (188 mg,
0.516 mmol) in Ar-degassed acetone (5 cm3) was stirred at
room temperature under Ar for 2 h. The addition of diethyl
ether afforded a dark precipitate which was filtered off, washed
with diethyl ether and dried. Purification was effected by
3618
J. Chem. Soc., Dalton Trans., 1999, 3617–3625

View Article Online
precipitations from acetone–aqueous NH4PF6 and then from
acetone–diethyl ether to afford a dark purple solid: yield 137
mg (32%). δH(CD3COCD3) 10.05 (2 H, s, C14H8N2), 9.83 (2 H, s,
C14H8N2), 8.54 (2 H, d, J 9.2, C14H8N2), 8.48 (2 H, d, J 9.2 Hz,
C14H8N2), 4.89 (3 H, s, Me), 3.59 (3 H, s, trans-NH3) and 2.81
(12 H, s, 4 × cis-NH3) (Found: C, 21.10; H, 3.08; N, 11.49. Calc.
for C15H26F18N7P3Ru: C, 21.44; H, 3.12; N, 11.67%).
trans-[RuIII(SO4)(NH3)4(Mebpe؉)]Cl2 7. This was prepared
in identical manner to that of salt 6 by using [Mebpeϩ]I
(214 mg, 0.660 mmol) in place of [Medapϩ]Cl to afford a
golden solid: crude yield 135 mg (77%).
trans-[RuII(NH3)4(py)(Medap؉)][PF6]3 8. A solution of trans-
[RuIII(SO4)(NH3)4(py)]Cl (108 mg, 0.284 mmol) in water (5 cm3)
was reduced over zinc amalgam (5 lumps) with argon agitation
for 15 min. This was filtered under Ar into a flask containing
[Medapϩ]Cl (360 mg, 1.41 mmol) and the solution was stirred
at room temperature in the dark under Ar for 6 h. The addition
of acetone (100 cm3) to the deep blue solution gave a dark
precipitate which was filtered off, washed with acetone and
dried (crude trans-[RuII(NH3)4(py)(Medapϩ)]Cl3). This material
was purified by precipitation from water–acetone and then
[RuII(NH3)5(PymQ؉)][PF6]3 2. This was prepared in identical
manner to salt 1 by using [RuII(NH3)5(H2O)][PF6]2 (100 mg,
0.202 mmol) and [PymQϩ]PF6 (77 mg, 0.203 mmol) in place
of [Medapϩ]PF6. Purification was effected by several precip-
itations from acetone–diethyl ether to afford a dark blue solid:
yield 75 mg (42%). δH(CD3COCD3) 10.17 (2 H, d, J 7.5,
C5H4N), 9.28–9.22 (4 H, m, C5H4N ϩ C4H3N2), 8.94 (2 H, d,
J 7.5 Hz, C5H4N), 8.03–7.96 (3 H, m, C5H4N ϩ C4H3N2), 3.75
(3 H, s, trans-NH3) and 2.74 (12 H, s, 4 × cis-NH3) (Found:
C, 19.15; H, 3.14; N, 13.76. Calc. for C14H26F18N9P3Ruؒ2H2O:
C, 18.84; H, 3.39; N, 14.13%).
Ϫ
metathesized to its PF6 salt by precipitation from water–
aqueous NH4PF6. Further purification was effected by several
precipitations from acetone–diethyl ether to afford a dark blue
solid: yield 99 mg (39%). δH(CD3COCD3) 10.15 (2 H, s,
C14H8N2), 9.99 (2 H, s, C14H8N2), 9.00 (2 H, d, H2,6), 9.12 (2 H,
d, J 9.1, C14H8N2), 8.63 (2 H, d, J 9.2 Hz, C14H8N2), 8.03 (1 H,
t, H4), 7.65 (2 H, t, H3,5), 5.01 (3 H, s, Me) and 2.83 (12 H, s,
4NH3) (Found: C, 26.78; H, 3.09; N, 10.61. Calc. for C20H28-
F18N7P3Ru: C, 26.62; H, 3.13; N, 10.86%).
[RuII(NH3)5(Mebpe؉)][PF6]3 3. This was prepared in similar
manner to salt 2 by using [Mebpeϩ]I (79 mg, 0.244 mmol)
in place of [PymQϩ]PF6 and 1:1 water–acetone (5 cm3) in
place of acetone. The addition of aqueous NH4PF6 afforded a
dark precipitate which was filtered off, washed with water and
dried. Purification was effected by precipitations from acetone–
trans-[RuII(NH3)4(py)(PymQ؉)][PF6]3 9. This was prepared
in identical manner to that of salt 8 by using trans-
[RuIII(SO4)(NH3)4(py)]Cl (103 mg, 0.271 mmol) and [PymQϩ]Cl
(367 mg, 1.36 mmol) in place of [Medapϩ]Cl. Purification was
effected by precipitation from acetone–diethyl ether to afford
a dark blue solid: yield: 172 mg (69%). δH(CD3COCD3) 10.28
(2 H, d, J 7.4, C5H4N), 9.27 (2 H, d, J 4.9, C4H3N2), 9.26 (2 H,
d, J 7.0, C5H4N), 8.99–8.93 (4 H, m, C5H4N ϩ H2,6), 8.19 (2 H,
d, J 7.0 Hz, C5H4N), 8.06–7.98 (2 H, m, H4 ϩ C4H3N2), 7.63
(2 H, t, H3,5) and 2.88 (12 H, s, 4NH3) (Found: C, 25.16; H, 3.10;
N, 13.32. Calc. for C19H28F18N9P3Ru: C, 24.85; H, 3.07; N,
13.73%).
diethyl ether, acetone–NBun Cl and finally water–aqueous
4
NH4PF6 to afford a dark purple solid: yield 58 mg (35%).
δH(CD3COCD3) 8.94 (4 H, d, J 6.9, C5H4N), 8.33 (2 H, d, J 7.0,
C5H4N), 8.02 (1 H, d, J 16.4, CH), 7.89 (1 H, d, J 16.4, CH),
7.54 (2 H, d, J 6.8 Hz, C5H4N), 4.49 (3 H, s, Me), 3.41 (3 H, s,
trans-NH3) and 2.65 (12 H, s, 4 × cis-NH3) (Found: C, 19.28;
H, 3.18; N, 11.68. Calc. for C13H28F18N7P3Ru: C, 19.08; H, 3.45;
N, 11.98%).
