Photoinduced Charge Separation in a Donor-Spacer-Acceptor Dyad with N-Annulated Perylene Donor and Methylviologen Acceptor

Article abstract of DOI:10.1002/cphc.201500615

The first donor-acceptor species in which a strongly emissive N-annulated perylene dye is connected to a methylviologen electron acceptor unit via its macrocyclic nitrogen atom, is prepared by a stepwise, modular procedure. The absorption spectra, redox behavior, spectroelectrochemistry and photophysical properties of this dyad and of its model species are investigated, also by pump-probe fs transient absorption spectroscopy. Photoinduced oxidative electron transfer from the excited state of the dyad, centered on the N-annulated perylene subunit, to the appended methyviologen electron acceptor takes place in a few ps. The charge-separated species recombines in 19 ps. Our results indicate that N-annulated perylene can be connected to functional units by taking advantage of the macrocyclic nitrogen, an option never used until now, without losing their properties, so opening the way to new designing approaches. Time for a NAP: For the first time a strongly emissive N-annulated perylene (NAP) dye is connected to an electron acceptor subunit by its macrocyclic nitrogen. This molecular motif does not alter the intrinsic properties of the NAP species, and photoinduced charge separation is obtained with a time constant of 5 ps, with the charge-separated state recombining in 19 ps. New ways for designing light-active multicomponent species containing NAP are proposed.

Full text of DOI:10.1002/cphc.201500615

DOI: 10.1002/cphc.201500615
Communications
Photoinduced Charge Separation in a Donor–Spacer–
Acceptor Dyad with N-Annulated Perylene Donor and
Methylviologen Acceptor
Marie-Pierre Santoni,* Antonio Santoro, Tania M. G. Salerno, Fausto Puntoriero,*
Francesco Nastasi, M. Letizia Di Pietro, Maurilio Galletta, and Sebastiano Campagna*[a]
The first donor–acceptor species in which a strongly emissive
N-annulated perylene dye is connected to a methylviologen
electron acceptor unit via its macrocyclic nitrogen atom, is pre-
pared by a stepwise, modular procedure. The absorption spec-
tra, redox behavior, spectroelectrochemistry and photophysical
properties of this dyad and of its model species are investigat-
ed, also by pump–probe fs transient absorption spectroscopy.
Photoinduced oxidative electron transfer from the excited
state of the dyad, centered on the N-annulated perylene subu-
nit, to the appended methyviologen electron acceptor takes
place in a few ps. The charge-separated species recombines in
19 ps. Our results indicate that N-annulated perylene can be
connected to functional units by taking advantage of the mac-
rocyclic nitrogen, an option never used until now, without
losing their properties, so opening the way to new designing
approaches.
the NAP dye has been connected to the other subunits of the
multicomponent systems (other light absorbers or electron ac-
ceptor mediators, in their turn linked to the metal oxide semi-
conductor electrode) by a substitution at the peri-position of
the macrocyclic nitrogen. Connection via the nitrogen atom of
the N-annulated cycle to other photo- and/or redox-active sub-
units (including metal oxide semiconductor surfaces) has not
been explored. In fact, substitution at the nitrogen atom has
only been used to improve solubility of NAP species, but not
to attach functional (e.g., photo- and/or redox-active) subunits.
Here we report the first example of a donor-bridge-acceptor
dyad, 1 (see structural formula in Figure 1) in which the donor
is an N-annulated perylene, connected to the bridge-acceptor
subunits via the nitrogen of the macrocyclic structure, and
show its absorption spectra, redox behavior, and photophysical
properties, including ultrafast photoinduced charge separation
and recombination. Our results indicate that the N-annulated
perylenes can conveniently be connected via the nitrogen
atom to other subunits for photoinduced processes, without
losing their properties, thus offering new ways to molecular
design of multicomponent systems suitable for light-powered
functions. Noteworthy, the peri-positions of NAP species are
therefore available for further substitution (e.g., to incorporate
other functional subunits) or tunability reasons.
