Evaluation of the Medicinal Potential of Two Ruthenium(II) Polypyridine Complexes as One- and Two-Photon Photodynamic Therapy Photosensitizers

Article abstract of DOI:10.1002/chem.201701392

Two [Ru(phen)₂dppz]²⁺ derivatives (phen=1,10-phenantroline, dppz=dipyrido[3,2-a:2′,3′-c]phenazine) with different functional groups on the dppz ligand [dppz-7,8-(OMe)₂ (1), dppz-7,8-(OH)₂ (2)] have been synthesi

Full text of DOI:10.1002/chem.201701392

A Journal of
Accepted Article
Title: Evaluation of the Medicinal Potential of Two Ruthenium(II)
Polypyridine Complexes as One- and Two-Photon Photodynamic
Therapy Photosensitizers
Authors: Jeannine Hess, Huaiyi Huang, Adrian Kaiser, Vanessa
Pierroz, Olivier Blacque, Hui Chao, and Gilles Gasser
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201701392
Link to VoR: http://dx.doi.org/10.1002/chem.201701392
Supported by

Chemistry - A European Journal
10.1002/chem.201701392
Evaluation of the Medicinal Potential of Two
Ruthenium(II) Polypyridine Complexes as One-
and Two-Photon Photodynamic Therapy
Photosensitizers
a,#
a,b,#
a
a
a
Jeannine Hess, Huaiyi Huang, Adrian Kaiser, Vanessa Pierroz, Olivier Blacque,
b,*
c,
Hui Chao, Gilles Gasser *
a
Department of Chemistry, University of Zurich, Winterthurerstrasse 190, CH-8057,
Zurich, Switzerland.
b
School of Chemistry, Sun Yat-Sen University, Guangzhou 510275, P. R. China.
c
Chimie ParisTech, PSL Research University, Laboratory for Inorganic Chemical Biology,
F-75005 Paris, France.
#
these authors have contributed equally to the work.
*
Corresponding authors. E-mail: ceschh@mail.sysu.edu.cn, Tel: 86 20 8411 0613; E-
mail: gilles.gasser@chime-paristech.fr Tel: +33 1 44 27 56 02, WWW :
www.gassergroup.com.
KEYWORDS: Cancer, Medicinal Inorganic Chemistry, Photodynamic Therapy (PDT),
Photosensitizers, Ruthenium(II) Complexes
1
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
Abstract: Two [Ru(phen)
dipyrido[3,2-a:2′,3′-c]phenazine) with different functional groups on the dppz ligand
dppz-7,8-(OMe) (1), dppz-7,8-(OH) (2)] have been synthesized, characterized and
2
dppz]2+ derivatives (phen = 1,10-phenantroline, dppz =
[
2
2
investigated as photosensitizers (PSs) for photodynamic therapy (PDT) against cancer.
1
Both complexes showed intense red phosphorescence and promising singlet oxygen ( O
2
)
quantum yields of 75 % (1) and 54 % (2), respectively, in acetonitrile. 1 (logPo/w = -0.52,
.4 nmol Ru per mg protein) was found to be more lipophilic, having also a higher cellular
2
uptake efficiency compared to 2 (logPo/w = -0.20, 0.9 nmol Ru per mg protein). 1 localized
evenly in HeLa cells whereas 2, was mainly visualized in the cell membrane by confocal
microscopy. In the dark, 1 (IC50 = 36.2 μM) was found to be more toxic than 2 (IC50 > 100
μM) on HeLa cells monolayer. Importantly in view of PDT applications, both complexes
were found to be non-toxic towards multicellular HeLa spheroids (IC50 > 100 μM). Upon
.
-2
one-photon irradiation (420 nm, 9.27 J cm ), 1 exhibited higher phototoxicity (IC50 = 3.1
M) than 2 (IC50 = 16.7 M) on HeLa cell monolayers. When two-photon irradiation (800

.
-2
nm, 9.90 J cm ) was applied, only 1 (IC50 = 9.5 μM) was found to be active toward HeLa
spheroids. This study demonstrates that the functional group on the intercalative ligand has
a strong influence on the cellular localization and anticancer activity of Ru(II) polypyridyl
complexes.
2
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
Introduction
Over the recent years, photodynamic therapy (PDT) has expanded the range of available
methods to treat certain types of cancer (i.e. lung, bladder, and urinary tumors).[1] PDT
works through the light activation of a photosensitizer (PS), which generates reactive
oxygen species (ROS) that are toxic. These ROS then ultimately lead to cancer cell death,
[
2]
with only minimal damage to normal tissues. More specifically, the light-activated PS
3
reacts with oxygen present in tissue ( O
2
) to generate ROS through two main types of
reaction pathways.[3] In type I reaction, the triplet-state of the PS directly transfers an
3
electron to O
2
or other oxygen-containing species and generates radical ions. In type II
reaction, the excited state of the PS reacts with ground-state molecular oxygen, namely
3O
(1
, via an excited-state energy transfer to generate extremely reactive singlet oxygen
2
1
O
2
). Due to its high reactivity, O
2
causes oxidative damage in different cellular
components (i.e. plasma membrane, endoplasmic reticulum, mitochondria, lysosome,
[
4]
nucleus, etc.). Interestingly, PDT can also induce vasoconstriction and blood flow stasis
resulting in anoxia and subsequent cell death inside tumours.[5] To date, the most
commonly used PS in clinical studies is Photofrin. Photofrin has been clinically approved
to treat bladder cancer, early stage lung cancer, oesophageal cancer and early non-small
cell lung cancer.[6] However, besides tedious synthesis and purification, as well as poor
water solubility, Photofrin is also associated with a pronounced and prolonged generalised
skin photosensitivity and only little initial selectivity.[6]
Although most of the approved PSs for PDT are porphyrin- or phtalocyanine-based organic
molecules, metal complexes can also be employed as excellent PS candidates.[7] Due to
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
their ideal physico-chemical properties (i.e. intense luminescence, large Stokes shifts, high
chemical stability and photo-stability, relatively long phosphorescent lifetimes, high water
1
[8]
solubility and high O
as a promising alternative to the current PSs. As a highlight, [Ru(dmb)
433, dmb = 4,4′-dimethyl-2,2′-bipyridine, IP-TT = 2-(2 ′,2″:5″,2′′′-terthiophene)-
imidazo[4,5-f][1,10]phenanthroline, Scheme 1) just entered into clinical trials for PDT
2
production), Ru(II) polypyridyl complexes have recently emerged
2
+
2
(IP-TT)] , (TLD-
1
[
9]
treatment of non-muscle invasive bladder cancer. Of note, in recent years, several DNA
intercalating Ru(II) polypyridyl complexes have been reported as effective DNA photo-
cleavage agents and photosensitizes for PDT.[10]
2
+
Scheme 1. Chemical structure of TLD-1433 ([Ru(dmb) (IP-TT)] , dmb = 4,4′-dimethyl-2,2′-
2
bipyridine, IP-TT = 2-(2′,2″:5″,2′′′-terthiophene)-imidazo[4,5-f][1,10] phenanthroline).