[RuII(NH3)5(Phbpe؉)][PF6]3 4. This was prepared in identical
manner to salt 2 by using [Phbpeϩ]PF6 (83 mg, 0.205 mmol)
in place of [PymQϩ]PF6. The product was purified as for 3 with
one further precipitation from acetone–diethyl ether to afford a
dark blue solid: yield 113 mg (62%). δH(CD3COCD3) 9.25 (2 H,
d, J 7.0, C5H4N), 8.97 (2 H, d, J 6.3, C5H4N), 8.51 (2 H, d, J 6.9,
C5H4N), 8.18 (1 H, d, J 16.2, CH), 8.04 (1 H, d, J 16.2, CH),
7.99–7.94 (2 H, m, Ph), 7.81–7.77 (3 H, m, Ph), 7.58 (2 H, d,
J 6.9 Hz, C5H4N), 3.44 (3 H, s, trans-NH3) and 2.65 (12 H, s,
4 × cis-NH3) (Found: C, 25.28; H, 3.52; N, 10.51. Calc. for
C18H30F18N7P3Ruؒ0.3C3H6O: C, 25.28; H, 3.57; N, 10.92%).
trans-[RuII(NH3)4(py)(Mebpe؉)][PF6]3 10. This was prepared
in similar manner to that of salt 8 by using 7 (135 mg,
0.253 mmol) in place of trans-[RuIII(SO4)(NH3)4(py)]Cl and
pyridine (0.1 cm3, 1.24 mmol) in place of [Medapϩ]Cl. The
solution was stirred for 3 h in the dark and addition of aqueous
NH4PF6 gave a dark precipitate which was filtered off, washed
with water and dried. Purification was effected by precipitation
from acetone–diethyl ether to afford a dark blue solid: yield
81 mg (36%). δH(CD3COCD3) 8.98–8.92 (4 H, m, C5H4N ϩ
H2,6), 8.88 (2 H, d, J 6.5, C5H4N), 8.36 (2 H, d, J 6.9, C5H4N),
8.09 (1 H, d, J 16.5, CH), 7.95 (1 H, t, H4), 7.93 (1 H, d, J 16.4,
CH), 7.74 (2 H, d, J 6.8 Hz, C5H4N), 7.56 (2 H, t, H3,5), 4.53
(3 H, s, Me) and 2.78 (12 H, s, 4NH3) (Found: C, 24.72; H, 3.50;
N, 10.85. Calc. for C18H30F18N7P3Ru: C, 24.56; H, 3.43; N,
11.14%).
trans-[RuIII(SO4)(NH3)4(mim)]Cl 5. A mixture of trans-
[RuIICl(NH3)4(SO2)]Cl (100 mg, 0.329 mmol) and 1-methyl-
imidazole (mim, 0.2 cm3, 2.51 mmol) was dissolved in water
(5 cm3) and heated at ca. 45 ЊC under Ar for 30 min. Acetone
(100 cm3) was added to the brown solution and a white precipi-
tate filtered off, washed with acetone and dried to afford
crude trans-[RuII(NH3)4(mim)(SO2)]Cl2 (124 mg, 98%). This
material was dissolved in water (5 cm3) and oxidized by the
addition of a 1:1 mixture of 30% aqueous H2O2–2 M HCl
(2 cm3). After 5 min at room temperature, acetone (200 cm3)
was added and the golden precipitate filtered off, washed with
acetone and dried: crude yield 111 mg (88%).
trans-[RuII(NH3)4(py)(Phbpe؉)][PF6]3 11. This was prepared
in identical manner to that of salt 8 by using trans-[RuIII-
(SO4)(NH3)4(py)]Cl (109 mg, 0.287 mmol) and [Phbpeϩ]Clؒ
2.25H2O (424 mg, 1.26 mmol) instead of [Medapϩ]Cl. Purifi-
cation was effected by precipitation from acetone–diethyl ether
to afford a dark blue solid: yield 163 mg (58%). δH(CD3COCD3)
9.30 (2 H, d, J 6.9, C5H4N), 8.98 (2 H, d, J 6.6, C5H4N), 8.90
(2 H, d, H2,6), 8.55 (2 H, d, J 7.0, C5H4N), 8.25 (1 H, d, J 16.2,
CH), 8.07 (1 H, d, J 16.2 Hz, CH), 8.00–7.93 (3 H, m, Ph ϩ H4),
7.82–7.78 (5 H, m, Ph ϩ C5H4N), 6.90 (2 H, t, H3,5) and 2.81
(12 H, s, 4NH3) (Found: C, 30.18; H, 3.39; N, 9.64. Calc. for
C23H32F18N7P3Ruؒ0.5C3H6O: C, 30.29; H, 3.63; N, 10.09%).
trans-[RuIII(SO4)(NH3)4(Medap؉)]Cl2 6. This salt was pre-
pared in identical manner to that of 5 by using [Medapϩ]Cl (167
mg, 0.656 mmol) in place of mim. The addition of acetone
(30 cm3) to the brown solution afforded a mauve precipitate
which was filtered off, washed with acetone and dried to yield
crude trans-[RuII(NH3)4(Medapϩ)(SO2)]Cl3 (151 mg, 82%). The
oxidation was carried out by using water (10 cm3) to dissolve
the SO2 salt and acetone (100 cm3) to precipitate the golden
product: crude yield: 108 mg (59%).
trans-[RuII(NH3)4(mim)(Medap؉)][PF6]3 12. This was pre-
pared in similar manner to that of salt 8 by using 6 (108
mg, 0.194 mmol) in place of trans-[RuIII(SO4)(NH3)4(py)]Cl and
mim (0.2 cm3, 2.51 mmol) in place of [Medapϩ]Cl. The solution
J. Chem. Soc., Dalton Trans., 1999, 3617–3625
3619

View Article Online
was stirred for 3 h in the dark and addition of aqueous NH4PF6
afforded a dark precipitate which was filtered off, washed with
water and dried. Purification was effected by sequential precipi-
acetonitrile)19 as an external reference. All measurements were
performed by using the 1064 nm fundamental wavelength of
an injection-seeded, Q-switched Nd-YAG laser (Quanta-Ray
GCR-5, 8 ns pulses, 7 mJ, 10 Hz). The use of dilute acetonitrile
solutions (10Ϫ5–10Ϫ6 mol dmϪ3) ensured a linear dependence
tations from acetone–NBun Cl, water–aqueous NH4PF6 and
4
finally from acetone–diethyl ether: yield 69 mg (39%). δH(CD3-
COCD3) 10.10 (2 H, s, C14H8N2), 9.88 (2 H, s, C14H8N2), 8.59
(2 H, d, J 8.8, C14H8N2), 8.53 (2 H, d, J 9.2 Hz, C14H8N2), 8.31
(1 H, s, C3N2H3), 7.51 (1 H, s, C3N2H3), 7.47 (1 H, s, C3N2H3),
4.93 (3 H, s, C14H8N2-Me), 3.95 (3 H, s, C3N2H3-Me) and 2.85
(s, 12 H, 4NH3) (Found: C, 26.02; H, 3.20; N, 11.98. Calc. for
C19H29F18N8P3Ruؒ0.3C3H6O: C, 25.90; H, 3.36; N, 12.14%).
2
of I2ω/Iω upon solute concentration, precluding the need for
Lambert–Beer correction factors. Samples were passed through
a 0.45 µm filter (Millipore), and checked for fluorescence.20,21
One-dimensional hyperpolarizability is assumed, i.e. β1064
β333, and a relative error of 15% is estimated.
