Sn(iv) Schiff base complexes: Triplet photosensitizers for photoredox reactions

Article abstract of DOI:10.1039/c4dt01427h

We present the synthesis and characterization of a series of four fluorescent Sn(iv) Schiff base complexes, which also possess long-lived triplet excited states. The complexes absorb visible light (λ_max = 420 to 462 nm) and the optical properti

Full text of DOI:10.1039/c4dt01427h

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Sn(IV) Schiff base complexes: triplet
photosensitizers for photoredox reactions†
Cite this: Dalton Trans., 2014, 43,
17754
Tod A. Grusenmeyer,a Albert W. King,b Joel T. Mague,a Jeffrey J. Rackb and
Russell H. Schmehl*a
We present the synthesis and characterization of a series of four fluorescent Sn(IV) Schiff base complexes,
which also possess long-lived triplet excited states. The complexes absorb visible light (λmax = 420 to
462 nm) and the optical properties are easily tunable without laborious synthetic elaboration. The triplet
excited states are not luminescent, but can be observed and followed using nanosecond transient
absorption spectroscopy. The lifetimes of the triplet excited states are on the order of 500 μs–10 ms in
PMMA matrices. The triplet state energies were estimated via energy transfer reactions with a series of
organic triplet acceptors. In addition, the photoexcited complexes react with electron donors and accep-
tors in solution. These results demonstrate the potential for the development of photosensitizers based
on main group elements with high spin orbit coupling constants.
Received 14th May 2014,
Accepted 9th July 2014
DOI: 10.1039/c4dt01427h
www.rsc.org/dalton
Os(II), Ir(III), Re(I) and Pt(II). A common criticism of artificial
photosynthetic systems involving these complexes has been
Introduction
There has been a great deal of work on exploiting transition the relative scarcity of these elements; all are among the nine
metal complex chromophores as sensitizers in artificial photo- rarest elements in the earth’s crust.
synthetic systems for conversion and storage of solar energy.
A clear challenge for the photochemistry community is the
The majority of such systems rely on the sensitizer to serve as development of complexes that have redox and photophysical
a single electron donor or acceptor following visible light exci- properties similar to those used in existing artificial photo-
tation. These chromophores are typically second and third row synthetic systems, but employ earth-abundant metals. Our
transition metal complexes because they have excited state life- intent is to utilize complexes that are easily prepared and have
times in the microsecond time regime, often exhibit stability visible electronic transitions. The objective for the metal is to
in multiple oxidation states, and are luminescent, allowing use an element with a large spin–orbit coupling matrix element
both the excited state energy and lifetime to be determined (similar to the noble metals) that forms complexes that are
conveniently. The desirable, relatively long excited state life- stable in multiple oxidation states. Also, as a pedagogical point
times result from the fact that the photoactive excited states of interest, we wanted to investigate molecules that may contain
are all of triplet spin multiplicity while the ground states are of non-phosphorescent long-lived excited states. In this regard, it
singlet multiplicity. The formation of triplet excited states is appeared that complexes of either Sn(II) or Sn(IV) might be appro-
facilitated by the presence of the metal due to large spin–orbit priate since Sn is relatively abundant in the earth’s crust and has
coupling matrix elements that enhance intersystem crossing. a spin–orbit coupling matrix element significantly greater than
The majority of published work focuses on complexes of Ru(II), that of Ru, which is widely used in metal complex sensitizers.1
There are a number of articles addressing the photoreactiv-
ity of Sn complexes in the literature. These reports mainly
aDepartment of Chemistry, Tulane University, New Orleans, Louisiana 70118, USA.
focus on the photochemistry of Sn(II) and Sn(IV) aquo, halide,
hydroxo, and cluster complexes,2–8 Sn porphyrins,9–28
E-mail: russ@tulane.edu
bDepartment of Chemistry and Biochemistry, Ohio University, Athens, Ohio 45701,
phthalocyanines,29–33 and corroles,34–37 and transition metal
USA
complexes utilizing organotin compounds as a ligand.38–47
The simple Sn(IV) complexes (halides, hydroxo, etc.) typically
absorb in the ultraviolet, do not luminesce and exhibit ir-
reversible photochemistry from LMCT excited states. Sn(II)
complexes mainly access metal based s→p type excited states,
some of which are luminescent at room temperature. Octa-
hedral Sn(II) clusters also access s→p metal based excited
†Electronic supplementary information (ESI) available: 1H NMR DATA, cyclic vol-
tammetry, room temperature luminescence, nanosecond transient absorption,
ultrafast transient absorption, Saltiel plots, ORTEP representations, illustrations
of calculated HOMO and LUMO orbitals, transient absorption decays, 1O2 phos-
phorescence, time correlated single photon counting data and the determi-
nation of intersystem crossing efficiencies are available as ESI. CCDC
1002911–1002913 and 1002904. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c4dt01427h
17754 | Dalton Trans., 2014, 43, 17754–17765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
states. These clusters show a red shift in their luminescence
due to metal–metal bonding interactions. Sn(II)/Sn(IV) halide
bridged dimers have also been investigated and are known to
exhibit MMCT transitions. Metal-to-ligand charge transfer
(MLCT) transitions are observed in Sn(II) bpy complexes and
porphyrins. The literature on Sn porphyrins, phthalocyanines
and corroles is extensive and includes reports of catalytic
hydrogen reduction, sensitization in organic solar cells, and
donor–acceptor arrays. Sn macrocycles are also known to
efficiently sensitize singlet oxygen and have been used in
schemes for pollutant remediation and photodynamic therapy;
here the excited states are largely localized on the macrocycle.
Given these Sn complexes as a basis and recognizing both the
rich redox chemistry and the extraordinarily large spin–orbit
coupling coefficient of Sn, recent reports of a series of fluo-
rescent Sn(IV) Schiff base derivatives with visible absorp-
tion48,49 appeared to be an excellent entrée for investigation of
triplet state formation and reactivity in Sn(IV) complexes.
Fig. 1 Left: Thermal ellipsoid plot (50% probability) of the complex
PhSn(Lnap)Cl2. Hydrogen atoms are omitted for clarity. A labeled ORTEP
representation and table of crystallographic information is presented in
ESI.† Right: The ligands reacted with Ph2SnCl2. (Top to bottom) HLnap,
HLsal, HLBr, HLdea.
([SnTPPCl2]) as well as the bond lengths and angles obtained
from the DFT geometry optimized singlet ground state are pres-
ented in Table 1. In crystals of [PhSn(LBr)Cl2], there are three
independent conformers in the asymmetric unit and they
differ primarily in the dihedral angle between the mean planes
of the phenyl group and the SnNICl2 unit (15.5°, 34.2° and
37.4°). The bond lengths and angles for these conformers are
given in Table 1 as [PhSn(LBr)Cl2] (1), [PhSn(LBr)Cl2] (2), and
[PhSn(LBr)Cl2] (3), respectively. The Sn–NI and the Sn–NQ
bond lengths are similar for all of the molecules in the series.
The shortest Sn–NI bond length of 2.167(1) Å and longest
Sn–NQ bond length of 2.242(2) Å both occur in [PhSn(Ldea)-
Cl2]. All of the other Sn–N bond lengths fall within this 0.06 Å
window and are approximately 0.1 Å longer than the Sn–N
bonds in [SnTPPCl2]. All of the Sn–C bonds are slightly shorter
than the Sn–N bonds. The Sn–C bonds in [PhSn(Lnap)Cl2],
[PhSn(Lsal)Cl2], and [PhSn(Ldea)Cl2] are all similar with
lengths of 2.146(2) Å, 2.151(2) Å, and 2.146(2) Å, respectively.
For the set of [PhSn(LBr)Cl2] conformers, the Sn–C bond
lengths are equivalent within experimental error, ranging from
2.137(3) Å to 2.145(3) Å. The Sn–O bond lengths are the short-
est of all of the bonds in the coordination environment and
are in the very narrow range 2.036(2) Å–2.050(2) Å. The Sn–Cl
bond lengths are significantly longer with values ranging from
2.4430(7) Å to 2.4996(4) Å. These are slightly longer than the
Sn–Cl bond lengths observed in [SnTPPCl2]. All of the crystallo-
graphically determined bond lengths are in good agreement
with those obtained from DFT calculations. The most notable
difference is in the determination of the Sn–Cl bond lengths
where gas phase DFT calculations yield bond lengths that are
roughly 2.5% longer than in the crystallized molecules.
