Substituent and solvent effects on the hyperporphyrin spectra of diprotonated tetraphenylporphyrins

Article abstract of DOI:10.1021/jp022046f

UV-visible spectra was studied for a series of p-substituted tetraphenylporphyrins (TPP) titrated with strong acid in various solvents. The longest wavelength absorption at 811 nm, however, was the greatest shift for any diprotonated substituted TPP. The magnitude of the shift was enhanced by a solvent effect. Diprotonated TAPP could be observed 10 fluoresce weakly in dichloromethane (DCM) solvent but not in dimethyl sulfoxide. DCM, which light exposure in the presence of a poryphyrin generated sufficient acid to significantly affect the spectrum. The effect was noted quantitatively by placing a dilute (4μM) TTP solution in a diode array spectrophotometer; an effective acid titration could be observed simply by exposing the sample to the monitoring beam over a period of 2 min.

Full text of DOI:10.1021/jp022046f

3486
J. Phys. Chem. A 2003, 107, 3486-3496
Substituent and Solvent Effects on the Hyperporphyrin Spectra of Diprotonated
Tetraphenylporphyrins†
James R. Weinkauf, Sharon W. Cooper, Aaron Schweiger, and Carl C. Wamser*
Department of Chemistry, Portland State UniVersity, Portland, Oregon 97207-0751
ReceiVed: September 11, 2002; In Final Form: NoVember 13, 2002
UV-visible spectra have been studied for a series of p-substituted tetraphenylporphyrins (TPPs) titrated with
strong acid in various solvents. Substituent effects on the Soret and Q(I) absorption peaks and the fluorescence
emission peaks have been treated by Hammett correlations. In general, there are only small effects with
electron-withdrawing substituents, but electron-donating substituents lead to lower energy transitions, with
especially strong effects observed in the case of the diprotonated porphyrins with good electron-donating
substituents (hyperporphyrin effects). The hyperporphyrin effects are attributed to the crossing of a π molecular
orbital on the substituted phenyl group above the usual porphyrin π highest occupied molecular orbital. For
the neutral TPPs, this crossing is estimated to occur with a substituent as electron-donating as p-methoxy,
and for the diprotonated TPPs, the crossing occurs at approximately unsubstituted TPP. Distinctive solvent
effects on the spectra and the Hammett correlations are observed.
8
Introduction
protonated TPP shows a nonplanar (saddle) form, in which the
four N-H bonds adopt positions alternately above and below
the mean porphyrin plane; this twisting allows the phenyl rings
to become more coplanar with the mean molecular plane. Thus,
it has been of some interest to determine the extent to which
peripheral substituents can interact with the central porphyrin
positions, where the charges are nominally located.
Porphyrins have a wonderful repertoire of optical and
electronic properties that allow them to play a variety of roles,
in nature and in technical applications. The spectroscopy of
1
porphyrins has received detailed treatment because of its
2
relevance to functionality as well as its fascinating variety. In
the ultraviolet-visible region, the characteristic spectrum
includes the intense Soret band around 400 nm and a series of
bands (Q bands) through the visible range. Different patterns
of peripheral substitutions, π conjugations, or central metal
Systematic effects of para substituents on the properties of
9,10
TPPs have been studied electrochemically and spectroscopi-
11-13
cally.
In general, electron-donating substituents lead to red
shifts for both the Soret (B) and Q bands, maintaining
2
incorporations lead to well-characterized spectral patterns.
1
3
approximately a constant B-Q splitting. The magnitude of
Specialized spectral features also appear based on interactions
the energy shifts can be correlated with Hammett substituent
3
between porphyrin units, such as H and J aggregates or
1
2
constants, either a mix of resonance and inductive components
interactions with other molecules based on charge, π stacking,
H-bonding, or hydrophobic effects.4 The Gouterman four
orbital model suffices to explain most of the fundamental
+
13
9,10
or σ values. Substituent effects on the redox potentials
,5
1
2
and pKa values also correlate with Hammett substituent
constants.
2
features of porphyrin and metalloporphyrin electronic spectra,
Nevertheless, several studies of TPPs substituted with highly
electron-donating groups have shown unusual spectroscopic
effects. Studies with para-hydroxy TPP derivatives have il-
lustrated numerous spectral and structural features, complicated
by accompanying redox processes leading to paramagnetic
but there are increasingly common examples of unusual spectral
features that require additional interpretation.
Acid-base behavior of porphyrins has also been well-studied.
The tetrapyrrole skeleton allows for loss of two N-H protons
to form a dianion in strong base or gain of two protons to form
a dication in acid. Each case leads to an increase in symmetry
and corresponding changes in the spectra, which are often
distinctive in their colors (e.g., purple tetraphenylporphyrins
14-16
species and quinonelike structures.