Despite the currently approved PDT agents are excited by one-photon (OP) excitation
using a high energy laser beam, they have only a limited tissue penetration since short
wavelengths are employed. In order to tackle this drawback, two-photon PDT (TP-PDT)
agents have recently emerged as attractive alternatives to the currently approved PSs.[11]
Contrary to one-photon PDT (OP-PDT), TP-PDT uses low energy near-infrared laser
4
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
irradiation, which not only allows for deeper tissue penetration, but is also less invasive
towards normal cells. This results in reduced photo-bleaching of the PSs.[12] It should be
noted that OP-PDT uses higher photon energy but can be carried out with low power lasers
and LEDs, whereas TP-PDT uses lower photon energy but requires higher power lasers. In
view of TP-PDT applications, viable PSs must have a high efficiency for two-photon
absorption (TPA); a property that is quantified by the two-photon absorption cross-section
value ( ).
2
Theoretically speaking, the higher the TP cross-sections is, the better the TP PDT efficiency
should be. However, one has also to consider the balance between hydrophilicity and
hydrophobicity, the singlet oxygen generation efficiency as well as the cellular uptake
efficiency of the PS. Although some conjugated porphyrins showed significant two-photon
absorption, the 
applications in TP-PDT.[13] Among the different PSs that have been tested for TP-PDT,
Ru(II) polypyridyl complexes ( values ≈ 62-250 GM) were found to be excellent
candidates, with 
values much higher than porphyrins.[10g, 14] However, the physical and
2
values of clinical used porphyrins are insufficient for adequate clinical
2
2
chemical properties of Ru(II) complexes can still be improved by lowering their dark
cytotoxicity or by increasing their photo-index (PI).[14a, 15]
Cancer cells divide and multiply much faster than normal cells. Targeting the DNA
replication machinery thus more severely affects cancer cells in relation to normal tissues,
and DNA targeting anticancer drugs such as cisplatin or doxorubicin are widely used in
chemotherapy.[16] Similarly, PSs preferentially localizing in the cell nucleus provide an
outstanding opportunity to target the replication system and selectively affect cancer cells.
5
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
2
+
[
Ru(phen) ] (phen = phenanthroline) was the first Ru(II) complex to be found to bind to
3
the minor groove of double helix DNA.[17] Subsequently, Barton et.al reported the
introduction of extended planar ligand systems, such as dppz (dppz = dipyrido[3,2-a:2′,3′-
2
+
c]-phenazine) in [Ru(bpy)
2
(dppz)] (bpy = 2,2'-bipyridine), which showed a significant
increase in DNA binding efficiency. Interestingly, the phosphorescence of
2+
[
Ru(bpy)
2
(dppz)] is quenched in aqueous solution. However, upon the addition of DNA,
the dppz ligand can rapidly intercalate between the base pairs of DNA and the Ru(II)
2
+
complex emits intense red phosphorescence. As a result, [Ru(bpy) (dppz)] is known as
2
an excellent DNA intercalative “light-switch” complex.[18]
It has been shown that the physico-chemical properties of Ru(II) polypyridyl complexes
can be rationally designed to be water soluble or water insoluble, photo-stable or photo-
unstable, or to have long or short triplet lifetimes.[19] For example, [Ru(bpy)
dppn] (dppn
2+
2
=
benzo[i]dipyrido[3,2-a:2,3-h]quinoxaline) exhibits much higher lipophilicity and dark
phen]2+ and [Ru(bpy)
dppz] .
2+ [20]
Worthy of note, McFarland
toxicity than [Ru(bpy)
2
2
reported that the same compound had a very low dark toxicity in a different cell line,
namely HL-60 promyelocytic cell line.[21] Moreover [Ru(phen)
dppz] presents enhanced
2+
2
light switching properties and stronger photoluminescence intensities than
2
+
[22]
[
Ru(bpy)
2
dppz] when binding to DNA. However, the impact on the photochemistry
and phototoxicity by changing the substituents on the dppz ligand has only been sparingly
2
+
studied. Some of us have reported [Ru(bpy) (dppz)] derivatives bearing different mono
2
functional groups on the dppz ligand as potential PSs in one-photon PDT[23] but there is
undoubtedly a lack of studies on this topic.
6
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
In this work, we report on the synthesis, characterization, and biological evaluation of two
2
+
substitutionally inert [Ru(phen)
2
(dppz)] derivatives with different functional groups on
the dppz ligand (dppz-7,8-(OMe)
2
(1), dppz-7,8-(OH) (2), Scheme 2) as PSs for one- and
2
two-photon PDT. The purpose of the conducted research was to investigate the influence
of small structural ligand modifications on the photochemical and biological behaviour of
Ru(II) complexes. The distribution coefficients (logPo/w), two-photon absorption cross-
section, singlet oxygen generation quantum yield, cellular distribution, cellular uptake
efficiency and stability in human plasma were investigated. Finally, the dark and light
cytotoxic activities of these two complexes was investigated on cervical cancer (HeLa) cell
monolayer and multicellular HeLa spheroids for OP-PDT and TP-PDT.
Results and discussion
Synthesis and characterization.