Results
Four tin complexes containing tridentate Schiff-base ligands,
[PhSn(Lnap)Cl2], [PhSn(Lsal)Cl2], [PhSn(LBr)Cl2], and [PhSn-
(Ldea)Cl2]; HLnap is 2-hydroxy-1-naphthaldehydene-8-amino-
quinoline, HLsal is salicylaldehydene-8-aminoquinoline, HLBr
is 5-bromosalicylaldehydene-8-aminoquinoline, and HLdea is
4-diethylaminosalicylaldehydene-8-aminoquinoline) were syn-
thesized and investigated. The ligands are prepared via an
imine forming condensation reaction of 8-aminoquinoline
with various salicylaldehyde derivatives and 2-hydroxy-
1-naphthaldehyde. The complexes were prepared in a one pot
process via the addition of Ph2SnCl2 to a solution of the
freshly prepared ligand in toluene. As the reaction progresses,
the tin complex precipitates from solution. HLnap was inde-
pendently prepared for characterization by reaction of the
2-hydroxy-1-naphthaldehyde with 8-aminoquinoline in absol-
ute ethanol as previously reported.50 The crystal structure of
one complex, [PhSn(Lnap)Cl2], and the other ligands reacted
with Ph2SnCl2 are shown in Fig. 1. Crystal structures have also
been obtained for the other complexes and ORTEP represen-
tations of each molecule are presented in ESI (Fig. S21–S24†),
each confirming the trans geometry of the chloride ligands.
The complexes were further characterized by 1H NMR spec-
troscopy; the spectra clearly indicate 119Sn (8.59%) and 117Sn
(7.68%) coupling to the ortho protons of the tin bound phenyl
ring and the imine proton of the various NNO ligands. Other
coupling assignments were made using 1H–1H COSY. Images
of 1H NMR spectra are included in ESI (Fig. S1–S5†).
X-ray crystallography
The Cl–Sn–Cl bond angle is significantly less than 180°
with the Cl atoms displaced away from the phenyl ring towards
the imine nitrogen of the ligand such that the SnNICl2 unit is
nearly planar. The departure of the Cl–Sn–Cl angle from 180°
is dependent on the dihedral angle between the phenyl ring
and the mean SnNICl2 plane. In [PhSn(Lsal)Cl2], this angle is
Selected bond lengths and angles are presented in Table 1.
The two Sn coordinated nitrogen atoms are distinguished as
the quinoline nitrogen, NQ, and the imine nitrogen (linking
the quinoline and the aldehyde), NI. In addition, the Sn–N and
Sn–Cl bond lengths for tetraphenylporphinatodichlorotin(IV)
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 17754–17765 | 17755

View Article Online
Paper
Dalton Transactions
Table 1 Selected bond lengthsa
Complex
[PhSn(Lnap)Cl2]
2.196(2) [2.199]
2.233(2) [2.244]
[PhSn(Lsal)Cl2]
2.220(2) [2.227]
2.230(2) [2.248]
[PhSn(LBr)Cl2] (1) [PhSn(LBr)Cl2] (2) [PhSn(LBr)Cl2] (3) [PhSn(Ldea)Cl2]
[SnTPPCl2]74
2.098(2)
Sn–NI
2.230(2) [2.234]
2.230(2) [2.244]
2.214(2) [2.234]
2.236(2) [2.244]
2.205(2) [2.234]
2.235(3) [2.244]
2.167(1) [2.204]
2.242(2) [2.251]
(imine)
Sn–NQ
(quinoline)
Sn–C
2.146(2) [2.147]
2.045(2) [2.028]
2.151(2) [2.146]
2.036(2) [2.014]
2.145(3) [2.144]
2.033(2) [2.018]
2.137(3) [2.144]
2.050(2) [2.018]
2.140(3) [2.144]
2.036(2) [2.018]
2.146(2) [2.150]
2.036(1) [2.012]
Sn–O
Sn–Cl
2.4761(6) [2.5330] 2.4755(7) [2.5320] 2.4723(8) [2.5275] 2.4716(8) [2.5275] 2.4670(8) [2.5275] 2.4996(4) [2.5403] 2.420(1)
2.4676(6) [2.5298] 2.4430(7) [2.5264] 2.4625(8) [2.5232] 2.4632(7) [2.5232] 2.4424(8) [2.5232] 2.4691(4) [2.5342]
169.74(2) [165.71] 166.85(2) [164.11] 166.78(3) [163.97] 167.66(3) [163.97] 168.36(3) [163.97] 169.24(2) [165.14]
Cl–Sn–Cl
bond angle
O–Sn–NQ
bond angle
C–Sn–NI
159.15(7) [159.75] 161.49(7) [161.45] 159.56(8) [161.37] 160.95(8) [161.37] 161.43(9) [161.37] 163.55(5) [162.37]
170.60(8) [173.15] 170.05(8) [171.73] 173.0(1) [171.8]
172.99(9) [171.84] 172.9(1) [171.8]
172.14(6) [171.52]
bond angle
a The initial entries in the table are the bond lengths obtained from X-ray structures with the estimated standard deviation presented in
parenthesis. The values presented in square brackets are the bond lengths obtained from the DFT geometry optimized singlet ground state.
The phenyl ring in the DFT geometry optimized singlet ground state of [PhSn(LBr)Cl2] is 6.5 degrees from the Cl–Sn–Cl axis.
13.3°, resulting in a Cl–Sn–Cl bond angle of 166.85(2)°. In con- not in good agreement with observed values. All of the calcu-
trast, when this angle increases to 21.3° in [PhSn(Ldea)Cl2] lated values are at least 3° less than the actual bond angles.
and 28.0° in [PhSn(Lnap)Cl2] the Cl–Sn–Cl bond angle
increases to 169.24(2)° and 169.74(2)°, respectively. For the
series of [PhSn(LBr)Cl2] conformers observed in the asym-
metric unit, as the dihedral angle increases from 15.5° to 37.4°
the Cl–Sn–Cl bond angle increases from 166.78(3)° to
168.36(3)°. The distortions from this geometry seen in the
solid state structures can be seen to be due in part to contacts
between the o-hydrogens on the phenyl group and the axial
chlorine atoms. With the smallest dihedral angles these con-
tacts are in the range 2.60–2.76 Å which is about that of a
normal van der Waals contact (2.74 Å).51 The larger Cl–Sn–Cl
and dihedral angles involve H⋯Cl contacts in the range
2.8–3.0 Å. Although an exhaustive analysis of the packing has
not been performed, it appears that the observed Cl–Sn–Cl
and dihedral angles represent the best balance between mini-
mizing the intramolecular H⋯Cl contacts and intermolecular
contacts involving the phenyl ring which affect the dihedral
angle. The O–Sn–NQ bond angles vary from 159.15(7)° in
[PhSn(Lnap)Cl2] to 163.55(5)° in [PhSn(Ldea)Cl2]. As the di-
hedral angle between the plane of the phenyl ring and the
SnNICl2 plane increases in [PhSn(LBr)Cl2], the O–Sn–NQ bond
angle increases from 159.56(8)° to 161.43(9)°. There also is a
slight disparity in the C–Sn–NI bond angles across the series of
molecules. [PhSn(Lnap)Cl2] and [PhSn(Lsal)Cl2] have similar
C–Sn–NI bond angles of 170.60(8)° and 170.05(8)°. [PhSn(LBr)-
Cl2] and [PhSn(Ldea)Cl2] have slightly larger C–Sn–NI bond
angles of 172.9(1)° and 172.14(6)°. Variations in the phenyl
plane/SnNICl2 dihedral angle do not affect the C–Sn–NI bond
angle in the series of [PhSn(LBr)Cl2] conformers. The DFT cal-
culated values for O–Sn–NQ bond angles are all in very good
agreement with the values obtained from the crystal structures.