Para-amino TPP deriva-
tives have long been known to have unusual spectral properties,
in particular broader, lower energy bands for both the Soret and
1
7,18
the visible regions.
Protonation of these amino-substituted
(TPPs) change to green in acid). Moreover, peripheral substit-
derivatives leads to further red shifts and unusual red bands.
For example, tetrakis(p-dimethylaminophenyl)porphyrin (TD-
MAPP) can be titrated with acid to show protonated derivatives
uents may confer additional acidic or basic properties, a tactic
often used to control water solubility.4
The basic structure of the porphyrin framework can be
described as a planar aromatic system, but steric effects around
the periphery, in particular meso substituents that interact with
1
9
from +2 to +6. The diprotonated form shows an unusually
intense red band, described as a hyperporphyrin. A similar band
has been observed in long chain dialkylamino-substituted
6
â pyrrole positions, often lead to nonplanar forms of porphyrins.
2
0
porphyrins in thin films.
TPP has been observed in three different crystal forms, but in
each case, the meso phenyl groups are significantly twisted from
Hyperporphyrins have been defined as porphyrins that exhibit
extra bands in the UV-visible, either due to specific interactions
with metal orbitals or due to charge transfer interactions with
7
the plane of the porphyrin. The X-ray structure of doubly
2
†
substituents. Molecular orbital calculations supporting the
Part of the special issue “George S. Hammond & Michael Kasha
Festschrift”.
charge transfer nature of hyperporphyrin bands in diprotonated
1
0.1021/jp022046f CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/20/2002

Spectra of Diprotonated Tetraphenylporphyrins
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3487
mL of propionic acid and brought to reflux. Pyrrole (6.2 mL,
8
(
4
9 mmol) was added, followed by para-methoxybenzaldehyde
8.9 mL, 73 mmol), and the solution was stirred at reflux for
5 min. The reaction mixture was allowed to cool to room
temperature and then placed in an ice bath, yielding purple
crystals. Thin-layer chromatography (TLC) indicated two major
components, identified as TMPP and HM3PP. Several mil-
ligrams of pure HM3PP were isolated by flash chromatography
using toluene/chloroform solutions as eluting solvents. NMR
(
7
CDCl3): -2.75 ppm (s, 2H, NH), 4.09 ppm (s, 9H, OCH3),
.20 ppm (d, 2H, H-ArOH), 7.28 ppm (d, 6H, H-ArOCH3), 8.07
ppm (d, 2H, H-ArOH), 8.14 ppm (d, 6H, H-ArOCH3), 8.86 ppm
Figure 1. Tetrasubstituted porphyrins and abbreviations. TPP, X )
H; TTP, X ) CH
NH ; TDMAPP, X ) N(CH
TCMPP, X ) CO CH . Mixed substituent porphyrins: H
OH or OCH ; H PP, H PP, H PP. DMA PP, X ) H or
N(CH ; DMA PP, DMA PP, DMA PP.
3
; THPP, X ) OH; TMPP, X ) OCH
3
; TAPP, X )
; TCPP, X ) COOH;
PP, X )
(
7
broad singlet, 8H, pyrroles). ICR-MS: M + 1 ) 721.25 (calcd
2
3
)
2
; TNPP, X ) NO
2
21.297).
2
3
n m
M
H2M2PP and H3MPP. Although the above preparation yielded
3
M
1 3
2
M
2
M
3 1
n
3
)
2
1
2
3
traces of other porphyrins, improved yields of the other
substituted derivatives were obtained by using a different ratio
of the two benzaldehydes. The above procedure was modified
by using para-hydroxybenzaldehyde (2.5 g, 20 mmol), para-
methoxybenzaldehyde (4.5 mL, 37 mmol), and pyrrole (5.5 mL,
80 mmol). The reaction mixture did not yield crystals but a
thick tar, which was treated with toluene and filtered. The
toluene solution was evaporated, and the solid residue was
chromatographed several times on silica gel using chloroform
eluent and then 1-5% MeOH in chloroform. The cis and trans
isomers of H2M2PP were not separated.
TAPP have recently been presented.21 We have collected UV-
visible absorption and fluorescence spectroscopy data on a series
of substituted TPPs (Figure 1) to categorize the effects of
substituents on the spectra. We have also investigated the role
of solvent effects on the spectra.
Experimental Section
Materials. Solvents and reagents were typically the highest
grade commercially available and were stored under nitrogen
with molecular sieves. Methanol (MeOH), acetonitrile (ACN),
and tetrahydrofuran (THF) were Sigma-Aldrich high-perfor-
mance liquid chromatography (HPLC) grade, 99.9% with no
inhibitors. Dimethyl sulfoxide (DMSO) and dimethylformamide
H2M2PP. NMR (CDCl3): -2.75 ppm (s, 2H, NH), 4.12 ppm
(s, 6H, OCH3), 7.23 ppm (d, 4H, H-ArOH), 7.30 ppm (d, 4H,
H-ArOCH3), 8.09 ppm (d, 4H, H-ArOH), 8.14 ppm (d, 4H,
H-ArOCH3), 8.86 ppm (broad singlet, 8H, pyrroles). MALDI-
MS: M + 1 ) 707.27 (calcd 707.266).
