Electronic characterisation and significant second-order NLO response of 10,20-diphenylporphyrins and their ZNII complexes substituted in the meso position with π-delocalised linkers carrying push or pull groups

Article abstract of DOI:10.1002/ejic.200501004

This work describes the synthesis and the electronic characterisation by electronic absorption spectroscopy, cyclic voltammetry and dipole moments of diphenyl porphyrins and their Zn^II complexes substituted at the meso position by a pseudo-line

Full text of DOI:10.1002/ejic.200501004

FULL PAPER
DOI: 10.1002/ejic.200501004
Electronic Characterisation and Significant Second-Order NLO Response of
10,20-Diphenylporphyrins and Their ZnII Complexes Substituted in the meso
Position with π-Delocalised Linkers Carrying Push or Pull Groups
Tamara Morotti,[a] Maddalena Pizzotti,*[b] Renato Ugo,[b] Silvio Quici,[c]
Maurizio Bruschi,[d] Patrizia Mussini,[e] and Stefania Righetto[b]
Keywords: Nonlinear optics / Porphyrins / Cyclic voltammetry / UV/Vis spectroscopy
This work describes the synthesis and the electronic charac-
terisation by electronic absorption spectroscopy, cyclic vol-
tammetry and dipole moments of diphenyl porphyrins and
their ZnII complexes substituted at the meso position by a
pseudo-linear π-delocalised substituent carrying an electron-
donor or an electron-withdrawing group. The second-order
NLO response was investigated by the EFISH technique
The second-order NLO response has been discussed by com-
paring the nature of the substituent (electron donor or ac-
ceptor) and the kind of substitution (β pyrrolic or meso). This
comparison evidenced a significant increase of the electron-
donor properties of NBu2 or NMe2 when they are connected
to the meso position or to the β-pyrrolic position, probably
because of an auxiliary donor effect of the electron-rich por-
phyrin ring. When the substituent is the electron-with-
drawing NO2 group, the substituent position (meso or β pyr-
rolic) is influential, with the existance of a significant in-
crease in the second order NLO response when substitution
occurs at the meso position.
working with
a nonresonant incident wavelength of
1.907 µm. This work confirms the ambivalent role of the pol-
arisable porphyrin ring, which, already in the ground state,
acts as a donor or acceptor depending on the nature (ac-
ceptor or donor) of the substituent in the meso position, as
was pointed out in our previous work on tetraphenyl porphy- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
rins substituted at the β-pyrrolic position.
Germany, 2006)
able solubility, good thermal and chemical stability, and po-
tential processability as layers.[8]
Introduction
In addition, the architectural flexibility of porphyrins is
well exemplified by the large number of structures reported
in the literature, as well as by a large variety of substituents
that can be attached at the meso or β-pyrrolic positions[5,6]
of the porphyrin ring or to the axial position of a metal
complex.[9]
Starting from studies carried out by Suslick and cowork-
ers at the beginning of the 90s[4a,4b] and later by Therien
and coworkers,[5a,5b] and following the theoretical approach
of Marks, Ratner et al.,[10] who suggested that the presence
of a π spacer between an electron-donor or acceptor group
and the porphyrin ring together with the right ring substitu-
tion (β pyrrolic or meso) could produce chromophores with
a large second-order NLO response, we recently investi-
gated the second-order NLO response of push-pull porphy-
rinic chromophores with electron-donor or withdrawing
groups linked, through a π spacer, to the β-pyrrolic position
of the 5,10,15,20-tetraphenylporphyrin or its ZnII com-
plex.[6]
Over the last two decades molecular materials for nonlin-
ear optics (NLO) have been investigated for their promising
applications in optoelectronics and photonics.[1] Although
most studies deal with pseudo-linear 1D organic[2] or orga-
nometallic[3] push-pull chromophores with second-order
nonlinear optical response, more recent investigations have
been devoted to two-dimensional push-pull chromophores
based on highly π-delocalised macrocycles such as porphy-
rins[4–6] and phthalocyanines.[7] Porphyrins and their metal
complexes appear to be advantageous, given their accept-
[a] Dipartimento di Chimica Organica e Industriale,
Via Golgi 19, 20133 Milan, Italy
[b] Dipartimento di Chimica Inorganica Metallorganica e Anali-
tica dell’Università di Milano, Unità di Ricerca dell’INSTM,
Centro di Eccellenza CIMAINA dell’Università di Milano e
Istituto di Scienze e Tecnologie Molecolari del CNR (ISTM),
Via Venezian 21, 20133 Milan, Italy
Fax: +39-2-50314405
E-mail: maddalena.pizzotti@unimi.it
[c] Istituto di Scienze e Tecnologie Molecolari del CNR (ISTM),
via Golgi 19, 20133 Milan, Italy
Our experimental results, in agreement with the sugges-
tion of Marks, Ratner et al.,[10] have shown that the second-
order NLO response is enhanced when an electron-donor
substituent is attached at the electron-rich β-pyrrolic posi-
tion of the porphyrin ring. We also evidenced an ambivalent
[d] Dipartimento di Biotecnologie e Bioscienze, Università di Mi-
lano Bicocca,
Piazza della Scienza 2, 20126 Milan, Italy
[e] Dipartimento di Chimica Fisica ed Elettrochimica dell’Univer-
sità di Milano,
Via Golgi 19, 20133 Milan, Italy
Eur. J. Inorg. Chem. 2006, 1743–1757
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1743

T. Morotti, M. Pizzotti, R. Ugo, S. Quici, M. Bruschi, P. Mussini, S. Righetto
FULL PAPER
role of the porphyrin ring (in the ground state but also in also performed to evaluate the dipole moment of the op-
charge transfer processes), behaving as an electron donor timised structures of the investigated porphyrinic macro-
or as an electron-rich enhancer of the donor properties of cycles, using the BP86[12] functional and an all-electron val-
an amino group, according to the nature of the substituent ence triple-ξ basis set with polarisation functions on all
in the β-pyrrolic position.
atoms.[13]
Very few studies have been devoted to the experimental
investigation of the second-order NLO response of push-
pull porphyrinic chromophores according to the nature of
the substituents and their position on the porphyrin ring.
Thus, in an effort to produce additional evidence of the
ambivalent role of the porphyrin ring, we extended our in-
vestigation to the electronic characterisation and second-
Results and Discussion
Synthesis of Porphyrins and Their ZnII Complexes
The asymmetrical 10,20-diphenylporphyrins and their
order NLO response of a series of 10,20-diphenylporphy- ZnII complexes investigated in this work are reported in
rins (H2DPP) and their ZnII complexes substituted, at the Figure 1; the ZnII complexes 2 and 4 have already been syn-
meso position, by electron-donor or electron-withdrawing thesised by Anderson et al.[14]
groups connected through a π spacer.
The porphyrin 1 was synthesised with a modification of
In order to avoid resonance effects or interference re- the synthetic route described by Anderson et al.,[14] involv-
sulting from fluorescent emissions, the second-order NLO ing a Wittig condensation of 5-formyl-10,20-diphenylpor-
response was measured by the EFISH technique working phyrin with 4-nitrobenzyltriphenylphosphonium bromide in
with a nonresonant 1.907 µm incident wavelength.[5a] Cal- refluxing 1,2-dichloroethane, using solid NaOH 20–
culations based on density functional theory (DFT)[11] were 40 mesh beads as the base. This reaction afforded 1 in 49%
Figure 1. Asymmetrical 10,20-diphenylporphyrins and their ZnII complexes substituted at the meso position with electron-withdrawing
or electron-donor groups.
Scheme 1. i) 4-Nitrobenzyltriphenylphosphonium bromide, NaOH, CH2Cl2, reflux, 4 h; ii) Zn(OAc)2·2H2O, CHCl3/MeOH, reflux, 1 h;
iii) 4-(Dibutylamino)benzyltriphenylphosphonium chloride, NaOH, CH2Cl2, room temp., 4 h; iv) Zn(OAc)2·2H2O, CHCl3/MeOH, reflux,
1 h.
1744
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1743–1757

Electronic Characterisation and Significant Second-Order NLO Response
FULL PAPER
Scheme 2. i) NBS, CHCl3, room temp., 30 min.; Zn(OAc)2·2H2O, CHCl3/MeOH, reflux, 1 h; ii) catalyst (see Exp. Sect.), [4-(dimeth-
ylamino)phenyl]acetylene, pyridine, toluene, 80 °C, 1.5 h; iii) TFA, CH2Cl2, 18 h; iv) catalyst (see Exp. Sect.), (4-nitrophenyl)acetylene,
pyridine, toluene, 80 °C, 1 h; v) TFA, CH2Cl2, 18 h.
yield after purification by column chromatography and 10,20-diphenylporphyrin (H2DPP). The significant red shift
crystallisation from CH2Cl2/MeOH.
of both the Soret B band and the Qα band at lower energy,
Analogously, Wittig condensation of 5-formyl-10,20-di- with an increase in the intensity of this latter band suggests
phenylporphyrin with 4-(dibutylamino)benzyltriphenyl- an electronic conjugation between the porphyrin ring and
phosphonium chloride afforded 5 in 59% yield after purifi- the π linker. The spectra, obtained in CH2Cl2 solution in
cation by column chromatography.
the presence of 1% pyridine, refer to pentacoordinate spe-
The corresponding ZnII complexes 2 and 6 were obtained cies with a pyridine in the axial position.
in quantitative yield after refluxing 1 and 5 with ZnII ace-
Here we report (Table 1) and discuss the absorption and
tate in CH2Cl2/MeOH for 1 h[15] (Scheme 1).
emission spectra of the ZnII complexes 2 and 4 in CHCl3
5-Formyl-10,20-diphenylporphyrin was prepared in 80% solution, with and without pyridine addition, together with
yield by the Vilsmaier formylation of the CuII complex of those of the corresponding porphyrins 1 and 3.
10,20-diphenylporphyrin, obtained in 83% yield by reaction
A comparison of the absorption spectra of the porphy-
of 10,20-diphenylporphyrin with CuII acetate in refluxing rins 1 and 3 with those of 2 and 4 indicates a decrease in
CHCl3/MeOH, followed by quantitative demetallation of the number of Q bands, from four to two, with the complete
the CuII complex with a mixture of trifluoroacetic acid/con- disappearance of the Qα band at low energy (around 655–
centrated sulfuric acid (9:1), as reported by Susumu et al.[15] 660 nm) as expected for an increase of the microsymmetry
ZnII complexes 4 and 8 (85% and 59% yields, respec- upon coordination.[6] Compounds 2 and 4, but also 1 and
tively) were synthesised, as described by Anderson et al.,[14] 3, still show a significant red shift of the Soret B band and
by the coupling reaction of the ZnII complex of 5- of the Qα band at lower energy, which increases its intensity,
bromo-10,20-diphenylporphyrin with 4-nitrophenylacety- in comparison with that of [Zn(DPP)]·THF and H2DPP
lene and [4-(dimethylamino)phenyl]acetylene, respectively. (Table 1). The red shift of the Soret B band is more signifi-
The reaction was carried out in anhydrous toluene and pyri- cant for 4 than for 2, in agreement with the suggested[14]
dine at 80 °C in the presence of a Pd0 catalyst, prepared in better π conjugation of the triple bond with the π core of
situ by the reaction, at 70 °C, of [Pd2(dba)3] and PPh3 in the porphyrin ring.
freshly distilled Et3N in the presence of CuI.
Differently from 5,10,15,20-tetraphenylporphyrin and its
The free base porphyrins 3 and 7 were obtained in quan- ZnII complex substituted in the β-pyrrolic position by a
titative yields by controlled demetallation, at room tempera- similar series of π-delocalised substituents,[6] the Soret B
ture, of the ZnII complexes 4 and 8 with trifluoroacetic acid band of porphyrins 1 and 3 and of their ZnII complexes 2
working in CH2Cl2 (Scheme 2).
and 4 is quite symmetric, but with an increase of the band-
width of the Soret B bands, which is further evidence of
electronic conjugation[6] (Table 1).
Electronic Absorption and Emission Spectra
Both 2 and 4 show, in the presence of pyridine, an ad-
Anderson et al.[14] have already investigated the elec- ditional small red shift of the Soret B band, and a more
tronic absorption and emission spectra of the ZnII com- significant red shift of the Qα band at lower energy, which
plexes 2 and 4, which evidence significant changes in com- is expected for the axial coordination of pyridine[9]
parison with those of the complex [Zn(DPP)]·THF of (Table 1). The ZnII complex 4 is quite insoluble, suggesting
Eur. J. Inorg. Chem. 2006, 1743–1757
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1745

