Interaction of sulfonated ruthenium(ii) polypyridine complexes with surfactants probed by luminescence spectroscopy

Article abstract of DOI:10.1002/1522-2675(20010919)84:9<2708::AID-HLCA2708>3.0.CO;2-M

Novel anionic [RuL₂L′]^2- complexes, where L stands for (1,10-phenanthroline-4,7-diyl)bis(benzenesulfonate) (pbbs; 3a) or (2,2′-bipyridine)-4,4′-disulfonate (bpds; 3b), and L′ is N-(1,10-phenanthrolin-5-yl)tetradecanamide (pta; 2a) or N-(1,10-phenanthrolin-5-yl)acetamide (paa; 2b), were synthesized, and their interaction with the prototypical surfactants sodium dodecylsulfate (SDS), cetyl trimethyl ammonium bromide (CTAB), and Triton X-100 (TX-100) was investigated by electronic absorption, luminescence spectroscopy, emission-lifetime determinations, and O₂-quenching measurements. [Ru(bpds)₂(pta)]^2- (5a) displayed cooperative self-aggregation in aqueous medium at concentrations above 1.3 μM; the observed association was enhanced in the presence of either β-cyclodextrin or NaCl. This amphiphilic Ru¹¹ compound showed the strongest interaction with all the detergents tested: nucleation of surfactant molecules around the luminescent probe was observed below their respective critical micellar concentrations. As much as a 12-fold increase of the emission intensity and a 3-fold rise in the lifetime were measured for 5a bound to TX-100 micelles; the other complexes showed smaller variations. The O₂-quenching rate constants decreased up to 1/8 of their original value in H₂O (e.g., for [Ru(bpds)₂(pta)]^2- (6a) bound to CTAB micelles). Luminescence-lifetime experiments in H₂O/D₂O allowed the determination of the metal-complex fraction exposed to solvent after binding to surfactant micelles. For instance, such exposure was as low as 25% for pta complexes · CTAB aggregates. The different behaviors observed were rationalized in terms of the Ru¹¹ complex structure, the electrostatic/hydrophobic interactions, and the probe environment.

Full text of DOI:10.1002/1522-2675(20010919)84:9<2708::AID-HLCA2708>3.0.CO;2-M

2708
Helvetica Chimica Acta ± Vol. 84 (2001)
Interaction of Sulfonated Ruthenium(II) Polypyridine Complexes with
Surfactants Probed by Luminescence Spectroscopy
by David García-Fresnadillo and Guillermo Orellana*
Â
Dept. de Química Organica, Facultad de Ciencias Químicas, Universidad Complutense de Madrid,
E-28040 Madrid
Â
Dedicated to Professor Andre M. Braun on the occasion of his 60th birthday
Novel anionic [RuL2L']2 complexes, where L stands for (1,10-phenanthroline-4,7-diyl)bis(benzenesulfo-
nate) (pbbs; 3a) or (2,2'-bipyridine)-4,4'-disulfonate (bpds; 3b), and L' is N-(1,10-phenanthrolin-5-yl)tetrade-
canamide (pta; 2a) or N-(1,10-phenanthrolin-5-yl)acetamide (paa; 2b), were synthesized, and their interaction
with the prototypical surfactants sodium dodecylsulfate (SDS), cetyl trimethyl ammonium bromide (CTAB),
and Triton X-100 (TX-100) was investigated by electronic absorption, luminescence spectroscopy, emission-
lifetime determinations, and O2-quenching measurements. [Ru(pbbs)2(pta)]2 (5a) displayed cooperative self-
aggregation in aqueous medium at concentrations above 1.3 mm; the observed association was enhanced in the
presence of either b-cyclodextrin or NaCl. This amphiphilic RuII compound showed the strongest interaction
with all the detergents tested: nucleation of surfactant molecules around the luminescent probe was observed
below their respective critical micellar concentrations. As much as a 12-fold increase of the emission intensity
and a 3-fold rise in the lifetime were measured for 5a bound to TX-100 micelles; the other complexes showed
smaller variations. The O2-quenching rate constants decreased up to 1/8 of their original value in H2O (e.g., for
[Ru(bpds)2(pta)]2 (6a) bound to CTAB micelles). Luminescence-lifetime experiments in H2O/D2O allowed
the determination of the metal-complex fraction exposed to solvent after binding to surfactant micelles. For
instance, such exposure was as low as 25% for pta complexes ´ CTAB aggregates. The different behaviors
observed were rationalized in terms of the RuII complex structure, the electrostatic/hydrophobic interactions,
and the probe environment.
Introduction. ± The use of luminophores as ꢀreporterꢁ probes for sensing
(bio)chemical species and microenvironments has been widely documented because
of their high sensitivity and specificity [1]. Luminescent RuII complexes are outstanding
molecules in this regard, due to the tuneable photophysical properties of their metal-to-
ligand charge-transfer (MLCT) excited state. (Polypyridine)ruthenium(II) complexes
display strong absorption bands in the UV/VIS region, orange-red emission at room
temperature with a large Stokesꢁ shift (ca. 200 nm), moderate to good emission
quantum yields (0.01 ± 0.30), and long luminescence lifetimes (0.1 ± 10 ms). In addition,
they are thermally and (photo)chemically stable. To tailor applications, the structure of
bis- and tris(chelate) complexes of RuII can be designed by judiciously selecting the
nature of the polyazaheterocyclic ligands incorporated in the coordination compound,
allowing in this way the possibility of obtaining a whole range of photophysical features
for these complexes [2]. This feature makes homoleptic and heteroleptic RuII
complexes very attractive versatile molecules for the development of optical probes
and sensors since their photophysical parameters are strongly dependent on the
microenvironment around the complex [3]. Not surprisingly, the photophysics of
luminescent RuII complexes in microheterogeneous phases such as those provided by
organized media (micelles, bilayers, colloids), biopolymers (DNA and proteins), and

Helvetica Chimica Acta ± Vol. 84 (2001)
2709
organic/inorganic polymers (poly(vinylsulfate), poly(styrenesulfonate), dendrimers,
silica gel, glass, zeolites, etc.) has been an area of active research [4].
The interaction of a metal complex with these entities usually produces dynamic
binding with the luminophore displaying different spectroscopic properties depending
on its location. The distribution of luminescent probes in these microheterogeneous
environments is markedly influenced by the hydrophobic/hydrophilic character of the
surrounding medium and by the presence of electrostatically charged interfaces.
Moreover, restricted spaces with reduced (1 ± 2) or even fractal dimensionalities
have to be considered to account for dynamic quenching processes involving the
probe molecule [5]. Some of the photophysical and photochemical characteristics
of the probe that can be monitored are: absorption and emission spectral shifts,
luminescence intensities and lifetimes, emission anisotropy, dynamic or static
quenching reactions, sensitization or energy transfer to a second luminophore.
Many of these parameters can be strongly affected by the equilibrium constants of
the aggregate, the distribution of probe molecules, and the in/out rate constants
of the luminophore and other solutes within the timescale of the excited-state
deactivation.
RuII Complexes usually show a strong interaction with anionic, cationic, and neutral
surfactants, due to a combination of hydrophobic and electrostatic interactions [6].
Therefore, their luminescence quantum yields and lifetimes generally increase in the
presence of the structured microenvironment provided by the surfactants. This is due to
protection of the luminophore from quenchers, as well as the increased microviscosity
that slows down collisional deactivation of the excited species by restricting their
diffusive motions (e.g., quenching by oxygen). On the other hand, their emission
maxima remain constant or are slightly shifted depending on the type of complex
[4c][6b,c][7]. For instance, multi-exponential emission decays have been observed for
[Ru(bpy)3]2‡ (bpy ˆ 2,2'-bipyridine) in the presence of surfactants for which the
complex displays strong affinity, such as sodium dodecyl sulfate, even at concentrations
below its critical micelle concentration (cmc) [8].
In spite of the large number of studies on cationic metal complexes carried out by
Demas, DeGraff, and co-workers [6][7] and other authors, there is little information
about the behavior of anionic RuII complexes in the presence of surfactants. Our
research has aimed the synthesis of novel heteroleptic RuII complexes with amphiphilic
structures, as well as the study of their photophysics in the presence of anionic, cationic,
and neutral surfactants, compared to their behavior in homogeneous solution, to
evaluate their suitability as optical probes for detergents. In this regard, desired
characteristics for maximum sensitivity of the new luminophores are i) a strong binding
to surfactants via hydrophobic/electrostatic interactions, leading to aggregation of the
detergent around the luminescent complex; ii) an emission quantum yield and lifetime
enhancement in the presence of surfactant molecules. To fulfill these requirements, we
designed and prepared RuII complexes containing two polyazaheterocyclic ligands that
provide a polar anionic head around the cationic metal center, and a third chelating
ligand bearing a long alkyl chain to confer hydrophobicity to the metal complex. For
reference purposes, similar RuII species lacking the long alkyl chain were prepared too.
In the structures of these novel [RuL2L']2 complexes 5 and 6, L is (1,10-phenanthro-
line-4,7-diyl)bis(benzenesulfonate) (pbbs; 3a) or (2,2'-bipyridine)-4,4'-disulfonate

2710
Helvetica Chimica Acta ± Vol. 84 (2001)
(bpds; 3b), and L' is N-(1,10-phenanthrolin-5-yl)tetradecanamide (pta; 2a) or N-(1,10-
phenanthrolin-5-yl)acetamide (paa; 2b).
The spectroscopy and photophysics of these metal species were studied in aqueous
solution and in the presence of prototype anionic (sodium dodecyl sulfate ˆ SDS),
cationic (cetyltrimethylammonium bromide ˆ CTAB), and neutral (Triton X-100 ˆ
TX100) surfactants. The degree of exposition to H2O of the different heteroleptic
complexes bound to the corresponding micellar aggregates was also determined, to
reveal the nature and strength of the interaction between the RuII complex and the
surfactant, as well as the location of the luminophore bound to the micelle. The
tendency to self-aggregation expected for some of the new probes was tested in the
presence of additives such as b-cyclodextrin and NaCl [9]. Finally, dissolved O2-
quenching of the luminophores in the absence and in the presence of surfactant
micelles was used to confirm the location of the probe in the supramolecular entity.
Results and Discussion. ± Synthesis. The 1,10-phenanthrolin-5-amine (1) was
prepared in good yield from 5-nitro-1,10-phenanthroline via catalytic reduction with
hydrazine hydrate over Pd/C [10]. Attempts to obtain amine 1 via reduction with SnCl2
or Na2S2O4 provided poor reaction yields. The amide ligands 2a and 2b were
synthesized from amine 1 following reported methods for the preparation of amides via
activated carboxylic acids [11], such as the mixed anhydride obtained from
tetradecanoic acid and ethyl carbonochloridate, or acetic anhydride. Reaction yields
were lower than 50%, due to the low nucleophilic properties of the heteroaromatic
amine 1.

