Theoretical study of electronic structure and absorption spectra of diacid and zinc species of series of meso-phenylporphyrins

Article abstract of DOI:10.1016/j.saa.2011.04.085

DFT and TDDFT calculations with density functionals (PBE1PBE, B3LYP, and PBEPBE) have been employed in a study of HCl-acidified diacids of a series of meso-phenylporphyrins. This study aims to clarify the influence of conformational distortion, meso-pheny

Full text of DOI:10.1016/j.saa.2011.04.085

Spectrochimica Acta Part A 79 (2011) 1449–1460
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
Theoretical study of electronic structure and absorption spectra of diacid and
zinc species of series of meso-phenylporphyrins
∗
Ying-Hui Zhang , Li-Hong Zhao, Wen-Juan Ruan, Yao Xu
Department of Chemistry, Nankai University, 94 Weijin Road, Tianjin 300071, China
a r t i c l e i n f o
a b s t r a c t
Article history:
DFT and TDDFT calculations with density functionals (PBE1PBE, B3LYP, and PBEPBE) have been employed
Received 18 January 2011
Received in revised form 26 April 2011
Accepted 29 April 2011
in a study of HCl-acidified diacids of a series of meso-phenylporphyrins. This study aims to clarify the influ-
−
ence of conformational distortion, meso-phenyl substituents, and counterion Cl on absorption spectra
of porphyrin derivatives. Calculations indicate that all three factors increase the MO’s level and decrease
the Gouterman HOMO–LUMO gap; this, further, brings about the redshift of absorption band. In compar-
ison with experimental methods, the PBE1PBE method produces a more credible description of UV–vis
spectra than other two methods. TDDFT calculation with the PBE1PBE method indicates that the elec-
tronic effect of the meso-phenyl group is dominant for the spectral redshift of porphyrin diacid series
as observed in zinc porphyrins. The redshift of the B band of porphyrin diacids is primarily caused by
an electron-donating effect of the meso-phenyl group, whereas the Q band is more sensitive to the ␲
electron delocalization effect. The counterion is indispensable in a theoretical study of electronic and
spectral structure of porphyrin diacids.
Keywords:
meso-Phenylporphyrins
Electronic structure
Absorption spectra
Redshift
©
2011 Elsevier B.V. All rights reserved.
1
. Introduction
tant insights, these studies focus only on a single aspect and,
therefore, lack comprehensive and systematic understanding of the
spectral features of porphyrin diacids.
Several spectroscopic methods have proved that peripheral
substitution has a significant influence on electronic and spec-
tral features of porphyrins [1–5]. good understanding of
In this article, we present a theoretical analysis of the influ-
ence of several typical factors (including conformational distortion,
meso-phenyl substituent, and counterion Cl ), on electronic struc-
A
−
structure–spectra relationship is very important in the design of
novel artificial materials. Absorption spectroscopy has been used
widely in the study of structural variation in porphyrin rings. As a
general phenomenon, the spectral shift has attracted much atten-
tion and has been ascribed to the change of electronic structure
induced by several factors, such as the electronic effect of sub-
stituent, out-of-plane distortion, and the change of bond length
and bond angle—called in-plane nuclear reorganization (IPNR) by
DiMagno and co-workers [6,7]. The combined action of several fac-
tors, frequently, confounds the identification of the primary cause
of spectral shift; therefore, this has given rise to certain debates in
the past decades [6–10].
Porphyrin diacids, as an important derivative, have attracted
much attention in the past decades because of their marked out-of-
plane distortion and novel spectral features with regard to neutral
porphyrin [11–15]. Studies on the electronic structure and spec-
tral variation of porphyrin diacids have been carried out by certain
groups of researchers [16–19]. Although providing certain impor-
ture and absorption spectra of a series of meso-phenylporphyrin
diacids (structure depicted in Fig. 1). Zinc complexes were also
studied because their 4-fold symmetry and near-planar geome-
try were expected to generate beneficial information. Meanwhile,
PBE1PBE, B3LYP, and PBEPBE methods were compared, particularly
in spectra simulation. Comparison with experimental methods
indicates that the PBE1PBE method produces a more credible
description of absorption structure, particularly for the Q band,
of porphyrin diacids than the B3LYP and PBEPBE methods. Fur-
ther, analysis indicates that the electronic effect of the meso-phenyl
group is the main cause for the absorption redshift of porphyrin
diacids series with increased meso-phenyl substitution.
2. Experimental
2.1. Chemicals and preparation of the zinc and diacid samples
Zinc porphyrins were synthesized by boiling corresponding
free-base porphyrin [20] with zinc acetate in a CH Cl /methanol
2
2
∗ Corresponding author. Fax: +86 22 2350 8841.
E-mail address: zhangyhi@nankai.edu.cn (Y.-H. Zhang).
solution (volume ratio: 2:1). The zinc porphyrin thus obtained
was purified on a silica gel column with chloroform as the eluant
1
386-1425/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2011.04.085

1
450
Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
C12 C13
of the acid used because of the close interaction between two
R3
R
4
moieties in solution as well as in the solid state; however, the
C11
C14
2+
study of H4P
without consideration of counterion has also
C10
C9
C
15
N23
X
Y
been reported [28]. Similarly, our calculations in the present
study also reveal marked difference in spectral structure between
C
16
C8
C
17
18
−
calculation with and without consideration of Cl ion for meso-
N22
M
N
24
phenylporphyrin diacids. The predicted spectra of the former
scenario coincide better with the experiment than the latter sce-
nario. The normal-coordinate structural decomposition (NSD) data
were obtained by online calculation utilizing the NSD Computation
Engine (version 3.0) provided on the Website of the Shelnutt group
C7
C
C6
N21
C5
C
20
C4
C1
R2
R1
[
29].
C3 C2
3
. Results and discussion
MTPP:
R
1
=R
=H, R
= R =H, R
= R = R =H, R
= R = R = R =H
2
=R
3
=R
4
=phenyl
= R =phenyl
= R =phenyl
=phenyl
3.1. The geometric structure of porphyrin diacids
MTrPP:
MDPP:
MMPP:
MP:
R
1
2
= R
3
4
Optimization of the geometric structure of zinc porphyrin by
application of the B3LYP method has been reported in a pre-
vious study [30]; the general conclusion was that the in-plane
distortion caused by the meso-phenyl substitution is significant,
whereas the out-of-plane distortion is slight. New calculation
with the PBE1PBE/6-31G(d)/LanL2DZ (see Table 1), B3LYP/6-
R
1
3
2
4
3
1G(d)/LanL2DZ (see Table S1), and PBEPBE/SVP methods (see
R
1
2
3
4
Table S2) generate similar results. The crystal data [31] and cal-
culation data, reported by another group of researchers [32], for
ZnTPP are also listed in Table 1 for comparison.
R
1
2
3
4
The structural parameters of HCl-acidified porphyrin diacids
optimized by the PBE1PBE/6-31G(d) method are summarized in
Table 2, wherein the X-ray data [33] and the calculated data of
H4TPP2+(Cl )2 [21] reported by other groups of researchers are
provided for comparison. The results of our calculation predict a
geometric structure similar to that reported in the published lit-
erature [21], with the exception of slight differences pertaining to
the Cl· · ·H–N hydrogen bond. For all porphyrin diacids, the sad-
dling structure is predicted to be the most stable, and this is in
accordance with the literatures [16–19]. Among all the structural
features of the porphyrin diacids, the most noticeable feature is
the significant out-of-plane distortion of the porphyrin ring that
increases successively with the extent of meso-phenyl substitution.
−
M=Zn, H
4
Fig. 1. Labeling diagram of porphyrins and polarization direction of molecular exci-
tation.
and then dissolved in chloroform for measurement of absorption
spectra. Diacid species were prepared by solving corresponding
free-base porphyrins with aqueous HCl (12 mol/L)/ethanol/CH Cl2
2
solution (volume ratio 1:10:50), or with a CF COOH/CH Cl solu-
3
2
2
2
+
−
tion (volume ratio 1:20). A UV–vis spectrometer was employed to
ensure complete conversion of free-base porphyrin to zinc com-
plexes or diacid species. The concentration of the sample used for
UV–vis absorption spectra measurement was 5 × 10 mol/L. Such
low concentrations do not usually generate measurable aggrega-
tion. All reagents (AR grade) were used without further treatment.
Absorption spectra were recorded on a Shimadzu UV-2450
absorption spectrophotometer.
The H DPP (Cl ) displays larger out-of-plane distortion than the
H MPP (Cl ) , which differs from the observation reported in the
4
2
2+
−
4
2
literature [19], where diarylporphyrin diacid is less distorted than
monoarylporphyrin diacid. This discrepancy could be ascribed to
the differences in molecular environment.
We estimated the NSD of zinc porphyrins and porphyrin diacids
on the basis of stable geometry optimized by PBE1PBE methods (see
Table S3). The total out-of-plane distortion (DOOP) of porphyrin
diacids (>1.5 A˚ ) is, in general, predicted to be larger than that of
zinc porphyrins (<0.40 A˚ ). Further, B2u modes (saddling vibration)
−
6
2
.2. Computational methods
are predicted to contribute significantly to DOOP for all porphyrins
studied here, whereas B1u modes (ruffling vibration) that usually
display marked out-of-plane displacement in ruffled porphyrin dis-
play negligible displacement in this study. In addition to the 1B2u
mode, the 2B2u mode also displays moderate out-of-plane displace-
ment (>0.2 A˚ ). Based on NSD analysis, out-of-plane distortion is
Structural optimization and TDDFT calculation were carried out
at B3LYP [21,22] and PBE1PBE [23] levels by using a 6-31G(d) basis
set for C, H, N, and Cl atoms, and LanL2DZ basis set for the Zn atom.
A similar theoretical level has been recently used in studies of the
hydrogen bond in porphyrin diacids [12] and electronic structure
of zinc porphyrins [24]. For comparison, calculations at the PBEPBE
expected to generate a larger spectral redshift for porphyrin diacids
than for zinc complexes.
[
25] level were also carried out by using the SVP basis sets for all
atoms. All calculations were performed with the Gaussian 03 suit
of programs [26]. Simulated UV–vis spectra were derived, from a
Gaussian output file using the SWizard program (revision 4.6 [27]),
Another noticeable feature of diacid species is the lower twisting
orientation of the meso-phenyl group relative to the mean-plane of
porphyrin ring. The dihedral (<43 ) between the meso-phenyl and
◦
−
1
in Gaussian line shape with a half-bandwidth of 1000 cm
.
the mean plane of the porphyrin ring, predicted to decrease with
In the literature [18], there have been reports that thorough
analysis of the structural and spectroscopic properties of por-
phyrin diacids should consider the influence of conjugated bases
the extent of meso-phenyl substitution, is much smaller than that
of zinc complexes (>62 ). The nearly co-planar orientation of the
meso-phenyl group with regard to porphyrin is believed to enhance
◦

Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
1451
Table 1
Calculated structural parameters of zinc porphyrins skeleton by the PBE1PBE/6-31G(d)/LanL2DZ method.
ZnTPP
Expa
ZnTrPP
ZnDPP
ZnMPP
ZnP
Calb
Calc
Bond length (Å)
C␤–C␤
1.349
1.361
1.360
1.360–1.361
1.361
1.364–1.365
1.362
C␣–C␤
C␣–N
C␣–Cm
N–Zn
C5· · ·C15
C20· · ·C10
Angle (degree)
1.440
1.376
1.400
2.037
1.446
1.376
1.408
2.055
6.926
6.926
1.442
1.368
1.407
2.059
6.905
6.905
1.439–1.445
1.365–1.368
1.394–1.409
2.058–2.060
6.931
1.439–1.446
1.366–1.367
1.395–1.408
2.058
6.954
6.801
1.439–1.445
1.364–1.367
1.394–1.409
2.057–2.058
6.903
1.442
1.365
1.396
2.057
6.852
6.854
6.827
∠
∠
∠
∠
C␣CmC␣
NC␣C␤
C␣C␤C␤
C␣NC␣
125.1
125.3
125.1–127.4
109.4–109.7
106.9–107.0
107.1–107.2
124.9–127.6
109.3–109.7
∼106.9
125.1–127.3
109.3–109.6
106.8–106.9
107.2–107.3
127.1
109.4
107.4
106.6
109.4
107.1
107.0
109.4
107.0
107.2
109.5
106.9
107.2
107.2
Dihedral (degree) and out-of-plane displacement (Å)
NC␣CmC␣
3.369
1.859–3.693
61.91–62.4
1.265
1.199–1.406
64.48
0.523
0.083–1.308
65.34
0.291
0.000
Ph-Por.d
62.42
1.513
0.146
|
ꢀtotal|
0.000
0.000
|
ꢁC␤|
0.085–0.153
0.048–0.054
0.003–0.051
a
b
c
Experimental data from Ref. [31].
Calculated data from Ref. [32].
Calculated data from the present study.
d
The dihedral between phenyl plane and the mean-plane of porphyrins.
the ␲-electron delocalization effect between two moieties in a por-
phyrin diacid.
3.2. Electronic structure of zinc complex and diacid species of
porphyrins
With regard to in-plane distortion, the variation upon meso-
phenyl substitution is similar to that observed in zinc porphyrins.
The largest variation was also found in the vicinity of meso-Cm
atoms.
Tables 3 and 4 summarize, respectively, the ground-state elec-
tronic level predicted by the PBE1PBE method for zinc porphyrins
and porphyrin diacids.
meso-Phenyl substitution also induces noticeable variation in
hydrogen bond configuration. The Cl· · ·H distance is increased and
the H–N bond is shortened; this indicates that the Cl· · ·H–N hydro-
gen bond is weakened by meso-phenyl substitution.
Both B3LYP and PBEPBE calculations generate similar geometric
variations for porphyrin diacids (see Tables S4 and S5) as compared
with the PBE1PBE calculation.
3.2.1. Zinc porphyrins
Review of the data presented in Table 3 reveals two notice-
able variations of frontier orbits with meso-phenyl substitution:
(1) the orbital level is elevated for HOMO, HOMO−1, and aver-
aged LUMO/LUMO+1 orbit by meso-phenyl substitution; this is
attributable to a higher energy level of the phenyl group than
Table 2
Optimized structural parameter of porphyrin diacids by PBE1PBE/6-31G(d) calculation.
H4TPP2+(Cl )2 (D2d)
−
H4TrPP2+(Cl )2 (C2)
−
H4DPP2+(Cl )2 (D2)
−
H4MPP2+(Cl )2 (C2)
−
H4P2+(Cl )2 (D2d)
−
Expa Calb
Calc
Bond length (Å)
C␤–C␤
C␣–C␤
C␣–N
C␣–Cm
1.366
1.369
1.365
1.364–1.365
1.433–1.437
1.363–1.365
1.394–1.413
1.054–1.056
2.002–2.022
3.053–3.065
6.845
1.368
1.364–1.365
1.432–1.437
1.362–1.364
1.391–1.409
1.055–1.057
1.994–2.011
3.040–3.049
6.815
1.364
1.436
1.363
1.395
1.056
1.999
3.040
6.880
6.880
1.431
1.390
1.414
1.435
1.373
1.414
1.074
1.973
3.044
1.432
1.364
1.408
1.054
2.020
3.069
6.926
6.926
1.433–1.436
1.363–1.364
1.394–1.408
1.055
2.006
3.050
N–H
H· · ·Cl
N· · ·Cl
C5· · ·C15
C20· · ·C10
Angle (degree)
6.766
7.042
6.988
6.971
∠
∠
∠
∠
∠
C␣CmC␣
NC␣C␤
C␣C␤C␤
C␣NC␣
NHX
123.2
124.1
124.0
124.0–127.6
107.4–107.6
107.4–107.6
109.8
123.9–128.3
107.4–107.6
107.7–107.8
109.8
124.4–128.0
107.4–107.6
∼107.6
127.5
106.1
108.7
110.3
165.0
107.5
107.7
109.6
174.3
107.5
107.6
109.8.
173.3
107.5
107.6
109.9
167.8
109.8–109.9
167.4–170.3
169.9–173.1
169.8
Dihedral (degree) and out-of-plane displacement (Å)
NC␣CmC␣
21.22
35.43
9.103
0.851
16.25–21.06
37.97–39.28
7.915
16.37–16.68
42.46
6.867
13.19–16.62
42.95
6.198
13.36
0
5.382
0.500
Ph-Por.d
|
ꢀtotal|
|
ꢁC␤|
0.93
0.838
0.667–0.805
0.633–0.650
0.482–0.661
a
b
c
Experimental data from Ref. [33].
Calculated data from Ref. [21].
Calculated data from the present study.
d
The dihedral between meso-phenyl plane and the mean-plane of porphyrins.