T. Morotti, M. Pizzotti, R. Ugo, S. Quici, M. Bruschi, P. Mussini, S. Righetto
Table 1. Absorption (λmax) and emission (λe) spectra in CHCl3 (Soret B band in bold).
FULL PAPER
Compound
Soret B
λmax [nm] (logε)
λe [nm][c]
Half bandwidth
∆ν1/2 [cm–1 [b]
]
H2DPP[a]
874
409 (5.49), 502 (4.27), 536 (3.74), 575 (3.78),
630 (3.18)
638, 695
[ZnDPP]·THF
718
407 (5.65), 536 (4.19), 571 (2.84)
579, 627
594, 646[d][h]
673.5, 760 (sh)
(1E)-5-[2-(4Ј-Nitrophenyl)ethenyl]-10,20-H2DPP (1)
(1E)-5-[2-(4Ј-Nitrophenyl)ethenyl]10,20 ZnDPP (2)
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-H2DPP (3)
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-ZnDPP (4)
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-H2DPP (5)
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-ZnDPP (6)
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-H2DPP (7)
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-ZnDPP (8)
2606
2629
1521
n.d.[e]
2550
n.d.[f]
2514
n.d.[g]
422 (5.27), 518 (4.07), 564 (4.11), 594 (3.89),
655 (3.65)
422 (5.23), 552 (4.19), 599 (4.04)
428 (5.05), 566 (3.72), 626 (3.89)[d]
436 (5.15), 529 (3.91), 573 (4.32), 602 (3.79),
661 (3.94)
439 (5.16), 560 (3.68), 608 (4.07)
449 (5.17), 575 (3.92), 634 (4.47)[d]
410 (5.06), 521 (4.00), 579 (4.15), 665 (3.76),
747 (3.52)
647(sh), 692
671[d]
665, 765 (sh)
617, 667 (sh)
700[d]
660 (sh), 702
408 (5.26), 554 (4.27), 606 (4.32)
696
695[d]
428 (5.20), 563 (4.06), 613 (4.09)[d]
425 (5.02), 441 (sh), 523 (3.93), 587 (4.32), 673 697, 746 (sh)
(3.94)
428 (4.97), 448 (4.98), 565 (4.02), 615 (4.24)
640, 688
654, 702[d]
451 (5.14), 579 (3.87), 632 (4.30)[d]
[a] H2DPP = 10,20-diphenylporphyrin. [b] Values were obtained according to equation ∆ν1/2 = f/(4.33×10–9 ×ε) (F. L. Pilar, Elementary
Quantum Chem., McGraw-Hill Book Comp., 1968) where f is the oscillator strength and ε is the extinction coefficient of the maximum.
[c] Values obtained by irradiation at the Soret wavelength. [d] With addition of pyridine. [e] Not soluble enough for a reliable determi-
nation. [f] Highly asymmetric band. [g] Two Soret bands of relative intensities depending on the concentration. [h] In CH2Cl2 solution
from ref.[14]
strong association in the solid state.[16] In accordance it be-
comes soluble with the addition of pyridine.
When an E-ethylenic linker connects a basic dibu-
tylamino electron-donor group with the porphyrin ring as
in porphyrin 5 and its ZnII complex 6 (Table 1), all the Q
bands are notably red shifted, (notably for 5 the Qα band
at lower energy shifts from 630 to 747 nm), while the Soret
B band is not red shifted when compared with H2DPP and
[ZnDPP]·THF. Both compounds 5 and 6 show a signifi-
cantly asymmetric Soret B band, with the asymmetry
centred at lower energy (Figure 2). Interestingly the com-
pound with the same substituent bound to the β-pyrrolic
position of the 5,10,15,20-tetraphenylporphyrin or its ZnII
complex, shows a symmetric Soret B band.[6]
The ZnII complex 8 with a dimethylamino electron-do-
nor group connected through a π-acetylenic linker to the
porphyrin ring shows, in CHCl3 solution even at concentra-
tions about 10–6 , two well-separated Soret B bands, with
a relative intensity changing with the concentration (Table 1
and Figure 3, a), which is evidence of aggregation.[16]
The position of the two bands suggests that aggregation
occurs via the interaction of the donor dimethylamino
group of the π-delocalised substituent with the axial posi-
tion of another molecule, since the Soret B band at 448 nm
is typical of pentacoordination, while that at 428 nm may
be because of some unassociated species. Intermolecular ag-
gregation is also supported by the addition of traces of pyri-
dine, which produces only one symmetric Soret B band at
451 nm (Figure 3, b) that does not shift on dilution as
would be expected for an unaggregated pentacoordinate
Figure 2. (a) Electronic absorption spectrum of 5 in CHCl3,
(b) Electronic absorption spectrum of 6 in CHCl3.
species.[9]
1746
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1743–1757

Electronic Characterisation and Significant Second-Order NLO Response
FULL PAPER
Figure 3. (a) Electronic absorption spectrum of 8 in CHCl3,
(b) Electronic absorption spectrum of 8 in CHCl3 with the addition
of pyridine.
Figure 4. (a) Electronic absorption spectrum of 7 in CHCl3,
(b) Electronic absorption spectrum of 7 in CHCl3 with the addition
of trifluoroacetic acid.
Also the ZnII complex 6 shows a Soret B band with some
Both H2DPP and [ZnDPP]·THF, when irradiated at the
asymmetry (Figure 2, b), which disappears on addition of Soret B band wavelength in a CHCl3 solution, show two
pyridine suggesting a similar kind of aggregation. A slow well-separated emission bands, the one at lower energy be-
aggregation process of 6 is confirmed by an 1H NMR spec- ing slightly more intense. On the contrary the porphyrins 1,
troscopic investigation, but working at higher concentra- 3, 5 and 7 show only one major emission with shoulders at
tions (about 10–3 ). In fact, the 1H NMR spectrum of 6 in lower (1, 3, 7) or at higher energies (5). Also the ZnII com-
CDCl3 is initially the one expected for a monomeric species, plexes 2 and 4 in CHCl3 solution show one major emission,
but after one hour significant changes take place.
with much weaker emission bands at lower (4) or higher
A weaker intermolecular aggregation, probably by hy- energies (2). Only the ZnII complexes 6 and 8 show one
drogen bonding between the NH and NMe2 groups of two significant emission band and two well-separated emission
molecules, seems to occur for porphyrin 7, where two po- bands of comparable intensity (Table 1). The shift at lower
orly resolved Soret B bands are seen (Figure 4, a). On di- energy of the emission bands of 1–8 compared to those of
lution two isosbestic points were evidenced, as expected for H2DPP and [ZnDPP]·THF (Table 1) is further evidence of
the dissociation of a dimeric species, resulting in a mono- conjugation of the π core of the porphyrin ring by the π-
meric species only at about 10–6  concentration.
delocalised linker.
The addition of traces of trifluoroacetic acid to a 10–5

The ZnII complexes 2 and 4 in CHCl3 solution show,
solution of 7 in CHCl3 produced only one Soret B band after the addition of traces of pyridine, only one emission
(Figure 4, b) as expected for the protonation of the dimeth- band shifted at lower energy, while complex 6 shows the
ylamino group with disruption of the hydrogen bond and same unshifted emission band. Only the ZnII complex 8
therefore of the intermolecular aggregation involving the di- shows two well-separated emission bands as for [ZnTPP]·
methylamino group of the substituent.
THF, but at much lower energy (Table 1).
Interestingly compounds 1–3, with a π-delocalised sub-
In summary, we have added further evidence to the pro-
stituent carrying an electron-withdrawing nitro group, did posal of Anderson et al.[14] for a conjugation of the π core
not show such spectroscopic evidence for significant aggre- of the porphyrin ring, particularly when the π-delocalised
gation, for instance by π stacking,[16] at concentrations of linker carries a nitro electron-withdrawing group. In ad-
about 10–4–10–6  in CHCl3.
dition, when the linker carries a donor dibutyl or dimeth-
Eur. J. Inorg. Chem. 2006, 1743–1757
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1747

