Ultrafast Electron Transfers in Organometallic Supramolecular Assemblies Built with a NIR-Fluorescent Tetrabenzoporphyrine Dye and the Unsaturated Cluster Pd3(dppm)3(CO)2+

Article abstract of DOI:10.1021/acs.organomet.6b00050

The sodium 9,18,27,36-tetra-(4-carboxyphenylethynyl)tetrabenzoporphyrinatozinc(II) (TCPEBP) and sodium 5,10,15,20-tetra-(4-carboxyphenylethynyl)porphyrinatozinc(II) (TCPEP, for comparison purposes) salts were prepared to investigate the ionic driven host-guest assemblies made with the unsaturated redox-active cluster Pd₃(dppm)₃(CO)²⁺ ([Pd₃²⁺], dppm = Ph₂PCH₂PPh₂ as a PF₆^- salt). Nonemissive dye?[Pd₃²⁺]_x assemblies (x = 1-4) are formed in methanol with K_1x (binding constants) values of 83 200 (TCPEBP) and 70 400 M^-1 (TCPEP; average values extracted from graphical methods (Benesi-Hildebrand, Scott, and Scatchard), matching those obtained from fluorescence quenching experiments (static model)). These values are consistent with the more electron rich TCPEBP dye. This conclusion is corroborated by electrochemical data, which indicate a lower oxidation potential of the TCPEBP dye (+0.46 V) vs TCPEP (+0.70 V vs SCE) and by shorter calculated average Pd?O distances (DFT (B3LYP): 3.259 vs 3.438 ?, respectively). Using the position of the 0-0 component of the Q-bands and the electrochemical data, the excited-state driving forces for dye??[Pd₃²⁺]_x → dye^+??[Pd₃^+?][Pd₃²⁺]_x-1 are estimated for TCPEBP (+1.22 V vs SCE) and TCPEP (1.08 V vs SCE). The time scale for this process occurs within the laser pulse (fwhm <75-110 fs) during the measurements of the femtosecond transient absorption spectra. Conversely, the back electron transfers (dye^+??[Pd₃^+?][Pd₃²⁺]_x-1 → dye?[Pd₃²⁺]_x) occur well within 1 ps (respectively 650 and 170 fs for TCPEBP and TCPEP). Arguments are provided that the reorganization energy governs this difference.

Full text of DOI:10.1021/acs.organomet.6b00050

Article
pubs.acs.org/Organometallics
Ultrafast Electron Transfers in Organometallic Supramolecular
Assemblies Built with a NIR-Fluorescent Tetrabenzoporphyrine Dye
and the Unsaturated Cluster Pd3(dppm)3(CO)2+
Peng Luo,† Paul-Ludovic Karsenti,† Gessie Brisard,† Benoit Marsan,*,‡ and Pierre D. Harvey*,†
†Dep
́
artement de Chimie, Universite
artement de Chimie, Universite
́
́
de Sherbrooke, Sherbrooke, QC J1K 2R1, Canada
du Quebec a Montreal, Montreal, QC H2X 2J6, Canada
‡Dep
́
́
̀
́
́
S
* Supporting Information
ABSTRACT: The sodium 9,18,27,36-tetra-(4-carboxyphenyl-
ethynyl)tetrabenzoporphyrinatozinc(II) (TCPEBP) and sodium
5,10,15,20-tetra-(4-carboxyphenylethynyl)porphyrinatozinc(II)
(TCPEP, for comparison purposes) salts were prepared to
investigate the ionic driven host−guest assemblies made with
the unsaturated redox-active cluster Pd3(dppm)3(CO)2+ ([Pd32+],
−
dppm = Ph2PCH2PPh2 as a PF6 salt). Nonemissive dye···
[Pd32+]x assemblies (x = 1−4) are formed in methanol with K1x
(binding constants) values of 83 200 (TCPEBP) and 70 400 M−1
(TCPEP; average values extracted from graphical methods
(Benesi−Hildebrand, Scott, and Scatchard), matching those obtained from fluorescence quenching experiments (static
model)). These values are consistent with the more electron rich TCPEBP dye. This conclusion is corroborated by
electrochemical data, which indicate a lower oxidation potential of the TCPEBP dye (+0.46 V) vs TCPEP (+0.70 V vs SCE) and
by shorter calculated average Pd···O distances (DFT (B3LYP): 3.259 vs 3.438 Å, respectively). Using the position of the 0−0
component of the Q-bands and the electrochemical data, the excited-state driving forces for dye*···[Pd32+]x → dye+•···
2+
[Pd3+•][Pd3
]x−1 are estimated for TCPEBP (+1.22 V vs SCE) and TCPEP (1.08 V vs SCE). The time scale for this process
occurs within the laser pulse (fwhm <75−110 fs) during the measurements of the femtosecond transient absorption spectra.
2+
Conversely, the back electron transfers (dye+•···[Pd3+•][Pd3
]
x−1 → dye···[Pd32+]x) occur well within 1 ps (respectively 650 and
170 fs for TCPEBP and TCPEP). Arguments are provided that the reorganization energy governs this difference.
study on the photoinduced electron transfer. Moreover, the
title compound TCPEP (Chart 1) was previously reported in
the context of preparing MOFs (metal−organic frameworks;
note that in this case the 0−0 peak is reported at 687 nm in
DMF), but no photophysical study was performed.6 Another
approach to extend the π-system, and therefore move the
electronic spectra to the red region, is to fuse benzene rings
onto the porphyrin chromophore.7 By doing so, the 0−0
component moves to 737 nm (in THF). As the absorption
band desirably shifts to the NIR, the singlet-state energy
decreases proportionally, thus affecting the excited-state driving
force for electron transfer.
INTRODUCTION
■
The rates for photoinduced electron transfers from dyads
composed of carboxyphenyl-containing porphyrin dyes linked
to the surface of TiO2 nanoparticles are very fast (<100 fs up to
420 ps).1 This property bears an obvious incidence of the
overall efficiency of dye-sensitized solar cells (DSSCs).2
Knowing that thermodynamic parameters control the size of
these rates, the latter’s are unavoidably bound to their
molecular structures (i.e., singlet-state manifold energies,
oxidation potentials, and excited-state driving forces). As
more recently pointed out, absorbing more photons,
particularly exploiting the near-IR region (NIR; 750−950
nm), is another parameter sought in solar cells. Obviously,
shifting the absorption bands to the NIR is easily achieved by
extending the π-system of the chromophore. In comparison
with ZnTPP (5,10,15,20-tetraphenylporphyrinatozinc(II);
λmax(Q-band) ≈ 580 nm), a particularly efficient way to expand
the π-system is to introduce ethynyl groups between the Cmeso
and the phenyl groups.3 In this case (i.e., 5,10,15,20-tetra(4-
phenylethynyl)porphyrinatozinc(II)), the Q-band is shifted to
nearly ∼680−700 nm.3 Although this zinc(II)porphyrin-
containing motif is relatively common,4 and several photo-
physical and electrochemical investigations exist,5 there is no
In the context of this work, we recently reported assemblies
of the type dye···[Pd32+]x (dye = MCP, 5-(4-carboxylphenyl)-
10,15,20-tristolyl(porphyrinato)zinc(II), DCP, 5,15-bis(4-car-
boxylphenyl)-15,20-bistolyl(porphyrinato)zinc(II)), which ex-
hibit smaller binding constants (K1x = 19 300 (MCP) and
22 000 M−1 (DCP, x = 1, 2)).8 The choice of
Pd3(dppm)3(CO)2+ ([Pd32+], dppm = Ph2PCH2PPh2 as a
PF6 salt)9 as the electron acceptor is based on the low
−
Received: January 20, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Chart 1. Structures of [Pd32+], TCPEBP, and TCPEP
a
Scheme 1. Initial Synthetic Route of TCPEBP-2
a
This approach was unsuccessful.
a
Scheme 2. Synthetic Route for TCPEBP
a
Reagents and conditions: (i) p-TsOH, −40 °C to rt, 4 h, DDQ; (ii) Zn(OAc)2·2H2O, rt, overnight, overall yield for (i) and (ii) 5%; (iii) 180−200
°C, 10 min, 91%; (iv) NaOH, reflux, 5 h; (v) HCl, overall yield for (iv) and (v) 82%; (vi) NaOH, rt, overnight, 72%.
reduction potentials10 and its well-known host guest behavior
to bind carboxylates in a purely ionic manner.11,12 Noteworthy,
the photoinduced oxidative electron transfer occurs in a time
scale < 85 fs in these two cases. We now report the dye···
[Pd32+]x assemblies where dye is either TCPEBP or TCPEP.