1
452
Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
Table 3
Predicted orbital energy of zinc porphyrins by the PBE1PBE/6-31G(d)/LanL2DZ method (in eV).
ZnTPP (D2d)
ZnTrPP (C2)
ZnDPP (D2)
ZnMPP (C2)
ZnP (D4h)
LUMO+1
LUMO
HOMO
−2.1126 (e)
−2.1126 (e)
−5.1180 (b2)
−5.3755 (b1)
3.0054
−2.1045 (b)
−2.1221 (a)
−5.1794 (b)
−5.3910 (a)
3.0573
−2.0925 (b1)
−2.1276 (b2)
−5.2556 (b3)
−5.4055 (a)
3.1230
−2.1012 (a)
−2.1186 (b)
−5.3260 (b)
−5.4204 (a)
3.2074
−2.1134 (eg)
−2.1134 (eg)
−5.3995 (a2u)
−5.4395 (a1u)
3.2861
HOMO−1
HOMO–LOMO
the porphyrin moiety [18]. The largest elevation (∼0.282 eV) is
observed for the HOMO orbit, which reduces the HOMO–LUMO
gap from 3.286 eV of ZnP to 3.005 eV of ZnTPP; (2) LUMO/LUMO+1
orbits, that degenerated in ZnP and ZnTPP, become nondegenerate
and there is a substantially split in energy for asymmetric por-
phyrins. The greatest division is observed in ZnDPP (0.035 eV), and
this is closely related to the maximum discrepancy between the
configuration around H- and phenyl-suspended Cm observed in
ZnDPP.
employed. By checking the electron orbital map (Fig. S2), the HOMO
and HOMO−3 orbits of the PBE1PBE calculation are identified as the
Gouterman-type HOMO and HOMO−1 orbits (abbreviated as G-
HOMO and G-HOMO−1 orbits), respectively. The orbital ordering
predicted by the B3LYP and PBEPBE methods (Tables S8 and S9) dif-
fers from that obtained from the PBE1PBE calculation and, further,
is closely related to the meso-phenyl number.
There is another lower occupied MO in H4P2+(Cl ) ; this is
−
2
located primarily on Cm, N, and Cl atoms as the G-HOMO orbit
but has different nodal features. This orbit was predicted to con-
tribute predominantly to an excitation in the Q region with notable
oscillator strength; therefore, it is labeled as the G-HOMO* orbit in
These afore-described variations should be ascribed to the par-
ticipation of the meso-phenyl group in Gouterman Four Orbits,
particularly in the HOMO orbit (Fig. S1). Natural orbit population
analysis indicates that the Gouterman-type HOMO orbit centers
primarily on Cm and N atoms and, therefore, is readily disturbed
by meso-substitution. For example, the MO coefficient of Cm and
N atoms decreases from ∼0.58 and ∼0.27 in ZnP, respectively, to
−
this study. The mixing of Cl and porphyrin-based orbits within
the G-HOMO and G-HOMO* orbits reflects the interaction between
−
the Cl ion and the porphyrin ring that drives the formation of
the Cl· · ·H–N hydrogen bond. It is apparent that G-HOMO and
G-HOMO* orbits display a contrary change trend following meso-
phenyl substitution (Table 5 and Fig. S2). The G-HOMO orbit has a
0.46 and 0.21 in ZnTPP; however, the MO coefficient of the phenyl
group increases. The HOMO−1 orbit is primarily centered on C␣
−
and C atoms and, therefore, is subjected to less influence than
higher energy level and displays decreasing MO coefficient of Cl
␤
the HOMO orbit. The splitting of LUMO/LUMO+1 orbits observed
in asymmetric porphyrins (ZnMPP, ZnDPP, and ZnTrPP) should be
attributed to the different numbers of meso-phenyl groups involved
in these two orbits. LUMO and LUMO+1 orbits of ZnDPP involve zero
and two meso-phenyl groups, respectively, this induces the maxi-
mum discrepancy in orbital composition and, therefore, produces
the biggest division in energy for LUMO and LUMO+1 orbits.
The B3LYP calculation (Table S6) generates a similar electronic
structure as that with the PBE1PBE method, whereas the PBEPBE
method predicts a slightly higher energy for HOMO and HOMO−1
orbits, a marginally lesser energy for LUMO/LUMO+1 orbits and,
consequently, a smaller HOMO–LUMO gap (Table S7).
ion and increasing coefficient of Cm and N atoms; however, the con-
verse is true for the G-HOMO* orbit. This difference indicates that
orbital mixing is reduced markedly by the meso-phenyl substitu-
−
tion that produces an almost pure Cl and porphyrin-based orbit,
2+
−
respectively, for G-HOMO* and G-HOMO in H TPP (Cl ) ; this fur-
4
2
ther proves that the Cl· · ·H–N hydrogen bond is weakened by the
meso-phenyl substitution observed in structural analysis.
In general, the spectral shift correlates with the variation of
the HOMO–LUMO gap [18]. For porphyrin diacids, although the
level of the MO is uplifted by meso-phenyl substitution, the G-
2
+
−
HOMO–LUMO gap is reduced from 2.9563 eV of H P (Cl ) to
4
2
2+
−
2.5530 eV of H TPP (Cl ) ; this is consistent with the observed
4
2
redshift of electronic absorption (see Section 3.3).
3
.2.2. Porphyrin diacids
2+
−
2
The complexity of the electronic structure of H P (Cl ) and
4
2+
−
−
H TPP (Cl ) , caused by the insertion of the Cl lone pair orbit,
3.2.3. Analysis of some influences to MO level
4
2
has been revealed in the published literature [17,18] and was also
testified in our calculation in this study. Although LUMO/LUMO+1
orbits are nearly similar to that of zinc porphyrins, the ordering
of occupied frontier orbits is observed to be significantly different
from zinc porphyrins and is influenced by the calculation method
To evaluate the influence of conformational distortion, electron
effect of meso-phenyl, and electronic effect of the Cl ion on MO
level, theoretical calculations were performed on several artificially
adjusted porphyrin models (Table 6), and the obtained results are
discussed in the following sub-sections.
−
Table 4
Calculated orbital energy of HCl-acidified porphyrins by the PBE1PBE/6-31G(d) method (in eV)a,b.
H4TPP2+(Cl )2
−
H4TrPP2+(Cl )2
−
H4DPP2+(Cl )2
−
H4MPP2+(Cl )2
−
H4P2+(Cl )2
−
LUMO+1
LUMO
HOMO
2.8410
2.8410
5.3940 (G-HOMO)
5.8298 (p)
5.8298 (p)
−2.8710
−2.9161
−2.9648
−3.0255
−2.8756
−2.9245
−2.9670
−3.0255
−5.5431 (G-HOMO)
−5.8779 (p)
−5.8779 (p)
−6.1029 (G-HOMO−1)
−6.1532 (p)
−6.1717 (p)
−6.1760
−5.6946 (G-HOMO)
−5.9307 (p)
−5.9331 (p)
−6.1548 (G-HOMO−1)
−6.2220 (p)
−6.2269 (p)
−6.2277
−5.8328 (G-HOMO)
−5.9794 (p)
−5.9835 (p)
−6.2060 (G-HOMO−1)
−6.2723 (p)
−6.2745 (p)
−6.2818
−5.9818 (G-HOMO)
−6.3704 (p)
HOMO−1
HOMO−2
HOMO−3
HOMO−4
HOMO−5
HOMO−6
HOMO−7
−6.3704 (p)
6.0727 (G-HOMO−1)
−6.2737 (G-HOMO−1)
−6.3300 (p)
6.0942 (p)
6.1216 (p)
6.1216 (p)
6.1320 (G-HOMO*)
2.5530
−6.3300 (p)
−6.3439 (p)
−6.2155 (G-HOMO*)
2.6675
−6.3139 (G-HOMO*)
2.7701
−6.4132 (G-HOMO*)
2.8658
−6.5481 (G-HOMO*)
2.9563
(
G-HOMO)–LOMO
a
p in parentheses denotes the p orbit of Cl atom.
G-HOMO and G-HOMO−1 correspond to the Gouterman-type HOMO and HOMO−1 orbit, respectively, G-HOMO* orbit has similar structure with G-HOMO but centers
b
mainly on the Cl atom.

Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
1453
Table 5
The calculated molecular orbit composition of G-HOMO and G-HOMO* orbit by the PBE1PBE/6-31G(d) method.
Molecular orbit coefficienta
H4TPP2+(Cl )2
−
H4TrPP2+(Cl )2
−
H4DPP2+(Cl )2
−
H4MPP2+(Cl )2
−
H4P2+(Cl )2
−
G-HOMO
G-HOMO*
G-HOMOc
Cm
N
Ph
Cl
0.405
0.227
0.234
0.033
0.412
0.247
0.179
0.067
0.414
0.269
0.124
0.22
0.398
0.283
0.062
0.186
0.376
0.302
0.000
0.269
b
Cm
N
0.044
0.085
0.012
0.791
0.022
0.081
0.017
0.764
0.020
0.078
0.021
0.726
0.071
0.072
0.019
0.627
0.144
0.890
0.000
0.597
b
Ph
Cl
Cm
N
0.308
0.122
0.362
0.308
0.118
0.349
0.290
0.110
0.375
0.286
0.104
0.359
b
Ph
a
Normalized to 1.
Phenyl group.
The orbit of ad-porphyrin III series.
b
c
Table 6
The collection of artificially adjusted porphyrin modelsa,b
.
Porphyrin diacids
Cl ion
Zinc porphyrinsc
Phenyl group
Effectsd
Zinc ion
Phenyl group
Effectsd
A
Ad-porphyrin I
Ad-porphyrin II
Ad-porphyrin III
Ad-porphyrin IVe
Parent series
×
√
×
A
A + B
A + C
A + B + C
A + B + C
√
×
√
×
√
√
×
√
√
√
√
√
f
A + Cf
A + C
√
a
All porphyrin models are deduced from parent porphyrin without further optimization.
√
b
c
d
e
f
“
” indicates involved while “×” indicates eliminated.
For adjusted zinc porphyrins, only ad-porphyrin II and IV series are constructed.
−
A” indicates conformational effect; “B” indicates electronic effect of counterion Cl ; and “C” indicates electronic effect of meso-phenyl group.
“
In which phenyl plane is adjusted to a perpendicular orientation with respect to the mean plane of porphyrin ring.
Only electronic donating effect is involved.
(
1) Conformational distortion. The replacement of the meso-
formational distortion is retained. The electron orbital level of
ad-porphyrin I is presented in Table 7. In comparison with the
parent porphyrin diacid, the ad-porphyrin I displays a much lower
MO level, due to the exclusion of the meso-phenyl group and cen-
phenyl by the H atom with the simultaneous extrusion of the
−
Cl ion produces the first type of an adjusted porphyrin model
(
ad-porphyrin I series); in this model, only the influence of con-
Table 7
Calculated orbital energy of several adjusted porphyrin diacid series by the PBE1PBE/6-31G(d) method (in eV).
H4TPP2+(Cl )2
−
H4TrPP2+(Cl )2
−
H4DPP2+(Cl )2
−
H4MPP2+(Cl )2
−
H4P2+(Cl )2
−
␦a
Ad-porphyrin I
LUMO+1
LUMO
−9.6590
−9.6590
−12.7318
−12.8876
3.0728
−9.6383
−9.6582
−12.7633
−12.8860
3.1052
−9.6258
−9.6636
−12.7949
−12.8879
3.1313
−9.6329
−9.6552
−12.8205
−12.8923
3.1653
−9.6427
−9.6427
−12.8479
−12.8972
3.2053
G-HOMO
G-HOMO−1
(
G-HOMO)–LOMO
0.1325
Ad-porphyrin II
LUMO+1
LUMO
−3.0586
−3.0586
−5.9824
−6.2557
−6.3830
2.9237
−3.0399
−3.0608
−5.9862
−6.2560
−6.4309
2.9254
−3.0181
−3.0641
−5.9903
−6.2631
−6.4752
2.9262
−3.0208
−3.0467
−5.9845
−6.2672
−6.5114
2.9376
−3.0255
−3.0255
−5.9818
−6.2737
−6.5481
2.9537
G-HOMO
G-HOMO−1
G-HOMO*
(
G-HOMO)–LOMO
0.0299
0.8807
0.1558
Ad-porphyrin III
LUMO+1
LUMO
−8.3915
−8.3915
−10.7160
−11.6460
2.3245
−8.6031
−8.6836
−11.1164
−11.8374
2.4325
−8.8683
−9.0386
−11.5290
−12.3532
2.4855
−9.2162
−9.3084
−12.0308
−12.5999
2.7225
−9.6427
−9.6427
−12.8479
−12.8972
3.2053
G-HOMO
G-HOMO−1
(
G-HOMO)–LOMO
Ad-porphyrin IV
LUMO+1
LUMO
−2.8508
−2.8508
−5.6513
−6.0702
2.8005
−3.0255
−3.0255
−5.9818
−6.2737
2.9563
G-HOMO
G-HOMO−1
(
G-HOMO)–LOMO
a
The difference of (G-HOMO)–LOMO gap between the derivatives of H4TPP2+(Cl )2 and H4P2+(Cl )2 in each adjusted series.
−
−

1
454
Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
−
tral Cl ion. Following meso-phenyl substitution, the level of the
MO, in particular for the G-HOMO, increases successively due to
increase in out-of-plane distortion. However, the variation step of
increased nearly in a directly proportional manner with the exten-
sion of meso-phenyl number. However, this increase is significantly
obliterated in the ad-porphyrin III series, where the MO coefficient
of the meso-phenyl group is observed to be almost constant for all
porphyrin diacids.
2+
−
the G-HOMO–LUMO gap from the derivative of H P (Cl ) to that
of H TPP (Cl ) (0.1325 eV) is less than that of parent porphyrin
4
2
2
+
−
4
2
diacids (0.4033 eV); this indicates that the conformational distor-
tion is not the primary cause for the variation in the electronic
The electronic effect of the meso-phenyl group can be further
divided into an electron-donating and ␲-electron delocalization
effects. These two effects can be differentiated by comparison of
parent porphyrin diacids with ad-porphyrin IV series, wherein
the meso-phenyl group is adjusted to be perpendicular to the
porphyrin ring. This configuration maximally diminishes the ␲-
electron delocalization between the meso-phenyl and porphyrin
rings. A comparison of the ad-porphyrin IV with the parent por-
phyrin diacid predicts a contribution of 0.1259 eV for the electronic
donating effect; therefore, this spares 0.2475 eV for the delocaliza-
tion effect to the variation of the G-HOMO–LUMO gap (Table 7).
A similar comparison was also performed without consideration
structure of the porphyrin diacid series.
−
(
2) Electronic effect of central Cl ion. The replacement of only
the meso-phenyl group of the parent porphyrin by a hydrogen
atom produces the second type of artificially adjusted porphyrin
(
ad-porphyrin II); this porphyrin possesses conformational and
−
electronic effects of the Cl ion. By comparing ad-porphyrin II with
ad-porphyrin I, we clarify the influence of the Cl ion on the MO
−
level. In the calculation, ad-porphyrin II has a higher MO level, par-
ticularly G-HOMO orbit, than the ad-porphyrin I (Table 7); this is
−
primarily attributed to the much higher orbital level of the Cl ion
−
than that of the porphyrin unit [34]. However, the G-HOMO–LUMO
gap, as well as its variation step (0.0299 eV) with extension of
meso-phenyl substitution, is less than that in ad-porphyrin I. This
of Cl , that is, a comparison between ad-porphyrin III series
and its corresponding further-adjusted series with perpendicular
meso-phenyl groups; the comparison predicted a contribution of
0.3832 and 0.3650 eV, respectively, for electron-donating and -
delocalization effects. Therefore, we conclude that consideration
−
indicates that consideration of Cl in the calculation would dimin-
ish the variation step of the G-HOMO–LUMO gap because of the
−
−
reduced interaction between Cl and porphyrin ring with increase
of Cl in calculations would reduce the influence of the electron-
in the extent of meso-phenyl substitution.
donating effect of the meso-phenyl group more significantly than
(
3) Electronic effect of meso-phenyl group. The third type of
that of the delocalization effect, and this is attributed to the com-
petitive electron-donating effect of the Cl ion.
−
adjusted porphyrins, ad-porphyrin III, is constructed by omission
of the central Cl ion of the parent series. Both conformational and
−
electronic effects of the meso-phenyl group are involved in this
series.
3.3. Absorption spectra
Similar to the observation in zinc porphyrins, the meso-phenyl
substitution uplifts the MO level, particularly the G-HOMO level
3.3.1. Experimental spectra
Fig. 2 depicts experimental UV–vis spectra of zinc porphyrins in
chloroform solution, and spectral data are summarized in Table 8.
From Fig. 2 and Table 8, we infer that both B and Q bands undergo
successive redshift following meso-phenyl substitution. The total
shift distance for both B and Q0 band from ZnP to ZnTPP (25 nm) is
comparable to that of the corresponding free-base species [20].
The absorption spectra of HCl-acidified porphyrin diacids are
displayed in Fig. 3, and the observed spectral data accord well with
that reported in the published literature [18,19]. For comparison,
the spectra of CF3COOH-acidified porphyrin diacids were recorded
(Fig. S3). With the exception of slight differences in absorption
wavelength due to different acid and solvent used, the redshift
trend of B and Q bands with meso-phenyl substitution is observed
to be similar for these two series of porphyrin diacids. In general,
the redshift extent in the porphyrin diacid series that is caused by
meso-phenyl substitution is greater than that observed in the zinc
porphyrin series. For example, the redshift step of the B band of
HCl-acidified porphyrin, measuring 36.5 nm, is at least 1.5-times
larger than that of zinc porphyrins; however, the redshift step of
the Q band (60.4 nm for Q1 and 79.7 nm for Q0 band) is approxi-
mately 2.5 times larger than that of zinc porphyrins. In addition,
the Q0 band, which appears only as a weak shoulder in the right
(
Table 7), because of a higher MO level of phenyl than the por-
phyrin moiety [18]. The G-HOMO–LUMO gap of ad-porphyrin III,
although smaller than that of ad-porphyrin I, displays a larger
variation step (0.8807 eV) than that of ad-porphyrin I. The pure
electronic effect of meso-phenyl substitution can reasonably be
evaluated by comparing the variation of the G-HOMO–LUMO gap
in the ad-porphyrin I with that in the ad-porphyrin III series;
the difference of the variation step between ad-porphyrin I and
III series (0.7482 eV) could be ascribed to the electronic effect of
meso-phenyl group. This step is greater than the one observed in
ad-porphyrin I, which indicates that the electronic effect of the
meso-phenyl group has a much greater influence on electronic
structure than conformational distortion does for the porphyrin
diacid.
−
If considering the Cl in calculation, the electronic effect of
meso-phenyl substitution could be evaluated by comparison of the
ad-porphyrin II series with the parent porphyrin diacid; this pre-
dicts a contribution of 0.3734 eV for the electronic effect of the
meso-phenyl group to the variation in the G-HOMO–LUMO gap.
This value is approximately half of the 0.7482 eV value obtained in
−
the calculation without considering the Cl ion, and can be inter-
preted in terms of competition between counterion Cl and the
−
2+
−
side of strong Q1 band in H4P (Cl )2, is enhanced markedly by
meso-phenyl group in the electron-donating effect to the porphyrin
ring. The electron donating effect of the counterion decreases with
increase in the extent of meso-phenyl substitution because of a
weakened Cl· · ·H–N hydrogen bond. Therefore, we conclude that
the influence of each factor overlaps with that of other factors; this
accounts for the abnormal lesser variation of the G-HOMO–LUMO
gap (0.4033 eV) observed in the parent diacid series instead of the
meso-phenyl substitution with regard to B and Q1 band.
3.3.2. Theoretical simulation of absorption spectra
3.3.2.1. Zinc porphyrins. For zinc porphyrins, the excitation struc-
ture predicted by the PBE1PBE method could be explained well by
the Gouterman Four Orbit model. The first four excitations pre-
dicted involve almost pure electron transition occurring between
the Gouterman-type HOMO/HOMO−1 and LUMO/LUMO+1 orbits;
therefore, they are easily assigned to the Q and B bands (Table 8
and Tables S10–S14). The simulated redshift of B and Q band is
consistent with the experimental results.
0
.8807 eV of the ad-porphyrin III series. The overlapping of the
−
electronic effect of the meso-phenyl group and the counterion Cl
is also demonstrated by comparison of MO population between
the parent porphyrin diacid series and ad-porphyrin III series
(
Table 5). Data presented in Table 5 indicate that the MO coeffi-
Detailed analysis of PBE1PBE calculation reveals certain notable
features:
cient of the phenyl group in the G-HOMO orbit of the parent series is

Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
1455
Table 8
Calculated excitation energy, oscillator strength (f), and composition determined by the PBE1PBE/6-31G(d)/LanL2DZ method and experimental wavelength (in CHCl3) of B
and Q bands of zinc porphyrins.
Exp. (nm)
Cal.
Excitation nm (eV)
374.4 (3.31)
f
One-electron transition (weight)
ZnTPP
ZnTrPP
ZnDPP
ZnMPP
ZnP
B
419.0
584.0
B
1.4177
1.4186
0.0282
0.0283
H-1 → L + 1(+41%)
H-1 → L + 0(+41%)
H-0 → L + 1(+61%)
H-0 → L + 0(+61%)
H-0 → L + 0(22%)
3
74.4 (3.31)
535.8 (2.31)
35.8 (2.31)
H-0 → L + 1(+22%)
H-1 → L + 0(43%)
H-1 → L + 1(+43%)
Q0
Q
5
B
413.0
578.0
Bx
By
Qx
Qy
368.4 (3.37)
367.0 (3.38)
527.6 (2.35)
527.1 (2.35)
1.5495
1.1023
0.0249
0.0068
H-1 → L + 1(+40%)
H-1 → L + 0(+38%)
H-0 → L + 0(+60%)
H-0 → L + 1(+59%)
H-0 → L + 0(22%)
H-0 → L + 1(+24%)
H-1 → L + 1(+44%)
H-1 → L + 0(46%)
Q0
B
407.0
570.0
Bx
By
Qx
Qy
352.4 (3.44)
351.0 (3.46)
517.6 (2.40)
517.4 (2.40)
1.7088
0.7948
0.0180
0.0000
H-1 → L + 1(+38%)
H-1 → L + 0(+34%)
H-0 → L + 0(+59%)
H-0 → L + 1(+56%)
H-0 → L + 0(+23%)
H-0 → L + 1(26%)
H-1 → L + 1(46%)
H-1 → L + 0(+50%)
Q0
B
401.0
564.0
Bx
By
Qy
Qx
360.5 (3.44)
359.3 (3.45)
509.4 (2.43)
509.2 (2.44)
1.3279
0.8650
0.0004
0.0015
H-1 → L + 1(+34%)
H-1 → L + 0(+32%)
H-0 → L + 1(+54%)
H-0 → L + 0(+56%)
H-0 → L + 0(25%)
H-0 → L + 1(+26%)
H-1 → L + 0(51%)
H-1 → L + 1(+49%)
Q0
B
394.0
559.0
B
343.5 (3.61)
0.9664
0.9664
H-1 → L + 1(+23%)
H-1 → L + 0(+8%)
H-1 → L + 0(+23%)
H-1 → L + 1(8%)
H-0 → L + 1(+46%)
H-0 → L + 0(+7%)
H-0 → L + 0(+46%)
H-0 → L + 1(7%)
H-0 → L + 0(20%)
H-0 → L + 1(+7%)
H-0 → L + 1(+20%)
H-0 → L + 0(+7%)
H-1 → L + 0(45%)
H-1 → L + 1(+7%)
H-1 → L + 1(+45%)
H-1 → L + 0(+7%)
343.5 (3.61)
Q0
Q
501.7 (2.47)
01.7 (2.47)
0.0014
0.0014
5
(
(
1) The contribution weight of electronic transition from HOMO
and HOMO−1 orbit is different for the B and Q bands. The Q
band receives slightly more contribution from the HOMO orbit,
whereas the B band receives its contribution primarily from the
HOMO−1 orbit. This nonuniform transition can be ascribed to
the lifting of nearly degenerated HOMO and HOMO−1 orbits
that is caused by the meso-phenyl substitution.
2) The degenerated transition Bx/By and Qx/Qy (the polarized
direction of x and y excitation; Fig. 1) observed in ZnP and
ZnTPP are predicted to split in the asymmetric porphyrin, and
the split size of the B band (>1.0 nm) is greater than that of the
Q band (<0.5 nm). It is notable that calculation indicates that
the polarized direction of zinc and diacid derivatives of por-
phyrin dose not bisect the pyrrolic rings along either the N–N or
N–HH–N axis as observed in free-base porphyrins, but, instead,
goes through the meso-position.
dicts similar variation trends for B and Q bands with the extent of
meso-phenyl substitution. It should be emphasized that the direct
comparison between calculated absorption wavelength and exper-
imental data is impractical because of differences in molecular
environment; however, the study of redshift trends of absorp-
tion band is much more meaningful. The B3LYP method provides
a similar description of absorption spectra of zinc porphyrins
(Tables S15–S19) with the PBE1PBE method, with the exception
of a lesser overestimation of excitation energy.
The PBEPBE method predicts a markedly more complicated exci-
tation structure (Tables S20–S24): (1) there are several excitations
with moderate oscillator strength located in the B band region for
asymmetric zinc porphyrins; (2) the electron transition from orbits
other than G-HOMO and G-HOMO−1 orbits are observed to have
notable, or a major contribution, to these excitations. Conversely,
the case is simple for the Q band where only the Gouterman-type
excitation is predicted to have notable oscillator strength. Fig. 4,
further, indicates that the PBEPBE method, although producing a
much closer absorption wavelength to that of the experiment by
coincidence, predicts a less consistent variation with the exper-
imental results for both B and Q bands, in comparison with the
PBE1PBE and B3LYP methods.
The split of absorption can be attributed to the lifting of the
degenerated LUMO/LUMO−1 orbit caused by unsymmetrical meso-
phenyl substitution. For the Bx/By pair of asymmetric porphyrin, the
excitation component involving a greater number of meso-phenyl
groups in a polarized direction is predicted to possess lower exci-
tation energy; this coincides with the general redshift of the B band
upon meso-phenyl substitution. Consistent with this analysis, the
greatest division of the B band is observed in ZnDPP (ꢀ = 2.7 nm).
However, the case of Q band is ambiguous; there is no general
rule defining the ordering of the Qx/Qy pair. The predicted oscil-
lator strength also correlates with the meso-phenyl number for
both B and Q bands. The component involving more number of
meso-phenyl groups in the polarized direction is predicted to pos-
sess greater oscillator strength. Therefore, the ZnDPP displays the
biggest difference in oscillator strength between two polarized
components for both B and Q bands.
3.3.2.2. Porphyrin diacids. The electronic structure of porphyrin
diacid is described as deviating from the Gouterman Four Orbit
model in the above-described data. Similarly, the excitation struc-
ture predicted by DFT methods cannot be fluently described by
the Gouterman Four Orbit model (predicted excitation composi-
tion is summarized in Table 9 and Tables S25–S29 for the PBE1PBE
method, Tables S30–S34 for the B3LYP method, and Tables S35–S39
for the PBEPBE method). The B band is easily identified because of its
large oscillator strength, which is observed both in experiment and
in theoretical calculation. However, the assignment of the Q band
is confounded by the coexistence of two excitations, both of which
have moderate oscillator strength in this region. The excitation with
the higher excitation energy primarily involves the electron transi-
tion out of G-HOMO*/G-HOMO−1 orbits, and the other excitation
The dependence of calculated absorption wavelength on the
meso-phenyl number is compared with experimental data in
Fig. 4. From Fig. 4, it can be observed that the TDDFT calculation,
although systematically overestimating the excitation energy, pre-

1
456
Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
ZnP
H
3
3
3
00
00
00
400
400
400
500
500
500
500
600
300
400
400
400
500
500
500
600
600
600
700
700
700
ZnMPP
300
600
ZnDPP
300
600
ZnTrPP
300
400
500
600
700
3
00
400
600
ZnTPP
300
400
500
600
700
300
400
500
600
Wavelength (nm)
Wavelength (nm)
Fig. 3. Absorption spectra of HCl-acidified porphyrin (solid line: experimental spec-
tra; dot line: calculated spectra by PBE1PBE/6-31G(d) method). The absorbance of
spectra region (>500 nm) is multiplied by a factor of 3.
Fig. 2. Absorption spectra of zinc porphyrin (solid line: experimental spectra; dot
line: calculated spectra by PBE1PBE/6-31G(d)/LanL2DZ method). The absorbance of
spectra region (>480 nm) is multiplied by a factor of 4.

Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
1457
600
580
560
540
520
500
Q band
680
640
600
560
Q band
420
400
380
360
340
B band
440
400
360
B band
0
1
2
3
4
0
1
2
3
4
meso-phenyl number
meso-Phenyl number
Fig. 5. Dependence of absorption wavelength (nm) on meso-phenyl number for
HCl-acidified porphyrin diacids (triangle: experimental data in chloroform solution;
quadrangle: calculated data by B3LYP/6-31G(d) method; circle: calculated data by
PBE1PBE/6-31G(d) method; diamond: calculated data by PBEPBE/SVP method).
Fig. 4. Dependence of absorption wavelength (nm) on meso-phenyl number for
zinc porphyrins (triangle: experimental data in chloroform solution; circle: cal-
culated data by PBE1PBE/6-31G(d)/LanL2DZ method; quadrangle: calculated data
by B3LYP/6-31G(d)/LanL2DZ method; diamond: calculated data by PBEPBE/SVP
method).
sition weight for H P2+(Cl ) , H MPP (Cl ) , and H DPP (Cl ) ;
−
2+
−
2+
−
2
4
2
4
2
4
(
2) certain excitations involving electron transition from primarily
with a lower excitation energy primarily involves electron transi-
tion out of G-HOMO/G-HOMO−1 orbits. Both PBE1PBE and B3LYP
calculations predict that the latter excitation has greater oscillator
strength and should be assigned to the Q absorption. This assign-
ment is consistent with the general viewpoint with regard to Q
absorption of porphyrins. However, the PBEPBE calculation gen-
erates a contrary assignment, similar to that suggested by Rosa
HOMO−4/HOMO−5 to LUMO−2, not evident in the PBE1PBE and
B3LYP calculations, are predicted to be located in the B band region
with moderate oscillator strength. The spectral variation profile
predicted by the PBEPBE method, although possessing a slightly
larger slope, is similar to that predicted by the PBE1PBE and B3LYP
methods.
In conclusion, the PBE1PBE method provides a more credible
result than the B3LYP and PBEPBE methods in spectral simulation
of diacid derivatives of meso-phenylporphyrin.
[
18] that the Q band receives a large contribution from the Cl-to-
porphyrin transition.
The dependent profile of predicted absorption wavelength with
meso-phenyl number is displayed in Fig. 5, where the PBE1PBE cal-
culation predicts a profile that appears to be more consistent with
experimental results than the other two DFT methods. Therefore,
our discussion is mainly carried out on the basis of the PBE1PBE
calculation.
Table 9 presents data indicating that the electron transition
from the G-HOMO−1 orbit is predominantly for the B band, with
increasing weight following meso-phenyl substitution. In addition,
the G-HOMO* orbit also contributes to the B band, but with a lesser
and further decreasing transition with meso-phenyl substitution. In
the Q band, the electron transition from G-HOMO is predominant,
and the G-HOMO−1 orbit provides a moderate contribution.
From the zinc porphyrins, we observe that the PBEPBE method
predicts an absorption wavelength that approximates better with
the experimental results coincidentally. This was also observed for
porphyrin diacids. However, the PBEPBE method has a less consis-
tent variation profile when compared with the PBE1PBE method
3.3.3. Analysis of the redshift of absorption bands
In this section, we attempt to allocate the total redshift of
absorption to several important factors as well as to identify the key
factor. The quantitive evaluation in the present study is performed
on the basis of theoretical calculation of several artificially adjusted
porphyrin models (Table 6). The TDDFT calculation results are pre-
sented in Tables S40 (PBE1PBE method) and S41 (B3LYP method),
and the allocated results are summarized in Table 10.
3.3.3.1. Zinc porphyrin series. For zinc porphyrins, only the elec-
tronic and conformational effects are analyzed in this article; the
influence of the hybrid orbital deformation (HOD) effect [10] has
been ignored because of the negligible out-of-plane distortion. For
both types of adjusted porphyrin models employed, ad-porphyrins
II and IV, the central Zn2+ ion remained untouched. At first, the
electronic effect of the meso-phenyl group and conformational
effect could be differentiated by comparison of the TDDFT cal-
culation of ad-porphyrin II series with that of the parent zinc
porphyrins. Later, the electronic effect could be further differenti-
ated into ␲-electron-delocalization and electron-donating effects
(
Fig. 5). Further, the PBEPBE calculation provides a more com-
plicated transition structure (Tables S35–S39) than the PBE1PBE
and B3LYP methods: (1) in the B band, certain Cl-centered MOs,
rather than G-HOMO* and G-HOMO−1 orbits, have dominant tran-

1
458
Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
Table 9
Calculated excitation energy, oscillator strength (f), and composition determined by the PBE1PBE/6-31G(d) method and experimental wavelength of B and Q bands of HCl
acidified porphyrins.
Exp. (nm)
Cal.
Excitation Nm (eV)
397.4 (3.04)
f
One-electron transition (weight)
H4TPP2+(Cl−)2
B
444.0
B
1.2711
1.2709
0.1433
0.1433
1.4830
0.9587
0.0538
0.1083
1.7345
0.6287
0.0113
0.0834
1.3053
0.6981
0.0050
0.0246
H-3 → L + 1(+53%)
H-3 → L + 0(+53%)
H-0 → L + 0(+71%)
H-0 → L + 1(+71%)
H-3 → L + 0(+43%)
H-3 → L + 1(+43%)
H-0 → L + 0(+70%)
H-0 → L + 1(+70%)
H-3 → L + 0(+38%)
H-3 → L + 1(+38%)
H-0 → L + 0(+71%)
H-0 → L + 1(+70%)
H-3 → L + 0(+33%)
H-3 → L + 1(+33%)
H-0 → L + 0(+56%)
H-0 → L + 1(+66%)
H-0 → L + 0(+10%)
3
97.4 (3.04)
614.6 (2.17)
14.6 (2.17)
H-0 → L + 1(10%)
H-3 → L + 1(27%)
H-3 → L + 0(+27%)
H-0 → L + 1(+12%)
H-0 → L + 0(12%)
H-3 → L + 1(+25%)
H-3 → L + 0(26%)
H-0 → L + 1(+12%)
H-0 → L + 0(12%)
H-3 → L + 1(+25%)
H-3 → L + 0(29%)
H-7 → L + 1(13%)
H-7 → L + 0(+13%)
H-3 → L + 1(20%)
H-3 → L + 0(+24%)
Q0 661.9
436.5
Q0 639.9
429.0
Q0 622.9
418.5
Q0 605.0
Q
6
H4TrPP2+(Cl−)2
B
Bx
By
Qy
Qx
Bx
By
Qy
Qx
Bx
By
Qy
Qx
388.7 (3.19)
387.8 (3.20)
598.9 (2.07)
597.1 (2.08)
379.3 (3.27)
376.8 (3.29)
587.8 (2.11)
582.3 (2.13)
367.9 (3.37)
367.0 (3.38)
575.8 (2.15)
572.7 (216)
H-7 → L + 1(+6%)
H-7 → L + 0(6%)
H4DPP2+(Cl )2
−
B
H-7 → L + 1(+9%)
H-7 → L + 0(9%)
H4MPP2+(Cl−−)2
B
H-0 → L + 1(12%)
H-0 → L + 0(+11%)
H-1 → L + 1(+19%)
H-1 → L + 0(+8%)
H4P2+(Cl )2
−
B
407.4
B
353.1 (3.51)
0.8310
0.8310
H-3 → L + 1(+15%)
H-7 → L + 0(+9%)
H-3 → L + 0(+15%)
H-7 → L + 1(9%)
H-3 → L + 0(+14%)
H-7 → L + 1(9%)
H-3 → L + 1(14%)
H-7 → L + 0(9%)
H-0 → L + 1(+21%)
H-3 → L + 0(7%)
H-0 → L + 0(+21%)
H-3 → L + 1(+7%)
353.1 (3.51)
Q0 578.9
Q
566.6 (2.19)
66.6 (2.19)
0.0027
0.0024
H-0 → L + 0(53%)
H-3 → L + 1(18%)
H-0 → L + 1(+53%)
H-3 → L + 0(18%)
5
by comparison of the ad-porphyrin IV with parent series (or with
ad-porphyrin II). From Table 10, we infer that the electronic effect,
particularly the electron-donating effect, is the key factor in spec-
tral redshift (both in B and Q bands). The B3LYP calculation results
in a similar evaluation.
10.7 nm for the B band and 16.9 nm for the Q band, respectively.
Further, the conformational effect could be divided into in-plane
and out-of-plane distortion effects.
The contribution of in-plane distortion could be evaluated
by comparison of H4P2+(Cl ) derivatives in ad-porphyrin I
−
2
2+
−
with its further modulated derivative (labeled as i-H4P (Cl ) ),
2
3
.3.3.2. Porphyrin diacid series. The TDDFT calculation of sev-
wherein the partial structure surrounding the Cm atom, includ-
ing bond length and bond angle, is adjusted to that of the
H4TPP2+(Cl )2 derivative in ad-porphyrin I. The in-plane variation
of the pyrrolic ring is negligible and, therefore, is not considered
here. The TDDFT calculation on i-H4P2+(Cl )2 produces a redshift
of approximately 1.4 (B band) and 3.4 nm (Q band) with regard
to the H4P2+(Cl )2 derivative in ad-porphyrin I (Table 10 and
Table S40).
eral adjusted porphyrin diacid models is also summarized in
Tables S40 and S41; the redshift analysis, particularly with regard to
the electronic effect of meso-phenyl group and the conformational
effect is summarized in Table 10.
−
−
(
1) Conformational effect. The redshift induced by the confor-
−
mational distortion is evaluated by analysis of TDDFT calculations
for the ad-porphyrin I series (Tables S40 and S41). In PBE1PBE
calculations, the predicted total redshift in the ad-porphyrin I is
The redshift generated by the out-of-plane distortion could
be deduced by comparison of H4P2+(Cl ) derivatives in ad-
−
2
porphyrin I with its other further-adjusted derivative (labeled
Table 10
2+
−
as o-H4P (Cl ) ), wherein the nonplanar degree (denoted by
Predicted redshift (in nm) with meso-phenyl substitution for each effect.
2
2
+
−
C␣CmC␣N dihedral angle) is adjusted to that of the H TPP (Cl )
derivative in ad-porphyrin I. Calculation on the o-H4P (Cl ) pro-
4
2+
2
Effect
Redshift (in nm)
Zinc porphyrin
−
2
Porphyrin diacids
B
vides the 11.3 (B band) and 16.8 nm (Q band) redshift (Table 10 and
2
+
−
Table S40). The sum of the redshift deduced from i-H P (Cl )
4
2
B
Q
Q
and o-H4P2+(Cl ) is a slightly greater than that observed from
−
2
B3LYP method
Total
Electron effect
Donating
Delocalization
Conformational effect
In-plane
ad-porphyrin I; this indicates that in-plane and out-of-plane dis-
tortion, in general, become admixed to a certain extent. From the
above analysis, we conclude that the conformational effect is gen-
erated primarily by out-of-plane distortion for porphyrin diacids.
Fig. 6 depicts the variation of spectral redshift induced by con-
formational distortion with the extent of meso-phenyl substitution
as predicted by the PBE1PBE and B3LYP methods; the calculated
data are fitted by a polynomial function. From Fig. 6, it is evident
that the PBE1PBE and B3LYP calculations have nearly conform-
ing predictions with regard to the conformational effect for both
zinc complexes and diacid derivatives. In general, conformational
distortion produces a profile with a positive and increasing slope
for both B and Q bands; this is similar to the performance of out-
of-plane distortions observed in the published literature [9]. The
profile of porphyrin diacid series has a larger slope than that of
30.2
26.8
19.6
7.2
32.8
26.6
16.2
10.4
6.2
43.7
30.1
22.5
8.7
3.4
11.1
1.8
11.3
17.7
4.5
16.9
1.0
3.7
Out-of-plane
2.4
2.5
PBE1PBE method
Total
Electron effect
Donating
Delocalization
Conformational effect
In-plane
30.9
27.3
19.3
8.0
34.1
28.0
16.1
11.9
6.1
44.3
30.9
17.8
13.1
10.7
1.4
48.0
43.1
12.6
30.5
16.9
3.4
3.6
0.9
3.2
a
a
Out-of-plane
2.7
2.9
11.3 (16.9 )
16.8 (24.2 )
a
With p-H4P2+(Cl )2 as reference.
−

Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
1459
5
4
3
2
1
0
0
0
0
0
0
Q band
Q band
2
0
0
/
/
zinc por.
por. diacid
1
0
5
40
30
20
10
0
1
B band
B band
/
/
zinc por.
por. diacid
1
0
5
0
0
1
2
3
4
0
1
2
3
4
meso-phenyl number
meso-phenyl number
Fig. 7. Predicted dependence of redshift caused by electronic effect of meso-phenyl
group on meso-phenyl number for zinc porphyrin (quadrangle) and porphyrin
diacids (triangle). The open data are predicted by PBE1PBE method and solid data
are predicted by B3LYP method.
Fig. 6. Predicted dependence of redshift caused by conformational effect on meso-
phenyl number for zinc porphyrin (quadrangle) and porphyrin diacids (triangle).
The open data are predicted by PBE1PBE method and solid data are predicted by
B3LYP method.
simulation of absorption spectra should consider the influence of
the counterion.
zinc porphyrins; this is attributed to the larger out-of-plane dis-
tortion in the former series as compared with the latter series.
Therefore, the conformational effect can be inferred to contribute
more significantly to spectral redshift in diacids series than in the
zinc complexes series.
The electronic effect of the meso-phenyl group predicted by the
PBE1PBE method generates a profile with a positive but decreasing
slope for the B band (Fig. 7). The B3LYP method provides a sim-
ilar prediction. It is apparent that the profile of zinc porphyrins
approaches that of the porphyrin diacids series, and the slight dis-
crepancy observed can be attributed to differences in the extent of
the ␲-electron delocalization effect.
2
+
−
The H4P (Cl ) has undergone significant out-of-plane dis-
2
2+
−
tortion; therefore, application of H4P (Cl ) derivatives in the
2
ad-porphyrin I series as the reference would underestimate the
redshift induced by out-of-plane distortion. The selection of a pla-
nar porphyrin model as the reference would generate a more
credible evaluation. Table S40 presents the TDDFT calculation (in
The PBE1PBE and B3LYP methods have similar predictions for
the Q band of zinc porphyrins. However, this is not applicable for the
Q band of porphyrin diacids; the profile predicted by the PBE1PBE
method has a much larger slope than that predicted by the B3LYP
method, which indicates that the simulation of the Q band of por-
phyrin diacids is quite sensitive to the calculation method applied.
Further differentiation of the ␲-electron delocalization effect
from the electron-donating effect can be performed by com-
parison of ad-porphyrin IV with the parent porphyrin or with
ad-porphyrin II. The TDDFT calculation of ad-porphyrin IV by the
PBE1PBE method (Table S40) demonstrates that the Q band is more
sensitive to the orientation of the meso-phenyl group than B band;
therefore, it is significantly redshifted under the influence of the
␲-electron delocalization effect. Nonetheless, the differentiation
of the delocalization and electron-donating effects by B3LYP cal-
culation has failed due to the smaller absorption wavelength of
the Q band predicted in the ad-porphyrin IV series in compari-
son with the ad-porphyrin II series (Table S41); this generates a
negative value for the electron-donating effect of the meso-phenyl
group. This indicates that the B3LYP method is less credible in the
evaluation of a weakly interacting system.
2
+
−
PBE1PBE method) of a planar derivative of H P (Cl ) in the ad-
4
2
porphyrin I (labeled as p-H4P2 (Cl ) ), wherein the pyrrolic ring is
+
−
2
adjusted to be coplanar with the mean plane of the porphyrin ring.
Employing this as a planar model for reference ascribes an addi-
tional 5.6 (B band) and 7.4 nm (Q band) redshift to out-of-plane
distortions.
(
2) Electronic effect of meso-phenyl group. The comparison of
TDDFT calculations of parent porphyrin diacids with that of ad-
porphyrin II series (Tables S40 and S41) can generate an estimation
of the redshift induced by the electronic effect of the meso-
phenyl group. The shift distance predicted by the PBE1PBE method,
3
0.9 nm (B band) and 43.1 nm (Q band), for porphyrin diacids
is greater than that of zinc porphyrin; this is attributed to the
probable enhanced interaction between the porphyrin and phenyl
rings caused by intensified out-of-plane distortion. In addition,
we attempted to evaluate the electronic effect by comparing ad-
porphyrin III and I series. However, the calculation generated a
more complicated B band for the ad-porphyrin III series (see the
2
+
result of H DPP (Cl)2 listed in Table S42, for example); therefore,
(3) Electronic effect of counterion. The comparison of the ad-
porphyrin I series with II series of the diacid species reveals the
4
this confounds the evaluation, and proves again that the theoretical

1
460
Y.-H. Zhang et al. / Spectrochimica Acta Part A 79 (2011) 1449–1460
−
effect of the counterion Cl . Both PBE1PBE and B3LYP methods
predict a greater redshifted B band and a less redshifted Q band
for the ad-porphyrin II series than for the ad-porphyrin I series
[4] C.-Y. Lin, Y.-C. Wang, S.-J. Hsu, C.-F. Lo, E.W.-G. Diau, J. Phys. Chem. C 114 (2010)
87.
5] Q. Huang, C.J. Medforth, R. Schweitzer-Stenner, J. Phys. Chem. A 109 (2005)
0493.
6
[
1
(
Tables S40 and S41). The less red-shifted Q band can be attributed
[6] A.G. DiMagno, A.K. Wertsching, C.R. Ross, J. Am. Chem. Soc. 117 (1995)
8279.
−
to reduced interaction between the porphyrin ring and the Cl with
the extent of meso-phenyl substitution.
[
7] A.K. Wertsching, A.S. Koch, A.G. DiMagno, J. Am. Chem. Soc. 123 (2001)
932.
3
[
[
8] H. Ryeng, A. Ghosh, J. Am. Chem. Soc. 124 (2002) 8099.
9] R.E. Haddad, S. Gazeau, J. Pecaut, J.-C. Marchon, C.J. Medforth, J.A. Shelnutt, J.
Am. Chem. Soc. 125 (2003) 1253.
4
. Conclusion
[
10] Z. Zhou, C. Cao, Q. Liu, R. Jiang, Org. Lett. 12 (2010) 1780.
In this study, we employed theoretical calculations to compare
[11] V.N. Knyukshto, K.N. Solovyov, G.D. Egorova, Biospectroscopy 4 (1998) 121.
[12] Y.-P. Ma, S.-G. He, X.-L. Ding, Z.-C. Wang, W. Xue, Q. Shi, Phys. Chem. Chem.
Phys. 111 (2009) 2543.
structural and spectral variations with the extent of meso-phenyl
substitution between porphyrin diacids and zinc porphyrins. The
PBE1PBE method generates more acceptable results in the TDDFT
calculation of porphyrin diacids than the B3LYP and PBEPBE
methods. The TDDFT calculation indicates that both out-of-plane
distortion and meso-phenyl substitution intensify spectral redshift.
[
13] T. Honda, T. Kojima, S. Fukuzumi, Chem. Commun. (2009) 4994.
[14] G.D. Luca, A. Romeo, L.M. Scolaro, G. Ricciardi, A. Rosa, Inorg. Chem. 46 (2007)
5979.
[
15] T. Nakanishi, K. Ohkubo, T. Kojima, S. Fukuzumi, J. Am. Chem. Soc. 131 (2009)
77.
5
[
16] V.S. Chirvony, A. Hoek, V.A. Galievsky, I.V. Sazanovich, T.J. Schaafsma, D. Holten,
J. Phys. Chem. B 104 (2000) 9909.
−
The influence of the counterion Cl is complicated. On one hand,
−
[17] I.V. Avilov, A.Yu. Pannarin, V.S. Chirvony, Chem. Phys. Lett. 389 (2004) 352.
18] S. Juillard, Y. Ferrand, G. Simonneaux, L. Toupet, Tetrahedron 61 (2005)
489.
[19] A. Rosa, G. Ricciardi, E.J. Baerends, A. Romeo, L.M. Scolaro, J. Phys. Chem. A 107
2003) 11468.
the introduction of the Cl would reduce the HOMO–LUMO gap
[
−
because of the electron-donating effect of the Cl ion, which then
3
brings about an additional spectral redshift with regard to the cal-
culation without consideration of counterion. On the other hand,
(
[
[
20] Y.-H. Zhang, W.-J. Ruan, Z.-Y. Li, Y. Wu, J.-Y. Zheng, Chem. Phys. 315 (2005) 201.
21] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
−
the interaction between the Cl ion and the porphyrin ring is weak-
ened by meso-phenyl substitution; this reduces the redshift size
of the Q absorption. The PBE1PBE calculation indicates that the
electron-donating effect is the key factor in the redshift of B and Q
bands of zinc porphyrin series as well as in the redshift of the B band
of porphyrin diacid series; however, the ␲ electron delocalization
effect is predicted to be the primary cause of the redshift of Q band
of porphyrin diacids series. Meanwhile, out-of-plane distortion is
predicted to provide a moderate contribution to the spectral red-
shift of porphyrin diacids series, because of intensified out-of-plane
distortion.
[22] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[
[
23] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 78 (1997) 1396.
24] R. Ma, P. Guo, H. Cui, X. Zhang, M.K. Nazeeruddin, M. Gratzel, J. Phys. Chem. A
113 (2009) 10119.
[25] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865.
[26] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian,
J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J.
Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A.
Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels,
M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman,
J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham,
C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen,
M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian Suite Gaussian 03, Revision C.01,
Gaussian, Inc., Wallingford, CT, 2004.
Acknowledgment
We thank Professor Shelnutt for kindly helping in NSD calcula-
tion and interpretation.
[
27] (a) S.I. Gorelsky, SWizard Program, University of Ottawa, Ottawa, Canada, 2010,
http://www.sg-chem.net/;
Appendix A. Supplementary data
(b) S.I. Gorelsky, A.B.P. Lever, J. Organomet. Chem. 635 (2001) 187.
28] D.-M. Chen, X. Liu, T.-J. He, F.-C. Liu, Chem. Phys. 289 (2003) 397.
29] NSD calculation was carried out online at http://jasheln.unm.edu/
jasheln/content/nsd/NSDengine/start.htm.
[30] Y.-H. Zhang, W. Zhao, P. Jiang, L.-J. Zhang, T. Zhang, J. Wang, Spectrochim. Acta
A 75 (2010) 880.
[
[
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.saa.2011.04.085.
[
31] W.R. Scheidt, J.U. Mondal, C.W. Eigenbrot, A. Adler, L.J. Radonovich, J.L. Hoard,
Inorg. Chem. 25 (1986) 795.
References
[
32] K.A. Nguyen, P.N. Day, R. Pachter, J. Phys. Chem. A 103 (1999) 9378.
[
[
[
1] M.O. Senge, in: K.M. Kadish, K.M. Smith, R. Guilard (Eds.), The Porphyrin Hand-
book, vol. 1, Academic Press, Boston, 2000, p. 239.
2] J.A. Shelnutt, C.J. Medforth, M.D. Berber, K.M. Barkigia, K.M. Smith, J. Am. Chem.
Soc. 113 (1991) 4077.
3] L.D. Spaulding, C.C. Chang, N.-T. Yu, R.H. Felton, J. Am. Chem. Soc. 97 (1975)
[33] B. Cheng, O.Q. Munro, H.M. Marqeus, W.R. Scheidt, J. Am. Chem. Soc. 119 (1997)
10732.
[34] PBE1PBE calculation on (Cl )2 cluster that has similar structure with the coun-
−
2
+
−
−
terpart in H4TPP (Cl )2 indicates that the p orbit of (Cl )2 involved in G-HOMO
orbit has much higher level.
2
517.