(DMF) were Aldrich spectrometric grade, 99.9 and 99.8%,
H3MPP. NMR (CDCl3): -2.76 ppm (s, 2H, NH), 4.12 ppm
respectively. Dichloromethane (DCM) was from Mallinckrodt,
indicating 0.0007 mequiv/g titratable acid; however, this source
of DCM did not detectably protonate the porphyrins, a common
problem with most chlorinated solvents from most other sources.
Acids used for titrations were either trifluoroacetic acid (TFA)
or methanesulfonic acid (MSA). TFA was from Aldrich, 99+%
spectrometric grade, and MSA was from J. T. Baker, 98%.
Water was distilled water available in the lab, boiled before
use.
(s, 3H, OCH3), 7.23 ppm (d, 6H, H-ArOH), 7.31 ppm (d, 2H,
H-ArOCH3), 8.09 ppm (d, 6H, H-ArOH), 8.14 ppm (d, 2H,
H-ArOCH3), 8.88 ppm (broad singlet, 8H, pyrroles). MALDI-
MS: M + 1 ) 693.27 (calcd 693.250).
Spectroscopy. UV-visible absorption spectra were taken on
a Shimadzu model 260 using a 2 mm slit width. Data acquisition
was via a GPIB card and a PC using Shimadzu UV-265 software
(version 3.1). Fluorescence spectra were taken on a Spex
Fluorolog model 112 spectrofluorimeter using a 150 W xenon
light source and right angle detection with optically dilute
samples (<1 µM). For emission spectra, the excitation slit width
was 5 mm and the emission slit width was 0.5 mm. The reverse
was set for excitation spectra. For determination of relative
emission intensities, excitation was at the isosbestic point of
the neutral and diprotonated porphyrins to allow for equivalent
absorbance. All spectra were plotted and analyzed with Igor
Pro (version 3.01). Fluorescence lifetimes were measured for a
few selected samples at Jobin Yvon-Spex, using a Spex
Fluorolog τ fluorimeter. Ion cyclotron resonance mass spec-
trometry (ICR-MS) was performed at the Pacific Northwest
National Laboratories using electrospray ionization. Matrix-
assisted laser desorption-ionization mass spectrometry (MALDI-
MS) was performed at Washington State University.
All of the porphyrins used were tetrasubstituted in their meso
positions with para-substituted phenyl groups. The various
symmetrically substituted porphyrins used in this study are
shown in Figure 1. The porphyrins were either purchased
commercially or synthesized by standard techniques. TPP was
purchased from Strem Chemicals (Newburyport, MA), TCPP
and TCMPP were purchased from Porphyrin Products (now
Frontier Scientific, Logan, UT), and TAPP was purchased from
TCI America (Portland, OR). THPP was prepared by demethy-
22
lating TMPP using pyridinium hydrochloride. TTP, TMPP,
and TNPP were prepared from pyrrole and the corresponding
para-substituted benzaldehyde by the standard Adler-Longo
23
procedure, and TDMAPP was prepared by a slightly modified
1
9
procedure.
Porphyrins containing different substituents were prepared
in the cases of HnMmPP (hydroxy or methoxy substituents) and
DMAnPP (dimethylamino substituents or unsubstituted posi-
Titrations. Porphyrins were protonated by strong acid (either
MSA or TFA) by titrations in a UV cuvette while monitoring
the spectra between 300 and 900 nm. Neat or concentrated
solutions of acid were added in 2 µL aliquots using a Gilson
20 µL Pipetman micropipet. The original porphyrin solution
was 6 µM or less in a total volume of 3.0 mL; corrections were
not made for change in total volume during the titration
(typically less than 100 µL). After addition of each aliquot, the
cell was capped and mixed by inversion, and the spectrum was
1
9
tions). Syntheses of the DMAnPP series have been reported.
The HnMmPP series was prepared by the Adler-Longo proce-
dure with a mix of benzaldehydes and chromatographic separa-
tion of the various derivatives as follows.
Preparation of Hydroxy/Methoxy-Substituted TPPs.
HM3PP. para-Hydroxybenzaldehyde (2.0 g, 16 mmol), previ-
ously recrystallized from 3:1 acetone:toluene, was added to 250

3488 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Weinkauf et al.
Figure 2. Titration of 5.23 µM TPP with MSA in DMSO.