The calculated values for C–Sn–NI are also in good agreement
with the crystal structure values other than for [PhSn(Lnap)Cl2]
where the calculated bond angle is 3° larger than the observed
value. The calculated values for the Cl–Sn–Cl bond angles are
Steady-state absorption and luminescence
The complexes are all yellow or orange in solution. The elec-
tronic absorption spectra of the complexes in acetonitrile are
shown in Fig. 2 with maxima and extinction coefficients given
in Table 2. The HLnap ligand has two overlapping absorption
bands (450 nm and 473 nm) while the complex has a single
maximum at 452 nm. The HLnap ligand and all of the Sn(IV)
complexes exhibit luminescence in acetonitrile. The fluo-
rescence of [PhSn(Lnap)Cl2] is ∼100 times greater than the
HLnap ligand (Fig. S10†). It is also important to note that
[PhSn(Ldea)Cl2] is the only complex in the series that shows
fluorescence solvatochromism (Fig. S11†), exhibiting a red
shift with increasing solvent polarity and suggesting that the
singlet excited state of this complex has significant charge
Fig. 2 Absorption spectra of [PhSn(Lsal)Cl2] (---), [PhSn(LBr)Cl2] (—),
[PhSn(Lnap)Cl2] ( ), and [PhSn(Ldea)Cl2] ( ) in acetonitrile.
17756 | Dalton Trans., 2014, 43, 17754–17765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 2 Photophysical data for all chromophores. All data was collected in acetonitrile
Complex
[PhSn(Lnap)Cl2]
[PhSn(Lsal)Cl2]
[PhSn(LBr)Cl2]
[PhSn(Ldea)Cl2]
Absorption λmax (nm)
452
20 400
524
19 100
1100 100
2.47
0.14
1.25
−1.37
424
10 800
532
18 800
660 50
2.41
434
11 300
541
18 500
690 40
2.11
462
40 700
602
16 600
—
—
Extinction coefficient (M−1 cm−1
Fluorescence λmax (nm)
)
Fluorescence λmax (cm−1
)
Singlet lifetime (ps), from TCSPC
Durbin–Watson valuea (d)
Fluorescence quantum yield 10%b
E0 (+/0) (V)
0.04
1.34
0.06
1.38
0.01
0.83
E0 (0/−) (V)
−1.23
1.1 0.3
16 300
0.59
−1.27
0.57 0.02
16 200
0.72
−1.22
11.9 1.3
15 500
0.05
Triplet lifetime in PMMA (ms)
0.98 0.09
15 300
0.36
Triplet energy, (cm−1
)
300 cm−1
Intersystem crossing efficiency 10%
kisc c (s−1
kr c (s−1
knr c (s−1
)
2.6 × 108
7.5 × 107
3.6 × 108
2.1 × 109
0.43
7.4 × 108
5.0 × 107
4.6 × 108
2.1 × 109
0.52
9.0 × 108
7.5 × 107
2.8 × 108
2.2 × 109
0.66
3.1 × 108
6.3 × 107
5.9 × 109
2.3 × 109
—
)
)
Oxygen quenching rate constant
Quantum yield of 1O2 generation 20%d
a Durbin–Watson values were determined using the residual values obtained from a single exponential fit of deconvoluted luminescence
intensity decays. Each decay was analyzed over 200 data points. b Fluorescence quantum yield values for [PhSn(Lsal)Cl2], [PhSn(LBr)Cl2],
and [PhSn(Lnap)Cl2] were determined with tetracene1 as a reference and the value for [PhSn(Ldea)Cl2] was determined with [Ru(bpy)3](PF6)2 75 as
a reference. c Kinetic values determined using the long decay component obtained from the ultrafast transient absorption spectrum of each
chromophore. d The quantum yield of 1O2 was determined with [Ru(bpy)3](PF6)2 55 as a standard for all complexes.
transfer character. Fluorescence quantum yields of the com-
plexes in acetonitrile range from 0.14 for [PhSn(Lnap)Cl2] to
0.01 for [PhSn(Ldea)Cl2] suggesting the possibility of signifi-
cant intersystem crossing to the triplet manifold. There is no
evidence of phosphorescence in fluid solution or 77 K glass.
The fluorescence lifetimes for the complexes were also deter-
mined by TCSPC. The instrument response function (scatter),
raw luminescence decay, and the exponential fit of the decon-
voluted luminescence decay of [PhSn(Lsal)Cl2] are shown in
Fig. S34.† Collected lifetimes are given in Table 2 for all of the
complexes except [PhSn(Ldea)Cl2] which has a decay that is
faster than the resolution of our system (∼300 ps). The fluore-
scence lifetimes and quantum yields for our series of chromo-
phores are similar to values previously reported for closely
related [SnLCl3] complexes.48,49
Fig. 3 (Top) Transient absorption spectrum of [PhSn(Lnap)Cl2] col-
lected 1 μs, 5 μs, 10 μs, 25 μs, and 50 μs after excitation with a 450 nm
laser pulse. TA spectra of the remaining complexes are provided in ESI.†
(Bottom) UV-Visible absorption spectrum collected before (—) and after
Nanosecond transient absorption
Each chromophore has a unique, long-lived excited state
absorption in de-aerated acetonitrile, which was assumed to
be associated with the triplet state of the chromophore. The
transient absorption spectrum of [PhSn(Lnap)Cl2] at room
temperature along with the ground state absorption over the
same spectral window is shown in Fig. 3. The key features of
these spectra are bleaching of the ground state absorption and
excited state absorption at all wavelengths to the red of
500 nm. The spectrum has three isosbestic points and individ-
ual decays indicate a small (<5%) degree of permanent
product formation. The dashed spectrum of Fig. 3 indicates
the changes observed in the ground state absorption following
collection of the transient spectrum further indicating photo-
degradation in acetonitrile. Transient absorption spectra for
the other complexes are shown in Fig. S13–S15.† Solution tran-
sient decays were non-exponential, regardless of complex
(---) the collection of the TA spectrum.
concentration, and were best fit using equal concentration
second order kinetics, suggesting that triplet–triplet annihil-
ation dominates excited state relaxation in solution. Given the
observed solution behavior involving collisional reaction pro-
cesses, the excited state decay dynamics were examined in
solid poly(methyl methacrylate) matrices (PMMA). In PMMA,
the transient absorption decays are single exponential and
have lifetime values of 570 μs for [PhSn(LBr)Cl2], 1.1 ms for
[PhSn(Lsal)Cl2], 980 μs for [PhSn(Lnap)Cl2], and 11.9 ms for
[PhSn(Ldea)Cl2].
Detailed analysis of the solution kinetics required knowl-
edge of excited state molar concentrations. The excited state
absorptivities and yields for triplet state formation were
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 17754–17765 | 17757

View Article Online
Paper
Dalton Transactions
Fig. 4 Ultrafast transient absorption spectrum of [PhSn(Lnap)Cl2] in acetonitrile. Inset: Early time traces 0.300 ps (—), 0.500 ps ( ), 1 ps ( ), and
5 ps ( ). Evidence of the laser pulse has been omitted for clarity.
obtained from experiments involving triplet energy transfer to
an acceptor with a known excited state absorptivity (anthra-
cene) using solutions that were absorbance matched with [Ru-
(bpy)3]2+ at the excitation wavelength (allowing determination
of the total concentration of excited states from the absor-
bance of the [Ru(bpy)3]2+ triplet absorbance; see experimental).
The decay data was converted from absorbance units using the
triplet–triplet extinction coefficients for each chromophore.
The excited state decays were then fit using equal concentration
second order kinetics. The intrinsic decay rate constants of the
triplet Sn complex, kr and knr, are orders of magnitude smaller
than k[Sn*]2, even at the lowest complex concentrations able to
be used in the transient absorbance experiments. The equal
concentration second order kinetic rate constants are similar
for all chromophores and range from 1.60 × 1010 for [PhSn-
(Ldea)Cl2] to 2.01 × 1010 for [PhSn(Lsal)Cl2].
Table 3 Kinetic information obtained from femtosecond transient
absorption spectra
[PhSn(Lnap)- [PhSn(Lsal)- [PhSn(LBr)- [PhSn(Ldea)-
Complex Cl2]
Cl2]
Cl2]
Cl2]
τ1 (ps)
τ2 (ps)
τ3 (ps)
0.56 0.05
70.0 7.6
1390 96
0.75 0.20
40.8 8.5
758 66
0.57 0.12
40.4 17
767 42
0.13 0.02
1.05 0.20
170 17
four complexes have a kinetic component that is intermediate
in time (between sub-picosecond and the long time kinetic
components), ranging from 1 to 70 ps for the different com-
plexes. This kinetic phase is difficult to assign, but, for all the
complexes, a rise in optical density and a blue shift of excited
state absorption feature(s) is observed after the sub-picosecond
kinetic process.
Ultrafast transient absorption
Triplet yields and energies
The femtosecond transient absorption spectrum of [PhSn- Triplet state energies of the complexes were obtained by
(Lnap)Cl2] in N2 de-aerated acetonitrile at room temperature is energy transfer quenching with a series of triplet acceptors in
shown in Fig. 4. Spectra of the remaining complexes are acetonitrile. The collected energy transfer rate constants were
shown in Fig. S16–18.† A summary of the kinetic information used in evaluation of the free energy dependence of the rate
for the spectra is compiled in Table 2. In aggregate, all four constants using the classical treatment of energy transfer in
complexes feature tri-exponential kinetics, with the longest fluid solution.52–54 Free energy dependence plots are shown in
lived transient spectrum (∼3200 ps) resembling the earliest Fig. S19 and 20† with the resulting triplet state energies given
transient spectrum from the nanosecond studies, illustrating in Table 2. While the results only allow an estimate of the
continuity between the time regimes examined. For all four triplet energy to be obtained, a clear trend in behavior is easily
complexes, there is a fast sub-picosecond kinetic component noted. The [PhSn(Lsal)Cl2] and [PhSn(LBr)Cl2] have triplet
ranging from about 100 fs to 800 fs. We ascribe this to an energies of 16 500 cm−1 and 16 300 cm−1, respectively. The
internal conversion process of the initially formed excited triplet state energies of these complexes are approximately
state. The long time kinetic phase for the complexes ranges 500 cm−1 higher than [PhSn(Ldea)Cl2] and [PhSn(Lnap)Cl2].
from 15 ps (most rapid) to 2500 ps (most slow). Transient
As mentioned above, intersystem crossing yields for popu-
spectra over this time regime have isosbestic points and lation of triplet excited states were obtained by sensitization of
exhibit the loss of narrow peaks or structure to yield broad fea- a triplet with a known excited state absorptivity (anthracene)
tures. This kinetic component corresponds to relaxation of the and comparison with a known standard [Ru(bpy)3]2+ (see
singlet excited state to yield either ground state or triplet. The Experimental). This experiment results in a range of efficien-
lifetimes of this component (Table 3) match well with the cies: 0.72 for [PhSn(LBr)Cl2], 0.59 for [PhSn(Lsal)Cl2], 0.36 for
singlet lifetimes obtained by TCSPC (Table 2). In addition, all [PhSn(Lnap)Cl2], and 0.05 for [PhSn(Ldea)Cl2].
17758 | Dalton Trans., 2014, 43, 17754–17765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Singlet oxygen sensitization
the MV+• ion with strong absorption at 395 nm and a broad
absorption band with a maximum centered at 600 nm.
Quenching rate constants with methyl viologen are 7 × 108
M−1 s−1 for [PhSn(LBr)Cl2], 1 × 109 M−1 s−1 for [PhSn(Lsal)Cl2],
2 × 109 M−1 s−1 for [PhSn(Lnap)Cl2], and ∼1 × 1010 M−1 s−1 for
[PhSn(Ldea)Cl2]. Control experiments of excitation in the
absence of chromophore resulted in no transient absorbance
in the visible. Back electron transfer rate constants for all
chromophores are diffusion limited with both MV and MPTH.
Given the observation of long-lived triplet state formation
and photoredox reactivity for the Sn(IV) complexes, possible
energy and electron transfer photoreactivity for the free
ligands alone was examined. The transient absorption spec-
trum of the HLnap ligand in de-aerated acetonitrile as well as
transient absorption spectra of the ligand in the presence of
9 mM 9-bromoanthracene and 1 mM MPTH are shown in
Fig. S12.† A transient species clearly remains in the ligand
spectrum after 3 milliseconds; however, this species is not
quenched by either MPTH or 9-bromoanthracene.
The triplet states of these complexes are also capable of sensi-
1
tizing the formation of O2. The collection of excited state life-
time measurements in the presence of oxygen allowed for the
determination of a bimolecular quenching rate constants in
CH3CN, which were 2 × 109 M−1 s−1 for all complexes. Further,
the quantum yield of singlet oxygen formation was determined
from the near infrared 1O2 phosphorescence in air saturated
solutions of each of the chromophores and comparing them
to an absorbance matched air saturated [Ru(bpy)3]2+ standard
solution (Φ = 0.57 (ref. 55)). The quantum yields for singlet
oxygen sensitization are 0.43 for [PhSn(Lnap)Cl2], 0.52 for
[PhSn(Lsal)Cl2], and 0.66 for [PhSn(LBr)Cl2]. These values paral-
lel the observed triplet yield efficiencies. No 1O2 phosphor-
escence was observed from [PhSn(Ldea)Cl2] sensitization; this is
likely due to the extremely low triplet yield for the complex.
Photoinduced electron transfer reactions
A key question relating to this class of complexes is the ability
of the chromophores to serve as either excited state electron
donors or acceptors. An estimate of the excited state oxidizing
and reducing ability can be obtained from the energy of the
triplet excited state and the electrochemical potentials
Discussion
Ligand behavior
obtained from cyclic voltammetry. Cyclic voltammograms of Given the similarity of the absorption and fluorescence data of
the complexes indicate that neither oxidation nor reduction is the ligands and the Sn(IV) complexes, the possibility exists that
reversible at sweep rates up to the limit of our instrumentation the observed behavior for the complexes is essentially a simple
(2 V s−1; Fig. S6–S9†). Approximate potentials based upon extension of the behavior of the free ligand. A search of the lit-
anodic and cathodic peak maxima were determined and are erature does not yield any previous microsecond to milli-
given in Table 1. The reduction potentials for all of the com- second transient absorption results for any of the ligands or
plexes are similar and have approximate values of −1.3 V vs. complexes, although steady state and ultrafast experimental
Ag/AgCl. The oxidation potentials for [PhSn(LBr)Cl2], [PhSn- work has been done on HLnap. As a starting point it is useful
(Lsal)Cl2], and [PhSn(Lnap)Cl2] are all similar and occur to discuss the structurally related molecule salicylideneaniline
around 1.3 V. As expected, [PhSn(Ldea)Cl2] is much easier to (SalAn, made from the condensation of salicylaldehyde and
oxidize with a potential of 0.83 V. Given that the triplet excited aniline). An interesting feature of this and HLnap that differs
state energy for all chromophores is approximately 2 eV from the Sn(IV) complexes is contribution of a keto tautomeric
(16 100 cm−1), a crude estimate of the [PhSn(L)Cl2]*/− potential form which can exist in E and Z stereoisomeric forms with
is +0.70 for all chromophores. For excited state oxidation, the respect to the C–N imine linkage as shown in Fig. 5. SalAn
[PhSn(L)Cl2]+/* potential is −0.70 V for [PhSn(LBr)Cl2], [PhSn- does not absorb above 400 nm and its absorption maximum is
(Lsal)Cl2], and [PhSn(Lnap)Cl2] and −1.2 V for [PhSn(Ldea)Cl2] invariant with changes in solvent polarity. The UV absorption
vs. Ag/AgCl, suggesting the complexes can serve as modest bands in the SalAn spectrum are attributed to absorbance of
one-electron acceptors and donors. To provide a test of photo- the ground state enol tautomer. By contrast, there is a clear
redox reactivity, reductive quenching with 10-methyl-pheno- solvent dependence of the ground sate absorption spectrum of
thiazine (MPTH, E0 (+/0) = 0.7 V vs. Ag/AgCl) and oxidative another Schiff base, naphthylideneaniline (NapAn, made from
quenching with methyl viologen (MV, E0 (2+/+) = −0.44 V vs. the condensation of 2-hydroxynaphthaldehyde and aniline);
Ag/AgCl) was carried out in CH3CN solutions.
two absorption bands appear above 400 nm as the solvent
Transient spectra obtained in acetonitrile in the presence of polarity increases. This change in absorbance is ascribed to
MPTH and MV are shown in ESI (Fig. S31 and S32†) for all the the formation of keto tautomer in solution. It is clear that
complexes. For reductive quenching measurements with both the keto and enol form of NapAn are present in polar sol-
MPTH, the absorbance signal between 510 and 550 nm is vents.56 The ground state absorption spectrum of the HLnap
clearly indicative of the MPTH cation radical and is observed ligand features two strong absorption bands above 400 nm
for all the complexes. Rate constants for quenching with and the absorption spectrum is unchanged in solvents of
MPTH are 9 × 109 M−1 s−1 for [PhSn(LBr)Cl2], 7 × 109 M−1 s−1 varying polarity; this is attributed to a stabilization of the keto
for [PhSn(Lsal)Cl2], 6 × 109 M−1 s−1 for [PhSn(Lnap)Cl2], and form through hydrogen bonding with the quinoline nitrogen
1 × 109 M−1 s−1 for [PhSn(Ldea)Cl2]. Quenching with methyl (Fig. 5).57 The photochromic nature of the SalAn excited state
viologen also yielded transient spectra that strongly resemble was initially described in the 1960s in the crystalline state,58
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 17754–17765 | 17759

View Article Online
Paper
Dalton Transactions
S1 state. In addition, formation of the E NH isomer postulated
for the HLnap is highly unlikely in the Sn(IV) complexes.