=
Crystal structure determinations
trans-[RuII(NH3)4(mim)(PymQ؉)][PF6]3 13. This salt was
prepared and purified in identical manner to that of 12 by using
5 (109 mg, 0.285 mmol) in place of 6 and [PymQϩ]Cl (385 mg,
1.42 mmol) in place of mim. A dark blue solid was obtained:
yield 76 mg (28%). δH(CD3COCD3) 10.22 (2 H, d, J 7.5,
C5H4N), 9.27–9.24 (4 H, m, C5H4N ϩ C4H3N2), 8.95 (2 H, d,
J 7.4, C5H4N), 8.28 (1 H, s, C3N2H3), 8.08 (2 H, d, J 6.6,
C5H4N), 8.03 (1 H, t, J 4.1 Hz, C4H3N2), 7.50 (1 H, s, C3N2H3),
7.44 (1 H, s, C3N2H3), 3.95 (3 H, s, Me) and 2.78 (12 H, s, 4NH3)
(Found: C, 24.03; H, 3.46; N, 14.68. Calc. for C18H29F18N10-
P3Ruؒ0.3C3H6O: C, 24.18; H, 3.31; N, 14.92%).
Crystals of salts 8ؒ4MeCN and 16ؒ2MeCN were obtained by
slow diffusion of diethyl ether vapour into acetonitrile
solutions. A red-brown crystal of 8ؒ4MeCN of approximate
dimensions 0.1 × 0.1 × 0.15 mm and a red-orange crystal of
16ؒ2MeCN with dimensions 0.2 × 0.1 × 0.05 mm were chosen
for diffraction studies.
Data collection details are as follows: for 8ؒ4MeCN,
data were collected on a Nonius Kappa CCD area-detector dif-
fractometer controlled by the COLLECT software package.22
Images were processed by DENZO23 and the data corrected
for absorption by using the empirical method employed in
SORTAV24 from within the MAXUS suite of programs.25 For
16ؒ2MeCN, data were collected on a Siemens SMART CCD
area-detector diffractometer. An empirical absorption correc-
tion was applied and the data frames were integrated using
SAINT.26 The structure of 8ؒ4MeCN was solved by direct
methods and refined by full-matrix least squares on all Fo2 data
using SHELXS 9727 and SHELXL 97.28 The same approach
was used for 16ؒ2MeCN, but using Siemens SHELXTL 5.03.26
In both cases, all non-hydrogen atoms, including those of the
trans-[RuII(NH3)4(mim)(Mebpe؉)][PF6]3 14. This salt was
prepared and purified in identical manner to that of 12 by using
7 (104 mg, 0.195 mmol) in place of 6. A dark purple solid was
obtained: yield 64 mg (37%). δH(CD3COCD3) 8.94 (4 H, d, J
5.1, C5H4N), 8.35 (2 H, d, J 6.5, C5H4N), 8.18 (1 H, s, C3N2H3),
8.06 (1 H, d, J 16.4, CH), 7.91 (1 H, d, J 16.5, CH), 7.62 (2 H, d,
J 6.6 Hz, C5H4N), 7.45 (1 H, s, C3N2H3), 7.37 (1 H, s, C3N2H3),
4.51 (3 H, s, C5H4N-Me), 3.91 (3 H, s, C3N2H3-Me) and 2.65 (12
H, s, 4NH3) (Found: C, 23.44; H, 3.56; N, 12.41. Calc. for
C17H31F18N8P3Ru: C, 23.11; H, 3.54; N, 12.68%).
Ϫ
PF6 anions and the acetonitrile molecules of crystallization,
were refined anisotropically. Crystallographic data and refine-
ment details are presented in Table 3.
trans-[RuII(NH3)4(mim)(Phbpe؉)][PF6]3 15. This salt was
prepared in identical manner to that of 12 by using 5 (115 mg,
0.300 mmol) in place of 6 and [Phbpeϩ]Clؒ2.25H2O (445 mg,
1.33 mmol) in place of mim. Purification was effected by pre-
cipitation from acetone–diethyl ether to give a dark blue solid:
yield 97 mg (34%). δH(CD3COCD3) 9.26 (2 H, d, J 6.9, C5H4N),
8.96 (2 H, d, J 5.6, C5H4N), 8.52 (2 H, d, J 6.9, C5H4N), 8.23
(1 H, d, J 16.2, CH), 8.18 (1 H, s, C3N2H3), 8.04 (1 H, d, J 16.2,
CH), 8.00–7.96 (2 H, m, Ph), 7.81–7.78 (3 H, m, Ph), 7.68 (2 H,
d, J 5.9 Hz, C5H4N), 7.45 (1 H, s, C3N2H3), 7.39 (1 H, s,
C3N2H3), 3.94 (3 H, s, Me) and 2.67 (12 H, s, 4NH3) (Found: C,
28.51; H, 3.94; N, 11.08. Calc. for C22H33F18N8P3Ruؒ0.3C3H6O:
C, 28.56; H, 3.64; N, 11.64%).
CCDC reference number 186/1638.
See http://www.rsc.org/suppdata/dt/1999/3617/ for crystallo-
graphic files in .cif format.
Results and discussion
Molecular design and synthesis
This report constitutes an extension of our previous studies
in which we have found that complexes trans-[RuII(NH3)4(LD)-
(LA)]3ϩ in which LA is a N-R-4,4Ј-bipyridinium ligand exhibit
large β0 responses which increase in the order R = Me < Ph <
4-MeCOPh.9c The new complexes in the salts 1–4 and 8–15
were designed to probe the effects of several further molecular
structural changes on the electronic absorption and quadratic
NLO properties. These changes can be summarized as follows.
(i) The complexes of N-methyl-2,7-diazapyrenium (Medapϩ)
in 1, 8 and 12 feature fixed, coplanar LA rings and 4 additional
π electrons when compared with their 4,4Ј-bipyridinium
analogues. (ii) The 2-pyrimidyl ring in the complexes of N-(2-
pyrimidyl)-4,4Ј-bipyridinium (PymQϩ) in 2, 9 and 13 is con-
siderably more electron deficient than the N-aryl substituents
used previously. (iii) The complexes of N-methyl-4-[trans-2-(4-
pyridyl)ethenyl]pyridinium (Mebpeϩ) and N-phenyl-4-[trans-2-
(4-pyridyl)ethenyl]pyridinium (Phbpeϩ) (in 3, 4, 10, 11, 14, 15)
are similar to their predecessors containing N-methyl-4,4Ј-
bipyridinium (MeQϩ) or N-phenyl-4,4Ј-bipyridinium (PhQϩ),
but have planar and extended conjugated systems. The salt 16
is included in this report primarily to allow crystal structural
comparisons (see later).
trans-[RuII(NH3)4(py)(4,4Ј-bpy)][PF6]2 16. This salt was
prepared in similar manner to that of 8 by using trans-
[RuIII(SO4)(NH3)4(py)]Cl (200 mg, 0.527 mmol) and 4,4Ј-
bipyridine (2.06 g, 13.2 mmol) in acetone (10 cm3) in place of
[Medapϩ]Cl. The crude product was precipitated by addition of
aqueous NH4PF6/Na2CO3, followed by removal of the acetone
in vacuo. Two reprecipitations from acetone–diethyl ether
afforded a red, microcrystalline solid: yield 215 mg (59%).
δH(CD3COCD3) 9.00 (2 H, d, J 6.8, C5H4N), 8.88 (2 H, d, H2,6),
8.74 (2 H, d, J 6.1 Hz, C5H4N), 7.95–7.84 (5 H, c m, 2C5H4N
and H4), 7.55 (2 H, t, H3,5) and 2.81 (12 H, s, 4NH3) (Found: C,
25.95; H, 3.69; N, 13.94. Calc. for C15H25F12N7P2Ru: C, 25.95;
H, 3.63; N, 14.12%).
Hyper-Rayleigh scattering
Details of the hyper-Rayleigh scattering (HRS) experiment
have been discussed elsewhere,17 and the experimental pro-
cedure used was as described previously.18 β Values were
determined by using the electric-field-induced second har-
monic generation β1064 for p-nitroaniline (29.2 × 10Ϫ30 esu in
The salt [PymQϩ]Cl has been reported previously, but with-
out details.29 This compound is prepared by nucleophilic attack
of 4,4Ј-bipyridine upon 2-chloropyrimidine, and it was found
that initial melt reactions improve the yields of subsequent
reactions in refluxing ethanol. The ion Phbpeϩ was obtained
3620
J. Chem. Soc., Dalton Trans., 1999, 3617–3625

View Article Online
with those for the related salts of the ligands MeQϩ, PhQϩ,
N-(4-acetylphenyl)-4,4Ј-bipyridinium (4-AcPhQϩ) and N-(2,4-
dinitrophenyl)-4,4Ј-bipyridinium (2,4-DNPhQϩ) for purposes
of comparison.9b,c
The MLCT band maxima of the Medapϩ complexes in 1 and
12 are blue-shifted by 0.03 eV with respect to those of their
MeQϩ counterparts.9 The maxima of the PymQϩ complexes in
2 and 13 are red-shifted by ca. 0.3 eV with respect to those of
their Medapϩ analogues and red-shifted by 0.06–0.08 eV with
respect to those of their 4-AcPhQϩ analogues.9c Given an
almost constant HOMO energy (see later), this shows that
PymQϩ has the lowest LUMO energy of the 4,4Ј-bipyridinium
ligands studied, in keeping with the highly electron-deficient
nature of the 2-pyrimidyl group. The acceptor strength of LA
hence increases in the order MeQϩ < PhQϩ < 4-AcPhQϩ < 2,4-
DNPhQϩ < PymQϩ. The MLCT energies of the complexes
of Mebpeϩ (in 3 and 14) and Phbpeϩ (in 4 and 15) are very
similar to those of their analogues featuring MeQϩ or PhQϩ,
respectively.9
Comparison of the MLCT data for the pairs 3/4, 10/11 and
14/15 reveals shifts of ca. Ϫ0.1 eV on replacing an N–Me with
an N–Ph substituent. This shows that Phbpeϩ is a considerably
stronger acceptor than Mebpeϩ, in keeping with the previously
noted difference between PhQϩ and MeQϩ.9c With a given LA,
the MLCT energy decreases as the net donor strength of LD
increases, in the order py < NH3 < mim, reflecting a progressive
destabilization of the Ru-based HOMO.
Electrochemical studies
All of the new complex salts, except for 5–7, were studied
by cyclic voltammetry in acetonitrile and results are presented
in Table 1. All exhibit reversible or quasi-reversible RuIII/II
oxidation waves, together with generally irreversible ligand-
based reduction processes. By contrast, the related complexes
of MeQϩ, PhQϩ or 4-AcPhQϩ each show two reversible or
quasi-reversible ligand reduction waves.9a,c Selected electro-
chemical data for 1–4 and 8–15 and for previously reported
complex salts are included in Table 2.
III/II
The Ru
E
1/2
values for 1 and 12 are shifted by ca. ϩ40 mV
with respect to those of their MeQϩ analogues,9 due to the
lower basicity of Medapϩ caused by increased delocalization.
The accompanying blue shifts in the MLCT bands (see earlier)
can hence be attributed largely to a slight stabilization of the
from the condensation of 4-methyl-N-phenylpyridinium chlor-
ide with pyridine-4-carbaldehyde using piperidine as a base
catalyst. The yield of this reaction diminishes with increasing
reflux times, in similar fashion to the related, published prepar-
ation of Mebpeϩ.30 The new complex salts were synthesized and
purified by using previously described procedures,9 with some
minor modifications. In the syntheses of the salts 8–16 the
sequences of ligand complexations described are those which
give the optimum product yields and purities.
Ru-based HOMOs in the Medapϩ complexes. By contrast, the
III/II
Ru
E
values for 3, 4, 14 and 15 are shifted by ca. Ϫ40 mV
1/2
with respect to those of their MeQϩ or PhQϩ analogues,9 due to
the mildly electron-donating character of the ethylene units.
Comparison of the RuIII/II E1/2 values for the pairs 3/4, 10/11
and 14/15 reveals negligible shifts on replacing an N–Me with
an N–Ph substituent, whilst the ligand Epc values show large
positive shifts. The corresponding red shifts in the MLCT
bands (see earlier) are hence purely a result of stabilization of
the ligand-based LUMOs. With a given LA, the RuIII/II E1/2
values become less positive as the donor strength of LD
increases, in the order py < mim ≈ NH3, reflecting the HOMO
destabilization indicated by the MLCT data (see earlier).
Comparison of the ligand reduction potentials for 2 and 13
with those of the previously reported complexes confirms
that the PymQϩ ligand is a considerably stronger electron
acceptor than MeQϩ, PhQϩ or 4-AcPhQϩ, in keeping with the
MLCT data (see earlier). The accompanying variations in the
RuIII/II E1/2 values are only small, showing that the HOMO
energy is relatively insensitive to changes in the substituent at N.
Complex 9 also exhibits a relatively anodic reduction wave
attributed to the PymQϩ ligand.
Electronic spectroscopy studies
Electronic absorption spectra for all of the new complex salts,
except for the sulfatoruthenium() intermediates 5–7, were
recorded in acetonitrile and results are presented in Table 1.
Complexes 1–4 and 8–15 show intense, broad dπ(RuII)→π*(LA)
(LA = pyridinium-substituted ligand) MLCT bands in the
region 560–700 nm, the energies of which depend on the rel-
ative energies of the Ru-based HOMO and of the LA-based
LUMO.31 The MLCT band of 16 is found at considerably
higher energy due to the weaker electron-accepting nature of
4,4Ј-bpy compared with the pyridinium ligands. The pyridine
complexes in 8–11 and 16 also show less intense, high energy,
dπ(RuII)→π*(py) MLCT bands at ca. 385 nm, although this
absorption is obscured by a more intense UV band for 8 and
overlaps with the dπ(RuII)→π*(4,4Ј-bpy) band for 16. All of
the complexes also show one or more intense UV absorptions
due to intraligand π → π* excitations. Data for the visible
MLCT bands of 1–4 and 8–15 are collected in Table 2, together
Crystallographic studies
Single crystal structures were obtained for salts 8ؒ4MeCN
and 16ؒ2MeCN, and representations of the molecular struc-
J. Chem. Soc., Dalton Trans., 1999, 3617–3625
3621

View Article Online
Table 1 Electrochemical and UV/VIS data for complex salts in acetonitrile
E1/2/V vs. SCE (∆Ep/mV)a
λmax/nm
b
Complex salt
RuIII/II
Ligand waves
Ϫ0.98c
(ε/dm3 molϪ1 cmϪ1
)
Assignment
1 [RuII(NH3)5(Medapϩ)][PF6]3
0.50 (70)
581 (14 800)
397 (10 900)
376 (7800)
dπ → π*(Medapϩ)
π → π*
π → π*
335 (24 900)
241 (37 200)
673 (18 000)
285 (21 500)
595 (16 100)
312 (23 800)
628 (17 200)
329 (25 800)
564 (12 300)
397 (11 300)
378 (10 900)
334 (19 100)
242 (33 100)
644 (16 800)
385 (5800)
283 (24 500)
563 (14 800)
386 (5900)
311 (23 800)
591 (17 100)
387 (sh) (6900)
328 (30 000)
593 (16 700)
398 (12 000)
377 (9300)
337 (27 600)
242 (33 100)
698 (18 700)
285 (24 300)
604 (16 200)
310 (23 000)
638 (17 600)
329 (28 400)
464 (17 000)
378 (sh) (2800)
246 (17 600)
π → π*
π → π*
2 [RuII(NH3)5(PymQϩ)][PF6]3
3 [RuII(NH3)5(Mebpeϩ)][PF6]3
4 [RuII(NH3)5(Phbpeϩ)][PF6]3
8 trans-[RuII(NH3)4(py)(Medapϩ)][PF6]3
0.49 (100)
0.41 (70)
0.42 (90)
0.65 (75)
Ϫ0.45 (200)d
Ϫ0.85c
dπ → π*(PymQϩ)
π → π*
dπ → π*(Mebpeϩ)
π → π*
Ϫ0.66c
dπ → π*(Phbpeϩ)
π → π*
Ϫ0.95c
dπ → π*(Medapϩ)
π → π*
π → π*
π → π*
π → π*
9 trans-[RuII(NH3)4(py)(PymQϩ)][PF6]3
10 trans-[RuII(NH3)4(py)(Mebpeϩ)][PF6]3
11 trans-[RuII(NH3)4(py)(Phbpeϩ)][PF6]3
12 trans-[RuII(NH3)4(mim)(Medapϩ)][PF6]3
0.67 (80)
0.60 (70)
0.61 (90)
0.51 (70)
Ϫ0.41 (90)d
dπ → π*(PymQϩ)
dπ → π*(py)
π → π*
Ϫ1.10 (100)d
Ϫ0.75 (150)d
dπ → π*(Mebpeϩ)
dπ → π*(py)
π → π*
Ϫ0.64c
dπ → π*(Phbpeϩ)
dπ → π*(py)
π → π*
Ϫ0.97c
dπ → π*(Medapϩ)
π → π*
π → π*
π → π*
π → π*
13 trans-[RuII(NH3)4(mim)(pymQϩ)][PF6]3
14 trans-[RuII(NH3)4(mim)(Mebpeϩ)][PF6]3
15 trans-[RuII(NH3)4(mim)(Phbpeϩ)][PF6]3
16 trans-[RuII(NH3)4(py)(4,4Ј-bpy)][PF6]2
0.52 (90)
0.42 (80)
0.43 (70)
0.60 (80)
Ϫ0.41 (120)d
Ϫ1.17 (100)d
Ϫ0.86a
dπ → π*(PymQϩ)
π → π*
dπ → π*(Mebpeϩ)
π → π*
Ϫ0.66c
dπ → π*(Phbpeϩ)
π → π*
dπ → π*(4,4Ј-bpy)
dπ → π*(py)
π → π*
a Measured in solutions ca. 10Ϫ3 mol dmϪ3 in analyte and 0.1 mol dmϪ3 in NBun PF6 at a platinum-bead working electrode with a scan rate of 200 mV
4
s
Ϫ1. Ferrocene internal reference E1/2 = 0.41 V, ∆Ep = 90 mV. b Solutions (5–7) × 10Ϫ5 mol dmϪ3
ible process as evidenced by ipc ≠ ipa.
.
c Epc for an irreversible reduction process. d Irrevers-
Table 2 Electrochemical, visible MLCT absorption and HRS data for the salts trans-[RuII(NH3)4(LD)(LA)][PF6]3 in acetonitrile
E1/2/V vs. SCE
λmax[MLCT]/nm
LD
LA
RuIII/II
LA ϩ/0
(ε/dm3 molϪ1 cmϪ1
)
Emax[MLCT]/eV
β1064
β0
a
a
Salt
1
2
3
NH3
NH3
NH3
NH3
NH3
NH3
NH3
NH3
py
py
py
py
mim
mim
mim
mim
mim
mim
mim
Medapϩ
PymQϩ
Mebpeϩ
Phbpeϩ
MeQϩ
0.50
0.49
0.41
0.42
0.46
0.46
0.47
581 (14800)
673 (18000)
595 (16100)
628 (17200)
590 (15800)
628 (19300)
654 (18000)
660 (16900)
564 (12300)
644 (16800)
563 (14800)
591 (17100)
593 (16700)
698 (18700)
604 (16200)
638 (17600)
602 (16200)
642 (21500)d
666 (19500)
2.13
1.84
2.08
1.97
2.10
1.97
1.90
1.88
2.20
1.93
2.20
2.10
2.09
1.78
2.05
1.93
2.06
1.93
1.86
660
640
828
751
750
858
1112
871
579
774
904
936
756
818
857
1068
523
874
962
89
230
142
192
123
220
354
289
51
Ϫ0.45
4
b
Ϫ0.91
Ϫ0.75
Ϫ0.64
b
b
b
PhQϩ
4-AcPhQϩ
2,4-DNPhQϩ
Medapϩ
PymQϩ
Mebpeϩ
Phbpeϩ
Medapϩ
PymQϩ
Mebpeϩ
Phbpeϩ
MeQϩ
8
9
0.65
0.67
0.60
0.61
0.51
0.52
0.42
0.43
0.47
0.46
0.48
Ϫ0.41
Ϫ0.75
228
78
10
11
12
13
14
151
126
336
168
310
100
254
332
Ϫ0.41
15
c
Ϫ0.88
Ϫ0.73
Ϫ0.63
b
b
PhQϩ
4-AcPhQϩ
a The value of β1064 (×1030 esu) is the uncorrected first hyperpolarizability measured using a 1064 nm Nd-YAG laser fundamental (15% error); β0 is
the static hyperpolarizability estimated by using the two-level model.10 The quoted cgs units (esu) can be converted into SI units (C3 m3 JϪ2) by
dividing by a factor of 2.693 × 1020. b Ref. 9(c). c Ref. 9(a). d λmax slightly altered from value quoted in ref. 9(c).