Helvetica Chimica Acta ± Vol. 84 (2001)
2711
The N,N'-dioxide of the disodium salt of the (2,2'-bipyridine)-4,4'-disulfonate was
prepared from 2,2'-bipyridine [12]. Reduction of the N,N'-dioxide to 3b was improved
by using milder conditions (ammonium formate over 10% Pd/C) than those described
by Balicki [13].
The cis-dichlorobis(chelate)ruthenium(II) complexes 4a and 4b were synthesized
according to the general method reported by Sullivan et al. [14], i.e., by refluxing RuCl3
with a stoichiometric amount of ligand and a high excess of LiCl under inert
atmosphere. The high concentration of Cl anions in the reaction mixture largely
prevented incorporation of a third chelating ligand in the RuII coordination sphere. The
reaction time depended on the structure of the chelating ligand: It was only 5 h for pbbs
(3a), but 24 h for the bpds (3b), due to free rotation between the pyridine rings of 3b
that slowed down chelation. The bis-chelate complexes 4 were isolated together with
some amount of LiCl due to the low solubility of this salt in acetone.
There are several methods to prepare tris(chelate)ruthenium(II) complexes with
polyazaheterocyclic ligands [15]. Typical reaction conditions for obtaining homoleptic
complexes such as 7 and 8 involve heating of RuCl3 and an excess of ligand in a protic
solvent (alcohols or mixtures with H2O) under inert atmosphere. Complexation takes
place via stepwise incorporation of ligands accompanied by reduction of the metal
center by the warm solvent. The synthesis of the heteroleptic complexes 5 and 6
requires previous preparation of the cis-dichlorobis(chelate) complexes 4 (see above)
and further reaction with excess of the third ligand under similar conditions to those
used for the homoleptic complexes.
Photophysical and Electrochemical Characterization. UV/VIS Absorption spectra
of the novel heteroleptic complexes 5 and 6 are summarized in Table 1, together with
those of the homoleptic [Ru(pbbs)3]4 (7) and [Ru(bpy)3]2‡ (bpy ˆ 2,2'-bipyridine; 8)
for the sake of comparison. All these species display strong absorption in the VIS
region, corresponding to non-degenerated d ! p* transitions due to symmetry
considerations in both the heteroleptic and homoleptic RuII complexes [16]. The
strong UV bands at 215 ± 300 nm arise from intraligand p ! p* transitions. All the
investigated complexes display absorption shoulders at ca. 315 (5) or 350 nm (6), due to
metal-centered transitions. Complexes 6 show an additional band at ca. 265 nm that can
be attributed to a more energetic MLCT transition by comparison to the [Ru(bpy)3]2‡
(8) chromophore [2b]. The absorption maxima of complexes 6 are slightly red-shifted
compared to those of 5 and also of 8, due to the strong electron-withdrawing effect of
Table 1. Absorption Maxima of the RuII Complexes in H2O
[nm] a) (e [m 1cm 1])b)
max
l
abs
5a
b
6a
b
460c) (36600)d), 427c) (35650)d), 315 (sh, 40200), 274 (175800)d)
459 (22700), 435 (22550), 315 (sh, 21800), 275 (102000), 215 (98600)
466 (17500), 434 (15600), 350 (sh, 12800), 296 (67000), 266 (42300), 218 (72000)
466 (16800), 446 (16100), 350 (sh, 12700), 296 (64600), 264 (44800), 218 (71700)
462 (29300), 438 (29100), 310 (sh, 35800) 278 (121300), 216 (116400)
452 (13700), 344 (sh, 9900), 285 (78100), 243 (24900)
7
8e)
a) Experimental error Æ 1 nm. b) Experimental error Æ 3%. c) Experimental error Æ 3 nm. d) Experimental
error Æ 8%. e) [24].

2712
Helvetica Chimica Acta ± Vol. 84 (2001)
the SO3 substituents that increases the p-acceptor character of the bpds ligand and,
consequently, lower the energy of their p* orbitals. This effect is less important in the
pbbs ligand, the SO3 groups of which are placed at the distal end of a more extended
aromatic structure. The absorption coefficients of 6 are lower than those of 5 and
[Ru(pbbs)3]4 (7) due to the smaller absorption cross-section of the sulfonated ligands
included in the former complexes.
All the investigated RuII complexes display a broad emission band in the VIS region
of the electromagnetic spectrum at room temperature. Luminescence maxima in
different solvents are collected in Table 2. The actual position of the emission band
depends on the solvent. This result was to be expected since the electronically excited
RuII complexes are very sensitive to their environment [2]. As it is observed in
the absorption spectra, the luminescence of bpds complexes appears at longer
wavelengths than that of the corresponding pbbs complexes. Similar reasons to those
outlined above may account for such differences. The only exceptions are the spectra
recorded in DMF solvent, where all the RuII complexes display the same emission
maximum (Table 2).
Table 2. Emission Maxima of the RuII Complexes in H2O and Organic Solvents
[nm]a)
max
l
em
H2O
MeOH
EtOH
DMF
MeCN/H2O 1 : 1
5a
b
6a
b
7
8
624
629
650
652
632
620
622
645
645
616
623
637
640
±b)
636
636
638
638
637
635
624
631
649
648
±b)
618
630c)
620d)
608d)
620d)
a) Corrected for the instrumental response; estimated error Æ 3 nm. b) Not determined, due to the very low
solubility of the RuII complex. c) [34]. d) [35]. e) In MeCN [36].
The different nature of the pbbs and bpds ligands strongly influences the emission
quantum yield and lifetime of the corresponding complexes 5 and 6 in H2O. Table 3
collects the luminescence quantum yields Fem, the emission lifetimes under Ar (t0) and
air-equilibrated (t), the unimolecular deactivation rate constants kr (radiative) and knr
3
(non-radiative), and the fraction FTO of MLCTexcited states quenched by oxygen, for
2
all the studied complexes in H2O. Those with the pbbs ligand have Fem values larger
than 0.15 and t0 values around 3.8 ms. However, the RuII complexes containing bpds
have ca. 7-fold smaller luminescence quantum yields and lifetimes. The presence of
polypyridine ligands with Ph substituents at positions 4 and 7 significantly increases Fem
and t0 of the corresponding RuII complexes, as it has been reported for RuII
coordination compounds with substituted 2,2'-bipyridine and 1,10-phenanthroline
ligands [17]. This result can be explained by the larger knr values of the luminpophores 6
(Table 3). The quenching rate constants by oxygen of the investigated RuII complexes
1
are in the range 1 ± 4 ´ 109 m s 1 (see below), a value close to the diffusion limit of the

Helvetica Chimica Acta ± Vol. 84 (2001)
2713
kinetics in aqueous solution [18]. Nevertheless, kq constants are slightly lower for the
bpds complexes 6. The longer emission lifetime of the pbbs complexes 5 and 7
compared to that of the bpds chelates 6 (Table 3), together with the corresponding
quenching rate constants, may account for the significantly higher oxygen-quenching
probability of the pbbs (FTO ꢀ 0.75) vs. the bpds complexes (FTO ꢀ 0.15).
2
2
Table 3. Photophysical Parameters and Probability of Excited-State Quenching by Oxygen of the RuII Complexes
in Water a)
b
Fem
)
t0 [ms]b)
t [ms]c)
kr [104s 1]b)
knr [105s 1]b)
FOT c),d)
2
5a
b
6a
b
7
8
0.17
0.20
0.031
0.024
0.15
0.042f)
3.70
3.83
0.48
0.40
3.80
0.61g)
1.05e)
0.99
0.39
0.36
0.90
0.39
4.6
5.1
6.4
6.0
3.9
6.9
2.2
2.0
20
24
2.2
16
0.72
0.74
0.19
0.10
0.76
0.36
a) [RuII] ꢁ 10 mm except otherwise stated. b) Ar-Purged solutions; estimated error Æ 5% (Fem), Æ 2% (t0), Æ
d
10% (kr and knr). c) Air-equilibrated solutions; estimated error Æ 2% (t), Æ 10% (kq), Æ 10% (FOT ).
)
2
Fraction of luminophore 3MLCTexcited states quenched by oxygen. e) At 1.0 mm concentration; estimated error
Æ 5%. f) [19a]. g) [24].
The emission lifetimes of the novel luminophores in organic solvents are given in
Table 4. The pbbs complexes 5 display a longer emission lifetime in H2O than in organic
solvents. However, the photoexcited bpds complexes 6 live longer in organic solvents.
This behavior can be rationalized in terms of excited-state quenching due to reaction
with dissolved oxygen and/or due to the non-radiative deactivation pathway by energy
transfer by the solvent OH oscillators in the different media [2b][19]. The solubility of
O2 in air-equilibrated organic solvents is ca. 10 times higher than in air-equilibrated
H2O [20]. Those complexes that are efficiently quenched by oxygen (large FTO ,
2
Table 3), such as 5a and 5b, will show shorter luminescence lifetimes in organic media.
However, since oxygen quenching of complexes 6 is poor (small FOT ), the excited-state
lifetime is controlled by the fast non-radiative deactivation (Table23). In this way, the
less efficient energy transfer to the organic solvent OH oscillators compared to H2O
yields longer emission lifetimes in the former media (Table 4).
a
Table 4. Emission Lifetimes of the Novel RuII Complexes in Air-Equilibrated Organic Solvents
t [ms]b)
)
MeOH
EtOH
MeCN/H2O 1 : 1
5a
b
6a
b
0.30
0.32
0.45
0.44
0.38
0.39
0.60
0.54
0.43
0.47
0.50
0.49
a) [RuII] ꢁ 10 mm. b) Experimental error Æ 2%.