Excited states and kinetics of the Sn(IV) complexes
As mentioned above, the absorption and luminescence spectra
of HLnap and the [PhSn(Lnap)Cl2] are similar, suggesting a
strong correspondence of the orbitals and states involved. The
HOMO and LUMO obtained from DFT geometry optimized
structures of the Sn(IV) complexes, shown in Fig. S25–S28,†
clearly illustrate that neither the Sn(IV) nor the coordinated
phenyl contribute in any significant way to the HOMO or
LUMO of any of the complexes. Small contributions to the
HOMO from the coordinated Cl are seen for all the complexes
other than [PhSn(Ldea)Cl2].
The ultrafast transient spectral data for the complexes other
than [PhSn(Ldea)Cl2] have common spectral features that
differ from those of the nanosecond transients but evolve over
a few nanoseconds to a final spectrum that closely corresponds
to the longer time transients (>10 ns). Thus, the sub-nano-
second kinetics results in the formation of the long lived
triplet. Two of the three dynamic components of the decays of
the ultrafast transients can be classified: the subpicosecond
decay is likely due to vibrational relaxation of the singlet mani-
fold and the longest decay corresponds to relaxation of the
singlet excited state (to triplet and ground state). The inter-
mediate component varies from 70 ps for [PhSn(Lnap)Cl2] to
1 ps for [PhSn(Ldea)Cl2], with the other two complexes closer
to [PhSn(Lnap)Cl2]. The time for all but the Ldea complex is
much too long for solvation of the excited state and the expla-
nation for such a long process is not clear. It is possible that
the initial excitation may lead to population of two different
singlet states that decay with vastly different rate constants. In
support of this is the observation that the ultrafast spectrum
does not really change during the 40–100 ps time period over
which the intermediate component decays.
The luminescence lifetime values for the complexes
obtained from TCSPC measurements are in good agreement
with the long component of the decay of the ultrafast transient
absorption spectrum for all chromophores. These values
reflect the decay of the initially formed singlet excited state by
radiative, non-radiative, and intersystem crossing processes.
Using the experimentally obtained lifetime value, quantum
yield of intersystem crossing, and the quantum yield of fluore-
scence, the rate constants of all three relaxation processes can
be determined.
Fig. 5 Structural representations of possible tautomers and stereo-
isomers of SalAn and HLnap.
low temperature matrices,59 and solution.59,60 These initial
reports assign this photochromic state as a ground state E NH
structure (Fig. 5). Subsequent literature reports have examined
the
kinetic
evolution
of
the
excited
state
of
salicylideneaniline61–67 in fluid solution. Excitation of the enol
form of SalAn results in sub-picosecond intramolecular proton
transfer to form the fluorescent, keto, form followed by relax-
ation to a ground state species with either a cis-zwitterionic
(iminium ion/naphthoate) or the isomerized E NH structure.
This latter isomeric species has a lifetime of 400 μs in
acetonitrile.
The absorption, luminescence and femtosecond time-
resolved absorption spectrum of the HLnap ligand has also
been reported.68,69 The ultrafast spectroscopic behavior con-
sists of three spectral features of note: (1) two well defined
bleaching bands corresponding to both of the local maxima in
the ground state absorption spectrum at 450 nm and 475 nm
that decay via a multi-exponential path that includes a very
long lived component (>5 ns), (2) an intense, broad absorption
between 500–620 nm, containing two peaks at 525 nm and
580 nm that disappear over 30 ps, and (3) stimulated emission
at wavelengths to the red of 620 nm that decays with a rate con-
stant of 4 ps. The authors conclude that the behavior can be
explained by processes that involve excitation of the Z NH
(Fig. 5) ground state followed by decay of the initially formed
vibronic state (relatively high in the manifold) to a mix of the
S1 and S2 states of the Z NH conformer over 700 fs. Stimulated
emission of the S1 state occurs over the first 4 ps, during
which time a long lived state is formed; meanwhile the S2 state
decays to the ground state with a lifetime of 13 ps. The long
lived state (>5 ns) is attributed by the authors to formation of
either a Z NH triplet or possibly the E NH isomer. The key pos-
tulate relating to the two components of the ultrafast behavior
is the formation and relatively slow (13 ps) decay out of the S2
state directly back to the ground state without relaxation to the
The rates of intersystem crossing vary from 2.6 × 108 s−1 for
[PhSn(Lnap)Cl2] to 9.0 × 108 s−1 for [PhSn(LBr)Cl2]. The inter-
system crossing rate of [PhSn(LBr)Cl2] is 3 times greater than
that of [PhSn(Lnap)Cl2] and [PhSn(Ldea)Cl2], and only slightly
greater than that of [PhSn(Lsal)Cl2]. This implies that presence
of a second heavy atom (Br) does not have a large effect on the
observed rate of intersystem crossing. It is also important to
note that the intersystem crossing rates for chromophores of
this type are four orders of magnitude slower than those of
well-known ruthenium polypyridyl chromophores. The radia-
tive decay rates of the singlet state only differ by a factor of two
17760 | Dalton Trans., 2014, 43, 17754–17765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
across the series of four chromophores from 1.0 × 108 s−1 for complex and quencher (MPTH). Since MPTH has no visible
[PhSn(Lnap)Cl2] to 5.0 × 107 s−1 for [PhSn(Lsal)Cl2] and are in absorbance and the absorption spectra of the starting complex
the range of radiative decay rate constants of organic chromo- and the MPTH cation radical are known, the only remaining
phores with allowed π–π* transitions.
unknown is the spectrum of the one-electron reduced form of
The largest difference in rate constants of the singlet the Sn complex. The voltammetric data clearly indicate that
excited states is observed for the non-radiative decay rate of reduction of the complex is irreversible (probably ErCi
[PhSn(Ldea)Cl2], which is an order of magnitude faster than process), thus we were unable to obtain the spectrum of the
the non-radiative decay rates for all of the other molecules in reduced complex by spectroelectrochemistry. However, it
the series. It is perhaps not surprising that the ultrafast should be noted that the transient spectra obtained following
absorption spectrum is fundamentally different from the other quenching with MPTH (Fig. S31†) strongly resemble spectra
three complexes, lacking completely the absorption feature generated from the difference of the MPTH cation radical and
that appears immediately upon excitation between 350 and the Sn complex ground state absorption. The implication is
400 nm and remains. The solvent dependence of the absorp- the that the absorbance of the reduced Sn complex in the
tion and luminescence behavior indicates that the excited visible is minimal, possibly suggesting reduction of the metal
singlet state has significant charge transfer character and (to Sn(III)) rather than ligand localized reduction to make the
relaxation of this excited state is the fastest of the four com- ligand anion radical, which would very likely have strong
plexes (knr = 6 × 109 s−1). The effect of this is reflected in a rela- visible absorption. However, the DFT results (Fig. S25–S28†)
tively low quantum yield for fluorescence and a very low clearly suggest that the LUMO is ligand localized. The back
intersystem crossing efficiency.
reaction of the reduced complex with the MPTH+• is observed,
Examination of quenching of the long lived transients with but over the course of the acquisition of a transient absorption
a series of triplet quenchers (Fig. S19 and 20†) indicated a sys- spectrum the MPTH+• begins to accumulate in solution, indi-
tematic dependence of the quenching rate constant on the cating decomposition of the reduced Sn complex occurs on
triplet energy of the quencher. The data were fit assuming a the hundreds of microseconds time scale.