3622
J. Chem. Soc., Dalton Trans., 1999, 3617–3625

View Article Online
Table 3 Crystallographic data and refinement details for complexes
8ؒ4MeCN and 16ؒ2MeCN
8ؒ4MeCN
16ؒ2MeCN
Formula
M
C28H40F18N11P3Ru C19H31F12N9P2Ru
1066.69
776.54
Crystal system
Space group
Orthorhombic
Pna21
Monoclinic
C2/c
a/Å
b/Å
c/Å
12.8413(3)
23.9689(3)
14.0305(5)
90
4318.47(19)
4
16.296(4)
22.383(3)
8.683(1)
108.33(2)
3006.4(10)
4
Fig. 1 Structural representation of the complex cation in the salt
8ؒ4MeCN (50% probability ellipsoids).
β/Њ
U/Å3
Z
T/K
150(2)
0.589
173(2)
0.731
µ/mmϪ1
Reflections collected
Independent reflections (Rint
Final R1, wR2 [I > 2σ(I)]a
(all data)
45910
15164
)
8434 (0.1226)
0.0482, 0.0867
0.0972, 0.0979
3458 (0.0215)
0.0324, 0.0914
0.0374, 0.0940
a Structures were refined on Fo2 using all data; the value of R1 is given
for comparison with older refinements based on Fo with a typical
threshold of Fo > 4σ(Fo).
Fig. 2 Structural representation of the complex cation in the salt
16ؒ2MeCN (50% probability ellipsoids).
Table 4 Selected interatomic distances (Å) and angles (Њ) for salts
8ؒ4MeCN and 16ؒ2MeCN
phenothiazine) and trans-[RuII(NH3)4(PhQϩ)(PTZ)][PF6]3ؒ
Et2O,9c of 9.6 and 2.6Њ, respectively. The marked contrast with
16ؒ2MeCN can be ascribed to increased delocalization in the
latter two complexes due to the greater electron-withdrawing
capabilities of MeQϩ and PhQϩ with respect to 4,4Ј-bpy. The
bond distances Ru(1)–N(11) and Ru(1)–N(21) in 16ؒ2MeCN
are equivalent and ca. 0.07 Å shorter than the average of the
Ru–NH3 distances.
8ؒ4MeCN
Ru1–N2
Ru1–N1
Ru1–N4
2.046(3)
2.083(3)
2.127(4)
Ru1–N5
Ru1–N7
Ru1–N6
2.138(5)
2.147(5)
2.154(4)
N2–Ru1–N1
N2–Ru1–N4
N1–Ru1–N4
N2–Ru1–N5
N1–Ru1–N5
N4–Ru1–N5
N2–Ru1–N7
N1–Ru1–N7
177.2(2)
91.18(16)
87.37(16)
87.9(2)
89.7(2)
88.8(2)
N4–Ru1–N7
N5–Ru1–N7
N2–Ru1–N6
N1–Ru1–N6
N4–Ru1–N6
N5–Ru1–N6
N7–Ru1–N6
90.72(19)
179.4(2)
90.85(15)
90.68(16)
177.21(19)
93.18(19)
87.30(19)
The structure of the cation in salt 8ؒ4MeCN resembles
that in 16ؒ2MeCN in that the trans ligands are almost coplanar
[C1–N1–N2–C6 9.8Њ], with the co-ordinated rings
approximately bisecting the H3N–Ru–NH3 angles. However, in
8ؒ4MeCN the Ru1–N2 distance is ca. 0.05 Å shorter than
Ru1–N1, indicating more extensive π-back bonding to
Medapϩ than to the py ligand; Ru1–N2 is also ca. 0.03 Å
shorter than the Ru1–N21 distance in 16ؒ2MeCN. These
observations are in keeping with Medapϩ being a considerably
stronger π acceptor than either py or 4,4Ј-bpy. The ammine
91.7(2)
90.6(2)
16ؒ2MeCN
Ru(1)–N(11)
Ru(1)–N(21)
Ru(1)–N(21)
2.076(3)
2.077(3)
2.140(2)
Ru(1)–N(2)
Ru(1)–N(11)
Ru(1)–N(1)
2.140(2)
2.149(2)
2.149(2)
N(11)–Ru(1)–N(21) 180.0
N(21)–Ru(1)–N(11) 87.74(10)
Ϫ
ligands in 8ؒ4MeCN are hydrogen bonded to one PF6 anion
N(11)–Ru(1)–N(21)
N(21)–Ru(1)–N(21)
N(11)–Ru(1)–N(2)
N(21)–Ru(1)–N(2)
N(21)–Ru(1)–N(2)
N(11)–Ru(1)–N(11)
N(21)–Ru(1)–N(11)
90.54(5)
N(2)–Ru(1)–N(11)
N(11)–Ru(1)–N(1)
N(21)–Ru(1)–N(1)
N(21)–Ru(1)–N(1)
N(2)–Ru(1)–N(1)
92.27(10)
89.20(5)
90.80(5)
92.27(10)
87.74(10)
and four acetonitrile molecules of crystallization.
89.46(5)
90.53(5)
89.47(5)
178.93(10)
89.20(5)
90.80(5)
Salt 16ؒ2MeCN adopts the centrosymmetric space group
C2/c, whilst that of 8ؒ4MeCN (Pna21) is non-centrosymmetric.
However, crystal packing diagrams of 8ؒ4MeCN reveal that the
complex dipoles (as represented by the Ru(1)–N(3) vectors) are
orientated almost completely antiparallel. This precludes the
likelihood of this particular crystalline form of 8 exhibiting
macroscopic quadratic NLO effects, in spite of the very large β0
value determined in solution (see later).
N(11)–Ru(1)–N(1) 178.40(11)
Symmetry transformation used to generate equivalent atoms: 1 1 Ϫ x,
3
–
y, Ϫz ϩ 2.
tures of the cations are shown in Figs. 1 and 2, bond lengths
and angles in Table 4.