T. Morotti, M. Pizzotti, R. Ugo, S. Quici, M. Bruschi, P. Mussini, S. Righetto
FULL PAPER
ylamino group, we found evidence of significant aggrega- dation step. Continuing the potential scan beyond the first
tion by intermolecular processes involving donor amino oxidation peak, porphyrins 1 and 3 give a complex, ill-de-
groups or hydrogen bonds.
fined anodic pattern, with multiple maxima without a neat
second oxidation step, which on the contrary was easily de-
tected not only for 5,10,15,20-tetraphenylporphyrins carry-
ing the same substituents in the β-pyrrolic position,[6] but
also, although to a lesser extent, for the parent compound
H2DPP (Table 2).
Voltammetric Investigation
The electronic properties of the porphyrins 1, 3 and 5
and of their ZnII complexes 2, 4 and 6 were investigated by
cyclic voltammetry in CH2Cl2 solution. The relevant oxi-
dation and reduction peak potentials are reported in
Table 2. The investigation was not extended to 7 and 8, be-
cause of the previously discussed significant aggregation in
CHCl3 solution, in the concentration range used for the vol-
tammetric investigation (10–3 to 10–4 ).
The ZnII complex 2 shows the more usual pattern, with
two subsequent neat oxidation peaks (but with the corre-
sponding return peaks less distinguishable), while we have
been unable to finely characterise the voltammetric behav-
iour of the ZnII complex 4 because of its poor solubility.
The first oxidation peak potentials of the porphyrins 1
and 3 and of the ZnII complex 2 are similar to those of
the parent compounds H2DPP and [ZnDPP]·THF or of the
analogues carrying the same substituents in the β-pyrrolic
position of 5,10,15,20-tetraphenylporphyrin.[6] The oxidis-
ability sequence is the same, since in both series the E-
double bond of the π linker seems to negligibly transmit the
electron attractor effect of the nitro group, resulting in
slightly higher oxidisability when compared with the case
of the triple bond. This observation is an additional sup-
port to the suggested easier π conjugation of the triple bond
with the π core of the porphyrin ring.[14] The complexation
of 1 and 3 to ZnII results in the expected strong enhance-
ment of oxidisability since the charge density on the por-
phyrin ring increases (negative shift of the first oxidation
peak of 0.20 V for 2 compared with 1 and 0.17 V for 4 com-
Oxidation Processes
The anodic oxidation of the porphyrins 1 and 3 and their
ZnII complexes 2 and 4 does not take place by two monoe-
lectronic, chemically reversible steps as such oxidation does
for compounds carrying the same π-delocalised substituents
in the β-pyrrolic position of 5,10,15,20-tetraphenylporphy-
rin or its ZnII complex.[6] Porphyrins 1 and 3 show a partial
return peak of the first oxidation step that can clearly be
perceived only by reversing the potential scan immediately
after the peak and increasing the potential scan rate. This
behaviour, typical of the parent compound H2DPP, sug-
gests a short-lived radical cation formed in the first oxi-
Table 2. Oxidation and reduction peak potentials Ep/V(SCE) for porphyrins 1, 3, 5 and their ZnII complexes 2, 4, 6, and four reference
compounds (CH2Cl2 + 0.1  TBAP, glassy carbon electrode, potential scan rate: 0.2 V·s–1).
[a] H2DPP = 10,20-diphenylporphyrin. [b] Chemically reversible peaks (i.e. featuring a return peak) are underlined (in particular, a dotted
line indicates partial return peaks, which are more evident at higher scan rates). The number of exchanged electrons, when assignable, is
reported in parenthesis. Bold characters denote electrochemically reversible or quasi-reversible peaks; the corresponding half-wave poten-
i
iii
tials, E1/2, half-peak widths, (Ep – Ep/2), and distances between anodic and cathodic peak (Ep,a – Ep,c), are reported with , ii, and
superscripts, respectively. Italic characters denote shoulders or ill-defined peaks. [c] The voltammetric characterisation is poor because of
the low solubility.
1748
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1743–1757

Electronic Characterisation and Significant Second-Order NLO Response
FULL PAPER
pared with 3). This effect is only slightly lower than that to the meso position appears to enhance this donor
found for the parent compounds H2DPP and [ZnDPP]· effect.
THF and for the analogues carrying the same substituents
In conclusion, if the oxidation peaks (the first for 1, 2, 3
in the β-pyrrolic position of the 5,10,15,20-tetraphenylpor- and 4 and the second for 5 and 6) can be taken as a first
phyrin.[6] For porphyrins 1 and 3, the poor chemical revers- indication of the HOMO energy, we may conclude that its
ibility of the first step and the ill-defined anodic pattern perturbation by substitution in the β-pyrrolic position or
beyond the first oxidation peak seems to suggest instability, the meso is not so relevant when the π-delocalised substitu-
introduced by the substitution at the meso position, of the ent carries an electron-withdrawing nitro group; instead
radical cation produced by the first oxidation step. This in- such perturbation will be quite significant, leading to desta-
stability after the first electron loss is not evident when por- bilisation, if the substituent carries an electron-donor dibu-
phyrin 1 is complexed to ZnII with increased negative tylamino group, particularly when substitution occurs in
charge on the porphyrin ring, since complex 2 produces a the meso position.
more stable radical cation (possibly because it is generated
at a far less positive potential), which can evolve to the di-
Reduction Processes
cation upon loss of a second electron.
Turning to the cathodic section of the voltammetric
Porphyrin 5 and its ZnII complex 6, carrying a dibu- pattern (Table 2), the porphyrins 1 and 3 appear signifi-
tylamino electron-donor group bound through an E-ethyl- cantly more reducible than their analogues with the same
ene linker at the meso position, show a first chemically and kind of substitution in the β-pyrrolic position.[6] Complexes
electrochemically reversible oxidation peak at nearly the 2 and 4, although characterised by a lower and broader
same potential as their analogues with the same substituent pattern of the voltammetric cathodic portion, show a first
in the β-pyrrolic position of 5,10,15,20-tetraphenylporphy- reduction peak at nearly the same potential as the corre-
rin.[6] In this last case, the first oxidation step was assumed sponding porphyrins 1 and 3 (Table 2) and appear to be
to be centred out of the HOMO energies of the porphyrin slightly more reducible than their analogues with the same
π core, probably at the amino functionality.[6] This assump- substituent in the β-pyrrolic position.[6]
tion was justified by several experimental observations,
It must be noted that we cannot reliably compare the
which still hold for 5 and 6. For instance their ∆E(ox–red) potentials of the first reduction peak of 1–4 with those of
parameters are 1.83 V and 1.89 V, respectively [considering the reference, unsubstituted compounds H2DPP and
half-wave potentials, defined as (Ep,a + Ep,c)/2], which devi- [ZnDPP]·THF. In fact the cathodic portion of the voltam-
ate significantly from the expected Kadish relationship metric pattern of 1–4 is strongly influenced by the presence
[∆E(ox–red) = 2.25 0.15 V typical of the π porphyrin of the nitro group, which, when bound to a π-delocalised
core].[17] In addition, complexation is not so effective since aromatic system, is more reducible than an unsubstituted
the first oxidation peak of the ZnII complex 6 occurs at a porphyrin core {see Table 2 with potentials at –1.26 V and
potential that is only 0.070 V less positive than that of the –1.03 V for reference compounds H2DPP and [4Ј-(p-ni-
parent porphyrin 5.
trophenyl)-2,2Ј:6Ј,2ЈЈ-terpyridine], respectively}. Therefore,
If we assume for 5 and 6 that only the second oxidation when we have a nitro group interacting very closely with a
peak is centred on the porphyrin π core, as suggested by the reducible π-delocalised system, as in our case, it is particu-
acceptable agreement of their potentials with the Kadish larly hard to discriminate whether the first electron enters
relationship (see later), and as proposed for the analogues at the nitro group or the porphyrin core. The effect of the
with the same substitution in the β-pyrrolic position,[6] it nitro group on the reduction of the conjugated molecule is
seems more appropriate to discuss the oxidisability of their clearly perceived since the difference ∆E(ox–red) between the
porphyrin core by comparing the first oxidation peak po- first oxidation and the first reduction process, which, for a
tentials of unsubstituted reference compounds H2DPP strictly porphyrin ring-based redox process, should always
(1.07 V) and [ZnDPP]·THF (0.84 V) with their second oxi- be within 2.25 0.15 V (Kadish relationship),[17] for por-
dation peak potentials, i.e. 0.77 V for 5 and 0.59 V for 6 (in phyrins 1 and 3 is 1.96 V and 2.03 V, respectively. Interest-
which case the huge shift in the negative direction of the ingly, the above values of ∆E(ox–red) are significantly smaller
second peak results in a bielectronic peak accounting for than 2.18 V and 2.10 V, in quite good agreement with the
both the first and the second oxidation steps). For the ana- Kadish relationship, found for the analogues with the same
logues substituted in the β-pyrrolic position, the second oxi- substituents in the β-pyrrolic position of 5,10,15,20-tet-
dation peak potentials are 1.01 V and 0.79 V, respectively.[6] raphenylporphyrin.[6] Therefore it appears that in com-
Since the second oxidation of 5 and 6 occurs at a signifi- pounds 1–4, differently from their analogues with the same
cantly less positive potential (0.25–0.30 V) than the first substituents in the β-pyrrolic position, the nitro group
oxidation of the unsubstituted reference compounds, and strongly influences the energy of the conjugated system so
also those (0.20–0.24 V) of the analogues with the same that we have a significant shift to less negative potentials of
substitution in the β-pyrrolic position, we can conclude the first reduction peak with respect to Kadish’s rule, an
that the increased oxidisability suggests a significant do- indication of a reduction process that does not involve just
nor effect of the dibutylamino group together with a sig- the π porphyrin core. The complexity of the first reduction
nificant role of the position on the porphyrin ring (β pyr- step is, for instance, reflected by the observation that the
rolic or meso). Shifting the substituent from the β-pyrrolic first reduction peak of porphyrin 1 is bielectronic, unlike
Eur. J. Inorg. Chem. 2006, 1743–1757
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1749

T. Morotti, M. Pizzotti, R. Ugo, S. Quici, M. Bruschi, P. Mussini, S. Righetto
FULL PAPER
the classical monoelectronic first reduction peak of a nitro Dipole Moments
group bound to a π-delocalised aromatic system [see Table 2
for the properties of the first reduction peak of the reference
aromatic compound 4Ј-(p-nitrophenyl)-2,2Ј:6Ј,2ЈЈ-terpyri-
dine].
If the first reduction peak is an indication of the ten-
dency of the whole molecule to be reduced, and therefore
of the LUMO energy, we can suggest that while the shift
from the β-pyrrolic to the meso position of π-delocalised
substituents carrying a nitro group has little effect on the
HOMO energy, it has a marked effect on the LUMO en-
ergy.
When the substituent carries an electron-donor dibu-
tylamino group as in 5 and 6 or in their analogues substi-
tuted in the β-pyrrolic position,[6] the cathodic section of
the voltammetric pattern is quite similar and closely re-
sembles that of their unsubstituted reference compounds
H2DPP and [ZnDPP]·THF, with two neat monoelectronic
reduction steps, chemically and electrochemically reversible.
Considering the reduction potentials (Table 2 and ref.[6]) it
appears that the displacement of the substituent carrying a
dibutylamino group from the β-pyrrolic position to the
meso one produces a less reducible system and thus has a
marked effect on the LUMO energy, in agreement with the
previously proposed enhancement of the electron donor ef-
fect of the dibutylamino group. The same conclusion can
be applied to the reduction peaks of the ZnII complex 6,
which are also characterised by the expected shift to the
negative direction.
Differently from 5,10,15,20-tetraphenylporphyrins and
their ZnII complexes substituted in the β-pyrrolic posi-
tion,[6] the experimental determination by the Guggenheim
method[18] of the dipole moment of 10,20-diphenyl porphy-
rins and their ZnII complexes with the same substituents in
the meso position, was not always reproducible because of
insolubility or aggregation processes, the exceptions being
the porphyrins 1 and 3 (Table 3). Therefore best geometries
first, and then dipole moments, were theoretically calcu-
lated by an ab initio approach based on density functional
theory (DFT),[11,12] using an extended basis set[13] (see Exp.
Sect.).
As already suggested by the X-ray crystal structural de-
terminations or by molecular mechanics modelling using
the MM2 force field[14] our DFT calculations confirm that
the aromatic ring of porphyrin ZnII complexes with an E-
ethylenic linker, such as 2 or 6, is significantly twisted.
(Table 4).
Our calculated angle of twisting for 2 (53.1°) is slightly
higher than that obtained in the solid state by the X-ray
crystal structure determination of the analogous com-
pound, but without the nitro group in the para position
(44.7°).[14] In general our calculations have produced an an-
gle of twisting of 10,20-diphenylporphyrin ZnII complexes
with an E-ethylenic linker (about 52–54°) or an acetylenic
linker (about 10–14.5°) (Table 4) that is higher than that
calculated by the less sophisticated molecular mechanism
modelling approach using the MM2 force field (about 38°
The increased role of the donor amino group on the π for the E-ethylenic linker and about Ͻ0.1° for the acetylenic
system of the conjugated molecule is reflected, in the case linker[14]).
of substitution at the meso position, by the significant devi-
The twisting of the porphyrin and aromatic rings calcu-
ation from Kadish’s rule; for instance let us consider the lated by our DFT approach is much less significant when
second oxidation peak discussed above: ∆E(ox–red) is 1.90 V the same para-substituted aromatic rings are bound, by the
for 5 and 2.05 V for its analogue with the substituent in same linkers, to the β-pyrrolic position of 5,10,15,20-tet-
the β-pyrrolic position, compared with the regular value of raphenylporphyrin or its ZnII complex, since the optimised
2.34 V for the reference unsubstituted compound H2DPP. geometries show a twisting angle of about 20–30° for an
For the ZnII complex 6 the value of ∆E(ox–red) is 1.90 V, E-ethylenic linker (Table 4) or a much lower value for an
compared with 2.07 V for its analogue with the substituent acetylenic one.
in the β-pyrrolic position and 2.23 V for the unsubstituted
ZnDPP.
In conclusion, we have produced additional structural
evidence that the triple bond, because of the lower twisting
Table 3. Theoretical and experimental dipole moments (µ) in CHCl3 and EFISH quadratic hyperpolarisability (β1.907) measured in CHCl3
working with an incident wavelength of 1.907 µm.
[b]
Compound[a]
µtheor (µexp) [D]
β1.907 [10–30 esu][c]
(1E)-5[2-(4Ј-Nitrophenyl)ethenyl]-10,20-H2DPP (1)
(1E)-5-[2-(4Ј-Nitrophenyl)ethenyl]10,20 ZnDPP (2)
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-H2DPP (3)
7.70[d] (6.45)
7.73
67
–94
64
8.54[d] (7.26)
8.48
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-ZnDPP (4)
n.d.[e]
64
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-H2DPP (5)
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-ZnDPP (6)
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-H2DPP (7)
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-ZnDPP (8)
6.07[d]
5.89
87.5
99
112
5.77[d]
5.65
[a] H2DPP = 10,20-diphenylporphyrin. [b] ab initio DFT calculations (see Experimental Section). [c] Mean values of measurements done
at concencentrations 10–3 to 5×10–4 . [d] Mean values of the two isomers with different locations of the NH bonds. [e] not soluble
enough.
1750
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1743–1757

Electronic Characterisation and Significant Second-Order NLO Response
FULL PAPER
Table 4. Dihedral angle between the plane of the porphyrin ring and the plane of the aromatic ring (meso substituted porphyrinic
molecules).
Compound[a]
θ [º]
(1E)-5[2-(4Ј-Nitrophenyl)ethenyl]-10,20-H2DPP (1)
(1E)-5-[2-(4Ј-Nitrophenyl)ethenyl]10,20 ZnDPP (2)
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-H2DPP (3)
54.8
53.1
10.1
10.0
52.7
51.8
14.3
14.6
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-ZnDPP (4)
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-H2DPP (5)
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-ZnDPP (6)
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-H2DPP (7)
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-ZnDPP (8)
Dihedral angle between the plane of the porphyrin ring and the plane of the aromatic ring (porphyrinic molecules substituted in β-pyr-
rolic position)[6]
Compound[b]
θ [º]
(1E)-2-(4-Nitrophenyl)ethenyl-H2TPP
(1E)-2-(4-Nitrophenyl)ethenyl-ZnTPP
(1E,3E)-2-(4-Nitrophenyl)buta-1,3-dienyl-H2TPP
(1E,3E)-2-(4-Nitrophenyl)buta-1,3-dienyl-ZnTPP
(1E)-2-(4-Dibutylaminophenyl)ethenyl-H2TPP
(1E)-2-(4-Dibutylaminophenyl)ethenyl-ZnTPP
29.2
17.6
18.5
19.3
20.3
27.9
[a] H2DPP = 10,20-diphenylporphyrin. [b] H2TPP = 5,10,15,20-tetraphenylporphyrin.
of the aromatic ring should, in all cases, favour a more fac- bound to the β-pyrrolic position of 5,10,15,20-tetraphen-
ile electronic communication between the aromatic ring ylporphyrin or its ZnII complex.[6]
(and therefore the electron-withdrawing or electron-donor
groups) and the porphyrin ring.
The porphyrins 1 and 3, with a substituent carrying an
Determination of the Quadratic Hyperpolarisability by the
EFISH Technique
electron-withdrawing nitro group, and their ZnII complexes
2 and 4 show dipole moments that are higher, and com-
pletely opposite in polarity, than those of porphyrins 5 and
The determination of the second-order NLO response
7 and their ZnII complexes 6 and 8, with a substituent car- was carried out by the EFISH technique[19], which allows
rying an amino electron-donor group (Table 3), as reported the evaluation of the molecular quadratic hyperpolaris-
for porphyrins and their ZnII complexes with the same ability, βλ, from Equation (1) (see Exp. Sect.). In order to
series of substituents in the β-pyrrolic position of avoid resonance enhancement, we worked with an off reso-
5,10,15,20-tetraphenylporphyrin or its ZnII complex.[6]
nance incident wavelength, λ, of 1.907 µm. When applying
When there is excellent π conjugation between the por- Equation (1) to highly π-conjugated two-dimensional mole-
phyrin ring and the π linker, as when the linker is acetylenic, cules, such as porphyrins and phthalocyanines, the usual
the calculated dipole moments of the compounds with a approach that ignores the third-order electronic contri-
substituent carrying a nitro group are quite independent bution to γEFISH, could not be considered to be completely
(about 8.5–8.8 D) of the position of substitution on the por- reliable.[4c,20] However, for asymmetrically substituted
phyrin ring (meso or β pyrrolic).[6] However, when the nitro phthalocyanines, structurally related to the porphyrinic
or the dibutylamino groups are connected to the porphyrin chromophores investigated in this work, and carrying an
ring by an E-ethylenic linker, the calculated dipole moments aryl ethenyl or an aryl butadienyl spacer connecting the
are slightly higher when substitution occurs in the β-pyrro- phthalocyanine ring to a nitro group in the para position,
lic position (8.5–8.8 D compared with 7.70–7.73 D for a the third-order electronic contribution to γEFISH was evalu-
substituent carrying a nitro group and 6.07–5.89 D com- ated as being much smaller than the dipolar orientational
pared with 5.33–5.16 D for a substituent carrying a dibu- one and therefore negligible.[20] In accordance we reliably
tylamino group). The validity of our DFT calculated dipole neglected the third-order electronic contribution in the de-
moments is supported by the fairly good agreement be- termination of the quadratic hyperpolarisability β1.907 from
tween calculated values of porphyrins 1 and 3 and those γEFISH by Equation (1).
measured experimentally (Table 3).
According to our DFT calculations, the push-pull por-
In conclusion the DFT calculations also provide evi- phyrinic chromophores investigated in this work are charac-
dence that for compounds 1–8 the π core of the porphyrin terised by a dipole moment axis, parallel to the π linker
ring, already in the ground state, acts as an electron donor axis, as in 1D pseudo-linear organometallic push-pull chro-
if the π-delocalised substituent in the meso position carries mophores.[21] In this approximation βvec, which is the vecto-
a nitro group, while it acts as an electron acceptor if it car- rial part of the quadratic hyperpolarisability tensor, and
ries a dimethylamino or dibutylamino group. Such a trend EFISH β1.907, which is the projection of βvec along the di-
was already observed when the same substituents were pole moment axis, should coincide.
Eur. J. Inorg. Chem. 2006, 1743–1757
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1751