The synthetic route for complexes 1 and 2 is outlined in Scheme 2 (the ligand synthesis
can be found in the Supporting Information Figure S1). In an initial step, the two nitro
groups of 1,2,-dimethoxy-4,5-dinitrobenzene (1a) were reduced using Pd/C and H
2
to
obtain 1,2-dimethoxy-4,5-diaminobenzene (1b). The dppz-7,8-(OMe) (1c) intermediate
2
was then obtained after Schiff base condensation of 1b with 1,10-phenantroline in 73%
yield. In the next step, the first ruthenium complex 1 was isolated in 60% yield after
refluxing 1c with [Ru(phen)
2
Cl
2
]2H
2
O in EtOH/H
2
O. Attempts were made to synthesize
1
by reacting [Ru(phen) (phen-dione)
2
2
]2H O directly with 1b. However, the reaction yield
2
for the desired complex was overall low. In the last reaction step, the two methyl groups of
7
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
1
were cleaved using an aqueous solution of 48% HBr to finally afford complex 2 in
1
13
moderate yield (33%). Both 1 and 2 were unambiguously characterized by H and C
NMR spectroscopy, mass spectrometry and elemental analysis (see Supporting Information
Figures S2-S7).
Scheme 2. Reagents and conditions: (i) [Ru(phen)
2
Cl
2
]2H
2
O, EtOH/H O, reflux, o.n. 40%; (ii) Acetic
2
acid, aq. 48% HBr, o.n., 33%.
Crystallography.
The molecular structures of 1 and its planar ligand 1c were confirmed by single crystal X-
ray diffraction (Figure 1, Table S1). In the crystal structure of 1 the Ru(II) center adopts a
distorted octahedral coordination geometry with two phenanthroline and one dppz-7,8-
(
OMe) ligands acting as N,N-bidentate ligands. The Ru–N bond distances and the N–Ru–
2
N bond angles fall in normal ranges: 2.040(6) – 2.115(5) Å and 80.72(19) – 81.5(2)°,
respectively (a search in the Cambridge Database revealed an average Ru–N bond distance
of 2.070 Å with a standard deviation of 0.024 Å obtained on 46 structures containing the
Ru(dppz) fragment; similarly an average N–Ru–N bond angle of 79.2° was found in the
same reported structures with a standard deviation of 1.0°).
8
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
Figure 1. Molecular structure of 1 with an atomic numbering scheme (counter-ions, water molecules
and hydrogen atoms excluded for clarity).
2
The three η ligands are perpendicular to each other as indicated by the dihedral angles of
8
5.3(3), 89.1(3) and 89.3(3)° calculated between the mean planes. The presence of three
planar aromatic ligands is favourable to ring interactions in the solid state; some of them
with centroid-centroid distances smaller than 4 Å can be considered as short -
interactions. The smallest distance was found to be 3.558(4) Å between two ligands almost
parallel to each other (4.9(3)° between the mean planes). The three-dimensional framework
formed by the - interactions created large tunnels running along the a and b axes in which
isolated water molecules are located, and a narrower tunnel along the c axis where ClO -
4
…
ions are observed, linked to the Ru(II) molecules via weak C–H O hydrogen bonds (the
…
…
shortest H O and C O distances are 2.33 and 3.14(2) Å, respectively) (Figures S8-S10).
The crystal structure of 1c revealed that the free ligand dppz-7,8-(OMe) adopts the same
2
conformation with similar structural parameters as in the crystal structure of 1 where it is
coordinated to a Ru(II) centre (Figure S11). The molecule is nearly planar, including the
9
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
carbon atoms of the OMe groups. Solvent molecules of chloroform co-crystallized with the
…
…
main species. They are linked together via weak C–H N and C–H Cl hydrogen bonds. In
addition, the planar dppz molecules are packed along the c axis by -stacking interactions
and the shortest centroid-centroid distance found in the crystal structure is 3.488(2) Å
between two parallel molecules (the dihedral angle between the corresponding mean planes
is 6.11(14)°) (Figures S12-S13).
Photophysical properties.
The photophysical properties of the two complexes were evaluated to gain further insight
in their potential as PDT PSs. 1 and 2 showed a typical metal-to-ligand charge-transfer
(
MLCT) absorption band for Ru(II) complexes in the visible region (400-500 nm) (Figure
S14).[24] The absorption of 2 was slightly red-shifted compared with 1. Upon excitation
with 420 nm light, 1 and 2 showed intense emission bands around 620 nm in CH CN
Figure S15), similarly to other polypyridyl Ru(II) complexes.[23b] In PBS solution, the
emission signals were significantly quenched by water. The phosphorescence quantum
yields (em) evaluated in air-equilibrated CH CN (Table 1) were calculated by comparison
in methanol (em = 4.2 %).[25] The quantum yield values in acetonitrile of
(em = 2.8 %) and 2 (em = 1.7 %) were found to be comparable with the unsubstituted
3
(
3
2
+
with Ru(bpy)
3
1
2+
complex [Ru(phen) dppz] (em = 3.3 %), meaning that the functional group did not have
2
an influence on the excited state.[26] The phosphorescence emission lifetimes in degassed
CH CN were much longer than in aerated CH CN (Figure S16). 1 showed longer
luminescence lifetimes in both aerated (325 ns) and degassed (645 ns) CH CN solutions
3
3
3
1
0
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
than 2 (109 ns and 220 ns, respectively) (Table 1).
3
Table 1. Photophysical data for the complexes in CH CN at room temperature.

ex

em

em
Lifetime /ns

 /%
log P
/
nm
/nm
620
621
/
2.8
1.7
air
degassed
645
PBS
3
CH CN
1
2
420
420
325
109
3
5
75
54
-0.52
-0.20
220
The presence of oxygen therefore had a significant influence on the lifetime of the excited
state for both of the complexes. These results confirm that molecular oxygen in its ground
state is able to interact with the triplet excited state of Ru(II) complexes. The TPA cross-
section ( ) of 1 and 2 from 720 to 850 nm are shown in Figure 2 and the slope of
2
integrated emission intensity as a function of laser power was found to be around 2 (Figure
S17).
2
2
1
1
50
00
50
00
Ru1
Ru2
5
0
0
7
00 720 740 760 780 800 820 840 860
Wavelength/nm
Figure 2. TPA cross section of 1 and 2 in methanol.