2714
Helvetica Chimica Acta ± Vol. 84 (2001)
With the exception of [Ru(pbbs)2(pta)]2 (5a), the luminescence decay kinetics of
the new RuII complexes in H2O is strictly exponential. The former shows a single-
exponential decay profile at 1 mm concentration, but a bi-exponential fit is required in
more concentrated aqueous solutions. These results point out spontaneous self-
aggregation of the pbbs/pta coordination compound 5a. The structure of this species
may be described as an amphiphilic molecule having a bulky anionic head and a long
(hydrophobic) aliphatic tail. To investigate this phenomena and its possible occurrence
in the other new RuII complexes, especially those involving the pta ligand, their
absorption and emission features were investigated in the 1 ± 100 mm concentration
range (data not shown). These experiments established that all the complexes follow
the Lambert-Beer law in the concentration range studied, except [Ru(pbbs)2(pta)]2
(5a), for which the absorbance vs. concentration plot is linear just up to 50 mm. In every
case, the emission intensity increases linearly with the luminophore concentration in
the 1 ± 10 mm interval. Higher concentrations lead to classical inner-filter effects. Only
for RuII complex 5a, the emission decay kinetics display different profiles depending on
its concentration in H2O: at higher values than 3 mm, bi-exponential functions have to
be used to fit the emission decay curves (Table 5). To compare the luminescence
lifetimes of the probes in different media and at various concentrations, the so-called
pre-exponential weighted mean lifetime, tM (Table 5) was calculated according to tM ˆ
S(aiti)/Sai [21], where ai are the normalized pre-exponential factors and ti the
reciprocal of the time constants for each component of the multi-exponential fit. The
data of Table 5 show that tM increases almost two-fold with the concentration of
[Ru(pbbs)2(pta)]2 (5a) in H2O. Its luminescence lifetime levels off at 50 mm complex.
This behavior may be explained if different emitting species co-exist in solution. The
short-lived species display decay kinetics similar to that observed in 1 mm solutions (ca.
1.0 ms lifetime) and can be attributed to isolated molecules of the RuII complex. The
long-lived species (ca. 2.8 ms) would correspond to aggregates of [Ru(pbbs)2(pta)]2
(5a) molecules, whose contribution to the overall emission increases with concen-
tration. The longer luminescence lifetime of the aggregates would be due to a higher
protection of the excited state from oxygen quenching and deactivation by the solvent.
As can be seen in Fig. 1, since the tM value reaches a plateau above 50 mm complex, the
aggregates seem to have a defined size. This behavior by which individual molecules in
Table 5. Bi-exponential and Pre-exponential Weighted Emission Lifetimes tM of [Ru(pbbs)2(pta)]2 (5a) a)b)
Concentration [mm]
t1 [ms]
(%)c)
t2 [ms]
(%)c)
tM [ms]
1.0
3.0
7.0
10
25
50
1.05
0.91
1.02
1.07
1.04
0.91
0.94
0.93
100
91
88
72
68
54
54
54
±
±
9
1.1
1.0
1.2
1.4
1.5
1.8
1.8
1.8
2.3
2.7
2.3
2.6
2.9
2.7
2.8
12
28
32
46
46
46
75
100
a) Air-equilibrated aqueous solution. b) Estimated error in t and tM Æ 10%. c) Contribution to the overall decay
calculated as ti % ˆ (ai/Sai) ´ 100.

Helvetica Chimica Acta ± Vol. 84 (2001)
2715
Fig. 1. Emission lifetime of [Ru(pbbs)2(pta)]2 (5a) at different concentrations in water (*) and in the presence
of 10 mm b-cyclodextrin (~) or 10 mm NaCl (&)
equilibrium with aggregates are present in solution is that of classical amphiphilic
molecules [1].
To elucidate the importance of hydrophobic and electrostatic interactions in self-
aggregation of the novel RuII complexes, their luminescence lifetimes were measured in
the presence of b-cyclodextrin or NaCl. Metal complexes 5b, 6a, and 6b displayed again
no sign of association in the investigated additive concentration range (5 ± 200 mm b-
cyclodextrin, 0.01 ± 1.0m NaCl), since their excited-state lifetimes remained unchanged.
The possible reason for the lack of self-aggregation of [Ru(bpds)2(pta)]2 (6a) despite
its amphiphilic character could be the size and high negative charge of its polar head,
making much more important the electrostatic repulsion than the stabilizing hydro-
phobic interactions occurring in micelles. Unlike its bpds analogue, the emission
lifetime of [Ru(pbbs)2(pta)]2 (5a) is strongly increased upon addition of either b-
cyclodextrin (10 mm) or NaCl (10 mm) (see Fig. 1). A plateau at tM ꢀ 2.4 ms is reached
in the presence of those additives, pointing out that their 3MLCT excited state is more
efficiently protected from dissolved oxygen and/or solvent deactivation. The lumines-
cence lifetime of 5a in pure H2O levels off at 50 mm concentration. In the presence of b-
cyclodextrin or NaCl, the concentration of 5a at which a plateau in tM is reached
decreases to ca. 30 and 10 mm, respectively. This behavior can be interpreted in terms of
a higher tendency to self-association, together with a change of the aggregate structure,
since tM of the complex at the plateau is longer. NaCl is more efficient than b-
cyclodextrin in inducing self-association: a higher ionic strength decreases the

2716
Helvetica Chimica Acta ± Vol. 84 (2001)
electrostatic repulsion between the anionic heads allowing the formation of more
compact aggregates. Insertion of the pbbs ligand into the hydrophobic cavity of b-
cyclodextrin would also decrease the electrostatic repulsion between the polar heads of
[Ru(pbbs)2(pta)]2 (5a) to produce more compact aggregates, too.
Since self-association of [Ru(pbbs)2(pta)]2 (5a) in H2O occurs at mm concen-
trations (Table 5 and Fig. 1), the onset of the aggregation process was determined by
measuring the changes in the surface tension of H2O upon addition of the RuII complex
5a (Fig. 2). Corroborating the conclusions from emission lifetime data, aggregation of
the luminophore was found to occur at 1.35 Æ 0.07 mm. The high tendency to self-
binding of this metal complex results from its long alkyl chain and bulky polar head,
which has also some hydrophobic character that favors lipophilic interactions that are
little disturbed by electrostatic repulsions.
Redox Potentials. The electrochemical features of the ground and excited states of
the novel RuII complexes and the prototypes [Ru(pbbs)3]4 (7) and [Ru(bpy)3)]2‡ (8)
are collected in Table 6. Cyclic-voltammetry experiments show one oxidation and three
reduction waves in DMF at room temperature. These waves correspond to the quasi-
reversible monoelectronic redox processes typical for RuII complexes, since the
sulfonate and amide substituents are electrochemically stable [2b][22]. The redox
potentials of the anionic complexes in their ground state agree well with those observed
for the related homoleptic complexes [Ru(pbbs)3]4 (7) and [Ru(bpy)3]2‡ (8) in DMF
(Table 6). The corresponding redox potentials of the 3MLCT excited state were
obtained according to the method of Lin et al. and are very close to those calculated for
Fig. 2. Variation of the surface tension g of [Ru(pbbs)2(pta)]2 (5a) aqueous solutions as a function of the metal-
complex concentration

Helvetica Chimica Acta ± Vol. 84 (2001)
2717
Table 6. Redox Potentials (in V vs. SCE) of the Ground and Excited States of the RuII Complexes in DMF at
23 Æ 28
Oxidationa)
Reductiona)
b
c
E0ꢂ3‡=2‡†
E0ꢂ3‡=2‡
)
E0ꢂ2‡=‡†
E0ꢂ2‡
)
E0ꢂ‡=0†
E0ꢂ0=
†
E0-0 [eV]d)
†
=‡†
*
*
5a
b
6a
b
7
8
1.27
1.24
1.29
1.29
1.25
1.29
0.83
0.86
0.80
0.80
0.84
0.81
1.25
1.24
1.23
1.24
1.24
1.25
0.85
0.86
0.86
0.85
0.85
0.85
1.37
1.40
1.40
1.43
1.35
1.42
1.57
1.65
1.68
1.69
1.56
1.69
2.10
2.10
2.09
2.09
2.09
2.10
b
a) Estimated error Æ 0.03 V. ) Eꢂ03‡=2‡ ˆ E0ꢂ3‡=2‡† E(0-0)
.
c) E0ꢂ2‡
ˆ E0ꢂ2‡=‡† ‡ E(0-0)
.
d) Energy of the 0-0
†
=‡†
*
*
transition from the 3MLCT excited state to the ground state.
the *[Ru(bpy)3]2‡ complex [23]. The similarity, within experimental error, between the
ground- and excited-state oxidation/reduction potentials observed for all the com-
plexes is in agreement with the observed uniformity of their luminescence maxima in
DMF (636 ± 638 nm), demonstrating once again the coincidence between the redox and
spectroscopic orbitals of those species [24].
Luminescence Measurements in Surfactant Solution. The number of probes in the
micelle should not be large since too many interacting species disturb the micellar
structure [1d]. Therefore, we chose an occupation-number [1a] close to one RuII
complex per micelle. All the experiments were performed at surfactant concentrations
in the order of their respective critical micelle concentration (cmc) to follow the
changes in the photophysical properties of the investigated luminophore during the
micellization process. Therefore, the experiments with SDS (cmc ˆ 8.0 mm, aggregation
number 62) [1d] were carried out at 1 mm probe, for a 0 ± 12 mm surfactant-
concentration range. Measurements with CTAB (cmc ˆ 0.92 mm, aggregation number
60) [1d] were performed at 0.3 mm probe, for a 0 ± 15 mm detergent-concentration
range. The experiments with TX-100 (cmc ˆ 0.24 mm, aggregation number 140) [1d]
were carried out at 0.5 mm probe, for a 0 ± 0.35 mm surfactant concentration range. The
variations in the luminescence intensities and lifetimes of the anionic RuII complexes in
the presence of detergents are depicted in Figs. 3 ± 5.
Measurements in the Presence of SDS. No changes in the wavelength of the MLCT
absorption maxima of the [RuL2L']2 complexes are detected upon addition of SDS.
However, in the presence of an increasing concentration of the anionic surfactant,
[Ru(pbbs)2(pta)]2 (5a) displays a strong hyperchromic shift in its absorption
spectrum. At 4 mm SDS, the absorbance is twice as much as that of the complex in
the absence of detergent, and it keeps constant at higher surfactant concentrations. This
behavior can be attributed to light scattering due to pre-micellization of the surfactant
around the luminophore. The induced formation of micelles in the presence of an
added solute at surfactant concentrations below its ꢀnormalꢁ cmc was firstly suggested
to explain the spectral changes of pinacyanol in SDS [7c][25]. The variation in the
emission intensity of the complexes with the SDS concentration is shown in Fig. 3, a.
No shifts in the position of the luminescence maximum are detected for 5a, but it
displays a strong signal enhancement in the 0 ± 8 mm SDS range, with a very steep

2718
Helvetica Chimica Acta ± Vol. 84 (2001)
Fig. 3. a) Normalized emission intensity and b) lifetimes of the [RuL2L']2 complexes (1 mm in air-equilibrated
water) in the presence of SDS: [Ru(pbbs)2(pta)]2 (5a; *), [Ru(pbbs)2(paa)]2 (5b; &), [Ru(bpds)2(pta)]2
(6a; ~), and [Ru(bpds)2(paa)]2 (6b; !)

Helvetica Chimica Acta ± Vol. 84 (2001)
2719
Fig. 4. a) Normalized emission intensity and b) lifetimes of the [RuL2L']2 complexes (0.3 mm in air-equilibrated
water) in the presence of CTAB: [Ru(pbbs)2(pta)]2 (5a; *), [Ru(pbbs)2(paa)]2 (5b; &), [Ru(bpds)2(pta)]2
(6a; ~), [Ru(bpds)2(paa)]2 (6b; !)

2720
Helvetica Chimica Acta ± Vol. 84 (2001)
Fig. 5. a) Normalized emission intensity and b) lifetimes of the [RuL2L']2 complexes (0.5 mm in air-equilibrated
water) in the presence of Triton X-100: [Ru(pbbs)2(pta)]2 (5a; *), [Ru(pbbs)2(paa)]2 (5b; &),
[Ru(bpds)2(pta)]2 (6a; ~), and [Ru(bpds)2(paa)]2 (6b; !)