Figure 3. Titration of 5.84 µM THPP with TFA in DMSO.
retaken. In several cases, the reversibility of the protonation
was demonstrated by addition of 1.0 M tetrabutylammonium
hydroxide in MeOH.
to a red shift and weakening of the Soret band, disappearance
of most of the Q bands due to higher symmetry, and for the
Q(I) band, a small red shift (646-659 nm) with a substantially
increased intensity.
For tetraarylporphyrins with good electron-donating substit-
uents, a modified pattern is observed upon protonation, as
illustrated in Figure 3 for THPP. Effects on the Soret band and
the Q bands are relatively similar to what is observed for TPP,
except for the Q(I) band, which is more strongly red-shifted
and more strongly increased in intensity. A similar red shift is
Results
Porphyrin Absorption and Emission Spectra. For refer-
ence, the normal UV-visible spectroscopic characteristics of
the meso-tetraarylporphyrins and their diprotonated forms are
illustrated in Figure 2 for unsubstituted TPP. Protonation leads

Spectra of Diprotonated Tetraphenylporphyrins
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3489
Figure 4. Fluorescence spectra of 0.78 µM THPP in DMSO.
observed in the fluorescence spectrum, but the emission intensity
is substantially reduced, as illustrated in Figure 4 for neutral
and diprotonated THPP.
For even more strongly electron-donating substituents, in
particular amino substituents, the trends are even more striking.
The complete titration of TAPP with MSA in DMSO is
illustrated in Figure 5, revealing protonation states +2 through
donating or electron-withdrawing solvents, i.e., there is a break
in the Hammett plot at TPP. The electron-donating effect is
much stronger (F ) +1.2/-0.4 in DCM; +1.5/-0.4 in DMSO).
For the Q(I) band, the substituent effects are smaller, with no
significant shifts for electron-withdrawing substituents (F )
+0.4/0 in DCM; +0.7/0 in DMSO).
Substituent effects on the dication absorption and emission
+
+
6, completely analogous to what was observed by Ojadi et
bands correlate better with σ rather than σ values. If the data
19
al. for TDMAPP. The longest wavelength absorption at 811
nm, however, is the greatest shift we have found for any
diprotonated substituted TPP. The magnitude of the shift is
enhanced by a solvent effect, as discussed in the next section.
Diprotonated TAPP can be observed to fluoresce weakly in
DCM solvent but not in DMSO. Fluorescence from diprotonated
TDMAPP has not been detected in any solvent.
are plotted vs σ, dication Soret bands show a red shift for either
electron-donating or electron-withdrawing solvents, extremely
similar to what is observed for the neutral Soret data. Dication
+
Soret data are plotted vs σ using only the electron-donating
+
substituents, since electron-withdrawing substituents have σ
values very similar to their σ values. As compared to the F
+
values from plots vs σ, F values are smaller, but this is an
+
+
Substituent Effects. In an effort to better understand the role
of electron-donating substituents in the generation of hyper-
porphyrin spectra, a series of para-substituted porphyrins were
prepared and titrated with acid. The observed effects on the peak
positions of the Soret bands, the Q(I) bands, and the fluorescence
emission are summarized in Table 1.
artifact of the larger values of σ relative to σ (F ) +0.65 in
DCM; +0.61 in DMSO). For the Q(I) band, the effects are larger
+
(F ) +0.9 in both DCM and DMSO). However, the data for
diprotonated TAPP Q(I) in DMSO fall distinctly off the linear
correlation, with the red shift under those conditions much larger
than expected.
The data from Table 1 were plotted as a function of Hammett
substituent constant (4σ, since there are four substituents), using
either the standard σ values or σ values (Figures 6-9).
Summary results are collected in Table 2.
The Hammett correlations in Figures 6-9 include data for
the Soret and Q(I) bands for both the neutral and the diproto-
nated forms, in both DCM and DMSO solvents. Slopes (F
values) were obtained using energy as the y-axis, consistent with
the typical expression of Hammett correlations as linear free
energy relationships. Data in additional solvents are summarized
in the following section, but these two solvents are both
representative and distinctive, especially for the most extreme
Solvent Effects. To help clarify the role of solvent in the
observed spectra, each of the substituted TPP derivatives was
studied in a variety of solvents and Hammett correlations were
plotted for each solvent. Table 3 indicates the different solvent
parameters that were considered when attempting to correlate
the effects on the spectra. None were very effective at revealing
correlations with the observed spectral shifts, but the ET (30)
scale was considered the best as well as the most relevant, since
it is derived from a solvent-induced shift in the position of a
charge transfer absorption band. ET is a similar scale,
normalized to the full range of polarity observed for ET (30).
Figures 10 and 11 show the data for Soret and Q(I) band
positions for both neutral and diprotonated forms of TPP and
four derivatives with electron-donating substituents, in six
different solvents. The Soret bands show a slight trend toward
increasing energy in more polar solvents, although the scatter
is substantial. The Q(I) band correlations are remarkably flat,
with similarly large scatter. Noteworthy, however, are the data
for diprotonated TAPP, which scatter widely over a large energy
+
2
4
N
(amino-substituted) hyperporphyrin behavior. Peak fluorescence
emission wavelengths for the diprotonated forms are also
presented to get an alternative tracking of the lowest excited
states.