11
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
The largest TPA of 1 (245 GM) and 2 (93 GM) was found to be around 800 nm. These
values are comparable with other Ru(II) polypyridyl complexes previously reported for
TP-PDT.[10g, 14] Worthy of note, the TPA values of 1 and 2 were much higher than the
clinical approved photosensitizer H
TPP (< 20 GM at 800 nm).[13a]
2
Singlet oxygen generation.
1
1
Aquantitative evaluation of the singlet oxygen O
2
(  ) production upon irradiation at 420
g
nm was performed to assess the potential of these complexes as PDT PSs.[27] The formation
1
of O
2
was monitored in an assay based on its reaction with an imidazole derivative to form
a trans-annular peroxide adduct, which is able to quench the absorbance of p-
nitrosodimethyl aniline (RNO).[28] The measurements of O
1
generation quantum yield
2
were carried out in both PBS (with 10 mM histidine) and acetonitrile (with 12 mM
1
imidazole) solutions. The O
2
production quantum yields were evaluated by comparison
1
) = 95%).[29] The results obtained are
with the reference molecule phenalenone ( ( 
g
shown in Table 1. Both complexes exhibited 10 times higher singlet oxygen generation
efficiency in acetonitrile than in PBS. These results are comparable with previously
reported data on similar Ru(II) complexes.[11]
Stability of 1 and 2 in human plasma.
In order to assess the compatibility of complexes 1 and 2 in physiological conditions, their
stability in human plasma was evaluated using an adapted method performed previously
for other Ru complexes.[23a] Although 1 and 2 are structurally similar polypyridyl Ru(II)
II
1
2
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
complexes, their stability greatly differ. 1 was found to be stable in human plasma upon
incubation at 37 °C for 48 h when compared to the internal standard (diazepam) (Figure 3,
Figures S18-S19). In contrast, compound 2 was found to decompose significantly in
human plasma. After 24 h and 48 h incubation at 37 °C, only 32% and 19%, respectively,
of complex 2 remained intact (Figure 3, Figures S20-S21).
Figure 3. Ratio [%] of 1 and 2 to diazepam (internal standard) at different time intervals (0 h, 24 h, 48
h) in human plasma. The UPLC chromatograms (375 nm) of complexes 1 and 2 as well as the
chromatograms of 1 and 2 in the presence of the internal standard (diazepam) after 0 h, 24 h and 48 h
of incubation in human plasma can be found in the supporting information (Figure S17-S20).
These results are in line with previously performed human plasma stability experiments on
similar Ru(II) complexes. As investigated previously, a similar [Ru(II)(bipy)2-dppz-7-
2
+
[23a]
hxdroxyl] complex partially decomposed after 4 h of incubation in human plasma.
These results demonstrate that only slight structural modifications of the ancillary
functional groups can have a dramatic impact on the stability of Ru(II) polypyridyl
1
3
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
complexes.
Hydrophobicity Measurements.
The lipophilicity of a compound is well-known to be related to its cellular uptake and
anticancer efficiency.[30] Lipophilicity can be quantitatively evaluated by the determination
of its logPo/w value, where Po/w is the partition coefficient between octanol and water.
LogPo/w determination revealed that 2 (logPo/w = -0.20) was more lipophilic than 1 (logPo/w
=
-0.52) (Table 1). The logPo/w value of 2 was also higher than [Ru(bpy)
2
dppz-OH]2+
(logPo/w = -0.45) with bpy as the auxiliary ligand and the same mono-functionalized dppz
ligand, whereas the difference between [Ru(bpy)
dppz-OMe]2+ (logPo/w = -0.42) and
3
8
2
Ru(bpy)
dppz-OH] (logPo/w = -0.45) was minor.38
2
+
[
2
Dark- and photo-toxicity.
After having established the photo-physical and chemical properties of complexes 1 and 2,
their dark and photo-toxicity was initially tested on cervical cancer HeLa cell monolayers
and the non-cancerous MRC-5 cell line (Table 2). The cells were exposed to the different
compounds (1, 2, ALA, Cisplatin) for 4 h. This was followed by replacement with fresh
medium and incubation in the dark for 44 h.
1
4
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
Table 2. IC50 values (M) and phototoxic index (PI) of HeLa cancer cell monolayer, multicellular HeLa cancer cell spheroids and
non-tumorigenic control cell line MRC-5.
Compound
MRC-5 normal cells
2D HeLa cell monolayer
3D multicellular HeLa spheroids
a
a
a
b
OPc
32.5  6.8 3.2
TPd
Dark
OP
PI
Dark
OP
PI
Dark
104  7.9
100
PI
PI
1.6
5.8  0.9
36.5 ± 3.0
100
3.1 ± 0.6
9.5  2.5
100
1
2
39.2 
100
200
17.4 
6.7
11.7
11
24.5
2.5
> 4.1
n.a.
n.a.
16.7 ± 2.6
> 5.9
1
100
200
n.ae
n.ae
n.ae
n.ae
n.ae
154.8 
14.5
200
200
ALA
200
200
1.6
18.8  1.9
28.3  3.1
29.5  2.6
n.ae
65.6  7.6 71.4  8.2 n.ae
69.5  5.4
Cisplatin
a
b
-2
4
h drug exposure followed by 44 h incubation in the dark. 4 h drug exposure, changed with fresh medium followed by light irradiation (420 nm, 9.27 J cm ) and
c
-2
d
2
then 44 h incubation in the dark. 24 h drug exposure followed by irradiation (450 nm, 10.00 J cm ) and then 48 h incubation at 37 °C, 5% CO . 24 h drug exposure
-
2
e
2
followed by irradiation at 800 nm (9.90 J cm ) and then 48 h incubation at 37 °C, 5% CO . Not applicable.
Cisplatin was used as a positive control for the dark cytotoxicity. 1 was found to be more
cytotoxic (IC50 = 36.5 M) than 2 (IC50 > 100 M) for HeLa cells treated for 4 h in the
dark. Of note, it was impossible to obtain an exact IC50 value of 2 due to precipitation of
the complex at concentrations exceeding 100 M. For comparative purposes, the dark
toxicity of 1 (36.5 M) after 4 h incubation was significantly higher than the non-
2
+
functionalized Ru(II) complex [Ru(phen) dppz] (152.1 M) even after much longer drug
2
exposure (48 h).[30] The introduction of two –OMe groups on the dppz ligand thus clearly
results in an enhanced dark toxicity. In contrast, introducing –OH groups as in complex 2
did not result in a significant change in dark toxicity.