Helvetica Chimica Acta ± Vol. 84 (2001)
2721
increment in the 0 ± 2 mm SDS range, in agreement to the proposed pre-micellization of
the detergent around the RuII complex [6b]. The maximum luminescence intensity is
reached around 7 mm SDS, a value close to the detergent cmc (8.0 mm) [1d]. For SDS
concentrations above these value, a plateau in the emission intensity is reached. The
emission lifetimes of the complexes in the presence of SDS are depicted in Fig. 3, b. All
the probe molecules display strictly exponential emission kinetics. The results for 5a
parallel those obtained with the steady-state measurements. There seems to be
preconcentration of surfactant molecules around the luminophore at SDS concen-
trations below its cmc, shown by an increase of t, and a plateau value (1.3 ms) is reached
thereafter. These changes are due to some protection of the 3MLCTexcited state of the
complex by the interacting surfactant molecules. The luminescence intensity and
lifetime measured at SDS concentration above its cmc can be the consequence of
stabilization of the structure of the luminophore ´ surfactant aggregates, since the
experimental conditions were adjusted to maximize the number of singly-occupied
micelle (see above). The fractional exposure to H2O (f value; see Exper. Part) of
[Ru(pbbs)2(pta)]2 in the presence of SDS micelles was calculated from the emission
lifetime data in H2O and D2O (Table 7). The resulting f value, 0.72 (Table 8),
demonstrates a rather weak interaction with the detergent micelles since it would mean
that, on the average, ca. 2/3 of complex 5a are exposed to the solvent molecules. This
might be due to the large size of the RuII complex and to the absence of favorable
Table 7. Luminescence Lifetimes (t0/ms) of the Novel RuII Complexes in H2O and D2O Solution in the Presence
a
of SDS, CTAB, or TX-100 Micelles
)
t0/ms
No detergent
H2O
SDSb)
H2O
CTABc)
H2O
TX-100d)
H2O
D2O
D2O
D2O
D2O
5a
b
6a
b
3.7
3.8
0.48
0.40
5.4
5.6
0.95
0.79
4.8
4.8
0.46
0.41
6.8
6.4
0.92
0.81
3.4
5.4
0.71
0.47
3.7
6.2
1.1
5.6
4.9
0.56
0.40
6.5
6.2
1.0
0.82
0.79
a) Ar-purged solutions at 258; 10 mm complex concentration, except for 5a (ꢁ 1.2 mm); estimated error Æ 2%.
b) 20 < [SDS] < 33 mm. c) [CTAB] ˆ 13 mm, except for 5a (1.3 mm). d) [TX-100] ˆ 5 mm, except for 5a (16 ꢁ
[TX-100] ꢁ 33 mm).
Table 8. Fractional Exposure to H2O (f) of the Novel RuII Complexes in Micellar Aqueous Solutions of SDS,
CTAB, and TX-100 at 258a)
f Value
SDS
CTAB
TX-100
5a
b
6a
b
0.72 Æ 0.06
0.59 Æ 0.05
1.05 Æ 0.05
0.97 Æ 0.06
0.24 Æ 0.03
0.27 Æ 0.02
0.48 Æ 0.02
0.72 Æ 0.04
0.29 Æ 0.02
0.52 Æ 0.04
0.77 Æ 0.04
1.00 Æ 0.06
a) Calculated from data of Table 7 (see also Exper. Part).

2722
Helvetica Chimica Acta ± Vol. 84 (2001)
electrostatic interactions with the micelle, in addition to hydrophobic interactions with
the long alkyl chain. This result agrees nicely with the observed quenching rate constant
of the luminophore by oxygen: kq (Table 9) decreases slightly in going from aqueous
1
1
solution to SDS micelles (2.7 vs. 2.1 ´ 109 m s ), as expected for a weak protection of
the photoexcited probe.
Table 9. Quenching Rate Constants (kq/109
m
1s 1) of the Novel RuII Complexes by Oxygen in H2O and in
Micellar Aqueous Solutions of SDS, CTAB, and TX-100 at 258a)
1
kq/109 m 1s
H2O
SDS
CTAB
TX-100
5a
b
6a
b
2.7
3.0
1.9
1.1
2.1
2.2
1.8
1.1
1.8
1.9
0.25
±b)
1.2
2.3
1.2
0.8
a) Experimental error Æ 10%. b) Not determined.
[Ru(pbbs)2(paa)]2 (5b) displays a similar behavior to that of 5a as far as its
absorption and emission features is concerned, but to a smaller extent (Fig. 3). This
result indicates that the absence of a long alkyl chain negatively influences the
nucleation process of the surfactant around the luminophore molecule. The fractional
exposure of 5b to H2O in the presence of SDS micelles is calculated to be 0.59
(Table 8), indicating some protection of the MLCTexcited state in the supramolecular
entity, as shown by the emission intensity and lifetime data. This result is corroborated
by the 20% decrease of the oxygen-quenching rate constant (Table 9), as also observed
for complex 5a.
RuII Complexes 6a and 6b do not exhibit any hyperchromic shift of their absorption
spectra and negligible changes in their emission features in the examined SDS
concentration range, pointing out the lack of interaction with SDS. This finding is
further supported by the unity f values measured for 6a and 6b (Table 8), as well as by
identical oxygen-quenching rate constants determined in H2O and in SDS solution
(Table 9). Such behavior would be due to an important repulsive effect of the anionic
RuII complex, having a high charge density, and the anionic micelle, which is not
balanced by peripheral hydrophobic benzene rings such as those contained in the pbbs
ligand.
Measurements in the Presence of CTAB. No changes in the wavelength of the MLCT
transition of the [RuL2L']2 complexes are detected upon addition of CTAB. However,
the position of their emission band is not constant, and while exponential luminescence
decays are measured for complexes 5b, 6a, and 6b, this is not always the case for
[Ru(pbbs)2(pta)]2 (5a). Random variations in the absorbance of the RuII complexes
are observed in the presence of the cationic surfactant. This fact is due to a slow
appearance of [RuL2L']2 /CTAB precipitates after ion-pair formation. The variations
in the emission intensity of the complexes with the CTAB concentration are displayed
in Fig. 4,a. Since the RuII concentrations are lower than in the absorption experiments,
no precipitation is observed upon addition of the detergent. The complex
[Ru(pbbs)2(pta)]2 (5a) shows a 12-nm red shift in its emission maximum in the

Helvetica Chimica Acta ± Vol. 84 (2001)
2723
presence of CTAB. The observed bathochromic shift can be ascribed to a somewhat less
polar environment experienced by the RuII complex (see the emission maximum in
H2O compared to a non-protogenic solvent such as DMF, Table 2), as it might be
expected for a luminescent probe buried in the micelle. This observation is also
supported by the dramatic luminescence enhancement that occurs in the 0 ± 0.4 mm
CTAB range, indicating efficient pre-micellization of surfactant molecules around the
RuII complex, even at very low detergent concentrations. The maximum intensity is
reached at 0.4 mm CTAB, far below its cmc value (0.92 mm) [1d]. For CTAB levels
above 0.4 mm, a slight decrease occurs, and a plateau is reached. The variations in the
emission lifetimes of the probes with the CTAB concentration are shown in Fig. 4,b.
Single-exponential decays are observed for all the complexes, except for
[Ru(pbbs)2(pta)]2 (5a), for which bi-exponential fits are required. This complex
displays single-exponential decays at [CTAB] ꢃ 1.3 mm. Therefore, pre-exponential
weighed mean lifetimes are used in Fig. 4,b for complex 5a at low CTAB
concentrations. Its luminescence lifetime increases as minute amounts of CTAB are
added to the solution, and a plateau value (1.4 ms) is reached at only 0.4 mm CTAB. The
strong interaction between RuII complex 5a and CTAB is confirmed by its low
fractional exposure to H2O in the presence of CTAB micelles (f ˆ 0.24, Table 8), i.e.,
only about 1/4 of this luminescent probe is in contact with the aqueous environment.
The very efficient protection of the 3MLCTexcited state of 5a by CTAB micelles is also
supported by a lower oxygen-quenching rate constant in surfactant medium (Table 9).
[Ru(pbbs)2(paa)]2 (5b) also shows an increase in its emission intensity, and
lifetime (Fig. 4). The absence of a long aliphatic chain in 5b has an effect on the
nucleation of the surfactant molecules around the luminophore, since the pre-
micellization process starts at higher CTAB concentrations compared to 5a. The latter
complex also shows a 12-nm red shift of its emission maximum in the presence of
CTAB, demonstrating the strong interaction of both complexes with the detergent. The
low fraction of solvent exposure observed for 5b (f ˆ 0.27, Table 8) agrees with the
proposed efficient association with CTAB micelles, largely due to favorable electro-
static interactions. The lower oxygen-quenching rate constant observed (Table 9) also
supports this result.
The complexes [Ru(bpds)2(pta)]2 (6a) and [Ru(bpds)2(paa)]2 (6b) behave
differently. The former displays a slight emission enhancement and longer lifetimes at
CTAB concentrations close to its cmc due to electrostatic and hydrophobic
interactions, but shows a progressive 8-nm blue shift in its luminescence maximum
with increasing CTAB concentration. Unlike the pbbs complexes, the hypsochromic
shift has also been observed when the bpds chelates are dissolved in DMF compared to
3
their emission maxima in H2O (Table 2). The observed destabilization of the MLCT
excited state can be attributed to a less polar environment experienced by the charge-
dense RuII(bpds) species, as it would occur for a probe partially protected from the
H2O molecules by the cationic micelle. Complex 6b does not show any change in the
position of its emission band, indicating that it is mainly surrounded by an aqueous
environment. A decrease of ca. 50% in its emission intensity is observed, even at very
low CTAB concentration (Fig. 4, a) due to luminescence quenching in the ion pair. This
effect is also noticed in the emission lifetime. The different behavior of 6a and 6b can
be rationalized by the effect of the alkyl chain, i.e., the existence or absence of