Substituent effects on the neutral absorption bands correlate
+
better with σ rather than σ values. The Soret band shows a
red shift, relative to unsubstituted TPP, for either electron-

3490 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Weinkauf et al.
Figure 5. Titration of 6.0 µM TAPP with MSA in DMSO, revealing various stages of protonation.

Spectra of Diprotonated Tetraphenylporphyrins
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3491
TABLE 1: Absorption and Fluorescence Data for Substituted TPPs (First Entry in DCM Solvent/Second Entry in DMSO
Solvent, NA ) Data Not Available)
Absorption Data
Hammett
Soret (nm)
dication
Q(I) (nm)
porphyrin
subst
-NO
σ
σ+
neutral
shifta
neutral
dication
shifta
TNPP
TCMPP
TCPP
TPP
2
0.81
0.47
0.44
0.00
-0.14
-0.28
-0.22
-0.57
-0.63
0.79
0.48
424/424
418/419
NA/420
416/417
417/419
421/423
419/424
427/438
436/NA
447/453
444/447
NA/447
438/443
442/447
451/455
452/458
460/466
470/477c
23/29
26/28
NA/27
22/26
25/28
30/32
33/34
33/28
34/NA
646/645
645/644
NA/645
646/646
648/648
651/651
648/653
655/669
663/NA
655/657
655/656
NA/657
657/659
670/671
695/696
706/711
725/811
773/807c
9/12
10/12
-CO
2
CH
3
-COOH
-H
0.42
NA/12
11/13
0.00
TTP
-CH
3
-0.31
-0.78
-0.92
-1.3
-1.7
22/23
TMPP
THPP
TAPP
TDMAPP
-OCH
-OH
3
44/45
b
58/58
-NH
-N(CH )
3 2
2
70/142
110/NA
Fluorescence Data
Hammett
fluorescence peak (nm)
dication
Stokes shift (nm)
porphyrin
subst
σ
σ+
neutral
shifta
neutral
dication
TNPP
TCPP
TPP
-NO
2
-COOH
-H
0.81
0.44
0.00
-0.14
-0.28
-0.22
-0.57
-0.63
0.79
0.42
NA/650
NA/650
650/649
652/650
658/656
654/661
669/699
686/NA
NA/699
NA/700
684/699
689/709
730/743
740/760
782/d
NA/49
NA/50
34/50
37/59
72/87
86/99
113/d
d/d
NA/5
NA/5
4/3
NA/42
NA/43
27/40
19/38
35/47
34/49
57/d
0.00
TTP
-CH
3
-0.31
-0.78
-0.92
-1.3
-1.7
4/2
TMPP
THPP
TAPP
TDMAPP
-OCH
-OH
3
7/5
6/8
14/30
23/NA
b
-NH
2
-N(CH
3
)
2
d/d
d/d
a
b
Red shift between the corresponding bands of the neutral and dication porphyrin. A “universal” value, among many suggested for the hydroxy
group.2
5
c
In propylene carbonate rather than DMSO. No fluorescence detected.
d
Figure 6. Absorption and emission data in DCM vs Hammett σ constants.
range and correlate with a trend to lower energy in more polar
solvents. A tabulation of the data used for Figures 10 and 11
appears in the Supporting Information for this paper, along with
a tabulation of the results of Hammett correlations in the
different solvents.
acid to significantly affect the spectrum. The effect was noted
quantitatively by placing a dilute (4 µM) TTP solution in a diode
array spectrophotometer; an effective acid titration could be
observed simply by exposing the sample to the monitoring beam
over a period of 2 min. The amounts of acid used for intentional
titrations were not corrected for amounts photochemically
generated; typical spectra were taken with a single pass, quickly
An unexpected solvent effect was noted in DCM, in which
light exposure in the presence of a porphyrin generated sufficient

3492 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Weinkauf et al.
Figure 7. Absorption and emission data in DCM vs Hammett σ+ constants.
Figure 8. Absorption and emission data in DMSO vs Hammett σ constants.
enough that this should have been a minor effect. The release
of HCl from chlorinated solvents is a well-known problem, but
this photochemical generation may be a unique and potentially
useful means of controlled release.

Spectra of Diprotonated Tetraphenylporphyrins
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3493
Figure 9. Absorption and emission data in DMSO vs Hammett σ+ constants.
TABLE 2: Summary of Hammett Correlations (Boldfaced
in DMSO solvent, as summarized in Table 4. Because the
substituent effects of hydroxy and methoxy are very similar,
the observed data were also very similar. However, it seems
clear that the hydroxy substituent acts as the more electron-
donating substituent in the context of the spectroscopic trends
under consideration.