T. Morotti, M. Pizzotti, R. Ugo, S. Quici, M. Bruschi, P. Mussini, S. Righetto
FULL PAPER
Since our investigation of the electronic absorption spec- larger experimental error. Therefore, the values of EFISH
tra provided evidence, particularly for compounds carrying β1.907 reported in Table 3 are the mean values of many mea-
a donor amino group, of aggregation processes even at low surements carried out at concentrations about 10–4 , thus
concentrations in nondonor solvents of low polarity such discussions mainly concern their order of magnitude, rather
as CHCl3, we carried out a series of EFISH measurements than their absolute value.
in CHCl3 solutions of different concentrations of com-
The EFISH β1.907 values (Table 3) of the porphyrins 1
pound 2, which could aggregate at relatively high concen- and 3 are positive and higher than those of porphyrins car-
tration only by π-π stacking[16] and of compounds 5–8, rying the same substituents in the β-pyrrolic position of
which could aggregate by intermolecular hydrogen bonding, 5,10,15,20-tetraphenylporphyrin[6] (about 65×10–30 esu for
or by intermolecular interaction of the amino donor group 1 and 3 against 30–40×10–30 esu, respectively).
of the π-delocalised substituent with ZnII via the axial posi-
Unexpectedly, EFISH β1.907 of the ZnII complex 2 is
tion of another molecule. For the ZnII complexes 2, 4, 6 negative and with a significant absolute value, while the
and 8 the EFISH measurements were also carried out in value of EFISH β1.907 of the analogous ZnII complex with
the presence of an excess of pyridine, while for porphyrins the same substituent in the β-pyrrolic position of
5 and 7 in the presence of traces of trifluoroacetic acid in 5,10,15,20-tetraphenylporphyrin is positive and with a
order to suppress any intermolecular aggregation (Table 5). lower absolute value[6] (about –94×10–30 esu for 2 com-
In the presence of pyridine or trifluoroacetic acid, the pared with 30×10–30 esu for the analogous ZnII complex ).
product µβ1.907, obtained from γEFISH using Equation (1),
Both porphyrins 5 and 7 and their ZnII complexes 6 and
was shown not to be dependent on dilution (within the ex- 8 show, at concentrations about 10–4 , a positive and fairly
perimental error of EFISH measurements on diluted solu- high value of EFISH β1.907 (Table 3). As suggested by the
tions) while, without these additions, a significant increase significant increase of µβ1.907 by dilution (Table 5), the value
by dilution was observed as expected for an increased disso- of EFISH β1.907 of the monomeric species of 6–8 should be
ciation (Table 5). Such an effect is less relevant for the ZnII higher. For 5 and 6, the value of EFISH β1.907 is quite sim-
complex 2 confirming, as already suggested by the investi- ilar to that of the corresponding chromophores with the
gation on electronic absorption spectra, that strong aggre- same π-delocalised substituent in the β-pyrrolic position
gation occurs mainly when the π-delocalised substituent (64×10–30 esu for 5 and 87.5×10–30 esu for 6 compared
carries a donor amino group. Measurements on the ZnII with 75.7×10–30 esu and 127.5×10–30 esu).
complex 4 were done only in the presence of pyridine be-
In summary, with the exception of the anomaly of the
cause of the low solubility of the complex. Interestingly the sign of EFISH β1.907 of 2 and 4, the major differences be-
order of magnitude and the sign of the product µβ1.907 do tween the two classes of porphyrinic chromophores carry-
not change upon addition of pyridine or trifluoroacetic acid ing the same π-delocalised substituents in the β-pyrrolic[6]
(Table 5).
or meso positions is given by the higher absolute values of
Since the µβ1.907 values of the porphyrins 5 and 7 and of EFISH β1.907 when the substituent carrying a nitro electron-
the ZnII complexes 2, 6 and 8 are dependent on dilution, it withdrawing group is bound to the meso position. This lat-
is expected that monomeric species prevail at higher di- ter result is in accordance with the proposal of Ratner,
lutions (concentrations lower than 10–4 ). However, at Marks et al.,[10] who suggested that second-order NLO re-
these dilutions EFISH measurements are affected by a sponses should increase when π-delocalised substituents
Table 5. µβ1.907 values at different concentrations in CHCl3 and in CHCl3 with addition of pyridine or trifluoroacetic acid.
Compound[a]
Concentration
µβ1.907 [10–48 esu][b]
µβ1.907 [10–48 esu]
(1E)-5-[2-(4Ј-Nitrophenyl)ethenyl]-10,20-ZnDPP (2)
10–3
–630
–816
–960
n.d.[d]
–815[c]
–1015[c]
–1210[c]
–645[c]
–795[c]
525[e]
615[e]
660[e]
815[c]
930[c]
950[c]
750[e]
860[e]
940[e]
625[c]
755[c]
830[c]
5×10–4
10–4
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-ZnDPP (4)
10–3
5×10–4
10–3
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-H2DPP (5)
350
427
690
407
611
1450
514
629
1260
473
789
1630
5×10–4
10–4
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-ZnDPP (6)
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-H2DPP (7)
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-ZnDPP (8)
10–3
5×10–4
10–4
10–3
5×10–4
10–4
10–3
2×10–4
10–4
[a] H2DPP = 10,20-diphenylporphyrin. [b] In CHCl3. [c] In CHCl3 with the addition of pyridine. [d] Too insoluble for acceptable measure-
ments. [e] In CHCl3 with traces of trifluoroacetic acid.
1752
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1743–1757

Electronic Characterisation and Significant Second-Order NLO Response
FULL PAPER
with electron-acceptor groups are appended to meso elec- phores investigated by us up to now,[6] an enhanced donor
tron-deficient centres instead of β-pyrrolic electron-rich effect of the dibutylamino or dimethylamino group.
centres.
We previously suggested[6] that in porphyrinic chromo-
phores with a π-delocalised substituent carrying a nitro
group the second-order NLO response can originate mainly
Conclusions
This and our previous investigation[6] on push-pull chro-
mophores with electron-withdrawing groups connected to
the porphyrinic ring, either in the β-pyrrolic or meso posi-
tion by an arylethenyl or arylethynyl linker, provide clear
evidence for the significant π conjugation between the por-
phyrin π core and the π-delocalised substituent, as a pos-
sible origin of the second-order NLO responses by a charge
transfer process. Such conjugation seems to be more facile,
according to the voltammetric investigation, if the π-delo-
calised substituent is bound to the meso position instead of
the β-pyrrolic one. Therefore if the substitution occurs in
the meso position, the second-order NLO response is fav-
oured due to a more facile charge transfer and to the in-
creased push properties of the porphyrin ring if we assume
a rather simplified description of these new push-pull chro-
mophores as pseudo 1D push-pull systems.
Accordingly, the experimental values of EFISH β1.907 of
chromophores 1 and 3 are comparable with those of struc-
turally related organic push pull chromophores having, as
the push group, a para-dimethylamino phenyl moiety, and
are more significant than that having, as the push group, a
ferrocenyl or of a 5,10,15,20-tetraphenylporphyrin moiety
substituted in the β-pyrrolic position (Table 6).
from a charge transfer from the occupied π orbitals of the
porphyrin ring to the π*-antibonding orbitals of the π
linker in accordance with a significant π conjugation be-
tween the π core of the porphyrin ring and the π linker. In
this hypothesis the porphyrin ring acts as donor to the ac-
ceptor aromatic ring carrying the nitro group in the para
position and the charge transfer takes place in the same
direction as the dipole moment, in agreement with the posi-
tive value of the quadratic hyperpolarisability EFISH β1.907
.
For compounds 1–4, our analysis of the absorption elec-
tronic spectra failed to produce clear direct evidence of such
a charge transfer process. However, the voltammetric inves-
tigation not only has shown a significant effect on the
LUMO energy of the porphyrin ring when conjugated with
a π linker, but also revealed a more relevant effect when
substitution occurs in the meso position, supporting a shift
of the above proposed charge transfer process at lower en-
ergy, in agreement with the observed increased absolute
value of EFISH β1.907
.
However, the anomalous negative sign of EFISH β1.907
found for the ZnII complexes 2 and 4 cannot be explained
by such a charge transfer process, given that EFISH β1.907
remains positive for analogous ZnII complexes carrying the
same π-delocalised substituents in the β-pyrrolic position.[6]
At the moment only a detailed theoretical investigation,
which is underway, may produce an explanation for the Table 6. Quadratic hyperpolarisability β1.907 of some organic and
negative EFISH β1.907 found for the ZnII complexes 2 and
organometallic push-pull systems measured in CHCl3 by the EF-
ISH technique at an incident wavelength of 1.907 µm.
4, since the involvement of aggregation by π-π stacking,[16]
which occurs certainly more easily in derivatives of 10,20-
diphenylporphyrin than in those of 5,10,15,20-tetraphen-
ylporphyrin, can be discarded because the second-order
NLO response of 2 and 4 remains negative also in the pres-
ence of pyridine (Table 5), which prevents this kind of ag-
gregation.
The lack of a significant decrease of EFISH β1.907 when
the π-delocalised substituent, carrying a strong electron-do-
nor group, is moved from the electron-rich β-pyrrolic posi-
tion to the electron deficient meso position does not com-
pletely fit with the proposal of Ratner, Marks et al.[10]
.
As for the analogues of 5 and 6 with the same π-delocal-
ised substituent in the β-pyrrolic position,[6] we can tenta-
tively propose that the high and positive EFISH β1.907 values
of chromophores 5–8 originate from an enhancement of the
donor properties of the dibutylamino or dimethylamino
group, induced by the very electron-rich π system of the por-
phyrin, which produces a significant depletion of the electron
density on the donor group and thereby a reduction of the
ground state polarisation.[22] In this view there is no involve-
ment of the π core of the porphyrin ring in the excitation
process controlling the second-order NLO response.
[a] DPP = 10,20-diphenylporphyrin substituted in the meso posi-
tion; TPP = 5,10,15,20-tetraphenylporphyrin substituted in the β-
In agreement with this proposal, our voltammetric inves-
tigation has shown, in both classes of porphyrinic chromo- pyrrolic position. [b] Ref.[23] [c] Ref.[6] [d] Ref.[24]
Eur. J. Inorg. Chem. 2006, 1743–1757
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1753

T. Morotti, M. Pizzotti, R. Ugo, S. Quici, M. Bruschi, P. Mussini, S. Righetto
FULL PAPER
emission spectra were obtained in CHCl3 with a Jasco FP-777 spec-
trofluorimeter. Elemental analyses were carried out in the Analyti-
cal Laboratories of the Department of Inorganic, Metallorganic
and Analytical Chemistry of Milan University. Dipole moments
were determined in CHCl3 solution, according to the Guggenheim
method,[18] using a WTW-DM01 dipolemeter (dielectric constant)
coupled with a Pulfrich Zeiss PR2 refractometer (refractive index).
We also propose, when the substituent bound to the meso
or β-pyrrolic position carries an electron-donor group, that
the electron-rich and highly polarisable character of the
porphyrin ring may be the origin of an electron screening
of the ground-state polarisation, so that the electron-rich
ring system weakly withdraws excessive electron density
from the donor substituent through inductive effects,
thereby increasing its donor properties.[22] Thus this process
produces an enhancement of the second-order NLO re-
sponse controlled by the n Ǟ π* transition.
Cyclic Voltammetry
The cyclovoltammetric (CV) investigation was carried out using
an Autolab PGSTAT 12 potentiostat/galvanostat (EcoChemie, The
Netherlands) run by a PC with GPES software, correcting the
ohmic drop by the positive feedback technique,[27] with a glassy
carbon GC (Amel, radius 1.5 mm) as the working electrode, a plati-
num counter-electrode, and an aqueous saturated calomel electrode
In accordance with this hypothesis the value of EFISH
β1.907 of chromophores 5, 6 and 7, 8 increases by complex-
ation to ZnII (e.g. 6 Ͼ 5 and 8 Ͼ 7 see Table 3), which
produces a higher and more polarisable electron density on
the porphyrin ring as supported by the voltammetric inves- (SCE) as the operating reference electrode (the half-wave potential
of the ferrocinium|ferrocene redox couple, recommended by IU-
PAC for inter-solvent comparison of potential scales,[28] being
0.484 V when measured in CH2Cl2, added with 0.1  TBAP against
our aqueous SCE). The working solutions were made up in dichlo-
romethane CH2Cl2 (Merck, HPLC grade), added with 0.1  tetra-
butylammonium perchlorate TBAP (FLUKA) as the supporting
electrolyte, working in a cell thermostatted at 298 K. The solutions,
carefully deaerated by nitrogen bubbling, had concentrations rang-
ing from 5×10–4 to 7×10–4  (with the exception of the less soluble
porphyrin complex 4, which reaches saturation at concentrations
tigation.
Since the π-electron core of the porphyrin ring is not di-
rectly involved in the excitation process controlling the sec-
ond-order NLO response, it becomes irrelevant if the π-de-
localised substituent is appended to an electron-rich (β-pyr-
rolic) or electron-deficient (meso) position of the porphyrin
ring, as experimentally observed.
Despite the above reported agreement with a series of
experimental observations, our interpretation of the elec-
tronic origin of the second-order NLO response in the below 3×10–4 ). The optimised polishing procedure for the work-
ing GC electrode consisted of surface treatment with diamond
powder (Aldrich diameter 1 µm) on a wet cloth (DP-Nap, Struers).
series of compounds investigated in this and in our previous
work,[6] does not explain the negative sign of EFISH β1.907
of the ZnII complexes 2 and 4.
The electrochemical reversibility and electron number of each well-
defined CV peak were checked by classical tests[29] including analy-
sis of (a) the Ip vs. v1/2 characteristics; (b) the Ep vs. logv character-
istics; (c) the (Ep – Ep/2) vs. logv characteristics, and (d) the “sta-
tionary”, step-like waves obtained by convolutive analysis of the
original CV characteristics.
Probably, the two-dimensional highly polarisable π sys-
tem of the porphyrin core is far too complex for such a
relatively simple approach based just on considerations typ-
ical of traditional 1D organic or organometallic push-pull
chromophores. As already mentioned, ab initio time-de-
pendent DFT investigations are underway in our laboratory
in an effort to achieve a more profound understanding of
the electronic origin of the significant second-order NLO
response of these new push-pull chromophores based on
the porphyrin ring.
For comparison purposes, the CV characteristics of the reference
compounds H2DPP, [ZnDPP]·THF and 4Ј-(p-nitrophenyl)
-2,2Ј:6Ј,2ЈЈ-terpyridine, recorded under the same experimental con-
ditions, are reported in Table 2. The 4Ј-(p-nitrophenyl)-2,2Ј:6Ј,2ЈЈ-
terpyridine has been included as a model for the nitro group reac-
tivity.
In conclusion this and our previous work[6] have experi-
mentally confirmed an interesting role of the porphyrin ring
in push-pull chromophores, already theoretically suggested
by Ratner, Marks et al.[10]
EFISH Measurements
EFISH measurements were performed in CHCl3 solutions, as a
function of concentration, working at 1.907 µm incident wave-
lengths using a Q-switched Nd:YAG laser with 60 and 20 ns pulse
durations, manufactured by Atalaser. The 1.907 µm fundamental
wavelength was obtained by Raman shifting of the 1.064 µm emis-
sion of the Q-switched Nd:YAG laser in a high pressure hydrogen
cell (60 bar). A liquid cell with thick windows in the wedge configu-
ration was used to obtain the Maker fringe pattern (harmonic in-
tensity variation as a function of liquid cell translation).[19] In the
EFISH experiments the incident beam was synchronised with a DC
field applied to the solution containing the molecular species in
order to break the centrosymmetry of the solution. The apparatus
for the EFISH measurements was a prototype made by SOPRA
(France).
Experimental Section
All solvents and chemicals were of reagent-grade quality, purchased
commercially and used without further purification unless other-
wise stated. 10,20-diphenylporphyrin[25a] was prepared according to
literature methods from dipyrrylmethane.[25b] 4-Nitrobenzyltri-
phenylphosphonium bromide and 4-(dibutylamino)benzyltri-
phenylphosphonium chloride were prepared following the new pro-
cedure described in our previous work.[6] 5-Bromo-10,20-di-
phenylporphyrin was synthesised using the method described by
Yeung et al.[5c] 4-Nitrophenyl acetylene was prepared according to
Takahashi el al.[26] 1H NMR spectra were recorded with a Bruker
AC-300 spectrometer in CDCl3 as solvent; electronic absorption
spectra were obtained in CHCl3 with a Jasco V-530 spectrometer;
From the concentration dependence (10–3–10–4 ) of the harmonic
signal with respect to that of the pure solvent, the NLO response
1754
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1743–1757

Electronic Characterisation and Significant Second-Order NLO Response
FULL PAPER
βλ was determined (assumed to be real as the imaginary was ne-
glected) from the experimental value γEFISH, using Equation (1).
washed with MeOH to afford 248 mg (83%) of a red powder. UV/
Vis (CHCl3): λmax (logε) = 405 (5.63), 527 (4.23), 561 (3.65) nm.
C32H20CuN4 (523.5): calcd. C 73.35, H 3.82, N 10.70; found C
72.94, H 3.78, N 10.89.
(b) 5-Formyl-10,20-diphenylporphyrinatocopper(II): A solution of
dimethylformamide (1.1 mL, 14.2 mmol) and phosphoryl chloride
(1.1 mL, 12.0 mmol) was left at room temperature and under mag-
netic stirring for 30 min. After this time, a solution of 10,20-di-
phenylporphyrinatocopper() (114 mg, 0.218 mmol) in 1,2-dichlor-
oethane (50 mL) was added and the reaction mixture was stirred
at 60 °C for 2 h. The dark green solution was then diluted with a
saturated aqueous solution of CH3COONa (150 mL) and main-
tained at 60 °C whilst stirring for a further 2 h. The organic phase
was separated, washed with H2O (2×100 mL), dried with Na2SO4
and evaporated in vacuo. Purification of the residue by column
chromatography (silica gel, CH2Cl2/n-hexane, 7:3) afforded 96 mg
(80%) of pure product. UV/Vis (CHCl3): λmax (logε) = 417 (5.42),
547 (3.97), 591 (4.15) nm. C33H20CuN4O (551.5): calcd. C 71.80,
H 3.63, N 10.15; found C 72.30, H 3.76, N 10.30.
(1)
where γEFISH is the sum of a cubic electronic contribution γ(–2ω;
ω, ω, 0) and of a quadratic orientational contribution µβλ (–2ω; ω,
ω)/5kT; µ is the ground state dipole moment and βλ is the projec-
tion along the dipole moment direction of the vectorial component
βvec of the tensorial quadratic hyperpolarisability working with the
incident wavelength, λ.
Computational Methods
Geometry optimisation and population analysis of the compounds
1–8 were carried out in the framework of the density functional
theory (DFT) by using the BP86 functional[12] and an all-electron
valence triple-ζ basis set with polarisation functions on all atoms
(TZVP).[13] Values of the dipole moments moduli and components
were computed using the Turbomole suite of programs[30] in con-
nection with the resolution of identity (RI) approximation.[31]
(c) 5-Formyl-10,20-diphenylporphyrin:
A solution of 5-formyl-
10,20-diphenylporphyrinatocopper() (251 mg, 0.455 mmol) in a
mixture of H2SO4/TFA (1:10 v/v, 15 mL) was stirred for 10 min at
room temperature, and then quenched by adding an excess of ice.
The product was extracted with CH2Cl2 (2×50 mL), the organic
phase was washed with aqueous NaOH (20%, 50 mL) and with
H2O (2×50 mL), the solvent was evaporated to dryness and the
residue was washed with pentane to afford 220 mg (quantitative
Synthesis of Porphyrins
For the numering of atoms see below.
1
yield) of a violet powder. H NMR (300 MHz, CDCl3): δ = 12.62
(s, 1 H, CHO), 10.28 (s, 1 H, Hmeso), 10.10 (d, J = 4.98 Hz, 2 H,
Hpyrrolic), 9.30 (d, J = 4.44 Hz, 2 H, Hpyrrolic), 9.08 (d, J = 4.98 Hz,
2 H, Hpyrrolic), 8.91 (d, J = 4.44 Hz, 2 H, Hpyrrolic), 8.22 (m, 4 H,
Ho), 7.85 (m, 6 H, Hm,p), –2.41 (s, 2 H, NH) ppm. UV/Vis (CHCl3):
λmax (logε) = 420 (5.35), 520 (4.01), 559 (3.95), 593 (3.77), 648
(3.81) nm. C33H22N4O (490): calcd. C 80.82, H 4.49, N 11.43;
found C 80.65, H 4.36, N 11.60.
(1E)-5-[2-(4Ј-Nitrophenyl)ethenyl]-10,20-diphenylporphyrin (1):
A
suspension
0.484 mmol),
of
5-formyl-10,20-diphenylporphyrin
4-nitrobenzyltriphenylphosphonium
(238 mg,
bromide
(670 mg, 1.407 mmol) and solid NaOH 20–40 mesh beads (400 mg,
10 mmol) in anhydrous 1,2-dichloroethane (25 mL) was refluxed,
with magnetic stirring and under a nitrogen atmosphere for 4 h.
After this time the solvent was evaporated to dryness, the residue
was purified by column chromatography (silica gel CH2Cl2/n-hex-
ane, 7:3). The product obtained from the column chromatography
was first washed with MeOH and then crystallised from CH2Cl2/
MeOH to give 145 mg (49%) of pure 1. 1H NMR (300 MHz,
CDCl3): δ = 10.17 (s, 1 H, Hmeso), 9.89 (d, J = 15.87 Hz, 1 H,
=CH), 9.50 (d, J = 4.83 Hz, 2 H, Hpyrrolic), 9.30 (d, J = 4.65 Hz, 2
H, Hpyrrolic), 8.97 (d, J = 4.77 Hz, 4 H, Hpyrrolic), 8.43 (d, J =
8.76 Hz, 2 H, H**), 8.24 (m, 4 H, Ho), 8.05 (d, J = 8.76 Hz, 2 H,
H*), 7.80 (m, 6 H, Hm,p), 7.42 (d, J = 15.93 Hz, 1 H, CH=), –2.78
(s, 2 H, NH) ppm. UV/Vis (CHCl3): λmax (logε) = 422 (5.27), 518
(4.07), 564 (4.11), 594 (3.89), 655 (3.65) nm. C40H27N5O2 (609):
calcd. C 78.82, H 4.43, N 11.49; found C 78.65, H 4.30, N 11.48.
5-Formyl-10,20-diphenylporphyrin: This porphyrin was prepared by
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-diphenylporphyrin (3): Trifluoro-
acetic acid (2 mL) was added in four portions to a suspension of
5-[2-(4Ј-nitrophenyl)ethynyl]-10,20-diphenylporphyrinatozinc() (4)
(100 mg, 0.149 mmol) in CH2Cl2 (50 mL) and the reaction mixture
was stirred at room temperature for 18 h. After this time, water
(30 mL) was added and the mixture was carefully neutralised with
a saturated aqueous solution of NaHCO3. The organic phase was
separated, dried with MgSO4 and the solvent evaporated in vacuo
the following three steps:[15]
(a) 10,20-Diphenylporphyrinatocopper(II): A slightly warm solution
of Cu(OAc)2·H2O (125 mg, 0.627 mmol) in MeOH (15 mL) was
slowly added to a refluxing solution of 10,20-diphenylporphyrin
(264 mg, 0.570 mmol) in CHCl3 (60 mL). After two hours the red
reaction mixture was cooled and the solvents evaporated to dryness
in vacuo. The red residue was collected with MeOH, filtered and
Eur. J. Inorg. Chem. 2006, 1743–1757
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1755