The search for new highly luminescent species is a very attrac-
tive field for several reasons, including the development of
new dyes for 1) fluorescence imaging dealing with the bio-
chemical and medical realms,[1] 2) light-emitting dyad systems
for illumination,[2] and 3) design of new dye-sensitized solar
cells (DSCs).[3,4] In recent years, a new class of dyes exhibiting
quite intense and tunable luminescence has been identified,
namely N-annulated perylenes (NAPs).[5] Indeed, NAP species
exhibit strong visible absorption, efficient photoluminescence,
and redox reversibility, all properties that are also largely tuna-
ble (e.g. strong absorption up to 700 nm can be obtained), in-
troducing substituents at the peri-position near the amine
site.[6] Actually, NAPs have recently been exploited as part of
multicomponent systems within the field of DSCs, taking ad-
vantage of both their absorption properties and reversible oxi-
dation behavior.[7] To our knowledge, in all the reported cases
The synthetic, modular approach is shown in Figure 1. The
N-annulated perylene dye obtained via step b[8] is reacted with
a,a’-dibromo-p-xylene (step c) to yield the Br-substituted spe-
cies in which the NAP chromophore is connected to a phenyl-
[a] Dr. M.-P. Santoni, A. Santoro, T. M. G. Salerno, Prof. F. Puntoriero,
Dr. F. Nastasi, Dr. M. L. Di Pietro, Dr. M. Galletta, Prof. S. Campagna
Dipartimento di Scienze Chimiche
Università di Messina and
Centro Interuniversitario per la Conversione Chimica dell’Energia Solare
(SOLARCHEM, Sezione di Messina)
Via F. Stagno d’Alcontres 31
Messina (Italy)
E-mail: mpcc.santoni@gmail.com
Figure 1. Structural formulae and modular synthetic pathway and reaction
conditions for 1 and 2: a) 1,4-dioxane/water/HNO3 (24/9/6), reflux, 1 h;
b) P(OEt)3, N2, reflux, 2 h; c) a,a’-dibromo-p-xylene (3.8 equiv), KOH
(1.5 equiv), KI (0.05 equiv), THF, N2, reflux, 48 h; d) methyl-4,4’-bipyridinium
hexafluorophosphate (1 equiv), MeCN, N2, reflux, 12 h, NH4PF6; e) bromoben-
zyl (3 equiv), KOH (4 equiv), KI (0.12 equiv), THF, N2, reflux, 26 h. Full details
in the Supporting Information.
fpuntoriero@unime.it
campagna@unime.it
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cphc.201500615.
ChemPhysChem 2015, 16, 3147 – 3150
3147
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
ene unit. This allows the introduction of a spacer to keep the
donor and acceptor units separated in the final, designed
structure. Similarly, using bromobenzyl instead of a,a’-dibro-
mo-p-xylene as a reactant, compound 2 is obtained (step e),
which is used as a model for the spectroscopic and photophys-
ical properties of the donor species. Finally, the electron ac-
ceptor methylviologen unit (MV2+) is connected to the bromo-
substituted NAP species (step d), yielding the multicomponent
(NAP)–Ph–MV2+ species 1.
The absorption spectra of 1 and 2 in acetonitrile (Table 1,
Figure 2) are dominated by the typical strong absorption of
the NAP compounds, assigned to a spin-allowed p–p* transi-
tion. The absorption spectra of 1 and 2 overlap each other in
the visible region, indicating that the presence of the MV2+
electron acceptor does not perturb the ground-state proper-
ties of the NAP subunit in 1. The model species 2 exhibits
a quite intense, structured emission (Figure 2, Table 1), both in
a acetonitrile liquid solution at room temperature (RT) and in
a butyronitrile solid matrix at 77 K. On the basis of the struc-
tured spectral shape, lifetime and quantum yield and taking
into account the similarities between RT and 77 K spectra and
lifetimes, the emission is assigned to fluorescence from the
lowest-lying singlet p–p* state; this result is also in agreement
with the reported fluorescence of similar NAP species.[6] Such
an emission is absent in 1, both at RT and 77 K, indicating effi-
cient intramolecular quenching due to the MV2+ subunit, in all
the investigated conditions.