2724
Helvetica Chimica Acta ± Vol. 84 (2001)
hydrophobic interactions. It is likely that, despite the favorable electrostatic attraction
between the complexes and the polar heads of the CTAB molecules, the geometry of
the aggregates is different in these two cases. Therefore, it is not surprising that the
degree of exposure to solvent for 6a and 6b is different (f ˆ 0.48 and 0.72, resp.). While
the pta complex 6a interacts significantly with the CTAB micelle, its paa analogue 6b
experiences mainly an aqueous environment. The oxygen-deactivation constant of 6a
bound to CTAB micelles is 8-fold lower than its value in H2O (Table 9), due to very
effective protection of the luminophore from the quencher approach. The oxygen-
quenching rate constant for 6b could not be determined, due to the luminescence
quenching observed in the presence of CTAB (see above).
Measurements in the Presence of TX-100. No changes in the position of the MLCT
absorption band of the [RuL2L']2 complexes are detected upon addition of TX-100.
All of them, with the exception of [Ru(bpds)2(paa)]2 (6b), display a hyperchromic
shift in their absorption spectrum, due to pre-micellization of the surfactant around the
luminophore. The variation of their emission intensity with the TX-100 concentration is
depicted in Fig. 5, a. [Ru(pbbs)2(pta)]2 (5a) shows a very strong luminescence
enhancement in the 0 ± 0.2 mm TX-100 range. Its emission intensity levels off at this
concentration, near to the TX-100 cmc (0.24 mm) [1d]. Moreover, a 4-nm bath-
ochromic shift of the emission band is observed in the presence of the non-ionic
surfactant. Such behavior reveals the strong tendency to association between the
amphiphilic complex 5a and TX-100. The effect of this detergent on the emission
lifetimes of the RuII complexes is displayed in Fig. 5, b. The luminescence decay profiles
observed for all the probes fit to single-exponential functions. Once again
[Ru(pbbs)2(pta)]2 (5a) shows the largest variation (2.5-fold) in its excited-state
lifetime upon addition of TX-100, in agreement with the steady-state results. The value
of t reaches a plateau (2.1 ms) at about TX-100 cmc. The calculated fractional exposure
of 5a to H2O in the presence of TX-100 micelles is 0.29 (Table 8), indicating that only
one third of the complex lies in the aqueous phase, and pointing out a high protection of
its MLCT excited state. The oxygen-quenching rate constant of 5a bound to TX-100
micelles is about half of the value measured in H2O (Table 9), a result consistent with a
deeper location of the probe in the micellar surface.
Unlike 5a, neither the emission intensity nor the band position of
[Ru(pbbs)2(paa)]2 (5b) undergo any change in the presence of TX-100. This complex
exhibits a small emission-lifetime increase on addition of the surfactant, due to a weak
interaction, as the solvent-exposure fraction (0.52, Table 8) points out too. This result is
also supported by a kq value only slightly lower than that measured in H2O and similar
to that observed in the presence of SDS (Table 9). All these data indicate that the
hydrophobic alkyl chain is a key structural element for an efficient interaction and
nucleation of the neutral surfactant around the RuII complex, since binding of these
probes to non-ionic micelles has been reported to be controlled by non-electrostatic
interactions [6c].
[Ru(bpds)2(pta)]2 (6a) and [Ru(bpds)2(paa)]2 (6b) show a different behavior
compared to the pbbs-containing complexes. Both probes display a 30-nm blue shift in
their emission band on the first addition of TX-100 (0.1 mm final concentration). This
large hypsochromicity progressively decreases in the presence of more concentrated
TX-100 solutions, until a spectral position similar to that of the complexes in H2O is

Helvetica Chimica Acta ± Vol. 84 (2001)
2725
recovered at detergent concentrations above its cmc. Such a result can be explained if a
pre-micellar association takes place, since the weakly polar ethoxyalkyl groups of TX-
100 would provide less stabilization to the highly dipolar MLCTexcited state than when
the RuII complex is surrounded by H2O molecules. Higher concentrations of TX-100
result in stabilization of the surfactant micellar aggregate itself and weaker interaction
with the luminophore, leading to lower emission intensity and a red shift of the
emission peak, due to H2O solvation of the free complex. The emission lifetimes of
complexes 6a and 6b do not change significantly upon addition of TX-100, due to nil or
weak interaction between luminophore and surfactant, either above or below its cmc.
This result agrees with the measured fractional exposure to H2O of these complexes
(ꢃ0.8, Table 8) and is consistent with the similar oxygen-quenching rate constants
measured in the presence and in the absence of TX-100 micelles (Table 9).
Overall, our results are in agreement with the ample studies carried out by Demas
and co-workers on the interaction of RuII photosensitizers with surfactant media [4c].
For instance, they observed an f value of 0.53 for the highly charged complex
[Ru(pbbs)3]4 (7) in the presence of SDS, although they did not found any evidence of
pre-micellar aggregation. The absence of a hydrophobic ancillary ligand such as pta
would account for the difference with 5a. In the case of their [Ru(CN)4(phen)]2
(phen ˆ 1,10-phenanthroline), a complex with small anionic head, no binding to SDS
was detected. A similar situation occurs with our complexes 6a and 6b. When the
cationic metal center is well separated from the anionic substituents, electrostatic
binding can occur by intercalation of the surfactant head groups between the sulfonated
benzene rings. With the bpds complexes and the like, charge separation is small, and
interaction with the anionic surfactant is hindered. The anionic [Ru(CN)4(phen)]2
binds tightly (yet with a high solvent accessibility, f ꢀ 0.6) to the surface of the cationic
detergent CTAB due to strong electrostatic attraction [6d]. Microcrystallization of
probe ´ surfactant aggregates was also noticed, in the same way that occurs for
complexes 5 and 6 below the surfactant cmc. [Ru(pbbs)3]4 (7) also interacts strongly
with CTAB (f ꢀ 0.3) [4c], a result paralleled by our pbbs complexes. [Ru(pbbs)3]4 (7)
and [Ru(pbbs)(phen)2] have been shown to promote induced cmc of TX-100 [4c][7c]
because of the peripheral sulfonate groups, which make the complexes more
hydrophilic and decrease the binding strength. It is not a surprise that our pbbs
complexes display the highest change of their spectroscopic properties upon binding to
TX-100.
Conclusions. ± The detailed study on the interaction of a homogeneous family of
anionic RuII complexes with detergents allows identification of the most important
factors that control their binding and helps to design future luminescent probes for
optical-sensor development. Selectivity to the analyte may be enhanced through a
careful design of the metal-complex architecture. Electrostatic factors are of utmost
importance to boost or minimize surfactant binding. A high hydrophobicity of the
polypyridine chelating ligands ensures strong interactions with detergent molecules.
The combination of both factors within a coordinated moiety (such as pbbs (3a))
provides the highest sensitivity of the probe to its microenvironment at all surfactant
concentrations. A weak binding due to charge repulsion can be partially or largely
offset by hydrophobic interactions. In this way, (polypyridine)ruthenium(II) com-

2726
Helvetica Chimica Acta ± Vol. 84 (2001)
plexes may also be designed to respond non-selectively to a wide variety of detergent
molecules.
Experimental Part
General. Solvents: purified H2O was obtained from a Milli-Q system (Millipore). All the org. solvents used
(MeOH, EtOH, BuOH, AcOH, CHCl3, Et2O, MeCN, DMF) were of spectroscopic grade. Surfactants and other
chemicals: sodium dodecyl sulfate (SDS; Sigma), cetyltrimethylammonium bromide (CTAB; Fluka), Triton
X-100 (TX100; Sigma), b-cyclodextrin (Aldrich), NaCl (Merck), disodium (1,10-phenanthroline-4,7-diyl)bis-
‡
(benzenesulfonate) (pbbs ´ 2 Na ; 3a) (Fluka), tris(2,2'-bipyridine-kN,kN')ruthenium (2‡) ([Ru(bpy)3]2‡; 8;
Aldrich). Prep. thin layer chromatography (TLC): silica gel plates, 0.5-mm thick (Merck 60). Column
chromatography (CC): Sephadex-LH-20 (Pharmacia) stationary phase. M.p.: Gallenkamp-MFB-595 apparatus;
uncorrected. UV/VIS Absorption spectra: Perkin-Elmer Lambda-3 spectrophotometer; lmax [nm] (e); at 23 Æ
1
28. IR Spectra: Perkin-Elmer 761 or Pye-Unicam SP3-200 spectrometers; KBr pellets (Merck, Uvasol); in cm
.
1H- and 13C-NMR Spectra: Varian VXR-300S spectrometer; in (D6)DMSO, CDCl3, CD3OD, or D2O, > 99% D;
at 300 and 75 MHz, resp.; chemical shifts d in ppm, J in Hz. MS: Varian MAT-711 or Hewlett-Packard HP-5989A
spectrometers. Elemental analysis: Perkin-Elmer 2400 CHN analyzer.
Fluorescence, Quantum Yields, and Lifetimes. UV/VIS Emission spectra: Perkin-Elmer LS-50 spectro-
fluorometer; at 25 Æ 0.58. Luminescence quantum yields (Fem): in H2O, following Parker and Reesꢁ method [26],
with [Ru(bpy)3]2‡ (8) in H2O (Fem ˆ 0.042 Æ 0.002 at 258 under Ar) as reference luminophore [19a][27];
corrected for the different absorption of the reference and sample aq. solns. at the excitation wavelength [28]; at
25.0 Æ 0.28 kept with a Haake D8-GH circulator; Ar-purged (N50, Praxair) solns. Luminescence lifetimes (t):
Edinburgh-Instruments FL-900 time-correlated single photon counting (TC-SPC) apparatus, equipped with a
hypobaric N2 (N55, Praxair) discharge lamp pulsed at 20 or 40 KHz, and a Hamamatsu R955 photomultiplier
tube (190 ± 850 nm) cooled at 278 with a Peltier, connected to an Ortec 416A delay generator for the required
time windows (0.1 ± 10 ms); at 25.0 Æ 0.28 with air-equilibrated, Ar- or O2-purged (N40, Praxair) solns. Analysis
of the emission profiles without deconvolution of the instrumental response, with the original Edinburgh-
Instruments software; the decays were fitted, from the peak channel to the baseline of the experimental
datapoints, to an exponential function, by means of the least-squares non-linear Marquardt algorithm [29]; when
necessary, an increasing number of exponentials was added until the fit was statistically acceptable, as judged by
the c2 test, the appearance of the weighted residuals plot, and the Durbin-Watson parameter.
Cyclic Voltammetry and Tensiometry. Cyclic voltammetry: halfwave potentials of the novel [RuL2L']2
complexes were measured in DMF at 23 Æ 28 with an EG&G-Princeton-Applied-Research-Versastat system. The
three-electrode system consisted of a glassy carbon working electrode (Metrohm 6.0905.010), a Pt counter
electrode, and a standard calomel electrode (SCE). Tetrabutylammonium hexafluorophosphate (0.1m; Fluka)
was used as supporting electrolyte. Sample solutions contained 2.0 ´ 10 4 m [RuL2L']2 and were purged with Ar
Â
(C50; Carburos Metalicos). DMF was thoroughly dried over 3-Š molecular sieves (5% w/v; Scharlau) for at
least one week prior to sample preparation. [Ru(bpy)3]2‡ (8) was used as reference compound (E0ꢂ3‡=2‡† ˆ 1.24 V,
E0ꢂ2‡=‡†
ˆ
1.27 V [30]). Tensiometry: Krüss K10-ST digital tensiometer. Self-aggregation of dissolved
[Ru(pbbs)2(pta)]2 (5a) was evaluated by measuring the changes in the absolute surface tension of water
following Wilhelmyꢁs method [31]. This procedure allows calculation of the critical micelle concentration (cmc)
of surfactant molecules by measuring the force between the Pt-plate of the tensiometer and the sample soln.
[32]. Such a force is proportional to the surface tension of the surfactant soln. and decreases with the surfactant
concentration, until a plateau is reached above the cmc of the surfactant. The experiments were performed at
26 Æ 18 by addition of 20-ml aliquots of a 2.0 ´ 10 m aq. soln. of [Ru(pbbs)2(pta)]2 (5a) to the tensiometer
4
vessel filled with 20 ml of purified H2O. The experiment was performed three times and the results were
averaged.
Surfactant Solutions for Luminescence Measurements. Aq. solns. of the [RuL2L']2 complexes in the
presence of surfactants at concentrations of the order of their cmcs were prepared, with RuII complex/detergent
ratios providing low micellar occupancy numbers to avoid self-quenching of the luminophore.
The fractional exposure of the micelle-bound luminophores to the aq. phase (f) is a parameter that
quantitates the part of the metal-complex structure not shielded from the aq. phase by the surfactant molecules.
Calculation of this parameter f is based on efficient RuII excited-state quenching by the OH oscillators of the
H2O molecules, according to the method developed by Demas and co-workers [6b]: f ˆ [1/tH(m) 1/tD(m)]/[1/
tH(s) 1/tD(s)], where tH(m) and tD(m) are the luminescence lifetime of the probe in the presence of micelles in
H2O and D2O, resp., while tH(s) and tD(s) are the luminescence lifetimes in H2O and D2O, resp.