+
Values Represent the Better Linear Fit between G or G )
DCM solvent
DMSO solvent
F+
F+
F
F
e-donating/ e-donating e-donating/ e-donating
e-withdrawing
only
e-withdrawing
only
absorption
0.45
neutral, Soret
neutral, Qx(0,0)
dication, Soret
dication, Q(0,0)
1.2/-0.4
0.4/0
1.5/-0.4
0.7/0
0.57
0.24
0.61
Discussion
0.16
1.6 /-0.4
0.65
0.93
1.4 /-0.5
Hyperporphyrin spectra that are based on substituent effects
a
2.2/0
3.6/0
0.9
(rather than effects of central metal orbitals) are generally
emission
19
considered to arise from charge transfer transitions. Recent
molecular orbital calculations on TAPP and its protonated forms
suggest that the effect originates from π orbitals that are
dication, Q(0,0)
2.7
1.1
2.6
0.85
a
+
Excluding TAPP data; including those data, F ) 1.4, with a poorer
correlation.
21
primarily localized on the aminophenyl substituent. These
orbitals extend over the amino group as well as the phenyl π
system and are calculated to correspond to the highest occupied
molecular orbital (HOMO) and HOMO-1 for diprotonated TAPP
TABLE 3: Solvent Polarity Parameters
dielectric
constant
N
solvent
formula
HOH
E
T
(30)
E
T
bp (°C)
(presumably additional calculations would locate all four of these
H
2
O
63.1
55.4
45.6
45.1
43.8
40.7
37.4
1.000
0.762
0.460
0.444
0.404
0.309
0.207
78.5
32.6
36.2
47.6
36.7
8.9
100.0
64.7
81.6
orbitalssdiprotonated monoamino TPP shows just a single
aminophenyl orbital as its HOMO). The corresponding lowest
unoccupied molecular orbital (LUMO) in these cases is a π*
orbital extending only over the porphyrin core. Thus, the lowest
energy (hyperporphyrin) transition is assigned to π(phenyl) f
π*(porphyrin), which can be construed as a charge transfer
transition. The reason the π(phenyl) orbital appears as the
HOMO in these cases is a combination of the electron-donating
effect of the amino group and the depression of core porphyrin
π and π* orbitals upon diprotonation. The π(phenyl) orbital is
not calculated to be the HOMO or HOMO-1 in the cases of
unsubstituted diprotonated TPP, neutral TAPP, fully protonated
(+6) TAPP, or fully protonated (+3) monoamino TPP.21
Experimental data on the oxidation and reduction potentials
for many of these same TPP derivatives have been reported.9
MeOH
ACN
CH
CH
SO
HCON(CH
CH Cl
cyclo-C
3
OH
CN
3
DMSO (CH
3
)
2
189.0
153.0
40.1
DMF
DCM
THF
3 2
)
2
2
H
4 8
O
7.4
65.4
Effects of Mixed Substituents. Because all of the TPP
derivatives initially studied for this work were symmetrically
tetrasubstituted, it was considered desirable to investigate
additional substitution patterns. Studies of TPPs with various
numbers of dimethylamino substituents had been reported and
provided highly important information with respect to the roles
of the substituents. TPPs with mixes of hydroxy and methoxy
substituents were prepared, and their spectra were determined
19
,10

3494 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Weinkauf et al.
Figure 10. Solvent effects on the Soret band of neutral and diprotonated porphyrins.
Figure 11. Solvent effects on the Q(I) band of neutral and diprotonated porphyrins.
Increased electron donation leads the observed half-wave
potentials (oxidation and two reductions) to shift to less positive
potentials, all with slopes around 60 mV based on standard
Hammett σ constants. These trends are interpreted as simple
substituent effects on the energy levels of the porphyrin π orbital
HOMO (one electron oxidation) and porphyrin π* orbital
LUMO (one electron reduction). The substituent effects must
be mediated from the para position through the phenyl group
to the porphyrin core, even though for the neutral TPPs these
ring systems are distinctly noncoplanar. Because the observed
substituents effects on the HOMO and LUMO run parallel, with
comparable slopes, this implies a relatively constant transition
energy, as is observed for most TPP derivatives. Indeed,
Hammett plots for the Q(I) transitions of substituted TPPs show
very low slopes, as shown in Figures 6-9. However, oxidation
potentials for the two most electron-donating substituents (THPP

Spectra of Diprotonated Tetraphenylporphyrins
J. Phys. Chem. A, Vol. 107, No. 18, 2003 3495
TABLE 4: Absorption and Fluorescence Data for Mixed Substituent TPPs (All in DMSO Solvent)
Absorption Data
Soret (nm)
dication
Q(I) (nm)
porphyrin
neutral
shifta
neutral
dication
shifta
TPP
417
423
423
423
424
424
443
455
455
456
457
458
26
32
32
33
33
34
646
651
649
652
651
653
659
696
700
703
707
711
13
45
51
51
56
58
TMPP
1
H M
2
H M
3
H M
3
2
1
PP
PP
PP
THPP
Fluorescence Data
Stokes shift (nm)
fluorescence peak (nm)
isosbestic
fluorescence
porphyrin
neutral
dication
shifta
neutral
dication
excit (nm)
rel yield
lifetime (ns)
TPP
649
656
657
658
658
661
699
743
750
749
755
760
50
87
93
91
97
99
3
5
8
6
7
8
40
47
50
46
42
49
427
434
434
434
435
436
0.30
0.18
0.15
0.13
0.11
0.12
9.3
8.6
8.5
8.5
8.4
8.4
TMPP
H
1
H
2
H
3
M
M
M
3
PP
2
PP
1
PP
THPP
a
Red shift between the corresponding bands of the neutral and dication porphyrin.