T. Morotti, M. Pizzotti, R. Ugo, S. Quici, M. Bruschi, P. Mussini, S. Righetto
FULL PAPER
to afford 90 mg (quantitative yield) of porphyrin 3. 1H NMR
(300 MHz, CDCl3): δ = 10.21(s, 1 H, Hmeso), 9.76 (d, J = 4.83 Hz,
552 (4.19), 599 (4.04) nm; (CHCl3 with addition of pyridine): λmax
(log ε) = 428 (5.05), 566 (3.72), 626 (3.89) nm. C40H25N5O2Zn
2 H, Hpyrrolic), 9.30 (d, J = 4.62 Hz, 2 H, Hpyrrolic), 9.01 (d, J = (672.4): calcd. C 71.39, H 3.72, N 10.41; found C 71.50, H 3.61, N
4.80 Hz, 2 H, Hpyrrolic), 8.95 (d, J = 4.62 Hz, 2 H, Hpyrrolic), 8.41(d, 10.30.
J = 8.76 Hz, 2 H, H**), 8.23 (m, 4 H, Ho), 8.13 (d, J = 8.73 Hz, 2
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-diphenylporphyrin-
H, H*), 7.81 (m, 6 H, Hm,p), –2.54 (s, 2 H, NH) ppm. UV/Vis
atozinc(II) (6): This compound was obtained in quantitative yield
(CHCl3): λmax (logε) = 436 (5.15), 529 (3.91), 573 (4.32), 602 (3.79),
661 (3.94) nm. C40H25N5O2 (607): calcd. C 79.08, H 4,12, N 11.53;
found C 78.96, H 4.16, N 11.53.
as a violet powder. 1H NMR (300 MHz, CDCl3): δ = 10.06 (s, 1 H,
Hmeso), 9.61 (d, J = 4.65 Hz, 2 H, Hpyrrolic), 9.48 (d, J = 15.75 Hz, 1
H, CH=), 9.28 (d, J = 4.53 Hz, 2 H, Hpyrrolic), 8.99 (d, J = 4.53 Hz,
(1E)-5-[2-(4Ј-Dibutylaminophenyl)ethenyl]-10,20-diphenylporphyrin
(5): A suspension of 5-formyl-10,20-diphenylporphyrin (103 mg,
0.211 mmol), 4-(dibutylamino)benzyltriphenylphosphonium chlo-
ride (185 mg, 0.358 mmol) and solid NaOH 20–40 mesh beads
(337 mg, 8.436 mmol) in anhydrous CH2Cl2 (20 mL) was stirred at
room temperature, under a nitrogen atmosphere, for 4 h. After this
time the solvent was evaporated and the residue was purified by
column chromatography (silica gel, CH2Cl2/n-hexane, 7:3) to afford
87 mg (59%) of 5 as a violet powder. 1H NMR (300 MHz, CDCl3):
δ = 10.08 (s, 1 H, Hmeso), 9.56 (m, 3 H, CH, Hpyrrolic), 9.27 (d, J =
4.53 Hz, 2 H, Hpyrrolic), 8.96 (d, J = 4.53 Hz, 2 H, Hpyrrolic), 8.91
(d, J = 4.68 Hz, 2 H, Hpyrrolic), 8.26(m, 4 H, Ho), 7.82, (m, 8 H,
H*, Hm,p), 7.27 (d, J = 16.0 Hz, 1 H, =CH), 6.86 (d, J = 8.58 Hz,
2 H, H**), 3.43 (t, J = 7.45 Hz, 2 H, CH2Butyl), 1.72 (m, 2 H,
CH2Butyl), 1.46 (m, 2 H, CH2Butyl), 1.05 (t, J = 7.23 Hz, 3 H,
CH3Butyl), –2.62 (s, 2 H, NH) ppm. UV/Vis (CHCl3): λmax (logε) =
353 (4.34), 410 (5.06), 521 (4.00), 579 (4.15), 665 (3.76), 747
(3.52) nm. C48H45N5 (691): calcd. C 83.36, H 6.51, N 10.13; found
C 82.95, H 6.57, N 9.97.
4 H, Hpyrrolic), 8.96 (d, J = 4.65 Hz, 2 H, Hpyrrolic), 8.21 (m, 4 H,
Ho), 7.77 (m, 8 H, H*, Hm,p), 7.25 (d, J = 15.75 Hz, 1 H, =CH),
6.82 (d, J = 8.55 Hz, 2 H, H**), 3.40 (t, J = 7.52 Hz, 2 H, CH2Butyl),
1.67 (m, 2 H, CH2Butyl), 1.43 (m, 2 H, CH2Butyl), 1.02 (t, J =
7.28 Hz, 3 H, CH3Butyl) ppm. UV/Vis (CHCl3): λmax (logε) = 349
(4.35), 408 (5.26), 554 (4.27), 606 (4.32) nm; (CHCl3 with addition
of pyridine): λmax (logε) = 428 (5.20), 563 (4.06), 613 (4.09) nm.
C48H43N5Zn (754.4): calcd. C 76.35, H 5.70, N 9.28; found C 75.99,
H 5.68, N 9.41.
5-[2-(4Ј-Nitrophenyl)ethynyl]-10,20-diphenylporphyrinatozinc(II) (4):
The catalytic solution was prepared by heating Pd2(dba)3 (9.5 mg,
9.9·10–3 mmol), PPh3 (21.8 mg, 8.3·10–2 mmol) and CuI (7.8 mg,
4.1 × 10–2 mmol) in freshly distilled Et3N (25 mL) at 70 °C for
30 min under a nitrogen atmosphere. This solution was transferred
into a solution of 5-bromo-10,20-diphenylporphyrinatozinc()
(100 mg, 0.165 mmol) and (4-nitrophenyl)acetylene (74 mg,
0.50 mmol) in anhydrous toluene (25 mL) and pyridine (0.6 mL)
under a nitrogen atmosphere, and the mixture was heated at 80 °C
and stirred for 1 h. After this time the solution was cooled and
filtered through silica eluting with toluene and the solvent was
evaporated to dryness. The residue was washed several times with
small amounts of MeOH and CHCl3 alternately to afford 93 mg
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-diphenylporphyrin (7):
Trifluoroacetic acid (2 mL) was added in four portions to a suspen-
sion of 5-[2-(4Ј-dimethylaminophenyl)ethynyl]-10,20-diphenylpor-
phyrinatozinc() (8) (100 mg, 0.150 mmol) in CH2Cl2 (50 mL) and
the reaction mixture was stirred at room temperature for 18 h. Af-
ter this time, water (30 mL) was added and the mixture was care-
fully neutralised with a saturated aqueous solution of NaHCO3.
The organic phase was separated, dried with MgSO4 and the sol-
vent evaporated in vacuo to afford 90 mg (quantitative yield) of
porphyrin 7. 1H NMR (300 MHz, CDCl3): δ = 10.09 (s, 1 H,
Hmeso), 9.81 (d, J = 4.74 Hz, 2 H, Hpyrrolic), 9.24 (d, J = 4.65 Hz, 2
H, Hpyrrolic), 8.93 (d, J = 4.35 Hz, 2 H, Hpyrrolic), 8.92 (d, J =
4.20 Hz, 2 H, Hpyrrolic), 8.23 (m, 4 H, Ho), 7.91 (d, J = 8.76 Hz, 2
H, H*), 7.79 (m, 6 H, Hm,p), 6.87 (d, J = 8.82 Hz, 2 H, H**), 3.11
(s, 6 H, NCH3), –2.47 (s, 2 H, NH) ppm. UV/Vis (CHCl3): λmax
(log ε) = 425 (5.02), 441, 523 (3.93), 587 (4.32), 673 (3.94) nm.
C42H31N5 (605): calcd. C 83.31, H 5.12, N 11.57; found C 82.98,
H 5.17, N 11.46.
1
(85%) of pure 4. H NMR (300 MHz, CDCl3 + Pyridine d5): δ =
10.12 (s, 1 H, Hmeso), 9.74 (d, J = 4.59 Hz, 2 H, Hpyrrolic), 9.26 (d,
J = 4.47 Hz, 2 H, Hpyrrolic), 8.99 (d, J = 4.59 Hz, 2 H, Hpyrrolic),
8.92 (d, J = 4.44 Hz, 2 H, Hpyrrolic), 8.36 (d, J = 7.47 Hz, 2 H,
H**), 8.20 (m, 4 H, Ho), 8.08 (d, J = 7.47 Hz, 2 H, H*), 7.75 (m,
6 H, Hm,p) ppm. UV/Vis (CHCl3): λmax (logε) = 439 (5.16), 560
(3.68), 608 (4.07) nm; (CHCl3 with addition of pyridine):
λmax (logε) = 449 (5.17), 575 (3.92), 634 (4.47) nm. C40H23N5O2Zn
(670.4): calcd. C 71.60, H 3.43, N 10.44; found C 71.51, H 3.48, N
10.32.
5-[2-(4Ј-Dimethylaminophenyl)ethynyl]-10,20-diphenylporphyrinato-
zinc(II) (8): The catalytic solution was prepared by heating a solu-
tion of Pd2(dba)3 (9.5 mg, 9.9·10–3 mmol), PPh3 (21.8 mg,
8.3·10–2 mmol) and CuI (7.8 mg, 4.1·10–2 mmol) in freshly distilled
Et3N (25 mL) at 70 °C for 30 min under a nitrogen atmosphere.
This solution was transferred into a solution of 5-bromo-10,20-
diphenylporphyrinatozinc() (100 mg, 0.165 mmol) and [4-(dimeth-
ylamino)phenyl]acetylene (73 mg, 0.50 mmol) in anhydrous toluene
(25 mL) and pyridine (0.6 mL) under a nitrogen atmosphere, and
the mixture was heated at 80 °C and stirred for 1.5 h. After this
time the solution was cooled, filtered through silica gel, eluted with
toluene, and the solvent was evaporated to dryness. The residue
was washed several times with small amounts of MeOH and CHCl3
Synthesis of ZnII Complexes
ZnII complexes 2 and 6 were synthesised following the general pro-
cedure described in the literature.[32] In a typical preparation the
free porphyrin (100 mg) was dissolved in slightly warmed CHCl3
(20 mL) and a solution of Zn(OAc)2·2H2O (free porphyrin/metal
salt = 1:1.1) in MeOH (10 mL) was then added. The mixture was
refluxed for 1 h, was then evaporated to dryness and the residue
was collected with MeOH, filtered and washed with MeOH.
1
(1E)-5-[2-(4Ј-Nitrophenyl)ethenyl]-10,20-diphenylporphyrinatozinc(II
)
alternately to afford 65 mg (59%) of pure 8. H NMR (300 MHz,
(2): This compound was obtained in quantitative yield as a violet
powder. 1H NMR (300 MHz, CDCl3): δ = 10.10 (s, 1 H, Hmeso),
9.89 (d, J = 16.32 Hz, 1 H, =CH), 9.56 (d, J = 4.41 Hz, 2 H,
CDCl3 with addition of Pyridine d5): δ = 10.05 (s, 1 H, Hmeso), 9.83
(d, J = 4.56 Hz, 2 H, Hpyrrolic), 9.25 (d, J = 4.47 Hz, 2 H, Hpyrrolic),
8.95 (d, J = 4.92 Hz, 2 H, Hpyrrolic), 8.94 (d, J = 4.98 Hz, 2 H,
Hpyrrolic), 9.35 (d, J = 4.29 Hz, 2 H, Hpyrrolic), 9.04 (d, J = 3.99 Hz, Hpyrrolic), 8.23 (m, 4 H, Ho), 7.92 (d, J = 8.82 Hz, 2 H, H*), 7.77
4 H, Hpyrrolic), 8.43 (d, J = 7.59 Hz, 2 H, H**), 8.22 (m, 4 H, Ho), (m, 6 H, Hm,p), 6.89 (d, J = 8.88 Hz, 2 H, H**), 3.11 (s, 6 H,
8.05 (d, J = 7.74 Hz, 2 H, H*), 7.79 (m, 6 H, Hm,p), 7.41 (d, J = NCH3) ppm. UV/Vis (CHCl3): λmax (logε) = 428 (4.97), 448 (4.98),
16.24 Hz, 1 H, CH=) ppm. UV/Vis (CHCl3): λmax (logε) 422 (5.23), 565 (4.02), 615 (4.24) nm; (CHCl3 with addition of pyridine):
1756
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1743–1757