Helvetica Chimica Acta ± Vol. 84 (2001)
2727
1,10-Phenanthrolin-5-amine (1). To a suspension of 5-nitro-1,10-phenanthroline (2.06 g, 9.0 mmol;
> 98%; Sigma) and 10% Pd/C (0.27 g; Fluka) in 90 ml of 95% EtOH, hydrazine hydrate (2.0 ml, 42 mmol;
> 99%; Fluka) was added dropwise. The mixture was heated for 3 h at 65 ± 708 and then filtered hot through a
glass-fiber filter (Macherey-Nagel MN85/90) to remove the catalyst. The yellow soln. was concentrated to 25 ml
and left at 48 for 24 h. The precipitate was filtered off, washed with cold Et2O, and dried at 0.007 Torr over P2O5
for 24 h: 1.35 g (77%) of 1. Yellow crystals. M.p. 245 ± 2468 ([33]: 259 ± 2608). IR: 3420, 3330, 3230, 3060, 1645,
1
1620, 1600, 1500, 1440, 840, 750. H-NMR ((D6)DMSO): 9.06 (dd, J ˆ 4.2, 1.7, H C(2)); 8.69 (dd, J ˆ 4.2, 1.7,
H
C(9)); 8.69 (dd, J ˆ 8.3, 1.7, H C(4)); 8.05 (dd, J ˆ 8.3, 1.7, H C(7)); 7.74 (dd, J ˆ 8.3, 4.2, H C(3)); 7.51
(dd, J ˆ 8.3, 4.2, H C(8)); 6.88 (s, H C(6)); 6.13 (s, NH2). 13C-NMR ((D6)DMSO): 177.6 (C(5)); 154.2 (C(2),
C(9)); 152.1 (C(10b)); 139.8 (C(6a)); 135.5 (C(4), C(7)); 129.0 (C(10a)); 125.1 (C(3), C(8)); 122.8 (C(4a)); 79.1
(C(6)).
N-(1,10-Phenanthrolin-5-yl)tetradecanamide (pta; 2a). To a soln. of tetradecanoic acid (ˆ myristic acid;
0.52 g, 2.0 mmol; Aldrich) in anh. CHCl3 (70 ml) cooled to 08, Et3N (0.3 ml, 2.0 mmol; Merck) was added
dropwise. To this mixture, ethyl carbonochloridate (0.2 ml, 2.0 mmol; Aldrich) was added. The cooled mixture
was stirred for 30 min, and 1 (0.40 g, 2.0 mmol) was added slowly until complete dissolution. The mixture was
left to reach r.t., and then it was refluxed for 2 d. The crude mixture was subsequently washed with dil. HCl soln.
at pH 4 (3 Â 30 ml), distilled H2O (2 Â 30 ml), sat. Na2CO3 soln. (2 Â 30 ml), and distilled H2O (2 Â 30 ml). The
combined yellowish org. phase was dried (MgSO4) and evaporated, the residue recrystallized from MeOH/H2O
and the product dried at 0.007 Torr over P2O5 overnight: 2a ´ 2 MeOH (0.37 g, 39%). White microcrystals. M.p.
79 ± 808. UV/VIS (CHCl3): 270 (19600), 240 (17400). IR: 3400, 3240, 3020, 2940, 2900, 2820, 1640, 1620, 1580,
1530, 1460, 730. 1H-NMR (CDCl3): 9.17 (d, J ˆ 3.9, H C(2)); 9.11 (d, J ˆ 3.9, H C(9)); 8.28 (d, J ˆ 8.1,
H
C(4)); 8.23 (s, NH); 8.18 (d, J ˆ 8.1, H C(7)); 7.75 (s, H C(6)); 7.63 (dd, J ˆ 8.1, 3.9, H C(3)); 7.61 (dd,
J ˆ 8.1, 3.9, H C(8)); 2.55 (t, J ˆ 7.0, 2 H C(2')); 1.83 (quint., J ˆ 7.0, 2 H C(3')); 1.29 (m, 20 H); 0.88 (t, J ˆ
6.6, Me(14')). 13C-NMR (CDCl3): 172.7 (CˆO); 149.5 (C(2), C(9)); 146.1 (C(10b)); 144.1 (C(10a)); 135.6
(C(4)); 130.5 (C(5)); 129.9 (C(7)); 128.0 (C(6a)); 124.1 (C(4a)); 123.2 (C(6)); 122.4 (C(3)); 120.1 (C(8)); 37.2
‡
(C(2')); 31.8 (C(12')); 29.5, 29.4, 29.2 (8C); 25.6 (C(3')); 22.5 (C(13')); 14.0 (C(14')). MS: 407 (1.4, [M ‡ 2] );
‡
‡
‡
406 (5.1, [M ‡ 1] ); 405 (16.5, M); 404 (3.2, [M 1] ); 390 (1.4, [M 15] ); 376 (1.9), 362 (1.9), 348 (1.9), 334
(1.9), 320 (1.4), 306 (1.4), 292 (1.6), 278 (1.2), 264 (1.4) (loss of CH2 fragments); 250 (5.6, McLafferty ‡ 13);
‡
237 (13.0, McLafferty); 196 (26.8); 195 (100.0, 1 ). Anal. calc. for C26H35N3O ´ 2 CH3OH: C 72.29, H 8.71, N 8.88;
found: C 71.61, H 9.03, N 8.95.
N-(1,10-Phenanthrolin-5-yl)acetamide (paa; 2b). To a soln. of Ac2O (1.1 ml, 12.0 mmol; Panreac) in AcOH
(1.8 ml, 30.3 mmol; Probus), 1 (0.40 g, 2.0 mmol) was added. The reddish soln. was stirred for 2.5 h at r.t. and
then refluxed for 24 h. The cooled crude was poured over distilled H2O and neutralized with concentrated
ammonia solution (Probus) to pH 8. The product was recrystallized from EtOH/H2O, filtered, washed, and
dried at 0.007 Torr over P2O5 for 24 h: 2b ´ H2O (0.18 g, 35%). White needles. M.p. 231 ± 2338. UV/VIS (CHCl3):
270 (24350), 244 (17200). IR: 3500, 3100, 3000, 1680, 1540, 1510, 1440. 1H-NMR (CDCl3): 8.98 (br. s, H C(2),
H
C(9)); 8.73 (s, NH); 8.31 (d, J ˆ 8.0, H C(4)); 8.02 (d, J ˆ 8.0, H C(7)); 7.92 (s, H C(6)); 7.51 (dd, J ˆ 8.0,
4.0, H C(3)); 7.41 (dd, J ˆ 8.0, 4.0, H C(8)); 2.29 (s, Me). 13C-NMR (CDCl3): 169.9 (CˆO); 149.8, 149.6
(C(2), C(9)); 146.1 (C(10b)); 144.2 (C(10a)); 135.8 (C(4)); 130.7 (C(5)); 130.3 (C(7)); 128.0 (C(6a)); 124.3
‡
‡
(C(4a)); 123.3 (C(6)); 122.5 (C(3)); 120.4 (C(8)); 24.0 (Me). MS: 238 (5.4, [M ‡ 1] ), 237 (31.3, M ), 195
‡
‡
(100.0, 1 ), 42 (42, CH2ˆCˆO ). Anal. calc. for C14H11N3O ´ H2O: C 64.12, H 5.19, N 16.14; found: C 64.50, H
5.09, N 16.47.
‡
Disodium (2,2'-Bipyridine)-4,4'-disulfonate (bpds ´ 2 Na ; 3b). To a suspension of 10% Pd/C (0.51 g; Fluka)
in anh. MeOH (45 ml), disodium [2,2'-bipyridine]-4,4'-disulfonate 1,1'-dioxide (4.60 g, 11.7 mmol), prepared
following the procedure by Anderson et al. [12], and anh. ammonium formate (8.00 g, 127 mmol; Fluka) were
added dropwise. The mixture was heated at 408 for 5.5 h and then filtered through a glass-fiber filter (Macherey-
Nagel MN85/90) and the filtrate evaporated. The product was redissolved in MeOH/H2O 4 :1, filtered hot
through a PTFE membrane (Millipore, Millex FGS; 0.2 mm pore size) and precipitated with EtOH. The
‡
resulting product was filtered, washed, and dried at 0.007 Torr over P2O5 overnight: 3b ´ 2 Na ´ 1.5H2O (1.50 g,
45%). White microcrystals. IR: 3080, 1680, 1580, 1550, 1460, 1240, 1050, 860, 770. 1H-NMR (D2O): 8.85
(d, J ˆ 5.1, H C(6), H C(6')); 8.44 (d, J ˆ 1.3, H C(3), H C(3')); 7.88 (dd, J ˆ 5.1, 1.3, H C(5), H C(5')).
13C-NMR (D2O): 158.5 (C(2), C(2')); 155.1 (C(6), C(6')); 153.5 (C(4), C(4')); 123.5 (C(5), C(5')); 121.1 (C(3),
C(3')). Anal. calc. for C10H6N2Na2O6S2 ´ 1.5H2O: C 30.87, H 2.27, N 7.19; found: C 31.01, H 2.34, N 7.24.
Tetrasodium Dichlorobis[(1,10-phenanthroline-4,7-diyl-kN,kN')bis(benzenesulfonato)(2 )]ruthenate(4
)
(4a) and Tetrasodium Bis[(2,2'-bipyridine)-4,4'-disulfonato(2 )-kN,kN']dichlororuthenate(4 ) (4b): General
‡
‡
‡
‡
Procedure. To a mixture of pbbs ´ 2 Na (3a ´ 2 Na ; Fluka) or bpds ´ 2 Na (3b ´ 2 Na ) (2.05 mmol) and RuCl3 ´