We then tested the photo-toxicity of both complexes towards cancer cells after 4 h
incubation in the dark, washing of the cells, replacement with fresh medium and then
1
5
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
-2
irradiation at 420 nm for 20 min (9.27 J cm ). The cells were further incubated for 44 h
before determining the cell viability. The clinically approved PDT PS 5-aminolevulinic
acid (ALA), which is a precursor in protoporphytin IX biosynthesis, was used as a positive
control.[31] Photo-toxicity experiments were carried out using the following protocol: 4 h
incubation, change to fresh medium, 20 min light irradiation and 44 h incubation in the
dark. The phototoxicity of 1 (IC50 = 3.1 M) on HeLa cell monolayers was found to be
higher than the one of 2 (IC50 = 16.7 M). Due to different experimental conditions used
for the determination of the dark cytotoxicity, we can only compare the IC50 values to Ru(II)
complexes with bpy as the auxiliary ligand and mono-functionalized group on dppz ligand.
2
+
[23a]
1
(IC50 = 3.1 M) was slightly more toxic than [Ru(bpy) dppz-OMe] (IC50 = 5.5 M).
2
The higher singlet oxygen generation quantum yield of 1 seemed to result in higher PDT
effect on cancer cells monolayer upon 420 nm light irradiation. Importantly, both Ru(II)
complexes were found to be more effective than 5-ALA (IC50 = 154.8 M, PI >1). The
relatively low PI value obtained for 5-ALA was due to lower concentrations used in this
assay if compared to the dose typically applied in clinical PDT treatments (about 1 mM),
at which 5-ALA has a higher PI value.[32] The highest concentration of 5-ALA tested in
[
27]
our assay was 200 M, following a previously used protocol, which was sufficient to
allow for a direct comparison to the investigated Ru(II) complexes.
In vitro cultured cancer cell monolayers have a different behavior compared with cancer
cells growing within solid tumors. The latter are generally less sensitive to
chemotherapeutics.[33] As an example, doxorubicin is highly toxic towards cancer cell
monolayers but relatively inactive on cancer cells present in solid tumors. This difference
1
6
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
reflects reduced drug penetration depth and pathophysiological differences or relatively
slower cell cycling in solid tumors.[33] In recent years, in vitro 3D multicellular tumor
spheroid (MCTS) cell culture systems have received increasing interest in anticancer drug
evolution. MCTSs can be used a cell model to study the tissue penetration of fluorescent
drugs. MCTSs can generate an extracellular matrix (ECM) and mimic metabolic and
proliferative gradients, hypoxic areas, as well as the multidrug resistance found in solid
tumors.[34] One of the most important advantages of TP light is its deeper tissue penetration
compared to OP light. In this respect, MCTSs function as a better cell model for TP
photodynamic therapy studies than cancer cell monolayer. Thus, HeLa MCTSs were
selected to investigate the TP-PDT anticancer activity of our Ru(II) polypyridyl
complexes.[35]
The IC50 values of 1 (104 M) and 2 (> 100 M) in the dark on 400 m HeLa MCTSs
-2
were found to be above 100 M. Upon TP irradiation (800 nm, 9.90 J cm ), the IC50 values
of 1 decreased to 9.5 M, approximately 3 times lower than the IC50 value obtained for OP
-2
irradiation (IC50 = 32.5 M, 450 nm, 10.0 J cm ). Though 2 generates higher singlet
oxygen quantum yield (5 %) in PBS than 1 (3 %), 2 was found totally inactive against
MCTSs (IC50 > 100 M, Table 2) under both OP and TP irradiation. These data
demonstrate a significant difference between the phototoxic properties of 1 and 2 on HeLa
cell monolayers and HeLa MCTSs. Thus, we carried out cellular localization and cell
uptake studies to obtain some more insights into the mechanism of cell death upon light
irradiation.
1
7
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
Intracellular localization.
To understand the phototoxic behaviour of the two Ru(II) complexes, we firstly quantified
the cellular amount of ruthenium by inductively coupled plasma mass spectrometry (ICP-
MS). The sensitivity of ICP-MS makes it a practical method to detect and identify metal-
based complex within cells.[36] After 4 h incubation, the amount of 1 (2.4 nmol Ru per mg
protein) within HeLa cells was higher than the one of 2 (0.9 nmol Ru per mg protein). This
difference in cellular localization can be regarded as a possible cause of the significant
difference in the dark cytotoxicity of 1 and 2.
In addition to ICP-MS, confocal microscopy imaging of 1 and 2 under OP (420 nm) and
TP (800 nm) excitation was also performed. The Ru complexes that were not internalized
at this time point were replaced with fresh medium. As shown in Figure S22, 1
homogeneously distributes within the cytoplasm and nucleus, while 2 locates mostly at the
outer surface of HeLa cells, as visualised under both OP and TP excitation. Subsequently,
the exact localization of 1 and 2 was determined by co-staining with commercial organelle
dyes. We incubated HeLa cells with complexes 1 and 2 (50 M, 4 h) and organelle dyes
(30 min) in the dark and then imaged them with a confocal microscope. As shown in Figure
4
, the red signal of 1 overlays with both the fluorescent signal of Hoechst 33342 (nucleus
dye) and of Mitotracker green (MTG, mitochondria dye). On the contrary, 2 was found to
locate at cell membranes. The luminescence of 2 overlays well with the one of the green
membrane cell dye Dio. No overlay of luminescence could be observed with Hoechst
3
3342.
1
8
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
Figure 4. Cellular localization of 1 and 2. HeLa cells were firstly incubated with 1 or 2 (50 μM, 2 h)
and then incubated with commercial dyes for another 30 min. Scale bar = 50 m. Excitation wavelengths:
4
6
3
58 nm (1 and 2), 488 nm (Dio), 488 nm (MTG green) and 405 nm (Hoechst 33342). Emission filter:
00  20 nm (for 1 and 2), 520  20 nm (for MTG), 500 ± 20 nm (Dio) and 460  20 nm (for Hoechst
3342).
Overall, the cellular distribution of 1 was found to be similar to the one of the Ru(II)
analogue [Ru(bpy)
dppz-OMe]2+ with a mono-functionalised dppz ligand previously
2
reported by our group. However, the cell membrane localization property of 2 was
relatively surprising since only Ru(II)-dppz complexes with alkyl ether chains have been
reported to image cell membranes.[37] In this previous study, the more hydrophobic
complex with long alkyl ether chains (R = C H13) was found to localise in cell
6
membranes.[37] Complex 2 seems to be stuck at the cell membrane due to strong interaction
with lipid bilayer, which result in reduced uptake inside cancer cells. Thus the higher dark
cytotoxicity of 1 compared to 2 may be due to a better cell uptake efficiency of 1 than 2.
1
9
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
In regards to cellular localization measured by fluorescent microscopy, a note of caution
must be applied since in view of their luminescent quenching properties in aqueous solution,
and 2 can only be visualized when present in hydrophobic environments, such as in the
1
cell membranes or when bound to proteins or DNA of HeLa cells. Therefore, it is not
excluded that these compounds are located in other compartments where they remain
undetected. Such an observation has been previously made by us and others.[38]
Apart from HeLa cell monolayers, we were also interested in the difference between the
OP and TP cell imaging properties of 1 and 2 on cancer cell spheroids. The cellular uptake
efficiency may be quite different between cancer cell spheroids and cancer cell monolayers
due to the lower cell penetration of a compound. As shown in Figures 5 and 6, the MCTS
penetration ability of 1 (50 M, 12 h) was much higher than the one of 2 (50 M, 12 h)
under the same experimental conditions.
Figure 5. (A) 1P and 2P images of 1 (50 M) after incubation with HeLa spheroids for 12 h. (B) Z-
2
0
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
stack images of the same HeLa spheroids captured every 5 μm along the Z-axis. Excitation wavelengths:
00 nm (1 and 2), Emission filter: 600  20 nm (for 1 and 2).
8
2
was found to only stain the outer sphere of HeLa spheroid as confirmed by Z-scan images
(Figure 6B). No signal from the centre of HeLa MCTSs could be observed. The low
penetration of 2 into MCTSs is coherent with its observed preferred localization on cell
membranes, thus making it a poor candidate for TP-PDT. Differently, 1 was capable of
penetrating deeply into the core of HeLa spheroids (up to 300 m), as can be seen in Figure
5
B. The internal structure of the spheroids was clearly illuminated. Moreover, 1 exhibited
a stronger and clearer phosphorescence in the deeper sections of the spheroids under TP
laser excitation compared with OP excitation, confirming the deeper tissue penetration and
enhanced PDT efficiency of TP light compared with OP light.
Figure 6. (A) 1P and 2P images of 2 (50 M) after incubation with HeLa spheroids for 12 h. (B) Z-
stack images of the same HeLa spheroids captured every 5 μm along the Z-axis. Excitation wavelengths:
2
1
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
800 nm (1 and 2), Emission filter: 600  20 nm (for 1 and 2).
Finally, we also investigated the cellular localization of Ru(II) complexes in HeLa cell
monolayers after light treatment. The cellular localization of 1 changed after light
irradiation, with the compound accumulating in the nucleus (Figure 7B). Similar re-
localization results had been reported for other Ru(II) PSs.[10g, 39] Such an observation could
be explained by the generation of singlet oxygen upon light irradiation, that damages the
nuclear membrane and hence enables the penetration of 1 into the nucleus. On the contrary,
2
was still found to localise in the cell membranes after light irradiation (Figure 7D).
Figure 7. Confocal microscopy images showing the cellular localization of 1 (50 μM, 4 h) without light
irradiation (A) and after irradiation (B). Confocal microscopy images showing the cellular localization
of 2 (50 μM, 4 h) without light irradiation (C) and after irradiation (D). Excitation wavelengths: 458 nm
(1 and 2), and 405 nm (Hoechst 33342). Emission filter: 600  20 nm (for 1 and 2), 460  20 nm (for
Hoechst 33342).
2
2
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
Conclusions
In this paper, we investigated the properties of two Ru(II) complexes in regard to their
potential as photosensitizers for one- and two-photon PDT. The correlation between
cellular localization and PDT efficiency was investigated. Although the structures of 1 and
2
are quite similar, significant differences in their biological properties were found. The
unveiled cell-membrane affinity of 2 significantly reduced its dark toxicity compared to 1,
which was found to spread throughout cytoplasm and nucleus by microscopy imaging. The
dark toxicity of 1 may thus arise from its ability to reach the nucleus, where it may interact
with the DNA, as shown for other nucleus targeting Ru(II) complexes exhibiting strong
anticancer activity.[19a] In the case of 2, due to its poor cellular uptake ability, it was only
detected at the cell membrane, where it may induce membrane damage on cancer cell
monolayers upon photoactivation. In the case of the cancer cell spheroids, 2 only reached
the outer shell of MCTSs, and irradiation did not result in oxidative damage to cells inside
the spheroids. The cellular uptake ability of 1 was better than 2 even though 1 was less
lipophilic than 2. 1 could penetrate into HeLa MCTSs, similarly to the lysosome targeting
Ru(II) polypyridyl complexes reported by our group before.[10] In this previous report, we
showed that the investigated Ru(II) polypyridyl complexes can induce lysosome damage
and re-localize towards the nucleus upon light irradiation. Similarly, we also observed
accumulation of 1 in the nucleus after light irradiation, illustrating that severe oxidative
damage occurred within the treated and irradiated cancer cells. Taken together, this study
illustrates that small changes of the functional groups in polypyridyl Ru(II) complexes have
a significant influence on their cellular localization as well as anticancer activity. This study
2
3
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
also confirms that MCTSs are a more ideal cancer cell model than cancer cell monolayer
to mimic cellular uptake and tissue penetration on solid tumour.