Non-linear optical studies
In salt 16ؒ2MeCN the torsion angle C(22)–N(21)–N(11)–
C(12) between the py ligand and the co-ordinated pyridyl ring
of the 4,4Ј-bpy is 12.2Њ, and the planes of the co-ordinated
rings approximately bisect the H3N–Ru–NH3 angles. Similar
arrangements have also been found in trans-[RuIII(NH3)4(Him)-
(isn)][CF3CO2]3ؒPriOH (Him = imidazole, isn = isonicotin-
amide)32 and trans-[RuIII(NH3)4(Him)2]Cl3ؒH2O,33 and it has
been suggested that the quasi-coplanarity of the trans rings
may allow π coupling through the Ru.32 As expected, steric
repulsions between the 2,2Ј-hydrogens cause the 4,4Ј-bpy
ligand in 16ؒ2MeCN to twist, with a C(32A)–C(31)–C(24)–
C(23) torsion of 38.5Њ. This compares with torsion angles
between the pyridyl rings of the 4,4Ј-bipyridinium ligands
in trans-[RuII(NH3)4(MeQϩ)(PTZ)][PF6]3ؒMe2CO9a (PTZ =
The first hyperpolarizabilities β of the new complex salts 1–4
and 8–15 were measured in acetonitrile by using the HRS
technique17,18 with a 1064 nm Nd-YAG laser fundamental.
Estimated static hyperpolarizabilities β0 were obtained by
application of the two-level model,10 and results are presented
in Table 2, together with data for a selection of previously
reported complex salts.9b,c It should be noted that the utility of
comparisons between these data is limited by the relatively large
experimental errors in the HRS β values ( 15%).
All of the new complexes show very large β and β0 values
which are comparable to those we have determined for related
complexes,9 and several of these values are amongst the largest
found for metal-containing chromophores. Comparison of the
data for the Medapϩ complexes in 1 and 12 with their MeQϩ
J. Chem. Soc., Dalton Trans., 1999, 3617–3625
3623

View Article Online
increase in the first hyperpolarizability. However, it should be
noted that this correlation is only partial, and this may be
explained by the fact that β0 is also influenced by a number of
other factors such as the oscillator strength of the charge-
transfer transition.
Conclusion
The MLCT absorption properties of dipolar ruthenium()
ammine complexes of pyridinium-substituted ligands show a
fine degree of tunability. The associated large molecular first
hyperpolarizabilities β0 of these complexes are also tunable,
although perhaps to a lesser extent. In general, β0 increases as
the MLCT absorption energy decreases, in keeping with the
two-level model. Notably, extension of the conjugated bridge in
these complexes neither significantly alters the MLCT energies
nor leads to increases in β0.
Fig. 3 Plot of the first hyperpolarizability against the inverse square of
the MLCT energy for the salts 1–4, 8–15 and other salts included in
Table 29b,c (᭜ = β1064; ᭿ = β0).
Acknowledgements
analogues shows that for LD = NH3 the Medapϩ complex has a
slightly smaller β0, whilst when LD = mim the reverse may be
true. However, the differences are small and so it is reasonable
to conclude that fixing the planarity of the 4,4Ј-bipyridinium
rings does not greatly affect β0. This is consistent with crystallo-
graphic studies which show that the MeQϩ ligand adopts a
quasi-planar configuration, and a similar situation is likely to
pertain in solution due to strong donor–acceptor electronic
coupling.9a
We are grateful to EPSRC for provision of a studentship (to
J. A. H.). This work was supported by research grants from The
Royal Society, from the Belgian National Fund for Scientific
Research (G.0308.96), from the Belgian government (IUAP-P4/
11) and from the University of Leuven (GOA/1/95). Thanks are
due to Johnson Matthey plc for a generous loan of ruthenium
trichloride.
Although the visible absorption and electrochemical data
clearly show that PymQϩ is the strongest electron acceptor
of the 4,4Ј-bipyridinium ligands studied (see earlier), this does
not result in the anticipated increases in β0 for the complexes
where LD = NH3 (in 2) or mim (in 13). The β0 values of 2
and 13 are, respectively, smaller than and equal to those of
their 4-AcPhQϩ analogues. At present, we cannot explain this
apparent anomaly, but tentatively suggest that the β0 values of
the 4-AcPhQϩ complexes are boosted by the presence of the
acetyl groups. Clearly, further experiments are required in order
to clarify this matter.
References
1 Nonlinear Optical Properties of Organic Molecules and Crystals,
eds. D. S. Chemla and J. Zyss, Academic Press, Orlando, 1987,
vols. 1 and 2; J. Zyss, Molecular Nonlinear Optics: Materials, Physics
and Devices, Academic Press, Boston, 1994; S. R. Marder, B.
Kippelen, A. K.-Y. Jen and N. Peyghambarian, Nature (London),
1997, 388, 845; T. Verbiest, S. Houbrechts, M. Kauranen, K. Clays
and A. Persoons, J. Mater. Chem., 1997, 7, 2175.
2 S. R. Marder, in Inorganic Materials, eds. D. W. Bruce and D.
O’Hare, Wiley, Chichester, 1992; D. R. Kanis, M. A. Ratner and
T. J. Marks, Chem. Rev., 1994, 94, 195; N. J. Long, Angew. Chem.,
Int. Ed. Engl., 1995, 34, 21; I. R. Whittall, A. M. McDonagh,
M. G. Humphrey and M. Samoc, Adv. Organomet. Chem., 1998, 42,
291.
3 W. M. Laidlaw, R. G. Denning, T. Verbiest, E. Chauchard and
A. Persoons, Nature (London), 1993, 363, 58; Proc. SPIE, Int. Soc.
Opt. Eng., 1994, 2143, 14; F. W. Vance and J. T. Hupp, J. Am. Chem.
Soc., 1999, 121, 4047.
4 J. Zyss, C. Dhenaut, T. Chauvan and I. Ledoux, Chem. Phys. Lett.,
1993, 206, 409; C. Dhenaut, I. Ledoux, I. D. W. Samuel, J. Zyss,
M. Bourgault and H. Le Bozec, Nature (London), 1995, 374, 339.
5 I. R. Whittall, M. G. Humphrey, A. Persoons and S. Houbrechts,
Organometallics, 1996, 15, 1935; R. H. Naulty, A. M. McDonagh,
I. R. Whittall, M. P. Cifuentes, M. G. Humphrey, S. Houbrechts,
J. Maes, A. Persoons, G. A. Heath and D. C. R. Hockless,
J. Organomet. Chem., 1998, 563, 137.
6 S. Houbrechts, K. Clays, A. Persoons, V. Cadierno, M. P. Gamasa
and J. Gimeno, Organometallics, 1996, 15, 5266; V. Cadierno, S.
Conejero, M. P. Gamasa, J. Gimeno, I. Asselberghs, S. Houbrechts,
K. Clays, A. Persoons, J. Borge and S. García-Granda, Organo-
metallics, 1999, 18, 582.
7 M. Tamm, T. Jentzsch and W. Werncke, Organometallics, 1997,
16, 1418.