The new (NAP)–Ph–MV2+ donor–bridge–acceptor 1 under-
goes a quasi-reversible oxidation process at +0.89 V vs. SCE
(Table 1), assigned to oxidation of the NAP subunit (2 under-
goes a roughly identical process), and a reversible reduction
process at À0.38 V, typical of MV2+ units.[9] To estimate the
driving force for photoinduced oxidative electron transfer in-
volving the excited NAP unit as the donor and the methylviol-
ogen electron acceptor, DGCS, Equations (1) and (2) can be
used:[10]
*
DGCS ¼ eð EoxÀEred
*
Þ
ð1Þ
ð2Þ
Eox ¼ EoxÀðE00=eÞ
In Equations (1) and (2), e is the electron charge, Eox and Ered
are the first oxidation and reduction potentials of 1, respective-
ly, *Eox is the oxidation potential of the excited state centered
in the NAP subunit, and E00 is the excited-state energy of 1, ap-
proximated to the 77 K emission maximum. In Equation (1),
the work term is neglected. Application of Equations (1) and
(2) yields a value of À1.59 eV for DGCS, indicating that the pho-
toinduced oxidative electron transfer in 1 is largely allowed by
a thermodynamic viewpoint, so justifying the quenching of
the NAP-based emission even at 77 K.
Table 1. Absorption, luminescence and redox data.[a]
To further investigate the excited-state quenching process,
pump–probe transient absorption spectroscopy is quite useful;
however, identification of expected signatures for radical cat-
ions and anions is a prerequisite for pump–probe experiments.
Figure 3 shows the absorption spectrum of the electrochemi-
cally generated, oxidized form of 1: the radical cation centered
on the NAP unit is characterized by a visible absorption in the
range 450–800 nm. The radical anion involving the MV2+ subu-
Absorption
l [nm]
Luminescence,
Redox data (V vs SCE)
l [nm][b]
t [ns]
Fem
E1/2(red)
E1/2(ox)
(e [MÀ1 cmÀ1])
1
2
422 (35450)
421 (36000)
–
–
–
0.9
À0.38
+0.89
+0.88
435
(432)[c]
5.0
(5.3)[c]
–
[a] Data in acetonitrile solution at RT, unless otherwise stated; [b] only the
highest-energy maximum is reported; [c] in butyronitrile rigid matrix at
77 K.
Figure 3. Spectral changes of 1 in acetonitrile obtained by applying a poten-
tial of +0.9 V vs. SCE in acetonitrile, so generating the radical cation of the
NAP subunit. Initial spectrum is the blue one, the red spectrum is that of
the electro-generated radical cation of 1.
Figure 2. Absorption of 1 and 2 and emission spectra of 2 in acetonitrile so-
lution (RT) and in butyronitrile rigid matrix (77 K). Full data in Table 1.
ChemPhysChem 2015, 16, 3147 – 3150
www.chemphyschem.org
3148
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
nit, obtained by applying a voltage of À0.38 V vs SCE to a solu-
tion of 1, is characterized by a less intense absorption band at
480–700 nm and by a broad absorption in the 800–900 nm
region (Supporting Information). These pieces of information
could be useful to investigate the photoinduced electron
transfer by pump–probe transient absorption spectroscopy
(see later).
Figure 4 shows the pump-probe transient absorption spec-
trum (TAS) of 1 in acetonitrile. The initial spectrum, essentially
coincident with that of 2 in the same conditions (Supporting
Information), and therefore assigned to the S1 (1p–p*) excited
state of the NAP unit, shows a bleach centered at about
460 nm, assigned to stimulated emission,[11] and a broad transi-
ent absorption in the region 460–750 nm. Then, whereas the
TAS of 2 decays monotonically to the ground state (Supporting
Information), for 1 an absorption peak at 490 nm appears after
few picoseconds.