Electronic Characterisation and Significant Second-Order NLO Response
FULL PAPER
[9]
a) M. Nappa, J. S. Valentine, J. Am. Chem. Soc. 1978, 100,
5075–5080; b) A. Antipas, J. W. Buchler, M. Gouterman, P. D.
Smith, J. Am. Chem. Soc. 1978, 100, 3015–3024.
I. D. L. Albert, T. J. Marks, M. A. Ratner, Chem. Mater. 1998,
10, 753–762.
λmax (logε) = 451 (5.14), 579 (3.87), 632 (4.30) nm. C42H29N5Zn
(668.4): calcd. C 75.40, H 4.34, N 10.47; found C 75.47, H 4.30, N
10.33.
[10]
Acknowledgments
[11]
[12]
[13]
[14]
a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. Lee,
W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
a) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100; b) J. P. Per-
dew, Phys. Rev. B 1986, 33, 8822–8824.
A. Schafer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100,
5829–5835.
T. E. O. Screen, I. M. Blake, L. H. Rees, W. Clegg, S. J. Bor-
wick, H. L. Anderson, J. Chem. Soc., Perkin Trans. 1 2002,
320–329.
K. Susumu, T. Shimidzu, K. Tanaka, H. Segawa, Tetrahedron
Lett. 1996, 37, 8399–8402.
W. I. White, The Porphyrins, vol. V, pp. 303–335 (Ed.: D. Dol-
phin), Acad. Press, New York, 1978.
J. H. Furhop, K. M. Kadish, D. G. Davis, J. Am. Chem. Soc.
1973, 95, 5140–5147.
E. A. Guggenheim, Trans. Faraday Soc. 1949, 45, 203.
This work was supported by the Ministero dellЈIstruzione, dellЈU-
niversità e della Ricerca [PRIN 1999 – Research Title: Molecole
per Materiali Funzionali Nanostrutturati: Composti organometallici
con proprietà ottiche non lineari (NLO) per materiali nanostruttur-
ati; FIRB 2001 – Research Title: Nano-organisation of hybrid or-
ganic-inorganic molecules with magnetic and non linear optical prop-
erties] and by the Consiglio Nazionale delle Ricerche. We thank
Mr. Pasquale Illiano for the NMR spectroscopy measurements.
[15]
[16]
[17]
[1] a) J. Zyss, Molecular Non Linear Optics: Materials, Physics and
Devices, Ed Academic Press, Boston 1993; b) B. E. A. Saleh,
M. C. Telch, Fundamentals of Photonics, Wiley: New York
1991.
[2] a) D. R. Kanis, M. A. Ratner, T. J. Marks, Chem. Rev. 1994,
94, 195–242; b) P. M. Burland, R. D. Miller, C. A. Walash,
Chem. Rev. 1994, 94, 31–75; c) T. J. Marks, M. A. Ratner, An-
gew. Chem. Int. Ed. Engl. 1995, 34, 155–173.
[3] a) I. R. Whittal, A. M. McDonagh, M. G. Humphrey, Adv. Or-
ganomet. Chem. 1998, 42, 291–362; b) S. Di Bella, Chem. Soc.
Rev. 2001, 30, 355–366; c) N. J. Long, Angew. Chem. Int. Ed.
Engl. 1995, 34, 21–38; d) B. J. Coe, Comprehensive Coordination
Chemistry II (Eds.: J. A. Meyer, T. J. McCleverty); Elsevier Per-
gamon: Oxford, U. K., 2004; vol. 9, pp. 621–687.
[4] a) K. S. Suslick, C. T. Chen, G. R. Meredith, L. T. Cheng, J.
Am. Chem. Soc. 1992, 114, 6928–6930; b) H. Chou, C.-T. Chen,
K. F. Stork, P. W. Bohn, K. S. Suslick, J. Phys. Chem. 1994, 98,
383–385; c) M. Pizzotti, R. Ugo, E. Annoni, S. Quici, I. Le-
doux-Rak, G. Zerbi, M. Del Zoppo, P. Fantucci, I. Invernizzi,
Inorg. Chim. Acta 2002, 340, 70–80.
[5] a) S. M. Le Cours, H. W. Guan, S. G. Di Magno, C. H. Wang,
M. J. Therien, J. Am. Chem. Soc. 1996, 118, 1497–1503; b) L.
Karki, F. W. Vance, J. H. Hupp, S. M. Le Cours, M. J. Therien,
J. Am. Chem. Soc. 1998, 120, 2606–2611; c) M. Yeung, A. C. H.
Ng, M. G. D. Drew, E. Vorpagel, E. M. Breitung, R. J. McMa-
hon, D. K. P. Ng, J. Org. Chem. 1998, 63, 7143–7150; d) M.
Pizzotti, E. Annoni, R. Ugo, S. Bruni, S. Quici, P. Fantucci,
M. Bruschi, G. Zerbi, M. Del Zoppo, J. Porphyrins Phthalocy-
anines 2004, 8, 1311–1324.
[6] E. Annoni, M. Pizzotti, R. Ugo, S. Quici, T. Morotti, M. Brus-
chi, P. Mussini, Eur. J. Inorg. Chem. 2005, 3857–3874.
[7] a) G. de La Torre, P. Vàsquez, F. Agullò-Lòpez, T. Torres, J.
Mater. Chem. 1998, 8, 1671–1683 and references cited therein;
b) G. de La Torre, T. Torres, F. Agullò-Lòpez, Adv. Mater.
1997, 9, 265–269 and references cited therein; c) G. de la Torre,
P. Vazquez, F. Agullò-Lòpez, T. Torres, Chem. Rev. 2004, 104,
3723–3750.
[18]
[19]
a) I. Ledoux, J. Zyss, Chem. Phys. 1982, 73, 203–213; b) D. J.
Williams, Angew. Chem. Int. Ed. Engl. 1984, 23, 690–703 and
references cited therein.
G. Rojo, G. de la Torre, J. Garcia-Ruiz, I. Ledoux, T. Torres,
J. Zyss, F. Agullo-Lopez, Chem. Phys. 1999, 245, 27–34 and
references cited therein.
[20]
[21] a) D. Roberto, R. Ugo, S. Bruni, E. Cariati, F. Cariati, P. Fan-
tucci, S. Quici, I. Ledoux, J. Zyss, Organometallics 2000, 19,
1775–1788; b) E. Lucenti, E. Cariati, C. Dragonetti, L. Man-
assero, F. Tessore, Organometallics 2004, 23, 687–692.
[22] I. D. L. Albert, T. J. Marks, M. A. Ratner, J. Am. Chem. Soc.
1997, 119, 6575–6582.
[23] L. T. Cheng, W. Tam, S. H. Stevenson, G. R. Meredith, G.
Rikken, S. R. Marder, J. Phys. Chem. 1991, 95, 10631–10643.
[24] D. R. Kanis, M. A. Ratner, T. J. Marks, J. Am. Chem. Soc.
1992, 114, 10338–10357.
[25] a) S. G. Di Magno, V. S. Y. Lin, M. J. Therien, J. Org. Chem.
1993, 58, 5983–5993; b) M. J. Plater, S. Aiken, G. Bourhill, Tet-
rahedron 2002, 58, 2405–2413.
[26] S. Takahashi, Y. Kuroyama, K. Sorogashira, N. Hagihara, Syn-
thesis 1980, 627–629.
[27] D. Britz, Electroanal. Chem. 1978, 88, 309–352.
[28] G. Gritzner, J. Kuta, Pure Appl. Chem. 1984, 56, 461–466.
[29] A. J. Bard, L. R. Faulkner, Electrochemical Methods. Funda-
mentals and Applications, Wiley, New York, 2002.
[30] R. Ahlrichs, M. Bar, M. Haser, C. Horn, C. Kolmel, Chem.
Phys. Lett. 1989, 162, 165–169.
[31] a) K. Eichkorn, O. Treutler, H. Öhm, M. Haser, R. Ahlrichs,
Chem. Phys. Lett. 1995, 240, 283–289; b) K. Eichkorn, F. Weig-
end, O. Treutler, R. Ahlrichs, Theor. Chem. Acc. 1997, 97, 119–
124.
[32] A. D. Adler, F. R. Longo, F. Kampas, J. Kim, J. Inorg. Nucl.
Chem. 1970, 32, 2443–2445.
[8] D. A. Offord, S. B. Sachs, M. S. Ennis, T. A. Eberspacher, J. H.
Griffin, C. E. D. Chidsey, J. P. Collman, J. Am. Chem. Soc.
1998, 120, 4478–4487.
Received: November 10, 2005
Published Online: March 6, 2006
Eur. J. Inorg. Chem. 2006, 1743–1757
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1757