reversible energy transfer model52–54 and the triplet energies
A similar behavioral trend was observed upon quenching
(Table 2) obtained indicate a relatively small range for the com- with an oxidative quencher, MV: the transient spectrum of the
plexes. Interestingly, the energy gaps between the singlet states long lived species resembled the difference between the MV
(taken from the fluorescence maxima) and the triplet states are cation radical absorbance and the ground state absorption of
3800 cm−1, 2500 cm−1, 2300 cm−1 and 1100 cm−1 for the the complex. All the complexes served as effective one-electron
Lnap, Lsal, LBr and the Ldea complexes, respectively. For the donors, with the Ldea complex being the most effective. While
first three the relative efficiencies for populating the triplet state we have no direct evidence, it is likely the oxidized complex
increase with decreasing S1–T1 gap, but the Ldea complex has a reflects the ligand cation radical (see the HOMO of DFT in
very low ηisc despite an apparently small S1–T1 gap. However, Fig. S25–S28†). As above, back electron transfer occurs, but
the observed fluorescence originates from a state with a large some decomposition of the oxidized Sn complex occurs on the
degree of charge transfer character and the triplet may well rep- time scale of the back reaction and reduced viologen gradually
resent a state that is more similar to the other complexes in accumulates in solution.
electron distribution. At this point, we have no data that relates
to the degree of charge transfer character in the triplet state of
Summary
the Ldea complex, but the low intersystem crossing efficiency The synthesis and characterization of a series of four fluore-
observed, despite the small apparent S1–T1 gap, clearly indicates scent Sn(IV) Schiff base complexes with reasonably high yields
a fundamental difference between this complex and the others.
of triplet state formation is described. The complexes absorb
The triplet energies of all the complexes are well above that in the visible (λmax = 420 to 462 nm) and their optical pro-
of singlet oxygen and near-infrared emission spectra of aerated perties are easily tunable without laborious synthetic elab-
solutions of all the complexes other than the Ldea complex oration. The complexes are fluorescent, with low emission
clearly showed the signature 1270 nm emission of singlet yields, and also form nonluminescent triplet excited states with
oxygen. Yields for oxygen formation were between 40 and 60% formation yields ranging from 5% to 70%. Ultrafast transient
in acetonitrile. The lack of sensitized singlet oxygen emission spectral studies provide lifetimes for the singlet excited states,
in the Ldea complex is attributed to the very low yield of triplet with decays that correlate well with luminescence followed by
states in this complex.
TCSPC. The triplets were observed and kinetically evaluated
Electron transfer quenching experiments with an electron using nanosecond transient absorption spectroscopy. The life-
donor and electron acceptor with clearly identifiable radical times of the triplet excited states are on the order of
ion absorbance in the visible (MPTH+• and MV+•) served to 500 μs–10 ms in PMMA matrices. Triplet–triplet annihilation
illustrate the potential of the complexes to serve as both processes occur at near diffusion limited rates in solutions of
excited state oxidants and reductants. For reductive quenching the chromophores. The triplet state energies were estimated
with MPTH, the transient absorption spectrum following elec- from the free energy dependence of energy transfer rate con-
tron transfer quenching should reflect the sum of the radical stants for reaction with a series of organic triplet acceptors.
ion species absorption less the absorption of the ground state All the complexes other than the Ldea complex were good
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 17754–17765 | 17761

View Article Online
Paper
Dalton Transactions
sensitizers for singlet oxygen emission. The photoexcited com- d, 4.08 Hz), 8.32 (1.65H, d, 7.48 Hz, 60.1 Hz), 8.14 (1.15H, d,
plexes react with both electron donors and acceptors in solu- 7.60 Hz), 7.97 (2H, m), 7.74 (2H, dd, 3.4 Hz, 7.08 Hz), 7.57 (3H,
tion, although decomposition of both the complex radical m), 7.37 (1H, d, 9.24 Hz), 6.48 (1H, d, 9.20 Hz), 5.96 (1H, s),
anion and radical cation competes to some degree with back 3.47 (4H, q, 7.12), 1.19 (6H, t, 7.08).
electron transfer. These results demonstrate the potential for
PhSn(Lsal)Cl2
the development of photosensitizers based on main group
elements with high spin orbit coupling constants.
90 μl salicylaldehyde (8.2 × 10−4 mol), 121.2 mg (8.4 × 10−4
mol) 8-aminoquinoline, and 288.2 mg (8.4 × 10−4 mol)
Ph2SnCl2 were used in the synthesis. 301.5 mg (71%) of yellow
solid was obtained. 1H NMR (CD3CN) δ ppm: 9.19 (1H, s,
44.0 Hz), 8.82 (1H, d, 8.04 Hz), 8.50 (1H, d, 4.24 Hz), 8.41 (1H,
d, 7.80 Hz), 8.29 (0.14H, 1.70H, 0.14H, d, 7.52 Hz, 59.5 Hz),
8.20 (1H, d, 8.28 Hz), 8.02 (1H, t, 8.12), 7.81 (1H, dd, 4.96 Hz,
8.28 Hz), 7.67 (1H, d, 7.12 Hz), 7.60 (5H, m), 6.98 (1H, t,
7.44 Hz), 6.84 (1H, d, 8.48 Hz).
Experimental details
Syntheses
Ph2SnCl2, 8-aminoquinoline, 2-hydroxy-1-naphthaldehyde,
salicylaldehyde, 5-bromosalicylaldehyde, and 4-diethylamino-
salicylaldehyde were all purchased from Sigma-Aldrich and
used as received. Acetonitrile, absolute ethanol, diethylether,
hexanes and toluene were obtained from Pharmco-AAPER and
used as received.
PhSn(LBr)Cl2
106.7 mg 5-bromosalicylaldehyde (5.3 × 10−4 mol), 74.2 mg
(5.1 × 10−4 mol) 8-aminoquinoline, and 177.3 mg (5.2 × 10−4
mol) Ph2SnCl2 were used in the synthesis. 235.6 mg (78%) of
2-Hydroxy-1-naphthaldehydene-8-aminoquinoline (HLnap)
1
The ligand 2-hydroxy-1-naphthaldehydene-8-aminoquinoline
was synthesized as previously reported.50 1 g (6.9 mmol) 8-amino-
quinoline and 1.19 g (6.9 mmol) 2-hydroxy-1-naphthaldehyde
were added to 40 ml absolute ethanol and allowed to stir for
12 hours. 2-Hydroxy-1-naphthaldehydene-8-aminoquinoline
precipitates as the reaction progresses. The solid was isolated
on fritted glass and washed with 25 ml cold absolute ethanol.
yellow solid was obtained. H NMR (CD3CN) δ ppm: 9.12 (1H,
s, 43.6 Hz), 8.83 (1H, d, 7.88 Hz), 8.50 (1H, d, 4.36 Hz), 8.39
(1.19H, d, 7.88 Hz), 8.27 (1.68H, d, 6.12 Hz, 62.0 Hz), 8.23 (1H,
d, 9.60 Hz), 8.11 (0.12H), 8.03 (1H, t, 7.84 Hz), 7.81 (2H, m),
7.60 (4H, m), 6.79 (1H, d, 8.76 Hz).
Spectroscopy
1
1H and 2D-COSY NMR spectra were recorded on a Varian
400 MHz NMR. UV-vis absorption spectra were obtained on a
Hewlett-Packard 8452A diode array spectrophotometer. Photo-
luminescence spectra were obtained using a PTi Quantamaster
1.92 g (93% yield) of orange solid obtained. H NMR (CDCl3)
δ ppm: 9.29 (1H, d, 11.1 Hz), 9.08 (1H, d, 2.8 Hz), 8.20 (d,
1H, 7.28 Hz), 8.00 (1H, d, 8.28 Hz), 7.77 (1H, d, 7.32 Hz), 7.68
(2H, m), 7.60 (2H, m), 7.51 (2H, m), 7.28 (1H, d, 7.48 Hz), 6.91
(1H, d, 9.44 Hz).
spectrophotometer fit with
detector.
a Hammamatsu R928 PMT
PhSn(Lnap)Cl2
66.3 mg (4.6 × 10−4 mol) 8-aminoquinoline and 99.4 mg
(4.6 × 10−4 mol) 2-hydroxy-1-naphthaldehyde were added to
15 ml toluene and refluxed for 15 h. The reaction was removed
from heat and allowed to cool to room temperature. 160 mg
(4.6 × 10−4 mol) Ph2SnCl2 was added to the reaction vessel.
The toluene was removed via rotary evaporation and the result-
ing orange precipitate was collected on fritted glass, washed
with hexanes (30 ml) and diethylether (15 ml), and dried
in vacuo. 229.5 mg (89% yield) of orange solid was obtained.
The synthesis of the salicylaldehyde containing derivatives
were carried out in identical fashion. 1H NMR (CD3CN) δ ppm:
9.90 (1H, s, 48.7 Hz), 8.85 (1H, d, 7.88 Hz), 8.55 (1H, d,
4.16 Hz), 8.49 (2.17H, m), 8.33 (1.65, d, 6.16 Hz), 8.17 (1.17H,
d, 8.24), 8.08 (2H, m), 7.87 (2H, m), 7.69 (1H, t, 7.84 Hz), 7.62
(3H, m), 7.48 (1H, t, 7.44 Hz), 7.01 (1H, d, 9.16).
Nanosecond time-resolved transient absorption
Nanosecond transient absorption measurements were per-
formed using a Qunatel Brilliant B Q-switched Nd:YAG laser-
pumped OPO (Opotek) as the pump source at a right angle to
the analyzing light source. Excitation pulses were <5 ns. An
Applied Photophysics LKS.60 laser flash photolysis spectro-
meter was used for detection; the instrument is equipped with
a 150 W pulsed Xe arc lamp as the analyzing light source, a
single grating monochromator (Applied Photophysics 0.25 m)
after the sample, and PMT detection (Hamamatsu R928). The
output was recorded on an Agilent Infinium transient digitizer.
Decays and spectra were acquired and analyzed with Applied
Photophysics LKS.60 software.
The quantum yield of triplet state formation was deter-
mined for the series of complexes using a [Ru(bpy)3]2+ stan-
dard and photosensitization of anthracene. The change in
optical density of a sample of [Ru(bpy)3]2+ at 370 nm resulting
PhSn(Ldea)Cl2
74.3 mg (5.2
×
10−4 mol) 8-aminoquinoline, 99.4 mg from a 10 mJ pulse of 450 nm laser light was determined.
(5.2 × 10−4 mol) 4-diethylaminosalicylaldehyde, and 189.5 mg Using the excited state molar absorptivity of [Ru(bpy)3]2+
(5.5 × 10−4 mol) Ph2SnCl2 were used in the synthesis. 246.8 mg at 370 nm,70 the concentration of triplet excited states can be
(81% yield) of red solid was obtained. 1H NMR (CD3CN) calculated. Samples of each chromophore were absorbance
δ ppm: 8.74 (1H, d, 7.84 Hz), 8.70 (1H, s, 50.5 Hz), 8.42 (1.14H, matched with the [Ru(bpy)3]2+ standard and quenched with a
17762 | Dalton Trans., 2014, 43, 17754–17765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
0.01 M solution of anthracene in acetonitrile. The resulting solutions containing 0.1 M TBAPF6. All samples were bubble
change in absorbance at 420 nm71 allows for the determi- degassed with argon for 10 minutes prior to collection. The
nation of the concentration of anthracene triplet formed. The presented sweep segments presented in ESI (Fig. S6–9†) were
concentration of excited anthracene is equal to the number of collected at a scan rate of 100 mV s−1
Sn complex triplet excited states created in the laser pulse. The
.
Computational details
ratio of Sn triplet excited states created to the number of
[Ru(bpy)3]2+ triplet excited states created gives the quantum
yield of triplet state formation.
DFT calculations were performed using the Gaussian 09 soft-
ware package.72 All calculations were performed at the B3LYP
level of theory, and utilized the LANL2DZ basis set with the
electron core potentials of Hay and Wadt for Sn73 and the
6-31G(d,p) basis set for all other atoms. No imaginary vibration
frequencies were observed, confirming the validity of the geo-
metry optimized structures.
Femtosecond time-resolved transient absorption
Femtosecond transient absorption measurements were col-
lected on an Ultrafast Systems HELIOS transient absorption
spectrometer. A Spectra Physics Solstice laser, a one-box regen-
erative amplifier containing a Mai Tai femtosecond oscillator
and Empower pump laser, was employed to produce 800 nm
pulses at a repetition rate of 1 kHz at 3.5 W average power and
a pulse width of <100 fs. From this unit, the beam is split
(50 : 50) with one beam directed to an optical parametric
amplifier (TOPAS, Light Conversion) and the other to the
HELIOS spectrometer (He-vis-3200) to create the pump
(TOPAS) and probe (HELIOS) sources, respectively. The 800 nm
probe beam passed through a CaF2 plate to generate a white
light continuum (∼330–700 nm). The spectrum was integrated
for 2 s each scan. The pump and probe beams were directed to
a 2 mm path length cuvette containing the sample where they
were spatially overlapped. The solution was vigorously stirred
in the 2 mm path length cuvette during data collection. Transi-
ent absorption data were corrected by subtracting spectral
background features that persisted from the previous pulse
and appeared prepulse as well as applying chirp correction
and T0 corrections using Surface Xplorer Pro 1.1.5 software
(Ultrafast Systems).
Acknowledgements
The authors wish to thank the U.S. Department of Energy,
Office of Chemical Sciences (grant DE-FG-02-96ER14617) for
support of this research. JR acknowledges support of the
National Science Foundation under Grants CHE 0809699,
0947031, and 1112250. TG wishes to thank the Louisiana
Board of Regents for a Graduate Fellowship.
References
1 S. L. Murov, I. Carmichael and G. L. Hug, Handbook of
Photochemistry, New York (NY), 1993.
2 K. Oldenburg, A. Vogler and O. Horváth, Inorg. Chim. Acta,
1997, 257, 149.
3 K. Oldenburg, A. Vogler, I. Miko and O. Horvath, Inorg.
Chim. Acta, 1996, 248, 107.
4 K. Oldenburg and A. Vogler, Z. Naturforsch., B: Chem. Sci.,
1993, 48, 1519.
5 H. Nikol, A. Becht and A. Vogler, Inorg. Chem., 1992, 31,
3277.
6 A. Volger and H. Nikol, Pure Appl. Chem., 1992, 64, 1311.
7 H. Kunkely and A. Vogler, Chem. Phys. Lett., 1991, 187, 609.
8 A. Vogler, A. Paukner and H. Kunkely, Coord. Chem. Rev.,
1990, 97, 285.
9 H. Kim, W. Kim, Y. Mackeyev, G. Lee, H. Kim,
T. Tachikawa, S. Hong, S. Lee, J. Kim, L. J. Wilson,
T. Majima, P. J. J. Alvarez, W. Choi and J. Lee, Environ. Sci.
Technol., 2012, 46, 9606.
Time correlated single photon counting (TCSPC)
Luminescence lifetimes were obtained using the time corre-
lated single photon counting technique. Each sample was
excited using an IBH NanoLED pulsed diode laser source
(441 nm excitation, 200 ps pulse width). The emitted light was
collected at right angle to the excitation pulse and collected by
an IBH Model TBX-04 cooled photomultiplier (PMT) detector.
The output of the PMT served as the input for the stop
channel of the time to amplitude converter (TAC, Tennelec
TC-863). Start pulses for the TAC were obtained from the syn-
chronous TTL output of the NanoLED laser source. The output
from the TAC was directed to the multichannel analyzer
(Ortec, Easy MCA) where the signal was accumulated. Laser 10 W. Kim, T. Tachikawa, T. Majima, C. Li, H. Kim and
light scattered from the nonluminescent solution was used to W. Choi, Energy Environ. Sci., 2010, 3, 1789.
accumulate instrument response profiles. The data were ana- 11 M. Martinez-Diaz, l. T. de and T. Torres, Chem. Commun.,
lyzed using the deconvolution routine in Origin 9.
2010, 46, 7090.
12 M. T. Indelli, C. Chiorboli, M. Ghirotti, M. Orlandi,
F. Scandola, H. J. Kim and H. Kim, J. Phys. Chem. B, 2010,
114, 14273.
Cyclic voltammetry
Cyclic voltammetry was performed on a CH Instruments
CH1730A Electrochemical Analyzer. The working electrode was 13 I. Seotsanyana-Mokhosi, T. Kresfelder, H. Abrahamse and
a glassy-carbon electrode, the counter electrode was a Pt wire, T. Nyokong, J. Photochem. Photobiol., B, 2006, 83, 55.
and the reference electrode was a Ag/AgPF6 in acetonitrile elec- 14 B. P. Rand, J. Xue, F. Yang and S. R. Forrest, Appl. Phys.
trode. All cyclic voltammetry data was collected in acetonitrile
Lett., 2005, 87, 233508.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 17754–17765 | 17763

View Article Online
Paper
Dalton Transactions
15 S. Wang, I. Tabata, K. Hisada and T. Hori, J. Porphyrins 43 T. Matsubara, Y. Saito, T. Yamakawa and S. Shinoda, J. Mol.
Phthalocyanines, 2003, 7, 199. Catal., 1991, 67, 175.
16 S. Wang, K. Enomoto, S. Tanizaki, I. Tabata, K. Hisada, 44 K. Nomura, Y. Saito and S. Shinda, J. Mol. Catal., 1989, 50,
T. Hori and S. Okubayashi, Sen’i Gakkaishi, 2003, 59, 48. 303.
17 S. Okubayashi, S. Ariga, H. Shosenji and T. Hori, Sen’i 45 R. R. Andrea, W. G. J. de Lange, D. J. Stufkens and
Gakkaishi, 1997, 53, 431. A. Oskam, Inorg. Chem., 1989, 28, 318.
18 C. Grewer, G. Schermann, R. Schmidt, A. Voelcker, 46 M. S. Holt, W. L. Wilson and J. H. Nelson, Chem. Rev.,
H. D. Brauer, A. Meier and F. P. Montforts, J. Photochem.
Photobiol., B, 1991, 11, 285.
1989, 89, 11.
47 J. D. Donaldson, Prog. Inorg. Chem., 1968, 8, 287.
19 A. R. Morgan, L. S. Cheng, D. Skalkos and G. M. Garbo, 48 K. Takano, M. Takahashi, T. Fukushima, M. Takezaki,
Photochem. Photobiol., 1990, 52, 987.
20 W. Szulbinski and J. W. Strojek, Inorg. Chim. Acta, 1986,
118, 91.
T. Tominaga, H. Akashi, H. Takagi and T. Shibahra, Bull.
Chem. Soc. Jpn., 2012, 85, 1210.
49 K. Takano and T. Shibahara, Chem. Lett., 2008, 37, 70.
21 J. Fuhrhop, W. Krüger and H. E. David, Liebigs Ann. Chem., 50 C. Jiang, J. Wang and F. He, Anal. Chim. Acta, 2001, 439,
1983, 1983, 204. 307.
22 P. Sayer, M. Gouterman and C. R. Connell, Acc. Chem. Res., 51 A. Bondi, J. Phys. Chem., 1964, 68, 441.
1982, 15, 73.
52 V. Balzani, F. Bolletta and F. Scandola, J. Am. Chem. Soc.,
1980, 102, 2152.
23 Y. Harel and J. Manassen, J. Am. Chem. Soc., 1977, 99, 5817.
24 M. Gouterman, F. P. Schwarz, P. D. Smith and D. Dolphin, 53 W. G. Herkstroeter and G. S. Hammond, J. Am. Chem. Soc.,
J. Chem. Phys., 1973, 59, 676. 1966, 88, 4769.
25 D. G. Whitten, J. C. N. Yau and F. A. Carroll, J. Am. Chem. 54 K. Sandros, Acta Chem. Scand., 1964, 18, 2355.
Soc., 1971, 93, 2291.
26 R. S. Becker and J. B. Allison, J. Phys. Chem., 1963, 67,
2669.
27 M. Ohtani, P. V. Kamat and S. Fukuzumi, J. Mater. Chem.,
2010, 20, 582.
28 J. Fortage, E. Goeransson, E. Blart, H. Becker,
55 A. A. Abdel-Shafi, D. R. Worrall and A. Y. Ershov, Dalton
Trans., 2004, 30.
56 S. R. Salman, S. K. Kanber and L. K. Arslian, Spectrosc.
Lett., 1990, 23, 870.
57 K. A. Abbas, S. R. Salman, S. M. Kana’n and Z. A. Fataftah,
Can. J. Appl. Spectrosc., 1996, 41, 119.
L. Hammarstroem and F. Odobel, Chem. Commun., 2007, 58 M. D. Cohen and G. M. J. Schmidt, J. Phys. Chem., 1962, 66,
4629. 2442.
29 C. Du, J. Yu, J. Huang and Y. Jiang, Energy Procedia, 2011, 59 M. D. Cohen, Y. Hirshberg and G. M. J. Schmidt, J. Chem.
12, 519. Soc., 1964, 2051.
30 M. Idowu and T. Nyokong, J. Photochem. Photobiol., A, 2008, 60 M. Ottolenghi and D. S. McClure, J. Chem. Phys., 1967, 46,
199, 282.
4620.
31 T. Nyokong, J. Porphyrins Phthalocyanines, 2008, 12, 1005.
32 B. Agboola, K. I. Ozoemena and T. Nyokong, J. Mol. Catal.
A: Chem., 2006, 248, 84.
33 N. Nensala, A. Nzimande and T. Nyokong, J. Photochem.
Photobiol., A, 1996, 98, 129.
34 I. Aviv-Harel and Z. Gross, Coord. Chem. Rev., 2011, 255,
717.
35 I. Aviv and Z. Gross, Chem. Commun., 2007, 1987.
61 R. Nakagaki, T. Kobayashi, J. Nakamura and S. Nagakura,
Bull. Chem. Soc. Jpn., 1977, 50, 1909.
62 P. F. Barbara, P. M. Rentzepis and L. E. Brus, J. Am. Chem.
Soc., 1980, 102, 2786.
63 R. S. Becker, C. Lenoble and A. Zein, J. Phys. Chem., 1987,
91, 3509.
64 W. Tuberville and P. K. Dutta, J. Phys. Chem., 1990, 94,
4060.
36 D. Walker, S. Chappel, A. Mahammed, B. S. Brunschwig, 65 T. Yuzawa, H. Takahashi and H. Hamaguchi, Chem. Phys.
J. R. Winkler, H. B. Gray, A. Zaban and Z. Gross, J. Por-
phyrins Phthalocyanines, 2006, 10, 1259.
37 B. Ventura, A. D. Esposti, B. Koszarna, D. T. Gryko and
L. Flamigni, New J. Chem., 2005, 29, 1559.
38 V. Pawlowski, H. Kunkely and A. Vogler, Inorg. Chim. Acta,
1995, 234, 55.
Lett., 1993, 202, 221.
66 K. Kownacki, A. Mordzinski, R. Wilbrandt and
A. Grabowska, Chem. Phys. Lett., 1994, 227, 270.
67 S. Mitra and N. Tamai, Chem. Phys. Lett., 1998, 282,
391.
68 P. Fita, E. Luzina, T. Dziembowska, D. Kopeć,
P. Piątkowski, C. Radzewicz and A. Grabowska, Chem. Phys.
Lett., 2005, 416, 305.
39 V. Pawlowski, H. Kunkely and A. Vogler, Coord. Chem. Rev.,
1994, 132, 23.
40 K. Makita, K. Nomura and Y. Saito, J. Mol. Catal., 1994, 89, 69 P. Fita, E. Luzina, T. Dziembowska, C. Radzewicz and
143. A. Grabowska, J. Chem. Phys., 2006, 125, 184508.
41 T. Matsubara, Y. Saito, T. Yamakawa and S. Shinoda, J. Mol. 70 U. Lachish, P. P. Infelta and M. Gratzel, Chem. Phys. Lett.,
Catal., 1993, 79, 29. 1979, 62, 317.
42 V. Pawlowski, H. Kunkely and A. Vogler, J. Am. Chem. Soc., 71 I. Carmichael and G. L. Hug, J. Phys. Chem. Ref. Data, 1986,
1993, 115, 7029.
15, 1.
17764 | Dalton Trans., 2014, 43, 17754–17765
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
72 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford,
CT, 2009.
F. Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, 73 W. R. Wadt and P. J. Hay, J. Chem. Phys., 1985, 82, 284.
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, 74 D. M. Collins, W. R. Scheidt and J. L. Hoard, J. Am. Chem.
K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar,
Soc., 1972, 94, 6689.
J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, 75 J. V. Caspar and T. J. Meyer, J. Am. Chem. Soc., 1983, 105,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
5583.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 17754–17765 | 17765