2728
Helvetica Chimica Acta ± Vol. 84 (2001)
3H2O (1 mmol; Aldrich) in DMF (6 ml; 99,5%; Panreac), LiCl (6.0 mmol; Panreac) was added. The soln. was
refluxed for 5 h (pbbs complex) or 24 h (bpds complex) under Ar. The mixture at r.t. was poured in a beaker and
kept inside a closed container under acetone atmosphere for 3 d. The respective products obtained were filtered
‡
off, washed with cold acetone, and dried at 0.007 Torr for 24 h over P2O5: 4a ´ 4 Na ´ 8H2O ´ 4 LiCl (1.37 g, 88%)
‡
‡
or 4b ´ 4 Na ´ 4H2O ´ 25 LiCl (0.85 g, 42%) as violet microcrystals. UV/VIS Data of 4a ´ 4 Na ´ 4H2O ´ 25 LiCl
(H2O): 465 (11600), 227 (61200). UV/VIS (DMF): 567 (8800), 476 (10300), 282 (70400). IR: 3060, 1640, 1190,
1040. Anal. calc. for C48H28Cl2N4Na4O12RuS4 ´ 8H2O ´ 4 LiCl: C 36.76, H 2.87, N 3.92; found: C 36.99, H 2.85, N
‡
3.59. Data of 4b ´ 4 Na ´ 4H2O ´ 25 LiCl (H2O): 516 (2740), 366 (8750), 301 (14600). (DMF): 564 (7700), 388
(9600), 303 (47100), 250 (24100). Anal. calc. for C20H12Cl2N4Na4O12RuS4 ´ 4H2O ´ 25 LiCl: C 11.51, H 0.87, N
3.22; found: C 11.87, H 1.00, N 2.88.
Heteroleptic Complexes 5 and 6 (Na2[RuL2L']): General Procedure. A mixture of 4a or 4b (1 mmol) and 2a
or 2b (2.0 mmol) in MeOH (40 ml) under Ar was refluxed for 24 h. Different purification procedures were
applied to the mixture (see below).
Disodium Bis[(1,10-phenanthroline-4,7-diyl-kN,kN')bis(benzenesulfonato)(2 )][N-(1,10-phenanthrolin-
‡
5-yl-kN,kN')tetradecanamide]ruthenate(2 ) (5a ´ 2 Na ): The complex was precipitated from the mixture with
acetone/H2O 9 :1, filtered off, washed with Et2O, and purified by prep. TLC (BuOH/AcOH/H2O 7:3 :4). The
‡
fraction containing the pure complex was desorbed with MeOH to yield 5a ´ 2 Na ´ 11 H2SiO3 (0.45 g, 19%).
Orange microcrystals. UV/VIS (H2O): 460 (36600), 427 (35650), 274 (175800). UV/VIS (MeOH): 457 (30000),
430 (28550), 278 (110000). IR: 2940, 2850, 1640, 1570, 1420, 1190, 1050, 810. 1H-NMR (CD3OD): 8.80 (d, J ˆ 7.5,
1 H); 8.63 (d, J ˆ 9.0, 1 H); 8.55 (br. s, 1 H); 8.44 ± 8.34 (m, 3 H); 8.30 ± 8.20 (m, 6 H); 8.04 (br. s, 2 H); 8.02 ± 7.90
(m, 6 H), 7.76 ± 7.60 (m, 16 H); 2.64 (t, J ˆ 6.8, 2 H C(2')); 1.80 (t, J ˆ 6.8, 2 H C(3')); 1.47 (m, 2 H C(4'));
1.26 (br. s, 18 H); 0.86 (t, J ˆ 6.0, Me). Anal. calc. for C74H63N7Na2O13RuS4 ´ 11 H2SiO3: C 36.81, H 3.23, N 4.21;
found: C 37.15, H 3.48, N 4.10.
Disodium Bis[(1,10-phenanthroline-4,7-diyl-kN,kN')bis(benzenesulfonato)(2 )][N-(1,10-phenanthrolin-
‡
5-yl-kN,kN')acetamide]ruthenate(2 ) (5b ´ 2 Na ). The complex was precipitated from the mixture with
acetone/Et2O 9 :1. The precipitate was filtered off, washed with Et2O, and then precipitated again from MeOH/
‡
Et2O. The product was filtered, washed with CHCl3, and dried at 0.007 Torr over P2O5 overnight: 5b ´ 2 Na ´
8H2O (0.75 g, 50%). Orange microcrystals. UV/VIS (H2O): 459 (22700), 435 (22550), 275 (102000), 215
(98600). IR: 1630, 1550, 1420, 1180, 1050, 840. 1H-NMR (CD3OD): 8.83 (d, J ˆ 8.3, 1 H); 8.64 (d, J ˆ 8.3, 1 H);
8.56 (br. s, 1 H); 8.46 ± 8.34 (m, 3 H); 8.32 ± 8.23 (m, 6 H); 8.12 ± 7.96 (m, 8 H); 7.88 ± 7.58 (m, 16 H); 2.36 (br. s,
Me). Anal. calc. for C62H39N7Na2O13RuS4 ´ 8H2O: C 49.30, H 3.46, N 6.02; found: C 49.33, H 3.64, N 6.29.
Disodium Bis[(2,2'-bipyridine)-4,4'-disulfonato(2 )-kN,kN'][N-(1,10-phenanthrolin-5-yl-kN,kN')tetra-
‡
decanamide]ruthenate(2 ) (6a ´ 2 Na ). The mixture was filtered and the solid product washed with MeOH,
dried at reduced pressure, and recrystallized from MeOH. Further recovery of the product from the mixture was
achieved by CC (MeOH) followed by reprecipitation with MeOH/MeCN. The product was filtered off, washed
‡
with cold MeCN, and dried at 0.007 Torr over P2O5 for 24 h: 6a ´ 2Na ´ 9 H2O (0.55 g, 41%). Orange
microcrystals. UV/VIS (H2O): 466 (17500), 434 (15600), 296 (67000), 266 (42300), 218 (72000). IR: 2900, 2840,
1620, 1530, 1420, 1220, 1060, 760. 1H-NMR (CD3OD): 9.00 (d, J ˆ 1.5, 2 H C(3) of bpds); 8.95 (s, J ˆ 1.5,
2 H C(3) of bpds); 8.77 (dd, J ˆ 8.4, 1.1, H C(4)); 8.63 (dd, J ˆ 8.4, 1.1, H C(7)); 8.52 (s, H C(6)); 8.20 (dd,
J ˆ 5.4, 1.1, H C(2)); 8.13 (d, J ˆ 5.6, 2 H C(6) of bpds); 8.11 (dd, J ˆ 5.4, 1.1, H C(9)); 7.98 (s, 0.5 H,
CONH); 7.96 (s, 0.5 H, CONH); 7.86 (dd, J ˆ 8.4, 5.4, H C(3)); 7.84 (dd, J ˆ 5.6, 1.5, 2 H C(5) of bpds); 7.80
(dd, J ˆ 8.4, 5.4, H C(8)); 7.73 (d, J ˆ 5.6, 2 H C(6) of bpds); 7.58 (dd, J ˆ 5.6, 1.5, 2 H C(5) of bpds); 2.64 (t,
J ˆ 7.0, 2 H C(2')); 1.82 (quint., J ˆ 7.0, 2 H C(3')); 1.44 (m, 20 H); 0.89 (t, J ˆ 6.3, Me). Anal. calc. for
C46H47N7Na2O13RuS4 ´ 9H2O: C 41.01, H 3.98, N 7.36; found: C 41.13, H 3.88, N 7.30.
Disodium Bis[(2,2'-bipyridine)-4,4'-disulfonato(2 )-kN,kN'][N-(1,10-phenanthrolin-5-yl-kN,kN')acet-
‡
amide]ruthenate(2 ) (6b ´ 2 Na ). The mixture was dissolved in MeOH/H2O 10 :1 and filtered hot to remove
unreacted ligand 2b. The filtrate was evaporated, and the residue reprecipitated from EtOH/H2O. The product
‡
was dried at 0.007 Torr over P2O5 overnight: 6b ´ 2 Na ´ 9H2O (0.93 g, 79%). Orange microcrystals. UV/VIS
(H2O): 466 (16800), 446 (16100), 296 (64500), 264 (44800), 218 (71700). IR: 1620, 1540, 1400, 1220, 1040, 760.
1H-NMR (CD3OD): 8.99 (d, J ˆ 1.5, 2 H C(3) of bpds); 8.95 (d, J ˆ 1.5, 2 H C(3) of bpds); 8.82 (dd, J ˆ 8.4,
1.2, H C(4)); 8.63 (dd, J ˆ 8.4, 1.2, H C(7)); 8.54 (s, H C(6)); 8.19 (dd, J ˆ 5.1, 1.2, H C(2)); 8.11 (d, J ˆ 5.9,
2 H C(6) of bpds); 8.10 (dd, J ˆ 5.1, 1.2, H C(9)); 7.97 (s, 0.5 H, CONH); 7.95 (s, 0.5 H, CONH); 7.86 (dd, J ˆ
8.4, 5.1, H C(3)); 7.83 (dd, J ˆ 5.9, 1.5, 2 H C(5) of bpds); 7.79 (dd, J ˆ 8.4, 5.1, H C(8)); 7.72 (d, J ˆ 5.9,
2 H C(6) of bpds); 7.57 (dd, J ˆ 5.9, 1.5, 2 H C(5) of bpds); 2.36 (s, Me). Anal. calc. for C34H23N7Na2O13RuS4 ´
9H2O: C 34.60, H 3.01, N 8.25; found: C 34.75, H 3.12, N 8.34.