The new compound TCPEBP is a NIR emitter and exhibits the
largest K1x of this series. The dye···[Pd32+]x assemblies are
found nonemissive as well, and the rates for electron transfer
compared to 4,5,6,7-tetrahydroisoindole or 4,7-dihydroisoin-
dole.13 On the basis of the typical method to incorporate
acetylene groups into porphyrinic macrocycle, the synthesis
approach illustrated in Scheme 1 was initially designed to
synthesize TCPEBP-2 as the key intermediate.14 However, this
approach leads to scrambling due to the high reactivity of 3-
trimethylsilylpropynal, which tends to lose the trimethylsilyl
groups under acid conditions and induces undesired polymer-
ization. The MALDI-TOF spectra, indeed, confirmed the
presence of TCPEBP-TMS1, but numerous unexplained
secondary signals were also observed. Consequently, another
synthetic strategy was employed.
→ dye+•···[Pd3+•][Pd3
]x−1) also occur
2+
2+
(dye*···[Pd3
]
x
within the laser pulse (fwhm < 75−110 fs).
RESULTS AND DISCUSSION
■
Inspired by the first and sole synthesis of tetraethynyl-
substituted tetrabenzoporphyrine, where direct condensation of
1 and phenylpropynal were applied,7 the synthesis of the target
dye was successfully achieved by utilizing aldehyde Ar-CHO,
Synthesis. 4,7-Dihydro-4,7-ethano-2H-isoindole (1) has
recently been reported to be an attractive synthon for the
synthesis of tetrabenzoporphyrins because it could be easily
oxidized to provide the target macrocycles in high yield
B
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
which can avoid the use of the “vulnerable” trimethylsilyl
groups and introduce both ethynyl and carboxyl groups
simultaneously (Scheme 2). The porphyrin macrocycle is
generated using the classical acid catalysis condensation.15
Then Zn(OAc)2·2H2O was employed to incorporate the zinc
metal inside porphyrin to get TCPEBP-1. p-Toluenesulfonic
acid (p-TsOH) used as a nonoxidizing organic acid gave a
higher yield (5%) than common inorganic acids such as
trifluoroacetic acid (1%) and boron trifluoride etherate (2%). It
was expected that this condensation would show a lower yield
compared to typical porphyrin condensation yields (20−30%)
because of the delocalization of electronic density over
aldehyde and ethynyl groups, which could increase the
reactivity of aldehyde group and cause more side reactions.
The intermediate TCPEBP-1 was not stable, and TCPEBP-2
can be prepared by retro-Diels−Alder reaction in the solid state
upon heating. The intermediate TCPEBP-2 was then hydro-
lyzed, and sodium salt was introduced by the neutralization of
the acid groups using NaOH. The desired functional group
2.61 and 3.59 ns for TCPEP in MeOH at 298 K and MeOH/
2MeTHF (1:1) at 77 K, respectively (Table S1). The
fluorescence quantum yields, ΦF, for TCPEBP and TCPEP
in MeOH are 0.033 and 0.039 based on H2TPP (ΦF = 0.11).16
These photophysical parameters are also reminiscent of that of
the zinc(II)porphyrin chromophore, except that the TCPEBP
chromophore emits in the NIR with maxima of the 0−0
component in the vicinity of 760 nm. The spectra signatures
and τF data of TCPEBP are very similar to those of TCPEBP-2
(SI).
Cyclic Voltammetry and Driving Force for Electron
Transfer. Both dyes exhibit irreversible oxidation and
reduction waves in the cyclic voltammograms (Figure 2), and
the oxidation potentials are rather low (Table 2).
By using the 0−0 peaks of the Q-bands (at 77 K for more
accuracy; 740 nm (E0/* = 1.68 eV) for TCPEBP and 695 nm
(E0/* = 1.78 eV) for TCPEP) and the first oxidation peak
potentials (E0/+ = 0.46 and 0.70 V vs SCE for TCPEBP and
TCPEP, respectively), the excited-state driving forces for
oxidative electron transfer, E*/+, can be evaluated (+1.22 and
+1.08 V vs SCE for TCPEBP and TCPEP, respectively;
modified Latimer diagram in Figure 2, right). Concurrently,
[Pd32+] exhibits a reduction potential E0/− = −0.50 V vs SCE.17
The thermodynamic outcome is that the reactions dye*···
[Pd32+]x → dye+•···[Pd3+•]x are favorable (ΔE = +0.72 and
+0.58 V vs SCE for dye = TCPEBP and TCPEP, respectively).
Concurrently, using the reduction data for TCPEBP and
TCPEP (Table 2) and the oxidation potential for [Pd32+] (E0/+
= +0.95 V vs SCE)18 the excited-state driving forces for
reduction electron transfer, E*/−, can be evaluated (−0.70 and
−0.67 V vs SCE for TCPEBP and TCPEP, respectively) as
well as the ΔE for the reaction dye*···[Pd32+]x → dye−•···
[Pd33+•]x (−0.70 + 0.95 = +0.25 and −0.67 + 0.95 = +0.28 V).
These data indicate that oxidation quenching of the dye in the
S1 state is far more favorable than for the reduction. For
comparison purposes, the reported ΔE values for donor*···
[Pd32+] → donor+•···[Pd3•+] (+0.67; MCP, +0.72 V vs SCE;
DCP)8 are similar to those for TCPEBP and TCPEP (i.e.,
respectively +0.72 and +0.58 V vs SCE). This observation is
expectedly consistent with the fact that the extension of the π-
system decreases both the singlet-state energy and oxidation
potential almost proportionally.
−
−CO2 promotes the desired ionic interactions with the
[Pd32+] cluster.
Photophysical Characterization of the Dyes. Prior to
describing the properties of the targeted assemblies, some basic
properties are provided. The electronic spectra of TCPEBP
and TCPEP in methanol exhibit the typical ππ* signature of
the porphyrin chromophore (Figure 1 and Table 1), which is
DFT and TDDFT Computations. The nature of the singlet
excited states for both dyes have been addressed computation-
ally. The frontier MOs exhibit the classic π-orbitals expected for
porphyrins with extension of the atomic contributions into the
ethynyl and fused benzene fragments (Figure 3). Further
confirmations are obtained by calculating the positions of the
spin-allowed transitions using TDDFT. The 100th transitions
are placed in the SI, but the first four are included in Table 3.
The positions of the lowest energy (doubly degenerate)
transitions are computed at 679 and 713 nm for TCPEP and
Figure 1. Top: Absorption (black), fluorescence (red), and excitation
(blue) spectra of TCPEBP in MeOH at 298 K (left) and in MeOH/
2MeTHF (1:1) at 77 K (right). Bottom: Absorption (black), emission
(red), and excitation (blue) spectra of TCPEP in MeOH at 298 K
(left) and in MeOH/2MeTHF (1:1) at 77 K (right).
confirmed by DFT and TDDFT computations below. The
fluorescence lifetimes, τF, are 1.26 and 3.77 ns for TCPEBP and
Table 1. Absorption, Emission, and Excitation Data for Porphyrinic Salts TCPEBP and TCPEP
a
b
T (K)
absorption λmax (nm) (ε (×103 M−1·cm−1))
emission λmax (nm)
excitation λmax (nm)
TCPEBP
TCPEP
298
77
529 (273.1), 676 (24.3), 727 (13.9)
544, 686, 740
760
756
699
696
530
545
482
492
298
77
480 (248.8), 628 (5.4), 681 (25.6)
491, 642, 695
a
λexc = 530 nm for TCPEBP at 298 K (in MeOH) and 540 nm at 77 K (in a MeOH/2MeTHF (1:1) mixture); 480 nm for TCPEP at 298 K and
b
500 nm at 77 K. λem = 800 nm for TCPEBP at 298 and 77 K; 760 nm for TCPEP at 298 and 77 K.
C
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. Left: Cyclic voltammograms of TCPEBP (red) and TCPEP (black) in MeOH (3.0 × 10−3 M) at 298 K containing 0.1 M TBAPF62+as
supporting electrolyte (scan rate = 50 mV/s). Right: Modified Latimer diagram for TCPEBP (A) and TCPEP (B) and Latimer diagram for [Pd3
]
(C).
Table 2. Oxidation (Epa) and Reduction (Epc) Peak
Potentials of TCPEBP and TCPEP in MeOH
Table 3. Computed Positions, Oscillator Strengths, and
Major Contributions for the First Four Electronic
Transitions for TCPEP and TCPEBP (MeOH Solvent Field
Applied)
oxidation Epa (i.e., E0/+
)
reduction Epc (i.e., E0/−
)
V vs SCE
V vs SCE
TCPEBP
TCPEP
+0.46, +0.63, +1.09
+0.70, +1.03, +1.43
−0.98
−1.11
λ
osc.
TCPEP (nm) strength
major contributions (%)
1
2
3
4
679
679
469
469
0.5959 H-1→L+1 (12), HOMO→LUMO (73),
HOMO→L+1 (13)
0.5961 H-1→LUMO (12), HOMO→LUMO (13),
HOMO→L+1 (73)
1.8906 H-1→LUMO (60), H-1→L+1 (20), HOMO→
L+1 (10)
1.8912 H-1→LUMO (20), H-1→L+1 (60), HOMO→
LUMO (10)
osc.
strength
TCPEBP λ (nm)
major contributions (%)
1
2
3
4
713
712
550
550
0.0119
0.0086
2.1583
2.176
H-1→LUMO (47), HOMO→L+1 (52)
H-1→L+1 (46), HOMO→LUMO (53)
H-1→LUMO (53), HOMO→L+1 (47)
H-1→L+1 (54), HOMO→LUMO (46)
bars in Figure 3, bottom) also compare favorably with the
experimental spectra. Thus, the computations confirm the ππ*
nature of the lowest energy S1 states.
Evolution of the Absorption Spectra upon Addition
of [Pd32+]. Additions of dyes into methanol solutions
containing [Pd32+] lead to spectral modifications (Figure 4).
For TCPEP, two isosbestic points are noted at 500 and 625 nm
and exhibit the typical host−guest behavior for this cluster in
the presence of a −CO2 group.11a Typically, the strong
−
[Pd32+] absorption shifts from 500 nm to 480 nm upon
additions. For TCPEBP, this sought isosbestic point is not
apparent because of the strong Soret band blurring the
spectrum. However, clear evidence is provided below
demonstrating the presence of this same host−guest behavior
described by dye + x[Pd32+] ⇌ dye···[Pd32+]x (x = 1−4).
Using three graphical methods, i.e., Benesi−Hildebrand,
Scott, and Scatchard,19 the binding constants K1x (x = 1−4)
have been determined (Table 4, and see graphs provided in
Figure S1). The similarity of the extracted values for each
method provides confidence in the data. Moreover, the fact that
these graphs are linear and exhibiting the same slope over a
large range of equivalents of [Pd32+], indicating that K11 ≈ K12
Figure 3. Top: Representations of selected frontier MOs of TCPEP
(left) and TCPEBP (right) as Na+ salts using a MeOH solvent field
(energies in eV; for more frontier MOs, see SI). Bottom: Experimental
UV−vis spectrum and oscillator strength for the 100th electronic
transition of TCPEP (left) and TCPEBP (right) as Na+ salts using a
MeOH solvent field.
TCPEBP, respectively, and compare favorably to the
corresponding experimental values at 681 and 727 nm (298
K). The calculated compositions of these transitions include the
LUMO−1, LUMO, HOMO, and HOMO−1 as expected.
Moreover, the computed positions of the 100th transition (blue
D
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 4. Left: Evolution of the absorption spectra of [Pd32+] (9.72 ×
10−5 M) in MeOH at 298 K upon addition of TCPEP (9.65 × 10−6
M). Curves A−G were obtained with successive addition of 0.1 mL of
the TCPEP solution. The arrows indicate the direction of absorption
change with increasing TCPEP. Right: Evolution of the absorption
spectra of [Pd32+] (1.08 × 10−4 M) in MeOH at 298 K upon addition
of TCPEBP (3.36 × 10−6 M). Curves A−G were obtained with
successive addition of 0.1 mL of the TCPEBP solution. The arrows
indicate the direction of absorption change with increasing TCPEBP.
Only certain initial spectra are shown for clarity.
2+
Table 4. Binding Constants for the Dye···[Pd3
Assemblies in MeOH at 298 K
]
x
b
binding constants K1x (M−1
)
Benesi−
a
substrate (average)
TCPEBP···[Pd32+]x (83 200 M−1
TCPEP···[Pd32+]x (70 400 M−1
Hildebrand
Scott
Scatchard
)
85 358
75 016
82 449
67 784
81 890
68 495
Figure 5. Top left: Evolution of the fluorescence spectra of TCPEBP
(9.71 × 10−6 M) upon adding [Pd32+] in MeOH (298 K). Curves A−J
were recorded with successive additions of 0.1 mL of [Pd32+] (6.08 ×
10−5 M). Top right: Decrease of the relative fluorescence intensity
(ΦF/ΦF°) as a function of the [Pd32+]/[TCPEBP]. Middle left: Plot
of (ΦF°/ΦF) vs [Pd32+] (i.e., Stern−Volmer plot). Middle right:
Graph of log[(ΦF° − ΦF)/ΦF] vs log[Pd32+]. Bottom left: graph of [1
− (ΦF/ΦF°)]/[Pd32+] vs (ΦF/ΦF°). The two to three points curving
out of the slope are due to the high concentration of clusters causing a
deviation of the Beer−Lambert law. Bottom right: Graph of ln(W) vs
[Pd32+] for the TCPEBP···[Pd32+]x assembly in MeOH (298 K).
)
a
The values in parentheses are the average from the three methods.
The uncertainties are ∼ 10% based on multiple measurements.
b
≈ K13 ≈ K14 for the assemblies dye···[Pd32+]x (x → 0, 1, 2, 3,
4), means that the binding is not sterically nor electronically
influenced by the sequential ionic anchoring of the clusters
onto the dye. These constants, respectively 70 400 and 83 200
M−1 for TCPEP and TCPEBP (average values), are consistent
with the tetrabenzoporphyrin being a more electron-rich
chromophore. These constants can be considered strong. The
evolution of the absorption spectra of the dyes upon addition of
[Pd32+] (Figure S2 in the SI) also exhibits spectral changes
where the cluster band first appears as a shoulder near 480 nm
but shifts to 500 nm when a large excess is added, meaning that
all the binding sites are occupied and the amount of free
[Pd32+] increases with further additions.
in the absence and presence of [Pd32+], Kb is the binding
constant, and n is the average number of binding sites.20 Values
of n = 3.85 (TCPEP) and 3.72 (TCPEBP) are evaluated
(Table 5), and the fact that they approached the saturation
value of 4 is fully consistent with the number of carboxylate
sites on the chromophores.
Fluorescence Quenching Measurements. These ab-
sorption spectral modifications are accompanied by a
fluorescence quenching (Figure 5 as an example and the SI).
However, the fluorescence lifetimes, τF (again 1.26 and 3.77 ns
for TCPEBP and 2.61 and 3.59 ns for TCPEP, respectively, in
MeOH at 298 K and MeOH/2MeTHF (1:1) at 77 K), remain
constant for all [Pd32+] concentrations (data in Table S1). This
behavior is strongly suggestive of a static quenching in both
singlet states (eq 1):
Table 5. Various Quenching Constants, n, KD, and V,
Extracted from the Fluorescence Quenching in MeOH at
298 K and in MeOH/2MeTHF (1:1) at 77 K
assembly (temperature)
n
KD (M−1
)
V (M−1
)
TCPEBP···[Pd32+]x (298 K)
TCPEP···[Pd32+]x (298 K)
TCPEBP···[Pd32+]x (77 K)
TCPEP···[Pd32+]x (77 K)
3.72
3.85
3.61
3.76
18 300
16 400
19 300
18 700
80 400
69 400
81 000
72 600
2+
dye* + x[Pd3
emissive
]
dye*···[Pd32+]x (x = 1−4)
nonemissive
⇌
In order to verify that the quenching is dominantly static, a
mixed dynamic−static model was also used.21 This graphical
approach stems from a sphere of action model using the
equation [1 − (ΦF/ΦF°)]/[Pd32+] = KD(ΦF/ΦF°) + (1 − W)/
[Pd32+], where W is the fraction of the excited-state quenching
occurring from a collisional process given by exp(−V[Pd32+]),
where V is the static quenching constant defining the volume of
the sphere of action, and KD is the dynamic quenching
constant. From a graph of [1 − (ΦF/ΦF°)]/[Pd32+] vs (ΦF/
(1)
In order to confirm this process, the fluorescence quenching at
298 K is also graphically analyzed (Figure 5 for TCPEBP and
SI for TCPEP). The absence of linearity in the Stern−Volmer
plots (see graph in middle left) indicates that the mechanism is
not dynamic quenching. The data are then analyzed using the
relationship log[(ΦF° − ΦF)/ΦF] = log(Kb) + (n log[Pd32+]),
where ΦF° and ΦF are the fluorescence intensities respectively
E
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
ΦF°), KD is obtained from the slope using a least-squares fit and
the intercept leads to the W values as a function of [Pd322++].
the same MO (Figure 7 for TCPEBP and the SI for TCPEP).
This feature indicates that no MO coupling between the two
fragments and that no electronic communication are favored.
This expected conclusion supports that the interactions
between the anion and cation are solely ionic in nature. The
HOMO, HOMO−1, LUMO+1, and LUMO+2 for TCPEBP···
[Pd32+] (Figure 7) and TCPEP···[Pd32+] (SI) are the same as
analogues illustrated for the free dyes (above), indicating that
these weak interactions do not alter the MO schemes of the
chromophores.
The LUMO is composed of the in-plane dσ*(Pd−Pd) MO
centered on the [Pd32+] cluster. This computational result is
consistent with previous DFT calculations on the free cluster22
and the cluster assembled with a dye.8 This MO “insertion” of a
cluster-localized MO within the four π and π* levels of the
porphyrin dye is relevant, as it adds the possibility of electronic
transitions between the filled frontier MOs of the dyes and this
LUMO, although these are expected to be forbidden due to
poor MO overlaps. As expected, the two lowest energy
electronic transitions involve this dσ*(Pd−Pd) MO, and two
π(dye*) → dσ*(Pd−Pd) transitions are computed (Table 7).
For the upper energy electronic transitions, the calculated
positions are very similar to that reported for the free dyes
(above).
Then, V is obtained from the slopes in the ln(W) vs [Pd3
]
plots. These graphs are provided in Figure 5 for TCPEBP at
298 K. The remainder of the plots are placed in the SI.
These values are KD = 18 300 and V = 80 400 for TCPEBP···
[Pd32+]x, KD = 16 400 M−1, and V = 69 400 M−1 for TCPEP···
[Pd32+]x. Because the V values are significantly larger than those
for KD, the static quenching dominates the overall mechanism.
In addition, the K1x values (determined from absorption
spectroscopy) are similar to those obtained for V obtained from
fluorescence quenching experiments as expected. In order to
confirm this conclusion, this same analysis was performed on
the assemblies’ fluorescence at 77 K, where only static
quenching is possible. The data are placed in the SI. Indeed,
the n, KD, and V values are very similar at both temperatures
(Table 5).
DFT Calculation Results for Dye···[Pd32+] Assemblies
and Their MO Representations. The geometries of the
dye···[Pd32+] assemblies have been optimized by DFT
computations. Only one anchored cluster was calculated for
size reasons (Figure 6). The optimized geometries feature the
A reasonable match is noticed when comparing the graphs of
the calculated positions and oscillator strengths of the first
electronic transitions with the experimental spectra for both
assemblies in MeOH (Figure 7). A blue-shift of ∼30−50 nm of
the calculated vs experimental positions of the pure electronic
transitions is observed. The presence of a weak trailing tail at
higher wavelengths from the Q-bands is obvious in the
experimental spectra. These features extend all the way to
800 nm and a bit more. Thus, this match between calculations
and experimental corroborates well the presence of assemblies
in solutions.
Relative Proportions of TCPEBP···[Pd32+]x and TCPEP···
[Pd32+]x. Because the ionic interactions are generally weak, as
corroborated by the binding constants in this work, the
sufficient generation of assemblies must be controlled to secure
reasonable amounts for the measurements of the transient
absorption spectra. Their relative quantities have been
estimated using both K1x (from absorption data, Table 4)
and V (from fluorescence quenching, Table 5) constants. These
relative proportions are provided in Table 8, and the calculation
procedure has been placed in the SI for convenience. In order
Figure 6. Optimized geometry of TCPEBP···[Pd32+] (top) and
TCPEP···[Pd32+] (bottom) assemblies in MeOH solvent field.
anticipated long Pd···O distances generally associated with ionic
interactions (selected data are placed in Table 6). The longer
average Pd···O and Pd···Zn separations for TCPEP compared
to TCPEBP are fully consistent with the smaller binding
constants for TCPEP···[Pd32+] assemblies.
The representations of the frontier MOs exhibit no atomic
contribution on both the porphyrin dyes and [Pd32+] cluster for
−
to obtain a 1:2 [CO2 ]/[Pd32+] ratio, it is necessary to use 1:8
−
2+
equiv of dyes and cluster. By using the 1:2 [CO2 ]/[Pd3
]
Table 6. Selected Distances for the Optimized Geometry of
the TCPEBP···[Pd32+] and TCPEP···[Pd32+] Assemblies in
the Ground State
ratio, >99% of the dyes are associated with at least one [Pd32+],
and the major component is the dye···[Pd32+]4 assemblies.
Transient Absorption Spectra (TAS). In the absence of
fluorescence in the assemblies, this time-resolved methodology
was used to evaluate the time scales for the singlet excited state
quenching of the dyes by [Pd32+]. The unassociated cluster and
dyes were first studied separately. The TAS of [Pd32+] were
previously reported by us (SI of ref 8; it is provided as Figure
S15 in the SI for convenience). It exhibit positive (bleach) and
negative (transient) signals, respectively, at ∼510 and ∼410
nm, and the decay trace lies in the short picosecond time scale.
No TAS signature of the cluster was observed in the spectra,
indicating that its signal is rather weak in comparison with that
of the dyes (i.e., strong bleached Soret band). The time
evolution of the TAS of TCPEP is shown in Figure 8A, where
an isosbestic point is depicted at ∼490 and ∼670 nm. The
distances (Å)
2+
TCPEBP···[Pd3
Pd−Pd
]
2.707, 2.695, 2.678 (av = 2.693)
Pd−P
2.433, 2.411, 2.407, 2.396, 2.395, 2.392 (av = 2.406)
Pd···O
1st O: 3.543, 3.213, 2.896 (av = 3.217); 2nd O:
3.696, 3.184, 3.023 (av = 3.301)
Pd···Zn
TCPEP···[Pd3
15.181, 14.887, 14.642 (av = 14.903)
2+
]
Pd−Pd
Pd−P
2.682, 2.675, 2.670 (av = 2.676)
2.443, 2.413, 2.409, 2.408, 2.399, 2.380 (av = 2.409)
Pd···O
1st O: 3.617, 3.438, 3.079 (av = 3.378); 2nd O:
3.868, 3.573, 3.056 (av = 3.499)
Pd···Zn
15.582, 14.998, 14.956 (av = 15.179)
F
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 7. Top: Representations of the frontier MOs of the TCPEBP···[Pd32+] assembly in a MeOH solvent field (energies in eV; see SI for those for
TCPEP···[Pd32+]). Bottom: Experimental UV−vis spectrum and oscillator strength for the 100th electronic transition of TCPEBP···[Pd32+] (left;
note the weak bar at 713 nm, detail shown in Table 7) and TCPEP···[Pd32+] (right) as Na+ salt using a MeOH solvent field. The 100th computed
transitions are shown in the SI.
Table 7. Computed Positions, Oscillator Strengths, and
Major Contributions for the First Four Electronic
Transitions for the TCPEBP···[Pd32+] and TCPEP···[Pd3
and Assemblies (MeOH Solvent Field Applied)
Table 8. Relative Percentage of Complexed Dyes in the 1:2
−
[CO2 ]/[Pd32+] Ratio Used for the Transient Absorption
2+
]
Spectra
[CO22−+]/
[Pd3
]
assembly
%: K1x, V
assembly
%: K1x, V
no.
λ (nm)
osc. strength
major contributions (%)
2+
1:2
TCPEBP···
87.8, 87.5 TCPEP···
86.3, 86.7
TCPEBP···[Pd3
]
2+
2+
[Pd3
]
[Pd3
]
4
4
1
2
3
4
792
761
713
713
0
HOMO→LUMO (100)
H-1→LUMO (100)
TCPEBP···
10.7, 11.0 TCPEP···
11.8, 11.6
1.62, 1.54
0.23, 0.19
2+
2+
0.0003
0.0154
0.0128
[Pd3
]
[Pd3
]
3
3
H-1→L+2 (44), HOMO→L+1 (49)
H-1→L+1 (44), HOMO→L+2 (48)
TCPEBP···
1.29, 1.38 TCPEP···
2+
2+
[Pd3
]
[Pd3
]
2
2
2+
TCPEBP···
0.17, 0.17 TCPEP···
TCPEP···[Pd3
]
2+
2+
[Pd3
]
[Pd3 ]
1
2
3
4
779
705
681
677
0.0015
0.0003
0.6985
0.5653
H-1→LUMO (91)
HOMO→LUMO (98)
species that is readily addressable decays with a 2.29 ns time
constant. This value matches well the τF of 2.61 ns and is
obviously associated with the S1 species that are not associated
with the cluster (Figure 8C, light green trace). The signal is
weak with respect to that of the T1 species, likely due to the
efficient intersystem crossing, as suggested by the fast
generation of this T1 species (i.e., 2−4 ps).
A third species, decaying at ∼11 ps (turquoise trace), is also
very weak and is tentatively assigned to an S1 species similar to
that encountered in solvent-induced vibrationally relaxed S1
species commonly encountered in ZnTPP-containing chromo-
phores.23 These nonemissive S1 species generally decay in the
tens of picoseconds. A fourth species, also exhibiting a weak
H-2→L+2 (13), HOMO→L+1 (86)
H-2→L+1 (14), HOMO→L+2 (86)
deconvolution of the spectra indicates the presence of at least
four species (Figure 8C). The easiest species to identify decays
on the nanosecond time scale clearly associated with a triplet
state, i.e., T1, of the dye. This value is not accurate because of
the delay line limited to about 8 ns. This T1 species is rapidly
generated in the 2−4 ps time scale, as noticed in the decay trace
of the bleached signal at 480 nm (Figure 8E; purple trace) after
exciting directly in the S1 state (i.e., λexc = 680 nm) to minimize
the number of possible species present in solution. Another
G
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
labeled by purple traces decay with similar kinetics can only be
coincidental. This major transient species (i.e., purple trace in
Figure 8D) is the porphyrin cation. The key question is what
the rise time is (i.e., rate of formation). The monitoring of this
strong signal at 490 nm (turquoise trace in Figure 8F) indicates
that a rise is occurring well within the excitation pulse.
Conclusively, the charge-separated state (dye*···[Pd32+]x →
2+
dye+•···[Pd3+•][Pd3
]x−1) is formed within 107 fs. All the
other species, whether they concern the dye in the absence or
in the presence of the cluster, are also formed within the pulse
or in the 2−4 ps time window for the triplet species. The decay
trace of the charge-separated species recovers with a kinetics of
2+
∼0.65 ps (i.e., back electron transfer (dye*···[Pd3
] →
x
2+
dye+•···[Pd3+•][Pd3
]x−1)).
The time evolution of the TAS of TCPEBP in the absence
and presence of [Pd32+] (A and B), the deconvoluted spectra of
various species (C and D), and corresponding kinetic traces (E
and F) are shown in Figure 9. These species behave somewhat
Figure 8. Top: Time evolution of the transient absorption spectra (λexc
= 680 nm; fwhm =107 fs) of free TCPEP (A) and the TCPEP···
[Pd32+]x (B) assemblies in MeOH at 298 K (λexc = 680 nm; fwhm =
107 fs). Middle: Deconvoluted spectra of transient species from the
2+
experimental spectra of free TCPEP (C) and the TCPEP···[Pd3
]
x
(D). Bottom: Rise times and decays measured at various wavelengths
−
of free TCPEP (A) and the TCPEP···[Pd32+]x (B). The [CO2 ]/
[Pd32+] stoichiometry for the TCPEP···[Pd32+]x assemblies is 1:2. The
positive signals are bleached bands, and the negative ones are transient
absorptions.
intensity and decaying on the femtosecond time scale, is also
deconvoluted, but its origin is not known with certainty. Its
presence and assignment do not interfere with the conclusion.
2+
The time evolution of the TAS of the TCPEP···[Pd3
]
x
assembly, in which the striking feature is a notable difference
with the TAS of the free dye (TCPEP), is depicted in Figure
8B. This observation indicates that new products are formed in
the presence of [Pd32+]. Quasi-isosbestic points are depicted at
∼500 and ∼670 nm, and the apparent final product is the
TCPEP triplet species, as clearly observed by the comparison
with the triplet species observed for the free dye (see red traces
in Figure 8A−D). Again the nanosecond lifetime is unreliable
due to the delay line. Because of the lower intensity of the
triplet signals for the same dye concentration (i.e., ∼3-fold),
one concludes that the other product is the dye in its ground
state. The two other species that are easily identified, based on
comparison between the TAS of the dye in the absence of
[Pd32+], are the S1 species (light green and turquoise traces in
Figure 8C and D). While the 11.2 and 8.63 ps decays compare
favorably, the decays for the 2.29 and 0.16 ns species do not.
We do not have an explanation for this except that the
extraction of each lifetime component in the time evolution in
the TAS is generally challenging. We cannot attribute this
difference to a quenching of the singlet state by electron
transfer because the rise time for the formation of the charge-
separated state should occur within the same time frame (i.e.,
0.16 ns; see below). This was not the case.
Figure 9. Top: Time evolution of the transient absorption spectra (λexc
= 720 nm; fwhm = 78 fs) of free TCPEBP (A) and the TCPEBP···
[Pd32+]x (B) assemblies in MeOH at 298 K (λexc = 720 nm). Middle:
Deconvoluted spectra of transient species from the experimental
spectra of free TCPEP (C) and the TCPEBP···[Pd32+]x (D). Bottom:
Rise times and decays measured at various wavelengths of free
−
2+
TCPEBP (A) and the TCPEBP···[Pd32+]x (B). The [CO2 ]/[Pd3
]
2+
stoichiometry for the TCPEBP···[Pd3
] assemblies is 1:2. The
x
positive and negative signals are bleached and transient bands,
respectively.
similarly to that for TCPEP above, and only the differences and
key features are described. The deconvoluted spectra of the free
TCPEBP dye exhibit one more species, notably on the
picosecond time scale, which disappear when [Pd32+] is added.
This behavior is consistent with the large binding constant for
the TCPEBP···[Pd32+]x assembly, and free dye is essentially
nonexistent (bottom of Table 8). The species with a 1.37 ns
decay (purple trace in Figure 9C), which belongs to the
unassociated TCPEBP, matches well the corresponding τF
Finally, the new species exhibits a significantly more intense,
larger, and red-shifted signal (purple trace in Figure 8D
compared to that in C) which features some fine structures in
the 530−660 nm window. The fact that these two species
value (1.26 ns). More importantly, the species associated with
the charge-separated state (dye+•···[Pd3+•][Pd3
]x−1; purple
2+
trace in Figure 9D), readily recognizable from the multiple
resolved features, exhibits the strongest signals, notably at 575
H
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
literature procedures. 3-(4-Methoxycarbonylphenyl)prop-2-ynal (Ar-
CHO)24e was synthesized using a modified method previously
reported. Carboxylate sodium salts TCPEBP and TCPEP were
further prepared from corresponding acid counterparts.25
and 720 nm (bleach). Again, the monitoring of the rise time
indicates that all transients occur within the excitation pulse,
here 78 fs. This is particularly evident for the signal at 720 nm
(red trace in Figure 9F). So the charge-separated state occurs
within this time frame, but also relaxes via a back electron
transfer rather quickly (i.e., ∼170 fs) in comparison to the
TCPEP dye (∼650 fs). At first glance, this result appears
curious since the back electron transfer should even be more
thermodynamically favorable for the TCPEP dye (ΔE = 1.20
(TCPEP) vs 0.96 V (TCPEBP) vs SCE; Figure 10), but the
3-(4-Methoxycarbonylphenyl)prop-2-yn-1-ol (Ar-OH; Reac-
tion Scheme S1 in the SI). A mixture of methyl 4-iodobenzoate
(4717 mg, 18.0 mmol), propargyl alcohol (2.1 mL, 36 mmol),
PdCl2(PPh3)2 (380 mg, 0.54 mmol), CuI (343 mg, 1.8 mmol), and
PPh3 (237 mg, 0.9 mmol) was refluxed under argon for 3 h in dry
triethylamine (100 mL). After cooling the reaction mixture, the solvent
was removed in vacuo and extracted with dichloromethane. The
combined extracts were washed with water and dried over anhydrous
MgSO4. After removal of solvent in vacuo, the product was purified by
silica column chromatography eluting with dichloromethane to give
the title compound (2670 mg, 78%). 1H NMR (400 MHz, CDCl3-d3):
δ (ppm) 8.01−7.96 (d, J = 8 Hz, 2 H, Ar), 7.52−7.47 (d, J = 8 Hz, 2
H, Ar), 4.52 (s, 2 H, CH2), 3.92 (s, 3 H, CH3). MS (ESI): calcd for
C11H10O3 190.0630, found 213.0522 (M + Na).
3-(4-Methoxycarbonylphenyl)prop-2-ynal (Ar-CHO; Reac-
tion Scheme S1 in the SI). Pyridinium chlorochromate (12.94 g,
60 mmol) was added into a solution of compound Ar-OH (3800 mg,
20 mmol) in dry dichloromethane (100 mL), and then the mixture
was stirred overnight at room temperature. Water was added to
quench the reaction, and the mixture was extracted with dichloro-
methane. The combined extracts were washed with water and dried
over anhydrous MgSO4. After removal of solvent in vacuo, the product
was purified by silica column chromatography eluting with dichloro-
methane to give compound Ar-CHO (1470 mg, 39%). 1H NMR (400
MHz, CDCl3-d3): δ (ppm) 9.45 (s, 1 H, CHO), 8.09−8.05 (d, J = 8
Hz, 2 H, Ar), 7.70−7.65 (d, J = 8 Hz, 2 H, Ar), 3.95 (s, 3 H, CH3). MS
(ESI): calcd for C11H8O3 188.0473, found 243.0629 (M + Na +
MeOH).
9,18,27,36-Tetra(4-carboxymethylphenylethynyl)-
3,6,12,15,21,24,30,33-octahydro-3,6;12,15;21,24;30,33-
tetraethanotetrabenzoporphyrinatozinc(II) (TCPEBP-1). (i) A
stirred solution of 1 (1450 mg, 10 mmol) and Ar-CHO (1880 mg, 10
mmol) in 1 L of dry dichloromethane was added to p-toluenesulfonic
acid monohydrate (190 mg, 1 mmol) under argon at −40 °C. After
stirring for 3 h, the mixture was allowed to warm to room temperature
and was stirred for an additional 1 h. Then 2,3-dichloro-5,6-
dicyanobenzoquinone (DDQ, 2500 mg, 11 mmol) and 10 mL of
triethylamine were added, and the reaction mixture was stirred for an
extra 2 h. The solvents were removed in vacuo, and the residue was
purified by column chromatography on alumina gel eluting with THF/
DCM (1:9). The green fraction was collected, and the solvents were
removed in vacuo. (ii) Zn(OAc)2·2H2O (2200 mg, 10 mmol) in 50 mL
of MeOH was added to 200 mL of DCM solution of the above free-
base porphyrin. The mixture was stirred overnight at room
temperature. The solution was washed with water and brine and
dried over anhydrous Na2SO4. The residue was purified by column
chromatography on alumina gel eluting with THF/DCM (1:9). The
crude product was collected as a green fraction and recrystallized with
DCM/MeOH to yield the title compound (164 mg, 0.12 mmol, 5%).
MS (MALDI-TOF): calcd for C84H60N4O8Zn 1316.3703, found.
1316.4673 (M), 1288.4589 (M − C2H4), 1260.3977 (M − C4H8),
1232.3964 (M − C6H12), 1204.3179 (M − C8H16).
Figure 10. Thermodynamic parameters for the processes: donor*···
[Pd32+] → donor+•···[Pd3+•] → donor···[Pd32+] (this format is adopted
2+
2+
instead of dye*···[Pd3
]
x
→ dye+•···[Pd3+•][Pd3
]
→ dye*···
x−1
[Pd32+]x to simplify the drawing).
reorganization energies associated with these processes are
bound to be different. Indeed, the presence of four fused
benzene rings onto the porphyrin chromophore permits the
positive charge to be spread over more C atoms than the
TCPEP skeleton, thus making the reorganization energy
(including both nuclear and solvent) smaller for TCPEBP.
Such a situation would indeed render the electron transfers in
TCPEBP assemblies more favorable.
CONCLUSION
■
By altering the structure of the dye going from TCPEP to the
near-IR emitter TCPEBP (i.e., adding supplementary π-
conjugation), unambiguously notable variations of the oxido-
reduction potentials, S1 energies, binding constants with
[Pd32+], and excited-state driving forces for the oxidative
2+
quenching of the fluorescence (dye*···[Pd3
]
→ dye+•···
x
2+
[Pd3+•][Pd3
]x−1), occur. However, for both assemblies, which
exhibit strong binding constants, this efficient process is
ultrafast (within the 75−110 fs time frame). The highly
thermodynamically favorable back electron transfers (dye+•···
2+
[Pd3+•][Pd3
]
→ dye*···[Pd32+]x) are also expectedly
x−1
ultrafast (∼170 and ∼650 fs for TCPEBP and TCPEP,
respectively). In the context of DSSCs, the fast injection of
electrons into the conduction band of the TiO2 nanoparticles
by porphyrin-containing dyes has been demonstrated in the
literature.1 This behavior is also clearly corroborated in this
work using the redox-active [Pd32+] cluster, but the noted
ultrafast back electron transfer (<1 ps) potentially represents an
obstacle not to be neglected among the many parameters for
the design of these solar cells. This work demonstrates that a
dye built upon the tetrabenzoporphyrin motif, and the
tetraethynylphenylporphyrin as well, exhibits a serious possible
thermodynamic/kinetic challenge to the photoinduced electron
transfer.
9,18,27,36-Tetra(4-carboxymethylphenylethynyl)-
tetrabenzoporphyrinatozinc(II) (TCPEBP-2). TCPEBP-1 (131
mg, 0.1 mmol) was heated in a sample tube under vacuum at 180−
200 °C for 10 min. Then the solid was purified by column
chromatography on silica gel eluting with dichloromethane and
recrystallized with DCM/MeOH to yield the title compound (110 mg,
0.091 mmol, 91%). 1H NMR (400 MHz, CD2Cl2-d2): δ (ppm) 8.55−
6.45 (m, 32 H, Ar and β-benzo), 4.16−3.87 (m, 12 H, CH3). MS
(MALDI-TOF): calcd for C76H44N4O8Zn 1204.2451, found
1204.2230 (M).
EXPERIMENTAL SECTION
■
Materials. All commercial reagents were used as received or
purified by standard procedures before use. The [Pd3(dppm)3(CO)]-
(PF6)2 cluster ([Pd32+]),24a 4,7-dihydro-4,7-ethano-2H-isoindole
(1),13a and 5,10,15,20-tetra(4-carboxyphenylethynyl)-
porphyrinatozinc(II) (TCPEP-H)6,24b−d were obtained according to
9,18,27,36-Tetra(4-carboxyphenylethynyl)tetrabenzo-
porphyrinatozinc(II) (TCPEBP-H). An aqueous NaOH solution (5
M, 1 mL, 5.0 mmol) was added to a THF/MeOH (2:1) solution of
TCPEBP-2 (120 mg, 0.1 mmol, 60 mL). The mixture was stirred
I
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
under reflux for 5 h. Then aqueous HCl solution (1 M) was carefully
added to adjust the solution pH to 6−7. The resultant precipitate was
filtered and washed with H2O and diethyl ether to afford the title
compound (94 mg, 0.082 mmol, 82%). 1H NMR (400 MHz, DMSO-
d6): δ (ppm) 9.92 (s, 4 H, CO2H), 8.80−7.11 (m, 32 H, Ar and β-
benzo). MS (MALDI-TOF): calcd for C72H36N4O8Zn 1148.1825,
found 1148.2194 (M).
Sodium 9,18,27,36-Tetra(4-carboxyphenylethynyl)-
tetrabenzoporphyrinatozinc(II) (TCPEBP). TCPEBP-H (115 mg,
0.1 mmol) was dissolved in 30 mL of THF under argon. Then 4.0
equiv of aqueous NaOH solution was added slowly, and the mixture
was stirred overnight at room temperature. The resultant precipitate
was filtered and washed with DCM, THF, and diethyl ether. The solid
was dissolved in 50 mL of MeOH and filtered; then 30 mL of toluene
was added into the filtrate. By carefully removing the MeOH in vacuo,
the porphyrin salt was precipitated out of the solution. The resultant
precipitate was filtered and washed with DCM and diethyl ether. After
drying under vacuum, the title compound was obtained (88 mg, 0.072
mmol, 72%). IR (KBr): 2189 cm−1 CC, 1571 cm−1 CO. 1H
NMR (400 MHz, CD3OD-d4): δ (ppm) 8.52−8.25 (m, 16 H, Ar),
8.08−7.80 (m, 16 H, β-benzo).
used without further purification. Ferrocene/ferrocenium (Fc/Fc+)
was used as internal standard before and after each measurement (less
than 1 h). Potentials were converted to values for saturated calomel
electrode (SCE) by addition of 0.16 V, which was calibrated using Fc/
Fc+.
Calculation Procedure. All density functional theory (DFT) and
time-dependent density functional theory (TDDFT) calculations were
performed with Gaussian 0926 at the Universite
́
de Sherbrooke with
the Mammouth supercomputer supported by Le Reseau Quebecois
́
́
́
De Calculs Hautes Performances. The DFT geometry optimizations as
well as TD-DFT calculations27−36 were carried out using the B3LYP
method. A 6-31G* basis set was applied to H, C, N, O, Na, and P
atoms in porphyrins, palladium cluster, and porphyrin-palladium
cluster assembly. Valence double ζ with SBKJC effective core
potentials were used for all Zn and Pd atoms.37−42 All calculations
were carried out in a methanol solvent field. The calculated absorption
spectra were obtained from GaussSum 2.1.43
Femtosecond Transient Absorption Spectroscopy. The fs
transient spectra and decay profiles were acquired on a homemade
system using the SHG of a Soltice (Spectra Physics) Ti-sapphire laser
(λexc = 398 nm; fwhm = 75 fs; pulse energy = 0.1 μJ per pulse,
repetition rate = 1 kHz; spot size ≈ 500 μm), a white light continuum
generated inside a sapphire window, and a custom-made dual CCD
camera of 64 × 1024 pixels sensitive between 200 and 1100 nm
(S7030, Spectronic Devices). The delay line permitted probing up to 4
ns with an accuracy of ∼4 fs. The results were analyzed with the
program Glotaran (http://glotaran.org), permitting extraction of a
Sodium 5,10,15,20-Tetra(4-carboxyphenyl)ethynyl-
porphyrinatozinc(II) (TCPEP). TCPEP-H (95 mg, 0.1 mmol) was
dissolved in 30 mL of DMF under argon. Then 4.0 equiv of aqueous
NaOH solution was added slowly, and the mixture was stirred
overnight at room temperature. The resultant precipitate was filtered
and washed with DMF, THF, and diethyl ether. The solid was
dissolved in 100 mL of MeOH and filtered. A 50 mL amount toluene
was added into the filtrate. By carefully removing the MeOH in vacuo,
the porphyrin salt was precipitated out of the solution. The resultant
precipitate was filtered and washed with THF and diethyl ether. After
drying under vacuum, the title compound was obtained (72 mg, 70%).
sum of independent exponentials (I(λ,t) = C1(λ) × e−t /τ + C2(λ) ×
1
e−t /τ + ...) that fits the whole 3D transient map.
2
ASSOCIATED CONTENT
■
1
IR (KBr): 2196 cm−1 CC, 1595 and 1541 cm−1 CO. H NMR
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00050.
(400 MHz, CD3OD-d4): δ (ppm) 9.79 (s, 8 H, β), 8.31−8.06 (m, 16
H, Ar).
Instruments. NMR spectra were collected on a Bruker DRX 400
spectrometer using deuterated solvent with tetramethylsilane as
internal standard. MALDI-TOF mass spectra were recorded on a
Waters Synapt MALDI HDMS TOF mass spectrometer (Waters
Canada, Ontario, Canada) powered by a solid-state laser with
dithranol as matrix. All samples were freshly prepared and measured
within 1 h. The IR spectra were acquired on a Bomem FT-IR MB
series spectrometer equipped with a baseline-diffused reflectance.
Absorption spectra were measured on a Varian Cary 300 Bio UV−vis
spectrometer at 298 K and on a Hewlett-Packard 8452A diode array
spectrometer with a 0.1 s integration time at 77 K. Steady-state
fluorescence and excitation spectra were acquired on an Edinburgh
Instruments FLS980 phosphorimeter equipped with single mono-
chromators. NIR emission was measured by a QuantaMaster 400
phosphorimeter from Photon Technology International, which was
excited by a xenon lamp and recorded with a NIR PMT-7-B detector.
All fluorescence spectra were corrected for instrument response.
Fluorescence lifetime measurements were made with the FLS980
phosphorimeter using a 378 nm picosecond pulsed diode laser (fwhm
= 90 ps) as an excitation source. Phosphorescence lifetime
measurements were acquired on the FLS980 using a microsecond
flashlamp. Data collection on the FLS980 system is done by a time-
correlated single photon counting system.
Electrochemistry. Electrochemical measurements were conducted
with a three-electrode potentiostat (Princeton, Applied Research
Corporation, model 273A) in solvents deoxygenated by purging with
purified Ar gas. Cyclic voltammograms were obtained by using a three-
electrode cell equipped with a glassy carbon disk (0.07 cm2) as the
working electrode, a platinum wire as auxiliary electrode, and a freshly
polished silver wire as pseudoreference electrode at 298 K. The
working electrode was polished with aluminum (0.03 μm) on felt pads
(Buehler) and treated ultrasonically for 1 min before each experiment.
The reproducibility of individual potential values was within 5 mV.
Tetra-n-butylammonium hexafluorophosphate (TBAPF6) was used as
supporting electrolyte, which was obtained from Sigma-Aldrich and
Figures and tables giving fluorescence lifetimes, binding
constants, UV−vis spectra evolutions, quenching con-
stants, DFT and TDDFT calculations, relative percent-
age of complexed dyes, NMR, and MALDI-TOF (PDF)
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: marsan.benoit@uqam.ca.
*E-mail: pierre.harvey@usherbrooke.ca.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by the Natural Sciences and
Engineering Research Council of Canada (NSERC), the Fonds
de Recherche du Queb
the Centre Quebecois des Mater
and the Centre d’Etudes des Mater
́
ec−Nature et Technologies (FRQNT),
iaux Fonctionnels (CQMF),
iaux Optiques et Photo-
́
de Sherbrooke (CEMOPUS).
́
́
́
́
́
niques de l’Universite
REFERENCES
■
(1) (a) Ye, S.; Kathiravan, A.; Hayashi, H.; Tong, Y.; Infahsaeng, Y.;
Chabera, P.; Pascher, T.; Yartsev, A. P.; Isoda, S.; Imahori, H.;
Sundstrom, V. J. Phys. Chem. C 2013, 117, 6066−6080. (b) Hart, A. S.;
̈
KC, C. B.; Gobeze, H. B.; Sequeira, L. R.; D’Souza, F. ACS Appl.
Mater. Interfaces 2013, 5, 5314−5323. (c) Imahori, H.; Kang, S.;
Hayashi, H.; Haruta, M.; Kurata, H.; Isoda, S.; Canton, S. E.;
Infahsaeng, Y.; Kathiravan, A.; Pascher, T.; Chabera, P.; Yartsev, A. P.;
Sundstrom, V. J. Phys. Chem. A 2011, 115, 3679−3690.
̈
J
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(2) Mathew, S.; Yella, A.; Gao, P.; Humphry-Baker, R.; Curchod, B.
F. E.; Ashari-Astani, N.; Tavernelli, I.; Rothlisberger, U.; Nazeeruddin,
3608−3610. (b) Liu, L.; Fortin, D.; Harvey, P. D. Inorg. Chem. 2009,
48, 5891−5900.
(17) Gauthron, I.; Mugnier, Y.; Hierso, K.; Harvey, P. D. Can. J.
Chem. 1997, 75, 1182−1187.
M. K.; Gratzel, M. Nat. Chem. 2014, 6, 242−247.
̈
́
(3) (a) Nowak-Krol, A.; Łukasiewicz, Ł. G.; Haley, J. E.; Drobizhev,
(18) Brevet, D.; Lucas, D.; Mugnier, Y.; Harvey, P. D. J. Cluster Sci.
2006, 17, 5−12.
(19) Connors, K. A. Binding Constants: The Measurement of Molecular
Complex Stability; Wiley: New York, 1987.
(20) Makarska-Bialokoz, M. J. Lumin. 2014, 147, 27−33.
(21) Acar, M.; Bozkurt, E.; Meral, K.; Arik, M.; Onganer, Y. J. Lumin.
2015, 157, 10−15.
(22) Provencher, R.; Harvey, P. D. Inorg. Chem. 1996, 35, 2113−
2118.
M.; Rebane, A.; Cooper, T. M.; Gryko, D. T. J. Porphyrins
Phthalocyanines 2014, 18, 998−1013. (b) Uetomo, A.; Kozaki, M.;
Suzuki, S.; Yamanaka, K.-i.; Ito, O.; Okada, K. J. Am. Chem. Soc. 2011,
133, 13276−13279. (c) Wilson, G. S.; Anderson, H. L. Synlett 1996,
11, 1039−1040.
(4) (a) Loew, N. L.; Dzyuba, E. V.; Brusilowskij, B.; Kaufmann, L.;
Franzmann, E.; Maison, W.; Brandt, E.; Aicher, D.; Wiehe, A.;
Schalley, C. A. Beilstein J. Org. Chem. 2012, 8, 234−245. (b) Hatano, J.;
Obata, N.; Yamaguchi, S.; Yasuda, T.; Matsuo, Y. J. Mater. Chem. 2012,
22, 19258−19263. (c) Samarat, A.; Kuninobu, Y.; Takai, K. Synlett
2011, 15, 2177−2180. (d) Berlicka, A.; Latos-Grazynski, L.; Lis, T.
Inorg. Chem. 2005, 44, 4522−4533.
(23) Yu, H.-Z.; Baskin, J. S.; Zewail, A. H. J. Phys. Chem. A 2002, 106,
9845−9854.
(24) (a) Manojlovic-Muir, L.; Muir, K. W.; Lloyd, B. R.; Puddephatt,
R. J. J. Chem. Soc., Chem. Commun. 1983, 22, 1336−1337.
(b) Anderson, H. L. Tetrahedron Lett. 1992, 33, 1104−1104.
(c) Sommer, J. R.; Shelton, A. H.; Parthasarathy, A.; Ghiviriga, I.;
Reynolds, J. R.; Schanze, K. S. Chem. Mater. 2011, 23, 5296−5304.
(d) Yamada, H.; Kushibe, K.; Okujima, T.; Uno, H.; Ono, N. Chem.
Commun. 2006, 4, 383−385. (e) Suchy, P.; Dvorak, D.; Havelkova, M.
Collect. Czech. Chem. Commun. 1999, 64, 119−129.
(25) (a) Du, B.; Stern, C.; Harvey, P. D. Chem. Commun. 2011, 47,
6072−6074. (b) Rochford, J.; Chu, D.; Hagfeldt, A.; Galoppini, E. J.
Am. Chem. Soc. 2007, 129, 4655−4665.
(5) (a) Bromby, A. D.; Keller, S. N.; Bozek, K. J. A.; Williams, V. E.;
Sutherland, T. C. J. Org. Chem. 2015, 80, 9401−9409. (b) Kuo, M.-C.;
Li, L.-A.; Yen, W.-N.; Lo, S.-S.; Lee, C.-W.; Yeh, C.-Y. Dalton Trans.
2007, 14, 1433−1439. (c) Screen, T. E. O.; Lawton, K. B.; Wilson, G.
S.; Dolney, N.; Ispasoiu, R.; Goodson, T., III; Martin, S. J.; Bradley, D.
D. C.; Anderson, H. L. J. Mater. Chem. 2001, 11, 312−320. (d) Spence,
J. D.; Lash, T. D. J. Org. Chem. 2000, 65, 1530−1539. (e) McEwan, K.
J.; Bourhill, G.; Robertson, J. M.; Anderson, H. L. J. Nonlinear Opt.
Phys. Mater. 2000, 9, 451−468. (f) Lacey, J. A.; Phillips, D.; Milgrom,
L. R.; Yahioglu, G.; Rees, R. D. Photochem. Photobiol. 1998, 67, 97−
100. (g) Anderson, H. L.; Wylie, A. P.; Prout, K. J. Chem. Soc., Perkin
Trans. 1 1998, 10, 1607−1611. (h) Milgrom, L.; Yahioglu, G.; Bruce,
D. W.; Morrone, S.; Henari, F. Z.; Blau, W. J. Adv. Mater. 1997, 9,
313−316. (i) Milgrom, L. R.; Yahioglu, G. Tetrahedron Lett. 1996, 37,
4069−4072.
(26) Frisch, M. J.; et al. Gaussian 09; Gaussian, Inc.: Wallingford, CT,
2004.
(27) Hohenberg, P.; Kohn, W. Phys. Rev. 1964, 136, B864−871.
(28) Hohenberg, P.; Kohn, W. J. Phys. Rev. 1965, 140, A1133−1138.
(29) Parr, R. G.; Yang, W. Density-Functional Theory of Atoms and
Molecules; Oxford Univ. Press: Oxford, 1989.
(30) Salahub, D. R.; Zerner, M. C. The Challenge of d and f
ElectronsAm. Chem. Soc.; American Chemical Society: Washington,
D.C., 1989.394ix10.1021/bk-1989-0394.pr001
(6) Matsunaga, S.; Endo, N.; Mori, W. Eur. J. Inorg. Chem. 2011, 29,
4550−4557.
(7) Shen, Z.; Uno, H.; Shimizu, Y.; Ono, N. Org. Biomol. Chem. 2004,
2, 3442−3447.
(31) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256,
454−464.
(8) Luo, P.; Karsenti, P.-L.; Brisard, G.; Marsan, B.; Harvey, P. D.
Chem. Commun. 2015, 51, 17305−17308.
(32) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R. J.
Chem. Phys. 1998, 108, 4439−4449.
(9) Harvey, P. D.; Mugnier, Y.; Lucas, D.; Evrard, D.; Lemaitre, F.;
Vallat, A. J. Cluster Sci. 2004, 15, 63−90.
(33) Stratmann, R. E.; Scuseria, G. E.; Frisch, M. J. J. Chem. Phys.
1998, 109, 8218−8224.
(10) (a) Salomon, C.; Dal Molin, S.; Fortin, D.; Mugnier, Y.; Boere,
R. T.; Juge, S.; Harvey, P. D. Dalton Trans. 2010, 39, 10068−10075.
(b) Dal Molin, S.; Mugnier, Y.; Fortin, D.; Harvey, P. D. Dalton Trans.
2010, 39, 8976−8981. (c) Lemaitre, F.; Lucas, D.; Groison, K.;
Richard, P.; Mugnier, Y.; Harvey, P. D. J. Am. Chem. Soc. 2003, 125,
5511−5522. (d) Lemaitre, F.; Lucas, D.; Mugnier, Y.; Harvey, P. D. J.
Org. Chem. 2002, 67, 7537−7540. (e) Brevet, D.; Lucas, D.; Cattey,
H.; Lemaitre, F.; Mugnier, Y.; Harvey, P. D. J. Am. Chem. Soc. 2001,
123, 4340−4341. (f) Gauthron, I.; Mugnier, Y.; Hierso, K.; Harvey, P.
D. Can. J. Chem. 1997, 75, 1182−1187.
(34) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter
Mater. Phys. 1988, 37, 785−789.
(35) Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett.
1989, 157, 200−206.
(36) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(37) Binkley, J. S.; Pople, J. A.; Hehre, W. J. J. Am. Chem. Soc. 1980,
102, 939−947.
(38) Gordon, M. S.; Binkley, J. S.; Pople, J. A.; Pietro, W. J.; Hehre,
W. J. J. Am. Chem. Soc. 1982, 104, 2797−2803.
́
(11) (a) Salomon, C.; Fortin, D.; Darcel, C.; Juge, S.; Harvey, P. D. J.
(39) Pietro, W. J.; Francl, M. M.; Hehre, W. J.; Defrees, D. J.; Pople,
J. A.; Binkley, J. S. J. Am. Chem. Soc. 1982, 104, 5039−5048.
(40) Dobbs, K. D.; Hehre, W. J. J. Comput. Chem. 1986, 7, 359−378.
(41) Dobbs, K. D.; Hehre, W. J. J. Comput. Chem. 1987, 8, 861−879.
(42) Dobbs, K. D.; Hehre, W. J. J. Comput. Chem. 1987, 8, 880−893.
(43) O’Boyle, N. M.; Tenderholt, A. L.; Langner, K. M. J. Comput.
Chem. 2008, 29, 839−845.
Cluster Sci. 2009, 20, 267−280. (b) Provencher, R.; Aye, K. T.; Drouin,
M.; Gagnon, J.; Boudreault, N.; Harvey, P. D. Inorg. Chem. 1994, 33,
3689−3699.
(12) Aly, S. M.; Ayed, C.; Stern, C.; Guilard, R.; Abd-El-Aziz, A. S.;
Harvey, P. D. Inorg. Chem. 2008, 47, 9930−9940.
(13) (a) Jeong, S.-D.; Min, B.; Cho, S. Y.; Lee, C.; Park, B. K.; An, K.-
S.; Lim, J. J. Org. Chem. 2012, 77, 8329−8331. (b) Filatov, M. A.;
Lebedev, A. Y.; Vinogradov, S. A.; Cheprakov, A. V. J. Org. Chem.
2008, 73, 4175−4185. (c) Filatov, M. A.; Cheprakov, A. V.; Beletskaya,
I. P. Eur. J. Org. Chem. 2007, 2007, 3468−3475. (d) Borek, C.;
Hanson, K.; Djurovich, P. I.; Thompson, M. E.; Aznavour, K.; Bau, R.;
Sun, Y.; Forrest, S. R.; Brooks, J.; Michalski, L.; Brown, J. Angew. Chem.
2007, 119, 1127−1130.
(14) Yen, W.-N.; Lo, S.-S.; Kuo, M.-C.; Mai, C.-L.; Lee, G.-H.; Peng,
S.-M.; Yeh, C.-Y. Org. Lett. 2006, 8, 4239−4242.
(15) Lindsey, J. S.; Schreiman, I. C.; Hsu, H. C.; Kearney, P. C.;
Marguerettaz, A. M. J. Org. Chem. 1987, 52, 827−836.
(16) (a) Silva, S.; Pereira, P. M. R.; Silva, P.; Paz, F. A. A.; Faustino,
́
M. A. F.; Cavaleiro, J. A. S.; Tome, J. P. C. Chem. Commun. 2012, 48,
K
DOI: 10.1021/acs.organomet.6b00050
Organometallics XXXX, XXX, XXX−XXX