TABLE 5: Oxidation-Reduction Data and Prediction of Transition Energies
Hammett
E(ox)a
E(red)a
calcdb
neutral
dication
porphyrin
subst
-CO CH
σ
(V)
(V)
∆E (eV)
Q(I) (eV)
Q(I) (eV)
TCMPP
TPP
TMPP
THPP
TAPP
2
3
0.47
0.00
-0.28
-0.38
-0.57
1.14
1.04
0.94
0.75
0.48
-0.92
-1.03
-1.08
-1.12
-1.18
2.06
2.07
2.02
1.87
1.66
1.93
1.92
1.90
1.90
1.85
1.89
1.88
1.78
1.74
1.53
-H
-OCH
-OH
3
-NH
2
a
Oxidation and reduction data are half-wave potentials measured in DMSO by cyclic voltammetry.10 b Calculated transition energies are simply
the difference between E(ox) and E(red).
Figure 12. HOMO and LUMO energy levels estimated from measured oxidation and reduction potentials of the neutral porphyrins.10 Corresponding
levels for diprotonated porphyrins are estimated (dashed lines) to illustrate the lower energy transitions observed in the dications.
and TAPP) are substantially lower (less positive) than predicted,
i.e., there is a HOMO higher in energy than predicted based
simply on substituent effects on the porphyrin. The interpretation
is that substituent effects in these cases are acting directly on
the phenyl ring to which they are attached, and they affect the
oxidation potential strongly because this orbital has become the
HOMO.
Table 5 shows measured oxidation and reduction potentials
for selected substituted TPPs in DMSO and calculates a
transition energy based simply on the difference between them.
Figure 12 illustrates these trends, assigning the oxidation
potential to the HOMO energy level and the reduction potential
to the LUMO energy level. In this illustration, the hyperpor-
phyrin effect can be seen as a crossing of the phenyl π orbital
over the porphyrin π orbital, creating a different HOMO and
thereby a charge transfer transition at lower energy. The crossing
apparently occurs at about the position of TMPP, corresponding
to the break in the Hammett plot observed for oxidation
potentials. An estimate of the energy level changes brought
about by diprotonation is illustrated with the dashed lines. On
the basis of the shifts observed in the hyperporphyrin cases, a
shift of about +0.2 eV from neutral to dication is applied to
the porphyrin orbitals but not the phenyl orbitals. If this is the
case, the crossing of the phenyl π orbital over the porphyrin π
orbital occurs at about the position of TPP, as is observed with
the Hammett plots of dication Q(I) bands (Figures 6-9).
1
0

3496 J. Phys. Chem. A, Vol. 107, No. 18, 2003
Weinkauf et al.
Figure 13. Comparison of observed Q(I) transition energies (in eV) with those calculated from oxidation and reduction potentials for neutral
porphyrins (as in Figure 12). The lines simply connect the points.
Figure 13 compares the calculated transition energies with
the observed Q(I) bands for both the neutral and the diprotonated
porphyrins. In the case of the dications, the calculated values
follow the observed trend very well, although the calculated
values are uniformly high by about 0.2 eV. In the case of neutral
TPPs, the calculation predicts a hyperporphyrin effect much
stronger than is observed (or earlier than is observed).
Press: New York, 1978; Vol. III, pp 1-165.
3) Akins, D. L.; Zhu, H.; Guo, C. J. Phys. Chem. 1996, 100, 5420-
425.
4) Hambright, P. Chemistry of Water-Soluble Porphyrins. In The
Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.;
Academic Press: New York, 2000; Vol. 3, pp 129-210.
(
5
(
(5) Clarke, S. E.; Wamser, C. C.; Bell, H. E. J. Phys. Chem. A 2002,
1
06, 3235-3242.
6) http://www-chem.ucdavis.edu/groups/smith/chime/Porph_Struct/
start_here.html, accessed August 2002.