Helvetica Chimica Acta ± Vol. 84 (2001)
2729
Tetrasodium
Tris[(1,10-phenanthroline-4,7-diyl-kN,kN')bis(benzenesulfonato)(2 )]ruthenate(4
)
‡
‡
(7 ´ 4 Na ). To a soln. of RuCl3 ´ 3H2O (0.02 g, 0.10 mmol; Aldrich) in EtOH/H2O (9 ml), 3a ´ 2 Na (0.2 g,
0.35 mmol; Fluka) was added under Ar, and the mixture was heated under reflux for 5 h. At r.t., the mixture was
filtered, the filtrate evaporated, and the residue reprecipitated as a colloid from EtOH/H2O. The slurry was
centrifuged at 4000 rpm, the orange solid washed thoroughly with EtOH, and the product dried at 0.007 Torr
‡
over P2O5 overnight: 7 ´ 4 Na ´ 12H2O (0.12 g, 64%). UV/VIS (H2O): 462 (29300), 438 (29100), 278 (121300),
216 (116400). UV-VIS (MeOH): 463 (32400), 438 (32100), 279 (137900), 219 (126600). IR: 1420, 1190, 1050.
Anal. calc. for C72H42N6Na4O18RuS6 ´ 12H2O: C 45.78, H 3.55, N 4.41; found: C 45.98, H 3.54, N 4.47.
We are grateful to the Spanish Government agency CICYT (R ‡ D National Plan) for financial support
through projects AMB98-1043-C02-01 and PPQ2000-0778-C02-01, as well as the Madrid Community
Government (CAM, project no. 07/0082/2000). D. G.-F. thanks the Complutense University for a pre-doctoral
grant. The authors thank Ana M. Castro and C. Knapp (UCM) for helpful assistance in the synthesis of the RuII
complexes.
REFERENCES
[1] a) N. J. Turro, M. Grätzel, A. M. Braun, Angew. Chem., Int. Ed. 1980, 19, 675; b) K. Kalyanasundaram,
ꢀPhotochemistry in Microheterogeneous Systemsꢁ, Academic Press, Orlando, 1987; c) ꢀPhotochemistry in
Organized and Constrained Mediaꢁ, Ed. V. Ramamurthy, VCH Publishers, New York, 1991; d) R. von
Wandruszka, Crit. Rev. Anal. Chem. 1992, 23, 187; e) ꢀTopics in Fluorescence Spectroscopyꢁ, Ed. J. R.
Lakowicz, Plenum Press, New York, 1994, Vol. 4.
[2] a) J. N. Demas, B. A. DeGraff, Anal. Chem. 1991, 63, 829A; b) A. Juris, V. Balzani, F. Barigelletti, S.
Campagna, P. Belser, A. Von Zelewsky, Coord. Chem. Rev. 1988, 84, 85.
[3] a) J. N. Demas, B. A. DeGraff, ꢀDesign and Applications of Highly Luminescent Transition Metal
Complexesꢁ, in ꢀTopics in Fluorescence Spectroscopyꢁ, Ed. J. R. Lakowicz, Plenum Press, New York, 1994,
Vol. 4, p. 71; b) G. Orellana, D. García-Fresnadillo, M. D. Marazuela, M. C. Moreno-Bondi, J. Delgado,
J. M. Sicilia, ꢀFibre-Optic Chemical Sensors: from Molecular Engineering to Environmental Analytical
Chemistry In the Fieldꢁ, in ꢀMonitoring of Water Quality: The Contribution of Advanced Technologiesꢁ,
Ed. F. Colin, P. Quevauviller, Elsevier, Amsterdam, 1998, p. 103.
[4] a) P. Serguievski, W. E. Ford, M. A. J. Rodgers, Langmuir 1996, 12, 348; b) G. Bünau, T. Wolff, Adv.
Photochem. 1988, 14, 273; c) S. W. Snyder, S. L. Buell, J. N. Demas, B. A. DeGraff, J. Phys. Chem. 1989, 93,
5265; d) A. Kirsch-De Mesmaeker, J. P. Lecomte, J. M. Kelly, Topics Curr. Chem. 1996, 177, 25; e) C.
Moucheron, A. Kirsch-De Mesmaeker, J. M. Kelly, Struct. Bonding 1998, 92, 163; f) E. Gicquel, N. Paillous,
P. Vicendo, Photochem. Photobiol. 2000, 72, 583; g) G. L. Duveneck, C. V. Kumar, N. J. Turro, J. K. Barton,
J. Phys. Chem. 1988, 92, 2028; h) R. Ramaraj, P. Natarajan, J. Polym. Sci. 1991, 29, 1339; i) H. D. Gafney,
Coord. Chem. Rev. 1990, 104, 113; j) P. V. Kamat, W. E. Ford, Photochem. Photobiol. 1992, 55, 159; k) K.
Maruszewski, D. P. Strommen, K. Handrich, J. R. Kincaid, Inorg. Chem. 1991, 30, 4579; l) M. C. Moreno-
Bondi, G. Orellana, N. J. Turro, Macromolecules 1990, 23, 910.
[5] a) ꢀMolecular Dynamics in Restricted Geometriesꢁ, Ed. J. Klafter, J. M. Drake, John Wiley and Sons, New
York, 1989; b) M. H. Gehlen, F. DeSchryver, Chem. Rev. 1993, 93, 199; c) P. Lianos, Heterogen. Chem. Rev.
1996, 3, 53.
[6] a) W. J. Dressick, B. L. Hauenstein, J. N. Demas, B. A. DeGraff, Inorg. Chem. 1984, 23, 1107; b) B. L.
Hauenstein, W. J. Dressick, S. L. Buell, J. N. Demas, B. A. DeGraff, J. Am. Chem. Soc. 1983, 105, 4251;
c) W. J. Dressick, B. L. Hauenstein, T. B. Gilbert, J. N. Demas, B. A. DeGraff, J. Phys. Chem. 1984, 83, 3337;
d) S. W. Snyder, J. N. Demas, B. A. DeGraff, Chem. Phys. Lett. 1988, 145, 434.
[7] a) W. J. Dressick, K. W. Raney, J. N. Demas, B. A. DeGraff, Inorg. Chem. 1984, 83, 875; b) B. L.
Hauenstein, W. J. Dressick, T. B. Gilbert, J. N. Demas, B. A. DeGraff, J. Phys. Chem. 1984, 88, 1902;
c) B. L. K. Krisnagopal Mandal, B. L. Hauenstein, J. N. Demas, B. A. DeGraff, J. Phys. Chem. 1983, 87, 328.
[8] a) U. Lachish, M. Ottolenghi, J. Rabani, J. Am. Chem. Soc. 1977, 99, 8062; b) J. H. Baxendale, M. A. J.
Rodgers, J. Phys. Chem. 1982, 86, 4906.
[9] a) J. I. Cline, W. J. Dressick, J. N. Demas, B. A. DeGraff, J. Phys. Chem. 1985, 89, 94; b) K. Gooding, K.
Johnson, L. Lefkowitz, B. M. Williams, J. Chem. Educ. 1994, 71, A8.
[10] R. Nasielski-Hinkens, M. Benedek-Vamos, D. Maetens, J. Heterocycl. Chem. 1980, 17, 873.
[11] D. H. R. Barton, F. Comer, D. G. T. Creig, P. G. Sammes, M. C. Cooper, G. Hewitt, W. G. E. Underwood, J.
Chem. Soc. C 1971, 3540.

2730
Helvetica Chimica Acta ± Vol. 84 (2001)
[12] S. Anderson, E. C. Constable, K. R. Seddon, J. E. Turp, J. E. Baggott, M. J. Pilling, J. Chem. Soc., Dalton
Trans. 1985, 2247.
[13] R. Balicki, Synthesis 1989, 645.
[14] B. P. Sullivan, D. J. Salman, T. J. Meyer, Inorg. Chem. 1978, 17, 3334.
[15] R. A. Krause, Struct. Bonding 1987, 67, 1.
[16] E. A. Seddon, K. R. Seddon, ꢀThe Chemistry of Rutheniumꢁ, in ꢀTopics in Inorganic and General
Chemistryꢁ, Ed. R. J. H. Clark, Elsevier, Amsterdam, 1984, Vol. 19, p. 1173.
[17] M. J. Cook, A. P. Lewis, G. S. G. McAuliffe, V. Skarda, A. J. Thomson, J. L. Glasper, D. J. Robbins, J. Chem.
Soc., Perkin Trans. 2 1984, 1303; P. C. Alford, M. J. Cook, A. P. Lewis, G. S. G. McAuliffe, V. Skarda, A. J.
Thompson, J. L. Glasper, D. J. Robbins, J. Chem. Soc., Perkin Trans. 2 1985, 705.
[18] D. García-Fresnadillo, Y. Georgiadou, G. Orellana, A. M. Braun, E. Oliveros, Helv. Chim. Acta 1996, 79,
1222.
[19] a) J. Van Houten, R. J. Watts, J. Am. Chem. Soc. 1976, 98, 4853; b) A. Pfeil, J. Am. Chem. Soc. 1971, 93,
5395.
[20] S. L. Murov, I. Carmichael, G. L. Hug, ꢀ Handbook of Photochemistryꢁ, 2nd edn., Marcel Dekker, New
York, 1993.
[21] E. R. Carraway, J. N. Demas, B. A. DeGraff, Anal. Chem. 1991, 63, 332; M. P. Xavier, D. García-
Fresnadillo, M. C. Moreno-Bondi, G. Orellana, Anal. Chem. 1998, 70, 5184; D. García Fresnadillo, M. D.
Marazuela, M. C. Moreno Bondi, G. Orellana, Langmuir 1999, 15, 6451.
[22] ꢀOrganic Electrochemistry: an Introduction and a Guideꢁ, Ed. M. M. Baizer, H. Lund, Marcel Dekker, New
York, 1983.
[23] C. T. Lin, W. Bötcher, M. Chou, C. Creutz, N. Sutin, J. Am. Chem. Soc. 1976, 98, 6536.
[24] G. Orellana, M. L. Quiroga, A. M. Braun, Helv. Chim. Acta 1987, 70, 2073.
[25] P. Mukerjee, K. Mysels, J. Am. Chem. Soc. 1955, 77, 2437.
[26] K. D. Mielenz, ꢀMeasurement of Photoluminescenceꢁ, in ꢀOptical Radiation Measurementsꢁ, Ed. F. Grum,
C. J. Bartleson, Academic Press, New York, 1982, Vol. 3.
[27] J. V. Caspar, T. J. Meyer, J. Am. Chem. Soc. 1983, 105, 5583.
[28] D. F. Eaton, Pure Appl. Chem. 1988, 60, 1107.
[29] S. W. Snyder, J. N. Demas, B. A. DeGraff, Anal. Chem. 1989, 61, 2704; D. V. OꢁConnor, D. Phillips, ꢀTime-
Correlated Single Photon Countingꢁ, Academic Press, London, 1984.
[30] C. M. Elliott, E. J. Hersenhart, J. Am. Chem. Soc. 1982, 104, 7519.
[31] ꢀGIT Fachzeitschrift für das Laboratoriumꢁ, ꢀMeasurement of Interfacial Tension and Surface Tension ±
General Review for Practical Manꢁ, GIT Verlag GmbH, Darmstadt, 1980, Vol. 24, p. 642 ± 648 and 734 ±
742; J. Höpken, M. Möller, Macromolecules 1992, 25, 1461.
[32] P. Mukerjee, K. J. Mysels, ꢀCritical Micelle Concentrations of Aqueous Surfactant Systemsꢁ, NSRDS-NBS
36, U.S. Dept. of Commerce, Washington D.C., 1971.
[33] E. Koft, C. H. Case, J. Org. Chem. 1962, 27, 865.
[34] C. de Dios, Ph.D. Thesis, Universidad Complutense de Madrid, 1994.
[35] M. Maestri, M. Grätzel, Ber. Bunsenges. Phys. Chem. 1977, 81, 504.
[36] Y. Kawanishi, N. Kitamura, Y. Kim, S. Tazuki, Sci. Pap. Inst. Phys. Chem. Res. 1984, 78, 212.
Received May 31, 2001