Experimental Section
1
13
Instrumentations and Methods, H and C NMR spectra were recorded in deuterated
solvents on Bruker AV2-401, AV2-400, AV-500 and AV-501 at room temperature. The
chemical shifts, δ, are reported in ppm (parts per million). The signals from the residual
protons of deuterated solvent have been used as an internal reference.[40] The abbreviations
for the peak multiplicities are as follows: s (singlet), d (doublet), dd (doublet of doublets),
t (triplet), q (quartet), m (multiplet), and br (broad). Thin layer chromatography (TLC) was
performed using silica gel 60 F-254 (Merck) plates with detection of spots being achieved
by exposure to UV light. Column chromatography was performed using Silica gel 60
(0.040-0.063 mm mesh, Merck). Eluent mixtures are expressed as volume to volume (v/v)
ratios. ESI mass spectrometry was performed using a Bruker Esquire 6000 spectrometer.
In the assignment of the mass spectra, the most intense peak is listed. Elemental
microanalyses were performed on a LecoCHNS-932 elemental analyser.
Synthesis of 1,2-Dimethoxy-4,5-diaminobenzene (1b). A solution of 1,2,-dimethoxy-4,5-
dinitrobenzene (0.30 g, 1.33 mmol) in MeOH (100 mL) was reduced by hydrogenation
using 5 bar of H
2
, 20 mmol % of Pd/C (32 mg, 0.30 mmol) and a few drops of acetic acid.
After 3 h, the catalyst was filtered off and the filtrated was concentrated in vacuo. The
2
4
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
crude product was purified by redissolving the residue in CH
2
Cl (50 mL) and extracted
2
with 1M HCl solution (100 mL). The two phases were separated and the aqueous pink layer
was adjusted to pH14 with solid NaOH pellets, which resulted in a colour change from
pink to green. The aqueous layer was then extracted with CH
2
Cl
2
(3  100 mL). The
combined organic layers were dried over MgSO , filtered and evaporated to obtain 1,2-
4
dimethoxy-4,5-diaminobenzene (1b) as a greenish oil, which was directly used without
further purification for the next reaction step. Yield 91% (0.20 g, 1.21 mmol). The
spectroscopic data match those previously reported by L. Perrin et al.[41]
7
,8-Dimethoxydipyrido[3,2-a:2`,3`-c]phenazine (dppz-7,8-(OMe) ) (1c). This ligand
2
was synthesized by Schiff base condensation using an adapted procedure described by
Aguirre et al.[42] A mixture of 1,2-dimethoxy-4,5-diaminobenzene (0.10 g, 0.59 mmol) and
1
4
,10-phenantroline (0.08 g, 0.40 mmol) was refluxed in EtOH (30 mL) for approximately
h. The reaction was allowed to reach room temperature. The solvent was evaporated to
leave only a few millilitres of EtOH and the residue was kept in the freezer until a brownish
precipitate was formed. The solids were filtered off and washed with cold EtOH and then
Et
2
O to obtain 7,8–dimethoxydipyrido[3,2-a:2`,3`-c]phenzine (1c). Yield: 73% (0.15 g,
0
.43 mmol). The spectroscopic data match those reported previously by Aguirre et al. [42]
Synthesis of [Ru(phen)
.18 mmol) and 7,8–dimethoxydipyrido-[3,2-a:2`,3`-c]phenazine (dppz-7,8-(OMe)
.075 g, 0.219 mmol) was refluxed under N in a thoroughly deareated EtOH:H O (1:1, 30
2
(dppz-7,8-(OMe)
2
)][(PF
6
)
2
] (1). [Ru(phen)
2
Cl
2
]2H
2
O (0.10 g,
0
0
2
(1c),
2
2
2
5
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
mL) solution overnight. The solvent was evaporated in vacuo. The solid was redissolved
in H O (5 mL) and NH PF was added to precipitate the crude complex as a PF salt. The
crude product was further purified by silica column chromatography with CH CN/aq.
KNO 0.4 M (10:1) as the eluent. The fractions containing the product were combined and
the eluent was evaporated in vacuo. The reddish/orange residue was redissolved in CH CN
10 mL) to dissolve the product, while the white, insoluble solid (excess of KNO ) was
removed by filtration. The solvent was removed in vacuo, the solid redissolved in H O (15
mL) and addition of NH PF allowed the precipitation of the product as a PF salt. The
precipitate was further washed with H O (3  30 mL) and freeze-dried to obtain
] (1) as a reddish powder. Yield: 60% (0.088 g, 0.110
CN): /ppm = 9.57-9.55 (m, 2H), 8.63-8.60 (m, 4H), 8.27
2
4
6
6
3
3
3
(
3
2
4
6
6
2
[
Ru(phen)
2
(dppz-7,8-(OMe)
2
6 2
)][(PF )
1
mmol). H NMR (400 MHz, CD
3
(s, 4H), 8.21-8.19 (m, 2H), 8.08-8.03 (m, 4H), 7.76-7.73 (m, 2H), 7.68-7.62 (m, 6H), 4.09
1
3
(
s, 6H). C NMR (125 MHz, CD
3
CN): /ppm = 156.9, 154.5, 154.2, 154.0, 150.9, 149.0,
1
1
48.9, 142.4, 138.6, 138.0, 137.9, 134.0, 132.1, 132.1, 131.9, 129.2, 129.1, 127.9, 127.0,
2+
26.9, 107.1, 57.5. ESI-MS: m/z (%) = 402.0 ([M-2PF
6
] , 100). Elemental Analysis: calcd.
for C44
H
30
F
12
N
8
O
2
P
2
Ru(H O)2.5: C, 46.41; H, 3.10; N, 9.84. Found: C, 46.11; H, 2.88; N,
2
9
.72.
Synthesis of [Ru(phen)
2
(dppz-7,8-(OH)
2
)][(PF
6
)
2
] (2). This complex was synthesized by
(dppz-
] (2, 0.05 g, 0.046 mmol) in glacial acetic acid (3 mL) was stirred at
0 °C for 30 min. An aqueous solution of 48% HBr (9 mL) was slowly added and the
an adapted synthetic procedure described by Perrin et al.[41] Asolution of [Ru(phen)
2
7
3
,8-(OH)
2
6 2
)][(PF )
2
6
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
reaction mixture was heated to reflux overnight. The next morning, an additional 48% HBr
solution (4.5 mL) was added and the reaction was kept at reflux for a further 12 h. The
reaction was allowed to cool to room temperature. The crude reaction was filtered through
a cotton plug and the solvent was reduced in vacuo. The residue was redissolved in CH
30 mL) and filtered again. The CH CN-solution containing the product was evaporated to
dryness. The solid was redissolved in H O (15 mL) and 10 mL saturated NH PF solution
was added to precipitate the product as a PF salt. The precipitate was further washed with
O (3  30 mL) and freeze-dried to obtain [Ru(phen) (dppz-7,8-(OH) )][(PF ] (2) as a
brownish solid. Yield: 33% (0.016 g, 0.015 mmol). H NMR (400 MHz, CD CN): /ppm
9.63-9.61 (m, 2H), 8.62-8.59 (m, 4H), 8.26 (s, 4H), 8.19-8.18 (m, 2H), 8.05-8.01 (m,
3
CN
(
3
2
4
6
6
H
2
2
2
)
6 2
1
3
=
1
3
4

1
2
2
H), 7.82 (s, 2H), 7.74-7.71 (m, 2H), 7.67-7.61 (m, 4H). C NMR (125 MHz, CD CN):
3
/ppm = 154.3, 154.2, 154.0, 153.7, 150.9, 149.0, 148.9, 142.2, 138.4, 137.9, 137.9, 134.0,
32.1, 132.1, 132.0, 129.1, 127.9, 127.0, 126.9, 110.0. ESI-MS: m/z (%) = 388.0 ([M-
2
+
PF
6
] , 100). Elemental Analysis: calcd. for C42
H
26
F
12
N
8
O
2
P
2
Ru(H O)2.5: C, 45.42; H,
2
.81; N, 10.09. Found: C, 45.33; H, 3.16; N, 10.27.
X-ray crystallography. Single-crystal X-ray diffraction data were collected at low
temperature (153(1) K for 1 and 183(1) K for 1c) on a SuperNova Atlas area-detector
diffractometer from Rigaku Oxford Diffraction[43] (formerly Agilent Technologies) using a
single wavelength Enhance X-ray source with Cu K radiation (λ = 1.54184 Å) from a
α
micro-focus X-ray source and an Oxford Instruments Cryojet XL cooler. The selected
suitable single crystals were mounted using polybutene oil on a flexible loop fixed on a
2
7
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
goniometer head and immediately transferred to the diffractometer. Pre-experiment, data
collection, data reduction and analytical absorption correction[44] were performed with the
Pro 1
[45]
program suite CrysAlis . The crystal structure of 1 was solved with SHELXS97 and
3
[46]
refined with SHELXL97 using the WinGX program system while the crystal structure
of 1c was solved with SHELXT[47] and refined with SHELXL2014[48] using the Olex2
crystallographic software.[49] The SQUEEZE[50] method included in PLATON[51] was used
to calculate the disordered solvent contribution to the calculated structure factors and to
determine the number of isolated water molecules in the solvent-accessible regions. For
more details about the data collection and refinements parameters, see Tables S1 and the
CIF files (CCDC-1502934 (for 1) and CCDC-1502935 (for 1c) contain the supplementary
crystallographic data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif).
Crystal data for 1.(ClO
4
)
2
.(H
2
O)32 (M = 1579.24 g/mol): tetragonal, space group I4 /a (no.
1
3
8
8), a = 30.4431(7) Å, c = 31.0134(9) Å, V = 28742.7(13) Å , Z = 16, T = 153(1) K,
-1
3
μ(CuKα) = 3.348 mm , Dcalc = 1.460 g/cm , 39218 reflections measured (4.1 ≤ 2Θ ≤
1
33.2), 12614 unique (Rint = 0.1204, Rsigma = 0.0816) which were used in all calculations.
The final R was 0.0975 (>2sigma(I)) and wR was 0.2979 (all data). Crystal data for
1
2
1
c.(CHCl
3
)
(M = 461.72 g/mol): orthorhombic, space group Pca2 (no. 29), a =
1
3
1
8.5556(5) Å, b = 15.5691(5) Å, c = 7.06046(15) Å, V = 2039.73(10) Å , Z = 4, T = 183(1)
-1
3
K, μ(CuKα) = 4.298 mm , Dcalc = 1.504 g/cm , 8546 reflections measured (5.7 ≤ 2Θ ≤
36.3), 2762 unique (Rint = 0.0183, Rsigma = 0.0184) which were used in all calculations.
The final R was 0.0433 (I > 2σ(I)) and wR was 0.1169 (all data).
1
1
2
2
8
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201701392
Photophysical property measurements. UV/Vis absorption spectra and extinction
coefficients were obtained on a Varian Cary 50 Scan UV/vis spectrophotometer using
standard quartz cells with 1 cm path length. Emission spectra were recorded on an
Edinburgh Instrument FLSP920 spectrometer equipped with a 450 W Xenon lamp, double
monochromators for the excitation and emission pathways, and a red-sensitive
photomultiplier (PMT-R928) as detector. The emission spectra were fully corrected by
using the standard corrections supplied by the manufacturer for the spectral power of the
excitation source and the sensitivity of the detector. The quantum yields were measured by
use of an integrating sphere with an Edinburgh Instrument FLSP920 spectrometer. The
absorbance of the samples was kept below 0.1 to avoid inner filter effects, except for the
concentration-dependent measurements, and all measurements were carried out at 293 K.
The luminescence lifetimes were measured by using a F900 pulsed 60W xenon
microsecond flash lamp with a repetition rate of 100 Hz and a multichannel scaling module.
The emission was collected at right angles to the excitation source with the emission
wavelength selected by using a double grated monochromator and detected by a R928-P
PMT. The instrument response function (IRF) was measured by using the blank solvent as
scattering sample and setting the monochromator at the emission wavelength of the
excitation beam. The resulting intensity decay is a convolution of the luminescence decay
with the IRF and iterative convolution of the IRF with a decay function and non-linear
least-squares analysis was used to analyze the convoluted data
2
9
This article is protected by copyright. All rights reserved.