7) Silvers, S. J.; Tulinsky, A. J. Am. Chem. Soc. 1967, 89, 3331-
3337.
(
Acknowledgment. The majority of this work comprised the
M.S. Thesis project of J.R.W. Additional experimental work
was done by S.W.C., sponsored by the Research Corporation
Partners in Science Program, by A.S. and Mary Ann Brennan,
who were sponsored by the Apprenticeships in Science and
Engineering Program, and by Nick Shults and HongDiem Thi
Nguyen. ICR-MS spectra were taken at the Pacific Northwest
National Laboratories Environmental Molecular Sciences Lab
Facilities through the cooperation of Dr. Gordon Anderson.
MALDI-MS spectra were taken at Washington State University
LBB2 Analytical Services Laboratory through the cooperation
of Dr. William Siems. Fluorescence lifetime measurements were
performed at the University of Texas Fast Kinetics Laboratory
through the cooperation of Professor Stephen Webber or at Spex
Industries through the cooperation of Dr. Sal Atzemi. Synthetic
assistance was provided by Dr. Peter Pessiki of The Evergreen
State College. Helpful correspondence and sharing of unpub-
lished data with Professor Henry Linschitz and Dr. Hong Wang
of Brandeis University and discussions with Professor Martin
Gouterman of The University of Washington are gratefully
acknowledged. Partial funding of the work came from the U.S.
Department of Energy Basic Energy Sciences/Solar Photochem-
istry Program, Grant No. DE-FG06-90ER14131, and from the
Portland State University Faculty Development Program.
(
(8) Stone, A.; Fleischer, E. B. J. Am. Chem. Soc. 1968, 90, 2735-
748.
9) Kadish, K. M.; Morrison, M. M. J. Am. Chem. Soc. 1976, 98, 3326-
328.
10) Ransdell, R. A.; Wamser, C. C. J. Phys. Chem. 1992, 96, 10572-
10575.
2
3
(
(
(11) Meot-Ner, M.; Adler, A. D. J. Am. Chem. Soc. 1972, 94, 4763-
764.
12) Meot-Ner, M.; Adler, A. D. J. Am. Chem. Soc. 1975, 97, 5107-
110.
13) Dalton, J.; Milgrom, L. R.; Pemberton, S. M. J. Chem. Soc., Perkins
4
5
(
(
II 1980, 370-372.
(14) Traylor, T. G.; Nolan, K. B.; Hildreth, R. J. Am. Chem. Soc. 1983,
1
05, 6149-6151.
15) Li, M.; Chen, X.-Y.; Zou, J.-Z.; Xu, Z.; You, X.-Z. Chin. J. Chem.
996, 14, 20-24.
16) Milgrom, L. R.; Yahioglu, G.; Jogiya, N. N. Free Radical Res.
(
1
(
1996, 24, 19-29.
(17) Datta-Gupta, N.; Bardos, T. J. J. Heterocycl. Chem. 1966, 3, 495-
5
1
02.
(18) Gunter, M. J.; Robinson, B. C. Aust. J. Chem. 1989, 42, 1897-
905.
(19) Ojadi, E. C. A.; Linschitz, H.; Gouterman, M.; Walter, R. I.;
Lindsey, J. S.; Wagner, R. W.; Droupadi, P. R.; Wang, W. J. Phys. Chem.
1
993, 97, 13192-13197.
20) Udal’tsov, A. V.; Kazarin, L. A. J. Photochem. Photobiol. A 1996,
9-107.
21) Vitasovic, M.; Gouterman, M.; Linschitz, H. J. Porphyrins Phtha-
(
9
(
Supporting Information Available: Tables of peak wave-
lengths and Hammett values. This material is available free of
charge via the Internet at http://pubs.acs.org.
locyanines 2001, 5, 191-197.
(22) Momenteau, M.; Mispelter, J.; Loock, B.; Bisagne, E. J. Chem.
Soc., Perkin Trans. 1 1983, 1, 189-196.
(23) Adler, A. D.; Longo, F. R.; Finarelli, J. D.; Goldmacher, J.; Assour,
J.; Korsakoff, L. J. Org. Chem. 1967, 32, 476-478.
References and Notes
(
24) Reichardt, C. SolVents and SolVent Effects in Organic Chemistry;
(
1) The Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard,
R., Eds.; Academic Press: San Diego, 2000.
2) Gouterman, M. Optical Spectra and Electronic Structure of Por-
phyrins and Related Rings. In The Porphyrins; Dolphin, D., Ed.; Academic
VCH: Weinheim, Germany, 1988.
(25) Exner, O. A Critical Compilation of Substituent Constants. In
Correlation Analysis in Chemistry; Chapman, N. B., Shorter, J., Eds.;
Plenum Press: New York, 1978; pp 439-489.
(