This peak is attributed to the formation of the radical cation
of the NAP unit, as it derives from a spectral combination due
to the formation of the radical cation band in the range 400–
750 nm, evidenced by the spectroelectrochemistry in Figure 3,
and the red tail of the bleaching of the ground-state absorp-
tion. Simultaneously, also the bleach due the stimulated emis-
sion at about 460 nm is fully recovered. To definitely assign the
fast quenching process to photoinduced electron transfer lead-
ing to the charge-separated state of 1, which can be represent-
ed as a (NAP+)ÀPhÀMV+ species, formation of the reduced
form of the acceptor MV2+ subunit would be also useful. How-
ever, spectroscopic identification of the reduced form of the
MV2+ subunit is a difficult task. In fact, the 480–700 nm region,
where the radical species MV+ absorbs (Supporting Informa-
C
tion), is obscured by the more intense absorption of the NAP
cation (Figure 3). The 800–900 nm region is experimentally
problematic due to laser pulse. The UV region, sometimes
used to identify the reduced methylviologen,[12] is not accessi-
ble to our pump–probe equipment for technical reasons. Al-
though identification of MV+.is therefore not feasible in our
case, we attribute the quenching process to photoinduced
electron transfer, since this process is the only one reasonable
by a thermodynamic viewpoint and identification of the radical
cation NAP strongly supports our hypothesis.[13]
From the kinetic traces in Figure 4, the time constant for the
formation of the charge-separated (NAP+)–Ph–MV+ species,
from the excited state of the NAP subunit is 5 ps (see also
Figure 5). The charge-separated state decays to the ground
state with a time constant of 19 ps (Figure 4), assigned to back
electron transfer (i.e., charge recombination). According to
Equations (1) and (2), and as illustrated in Figure 5, driving
force for the charge recombination, DGCR, is 1.29 eV. Compari-
son between forward and backward electron transfer rate con-
stants in 1, however, would require other pieces of informa-
tion, like reorganization energy, that are not available at the
moment and are outside the scope of this paper, so this
aspect will not be discussed here.
Figure 4. TAS of 1 in acetonitrile and its decay kinetics. Middle: complete 3D
TAS matrix representation; top: sketch evidencing the spectral evolution;
bottom: decay kinetics (early times in the inset). lexc. : 400 nm.
phore and a methylviologen unit.[14] For such a species, photo-
induced charge separation was obtained with a time constant
of 7.1 ps, and charge recombination had a lifetime of 59 ps.
This comparison further confirms that the NAP subunit has ex-
cited-state properties that can compete with those of the most
popular bodipy dyes.
Noteworthy, the kinetics of charge separation and recombi-
nation for 1 are similar to those recorded in a somewhat relat-
ed dyad made of a boron-dipyrromethene (bodipy) fluoro-
In conclusion, the macrocyclic nitrogen atom of an N-annu-
lated perylene species has been used to link such a dye, via
ChemPhysChem 2015, 16, 3147 – 3150
www.chemphyschem.org
3149
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
ence & Business Media, Berlin, 2012; b) S. A. Vinogradov, D. F. Wilson, in
Designing Dendrimers (Eds.: S. Campagna, P. Ceroni, F. Puntoriero),
Wiley, Hoboken, 2012, p. 463; c) S. Udenfriend, Fluorescence Assays in
Biology and Medicine, Vol. 2, Academic Press, San Diego, 2014; d) J. V.
Frangioni, Curr. Opin. Chem. Biol. 2003, 7, 626; e) Z. Guo, S. Park, J.
Yoon, I. Shin, Chem. Soc. Rev. 2014, 43, 16; f) T. Papalia, G. Siracusano, I.
Colao, A. Barattucci, M. C. Aversa, S. Serroni, S. Campagna, M. T. Sciorti-
no, F. Puntoriero, P. Bonaccorsi, Dyes Pigm. 2014, 110, 67.
[2] See, for example: a) H. Yersin, Top. Curr. Chem. 2004, 241, 1; b) N. Armar-
oli, G. Accorsi, M. Holler, O. Moudam, J.-F. Nierengarten, Z. Zhou, R. T.
Wegh, R. Welter, Adv. Mater. 2006, 18, 1313; c) C. A. Strassert, C.-H.
Chien, M. D. Galvez-Lopez, D. Kourkoulos, D. Hertel, K. Meerholz, L.
De Cola, Angew. Chem. Int. Ed. 2011, 50, 946; Angew. Chem. 2011, 123,
976; d) J. Kalinowski, Organic Light-Emitting Diodes: Principles, Character-
istics & Processes, CRC Press, Boca Raton, FL, 2014.
[3] a) B. O’Regan, M. Grätzel, Nature 1991, 353, 737; b) S. Ardo, G. J. Meyer,
Chem. Soc. Rev. 2009, 38, 115, and refs. therein.
[4] Luminescence is not a compulsory requisite for dyes to be used in DSC.
However, it is an indirect indication for relatively slow radiationless
decay of the dye excited state, and this is a convenient property for ef-
ficient DSCs.
Figure 5. State diagram and decays of 1.
[5] a) Y. Li, Z. Wang, Org. Lett. 2009, 11, 1385; b) Y. Li, J. Gao, S. D. Motta, F.
Negri, Z. Wang, J. Am. Chem. Soc. 2010, 132, 4208; c) C. Jiao, K.-W.