8 I.-Y. Wu, J. T. Lin, J. Luo, S.-S. Sun, C.-S. Li, K. J. Lin, C. Tsai,
C.-C. Hsu and J.-L. Lin, Organometallics, 1997, 16, 2038; I.-Y. Wu,
J. T. Lin, J. Luo, C.-S. Li, C. Tsai, Y.-S. Wen, C.-C. Hsu, F.-F. Yeh
and S. Liou, Organometallics, 1998, 17, 2188.
9 (a) B. J. Coe, M. C. Chamberlain, J. P. Essex-Lopresti, S. Gaines,
J. C. Jeffery, S. Houbrechts and A. Persoons, Inorg. Chem., 1997, 36,
3284; (b) B. J. Coe, J. P. Essex-Lopresti, J. A. Harris, S. Houbrechts
and A. Persoons, Chem. Commun., 1997, 1645; (c) B. J. Coe, J. A.
Harris, L. J. Harrington, J. C. Jeffery, L. H. Rees, S. Houbrechts and
A. Persoons, Inorg. Chem., 1998, 37, 3391.
The complexes of Mebpeϩ and Phbpeϩ, with the exception of
14, have β0 values indistinguishable from those of their 4,4Ј-
bipyridinium analogues, providing little evidence for enhance-
ment of the NLO response upon addition of a trans-ethylene
unit. This contrasts with the situation in related organics where
extension of conjugated bridges typically leads to increased first
hyperpolarizabilites.1 For example, the β0 values determined via
electric-field-induced second harmonic generation at 1907 nm
for 4-(dimethylamino)-4Ј-nitrobiphenyl and trans-4-(dimethyl-
amino)-4Ј-nitrostilbene are 40 and 55 × 10Ϫ30 esu, respec-
tively.34 However, in the latter case the addition of an ethylene
unit also causes a 0.3 eV red shift in the intramolecular charge-
transfer absorption band, whilst the same structural change
does not significantly affect the MLCT maxima of the present
complexes (see earlier).
The marked red-shifting of the MLCT bands due to the
greater acceptor strength of Phbpeϩ with respect to Mebpeϩ
(see earlier) also gives rise to increases in β0 between the pairs
3/4, 10/11 and 14/15. This is in keeping with the previously
noted difference between complexes of PhQϩ and MeQϩ.9c
For a given LA, the MLCT energy increases in the order LD =
mim < NH3 < py (see earlier), and the β0 values broadly
decrease in the opposite order.
According to the two-level model,10 the dominant com-
2
ponent of β is inversely proportional to Ege (the square of
the energy gap between the ground state and first charge-
transfer excited state). We have previously demonstrated a
10 J. L. Oudar and D. S. Chemla, J. Chem. Phys., 1977, 66, 2664; J. Zyss
and J. L. Oudar, Phys. Rev. A, 1982, 26, 2016.
11 B. J. Coe, S. Houbrechts, I. Asselberghs and A. Persoons, Angew.
Chem., Int. Ed., 1999, 38, 366; B. J. Coe, Chem. Eur. J., 1999, 5,
2464.
2 9c
good linear correlation between β0 and 1/Ege
,
and Fig. 3
shows that the data for the new complex salts fit reasonably well
with this trend. Hence, decreasing the MLCT energy of these
ruthenium() chromophores generally corresponds to an
3624
J. Chem. Soc., Dalton Trans., 1999, 3617–3625

View Article Online
12 J. C. Curtis, B. P. Sullivan and T. J. Meyer, Inorg. Chem., 1983, 22,
224.
13 A. J. Blacker, J. Jazwinski and J.-M. Lehn, Helv. Chim. Acta, 1987,
70, 1.
14 E. D. Bergmann, F. E. Crane, Jr. and R. M. Fuoss, J. Am. Chem.
Soc., 1952, 74, 5981.
15 P. J. Stang, D. H. Cao, S. Saito and A. M. Arif, J. Am. Chem. Soc.,
1995, 117, 6273.
25 S. Mackay, C. J. Gilmore, C. Edwards, M. Tremayne, N. Stewart
and K. Shankland, MAXUS, a computer program for the solution
and refinement of crystal structures from diffraction data,
University of Glasgow, UK, Nonius B.V., Delft, The Netherlands
and MacScience Co. Ltd., Yokohama, Japan, 1998.
26 SHELXTL 5.03 program system, Siemens Analytical X-Ray
Instruments, Madison, WI, 1995.
27 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467.
28 G. M. Sheldrick, SHELXL 97, Program for crystal structure
refinement, University of Göttingen, 1997.
16 S. Hünig, J. Groß, E. F. Lier and H. Quast, Liebigs Ann. Chem.,
1973, 339.
17 K. Clays and A. Persoons, Phys. Rev. Lett., 1991, 66, 2980; Rev. Sci.
Instrum., 1992, 63, 3285; E. Hendrickx, K. Clays, A. Persoons,
C. Dehu and J. L. Brédas, J. Am. Chem. Soc., 1995, 117, 3547.
18 S. Houbrechts, K. Clays, A. Persoons, Z. Pikramenou and
J.-M. Lehn, Chem. Phys. Lett., 1996, 258, 485.
19 M. Stähelin, D. M. Burland and J. E. Rice, Chem. Phys. Lett., 1992,
191, 245.
20 E. Hendrickx, C. Dehu, K. Clays, J. L. Brédas and A. Persoons,
ACS Symp. Ser., 1995, 601, 82.
21 I. D. Morrison, R. G. Denning, W. M. Laidlaw and M. A.
Stammers, Rev. Sci. Instrum., 1996, 67, 1445.
29 D. W. Clack, J. C. Evans, C. R. Morris and C. R. Rowlands,
J. Chem. Soc., Perkin Trans. 2, 1988, 1541.
30 M. Horner and S. Hünig, Liebigs Ann. Chem., 1982, 1183.
31 P. Ford, De F. P. Rudd, R. Gaunder and H. Taube, J. Am. Chem.
Soc., 1968, 90, 1187; C. R. Johnson and R. E. Shepherd, Inorg.
Chem., 1983, 22, 2439.
32 J. F. Wishart, X. Zhang, S. S. Isied, J. A. Potenza and H. J. Schugar,
Inorg. Chem., 1992, 31, 3179.
33 K. J. LaChance-Galang, P. E. Doan, M. J. Clarke, U. Rao,
A. Yamano and B. M. Hoffman, J. Am. Chem. Soc., 1995, 117, 3529.
34 L.-T. Cheng, W. Tam, S. R. Marder, A. E. Stiegman, G. Rikken and
C. W. Spangler, J. Phys. Chem., 1991, 95, 10643.
22 Collect, Data collection software, R. Hooft, Nonius B.V., Delft,
The Netherlands, 1998.
23 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276, 307.
24 R. H. Blessing, Acta Crystallogr., Sect. A, 1995, 51, 33; J. Appl.
Crystallogr., 1997, 30, 421.
Paper 9/05652A
J. Chem. Soc., Dalton Trans., 1999, 3617–3625
3625