Huang, J. Wu, Org. Lett. 2011, 13, 632.
a phenylene spacer, to an electron acceptor subunit, namely
a methylviologen moiety. This has allowed the preparation of
a donor–spacer–acceptor species by a design approach never
explored so far, as in all reported cases N-annulated perylene
dyes have been connected to other photo and/or redox-active
units exclusively by substitution at the peri-position of the
macrocyclic nitrogen. The spectroscopic, redox, and photo-
physical results indicate that in the so-formed, novel donor–
spacer–acceptor dyad the N-annulated perylene subunit main-
tains its remarkable photophysical and redox properties, allow-
ing to efficiently generate, by photoinduced oxidative electron
transfer, a charge-separated species with a time constant of
5 ps, which recombines in about 20 ps. Such results open fur-
ther perspectives to the molecular design of multicomponent
systems containing the quite interesting, new N-annulated per-
ylene chromophores and suitable for light-powered functions.
[6] L. Zhu, C. Jiao, D. Xia, J. Wu, Tetrahedron Lett. 2011, 52, 6411, and refs.
therein.
[7] a) J. Luo, M. Xu, R. Li, K.-W. Huang, C. Jiang, Q. Bi, W. Zeng, J. Zhang, C.
Chi, P. Wang, J. Wu, J. Am. Chem. Soc. 2014, 136, 265; b) Z. Yao, C. Yan,
M. Zhang, R. Li, Y. Cai, P. Wang, Adv. Energy Mater. 2014, 4, 1400244;
c) Z. Yao, H. Wu, Y. Ren, Y. Guo, P. Wang, Energy Environ. Sci. 2015, 8,
1438; d) Z. Yao, M. Zhang, H. Wu, L. Yang, R. Li, P. Wang, J. Am. Chem.
Soc. 2015, 137, 3799; e) L. Yang, Y. Ren, Z. Yao, C. Yan, W. Ma, P. Wang, J.
Phys. Chem. C 2015, 119, 980; f) L. Yang, Z. Zheng, Y. Li, W. Wu, H. Tian,
Z. Wang, Chem. Commun. 2015, 51, 4842.
[8] J. J. Looker, J. Org. Chem. 1972, 37, 3379.
[9] J.-P. Sauvage, J.-P. Collin, J.-C. Chambron, S. Guillerez, C. Coudret, V. Bal-
zani, F. Barigelletti, L. De Cola, L. Flamigni, Chem. Rev. 1994, 94, 993, and
references therein.
[10] V. Balzani, P. Ceroni, A. Juris, Photochemistry and Photophysics: Concepts,
Research, Applications, Wiley-VCH, Weinheim, 2014.
[11] In fact, although a steady-state emission is not recordable for 1 due to
electron transfer quenching, the initially-formed transient species, in-
volving excited NAP subunit, exhibits stimulated emission. The residual
emission is evidently too low to be recorded on a steady-state instru-
ment.
[12] R. Lomoth, T. Häupl, O. Johansson, L. Hammarstrçm, Chem. Eur. J. 2002,
8, 102.
Acknowledgements
[13] In principle, intersystem crossing from the S1 to the T1 of the NAP subu-
nit is, obviously, thermodynamically possible. However, there is not any
plausible reason because the presence of the electron acceptor subunit
promotes fast intersystem crossing in 1, whereas is not efficient in 2.
Moreover, such (eventual) triplet state would decay with a slow kinetic,
incompatible with the time constant of the produced state (19 ps).
[14] D. Frath, J. E. Yarnell, G. Ulrich, F. N. Castellano, R. Ziessel, ChemPhys-
Chem 2013, 14, 3348.
The authors thank MIUR (FIRB Nanosolar and PRIN Hi-Phuture
projects) and COST-Perspect H2O for financial support.
Keywords: charge-separated state
· femtosecond pump–
probe spectroscopy · luminescence · N-annulated perylene ·
photoinduced electron transfer
[1] See, for example: a) Applied Fluorescence in Chemistry, Biology and Medi-
cine (Eds.: W. Rettig, B. Strehmel, S. Schrader, H. Seifert), Springer Sci-
Manuscript received: July 30, 2015
Final Article published: August 31, 2015
ChemPhysChem 2015, 16, 3147 – 3150
www.chemphyschem.org
3150
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim