Neocuproine-extended porphyrin coordination complexes. 2. Spectroscopic properties of the metalloporphyrin derivatives and investigations into the HOMO ordering

Article abstract of DOI:10.1021/ic990914q

The synthesis of a porphyrin compound, 1, containing a 2,9-dimethyl-1,10-phenanthroline moiety that is fused at the β-pyrrole positions is reported. The absorption spectra of the free-base, copper(II), and zinc(II) derivatives have been studied. On the basis of absorption band intensities, the HOMO of the free base (H₂1) and its copper and zinc complexes (Cu1 and Zn1) was determined to be of a(1u) symmetry. Relative to H₂1, compounds Cu1 and Zn1 show enhanced spectral changes upon external metal ion binding. Although the HOMO is the same in all three compounds, the energy gap between the two highest occupied orbitals is greater for Cu1 and Zn1 than it is for the free-base compound. Several metal ions (Ni²⁺, Cu⁺, Cu²⁺, Zn²⁺, Li⁺) were examined in their binding to the phenanthrolinic group by measuring the resulting changes in the absorption spectra. It is shown that the observed changes in the absorption spectra are insensitive to the nature of the metal ion coordinated by the phenanthroline moiety. Significant differences in the absorption and emission spectra between Zn1 and [Zn(Zn1)₂]²⁺ clearly demonstrate that the porphyrin π-system is strongly affected by the binding of metal ions at the fused phenanthrolinic moiety.

Full text of DOI:10.1021/ic990914q

340
Inorg. Chem. 2000, 39, 340-347
Neocuproine-Extended Porphyrin Coordination Complexes. 2. Spectroscopic Properties of
the Metalloporphyrin Derivatives and Investigations into the HOMO Ordering
Tommaso A. Vannelli and Timothy B. Karpishin*
Department of Chemistry and Biochemistry, University of California, San Diego,
La Jolla, California 92093-0358
ReceiVed July 30, 1999
The synthesis of a porphyrin compound, 1, containing a 2,9-dimethyl-1,10-phenanthroline moiety that is fused at
the â-pyrrole positions is reported. The absorption spectra of the free-base, copper(II), and zinc(II) derivatives
have been studied. On the basis of absorption band intensities, the HOMO of the free base (H21) and its copper
and zinc complexes (Cu1 and Zn1) was determined to be of a1u symmetry. Relative to H21, compounds Cu1 and
Zn1 show enhanced spectral changes upon external metal ion binding. Although the HOMO is the same in all
three compounds, the energy gap between the two highest occupied orbitals is greater for Cu1 and Zn1 than it is
for the free-base compound. Several metal ions (Ni2+, Cu+, Cu2+, Zn2+, Li+) were examined in their binding to
the phenanthrolinic group by measuring the resulting changes in the absorption spectra. It is shown that the
observed changes in the absorption spectra are insensitive to the nature of the metal ion coordinated by the
phenanthroline moiety. Significant differences in the absorption and emission spectra between Zn1 and [Zn-
(Zn1)2]2+ clearly demonstrate that the porphyrin π-system is strongly affected by the binding of metal ions at the
fused phenanthrolinic moiety.
Introduction
filled MOs, either of which can be the HOMO depending on
the porphyrin substituents. A thorough understanding of the
orbital structure of the porphyrins is thus germane to one’s
ability to rationally design and optimize the properties of
multiporphyrin systems.
Porphyrins have played and continue to play an essential role
as building blocks for photoactive molecules and materials, and
to model natural photosynthesis. A plethora of methodologies
have been employed to couple porphyrin systems to each other
and to other molecules of interest such as quinones and
carotenoids.1-10 As multiporphyrin complexes have been ex-
tensively examined, numerous electron-transfer and energy-
transfer experiments have been undertaken to understand inter-
porphyrin communication in these systems. Desired properties
of many systems include enhanced electronic conjugation and/
or excitonic communication between the porphyrin units.
Recently, Strachan et al. have shown that the nature of the
porphyrin highest occupied molecular orbital (HOMO) signifi-
cantly affects the electron-transfer rates that are observed in
various systems.11 In most porphyrins, there are two high-energy
The research presented herein is part of an ongoing pursuit
in this laboratory to prepare and examine multiporphyrin
complexes assembled via coordination chemistry. Although
coordination chemistry has been a method used previously to
assemble porphyrin units,12-20 the majority of systems have
utilized a meso-pyridyl unit of a meso-tetrasubstituted porphyrin
as the ligating group. Because of steric interactions, a meso-
pyridyl group is oriented primarily orthogonal to the porphyrin
plane, and is thus somewhat decoupled from the π-system of
the porphyrin. The porphyrin building block in our system
comprises neocuproine (2,9-dimethyl-1,10-phenanthroline) as
the ligating group fused directly to the â-pyrrole carbons of a
tetraaryl porphyrin (Figure 1). This design was chosen to
maximize the delocalization of π electrons of the porphyrin into
* To whom correspondence should be addressed. Phone: (858) 534-
5265. Fax: (858) 534-5383. E-mail: tkarpish@ucsd.edu.
(1) Wasielewski, M. R. Chem. ReV. 1992, 92, 435-461.
(2) Sessler, J. L.; Capuano, V. L.; Harriman, A. J. Am. Chem. Soc. 1993,
115, 4618-4628.
(12) Crossley, M. J.; Burn, P. L.; Langford, S. J.; Prashar, J. K. J. Chem.
Soc., Chem. Commun. 1995, 1921-1923.
(3) Lin, V. S.-Y.; DiMagno, S. G.; Therien, M. J. Science 1994, 264,
1105-1111.
(13) Harriman, A.; Odobel, F.; Sauvage, J.-P. J. Am. Chem. Soc. 1995,
117, 9461-9472.
(4) Harriman, A.; Sauvage, J.-P. Chem Soc. ReV. 1996, 41-48.
(5) Officer, D. L.; Burrell, A. K.; Reid, D. C. W. Chem. Commun. 1996,
1657-1658.
(14) Hunter, C. A.; Hyde, R. K. Angew. Chem., Int. Ed. Engl. 1996, 35,
1936-1939.
(6) Wagner, R. W.; Johnson, T. E.; Lindsey, J. S. J. Am. Chem. Soc. 1996,
118, 11166-11180.
(15) Allesio, E.; Macchi, M.; Heath, S. L.; Marzilli, L. G. Inorg. Chem.
1997, 36, 5614-5623.
(7) Kawabata, S.; Yamazaki, I.; Nishimura, Y.; Osuka, A. J. Chem. Soc.,
Perkin Trans. 2 1997, 479-484.
(16) Flamigni, L.; Armaroli, N.; Barigelletti, F.; Balzani, V.; Collin, J.-P.;
Dalbavie, J.-O.; Heitz, V.; Sauvage, J.-P. J. Phys. Chem. B 1997, 101,
5936-5943.
(8) Osuka, A.; Shimidzu, H. Angew. Chem., Int. Ed. Engl. 1997, 36, 135-
137.
(17) Linke, M.; Chambron, J.-C.; Heitz, V.; Sauvage, J.-P. J. Am. Chem.
Soc. 1997, 119, 11329-11330.
(9) Jiang, B.; Yang, S.-W.; Barbini, D. C.; Jones, W. E., Jr. Chem.
Commun. 1998, 213-214.
(18) Amabilino, D. B.; Sauvage, J.-P. New J. Chem. 1998, 395-409.
(19) Drain, C. M.; Nifiatis, F.; Vasenko, A.; Batteas, J. D. Angew. Chem.,
Int. Ed. 1998, 37, 2344-2346.
(10) Taylor, P. N.; Wylie, A. P.; Huuskonen, J.; Anderson, H. L. Angew.
Chem., Int. Ed. 1998, 37, 986-989.
(11) Strachan, J.-P.; Gentemann, S.; Seth, J.; Kalsbeck, W. A.; Lindsey, J.
S.; Holten, D.; Bocian, D. F. J. Am. Chem. Soc. 1997, 119, 11191-
11201.
(20) Huck, W. T. S.; Rohrer, A.; Anilkumar, A. T.; Fokkens, R. H.;
Nibbering, N. M. M.; vanVeggel, F. C. J. M.; Reinhoudt, D. N. New
J. Chem. 1998, 165-168.
10.1021/ic990914q CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/30/1999

Neocuproine-Extended Porphyrin Complexes
Inorganic Chemistry, Vol. 39, No. 2, 2000 341
Figure 1. Schematic depiction of the molecules investigated.
was purified by eluting through a column of Alumina B Super I grade
purchased from Selecto Scientific. All NMR solvents were purchased
from Cambridge Isotope Laboratories and used without further purifica-
tion. Acetonitrile used for spectroscopy was UV-grade purchased from
Burdick and Jackson. Column chromatography was performed with
Silica Gel 60 purchased from EM Science. All other solvents and
chemicals were purchased from either Fisher or Aldrich and used
without further purification.
the neocuproine unit. Recently reported absorption and emission
studies on the free-base molecule, H21, have shown that there
is efficient delocalization of the porphyrin π-system into the
neocuproine moiety.21 This finding led to the investigation of
the spectroscopic effects observed upon “external” (i.e., neo-
cuproine) metal ion binding. Upon formation of the zinc(II) and
copper(I) complexes of H21, ([Zn(H21)2]2+ and [Cu(H21)2]+),
significant changes in the emission spectra were observed, as
well as changes in the shape and maxima of the absorption
bands.21 These experiments demonstrated that the porphyrin
π-system of H21 is sensitive to external metal binding. Thus,
metal coordination can be utilized not only to assemble the
porphyrin units, but also to control the photophysical properties
of the multiporphyrin assemblies.
This work describes investigations of the photophysical
properties of the copper(II) and zinc(II) metalloporphyrin
derivatives of H21. As expected, upon “internal” metalation of
H21, there are changes in the absorption and emission spectra
of the resultant porphyrins. The spectral effects that are observed
upon the binding of metal ions externally to Cu1 and Zn1 are
dramatically different from the effects that were reported with
the free base, H21. To better understand these differences, the
nature of the porphyrin HOMOs in our systems has now been
investigated. An effective method that can be used to assign
the HOMOs is through analysis of the absorption spectra in a
series of molecules.22,23 Determination of the HOMOs in our
molecules allows the photophysical differences observed be-
tween Cu1/Zn1 and H21 to be rationalized. Investigation of the
external binding of a number of different metal ions has been
undertaken to determine if the nature of the external metal ion
is important. In addition, the full synthetic procedures used to
prepare these novel neocuproine-extended porphyrin complexes
are now reported.
Instrumentation. Absorption spectra were collected on a Hewlett-
Packard 8452A diode array spectrophotometer. All NMR spectra were
obtained on Varian Mercury instruments. Emission spectra were
obtained on a Perkin-Elmer LS 50 B luminescence spectrometer.
Titrations. Titrations were performed in acetonitrile. A stock solution
of Cu1 was prepared and its concentration determined spectroscopically
using the extinction coefficient at 558 nm. Stock solutions of the metal
salts in acetonitrile were prepared, and their concentrations were
accurately determined using atomic absorption spectroscopy. The metal
solution was then added in 0.1 equiv amounts, and absorption spectra
were collected after the mixture was stirred for 2 min. This procedure
was continued until 2 equiv had been added. At that point, the metal
solution was added in 0.5 equiv amounts up to 4 equiv total.
meso-Tetramesitylporphyrin (H2TMP). This procedure was adapted
from a published protocol24 that utilizes Lindsey conditions.25 Under a
nitrogen atmosphere with oven-dried glassware, boron trifluoride diethyl
etherate (0.244 mL, 1.98 mmol) was added to a rapidly stirring solution
of mesitaldehyde (4.0 mL, 27.1 mmol) and pyrrole (1.9 mL, 6.01 mmol)
in CHCl3 (500 mL). The reaction flask was covered in aluminum foil
and stirred at room temperature for 1 h. Triethylamine (0.334 mL, 2.40
mmol) was added followed by 2,3-dichloro-5,6-dicyano-1,4-benzo-
quinone (1.026 g, 4.52 mmol). The resultant mixture was then heated
at reflux for 1 h. The crude reaction mixture was evaporated to dryness,
and the solid residue was triturated with diethyl ether. The ether was
evaporated, the solid was dissolved in a minimal amount of CH2Cl2,
and the mixture was purified on silica (eluent CH2Cl2). The leading
dark red band was collected. This fraction was evaporated to dryness,
resulting in 441 mg (0.563 mmol) of H2TMP as a purple microcrys-
talline solid (38% yield). UV-vis [λmax (nm) (log(ꢀ), M-1 cm-1)]: CH2-
Cl2, 418 (5.63), 514 (4.25), 546 (3.74), 590 (3.72), 646 (3.45). 1H NMR
(CDCl3, 400 MHz) [δ (ppm)]: -2.53 (2H, s, NH), 1.83 (24H, s, o-aryl-
C-H3), 2.60 (12H, s, p-aryl-CH3), 7.25 (8H, s, m-aryl-H), 8.60 (8H, s,
pyrrole-â-H).
Experimental Section
Materials. Chloroform was freshly distilled from K2CO3, and the
first 10 mL was discarded to remove any ethanol as the CHCl3/ethanol
binary azeotrope. Pyrrole was vacuum distilled from CaO immediately
prior to use. Methylene chloride used in reactions and for spectroscopy
meso-Tetramesitylporphyrinatozinc(II) (ZnTMP). A saturated
methanolic solution of Zn(CH3CO2)2‚2H2O (1 mL) was added to a
refluxing solution of H2TMP (30 mg, 0.038 mmol) in 30 mL of CHCl3.
(21) Vannelli, T. A.; Karpishin, T. B. Inorg. Chem. 1999, 38, 2246-2247.
(22) Spellane, P. J.; Gouterman, M.; Antipas, A.; Kim, S.; Liu, Y. C. Inorg.
Chem. 1980, 19, 386-391.
(23) Binstead, R. A.; Crossley, M. J.; Hush, N. S. Inorg. Chem. 1991, 30,
1259-1264.
(24) Bookser, B. C.; Bruice, T. C. J. Am. Chem. Soc. 1991, 113, 4208-
4218.
(25) Lindsey, J. S.; Wagner, R. W. J. Org. Chem. 1989, 54, 828.

342 Inorganic Chemistry, Vol. 39, No. 2, 2000
Vannelli and Karpishin
The reaction was monitored by absorption spectroscopy and silica TLC
(eluent 50% CH2Cl2, 50% hexanes). Total conversion to the desired
product was observed after 5 min. The solution was cooled to room
temperature, poured over a silica gel plug, and eluted with 1 L of CH2-
Cl2. After evaporation of the solvent, 30 mg (0.035 mmol) of a pink/
purple solid was obtained (92% yield). UV-vis [λmax (nm)]: CH2Cl2,
420, 512, 550, 586.
meso-Tetramesitylporphyrinatocopper(II) (CuTMP). The free-
base porphyrin (H2TMP) (0.738 g, 1.28 mmol) was dissolved in 400
mL of CHCl3, and the solution was brought to reflux. To this solution
was added 35 mL of a solution of Cu(CH3CO2)2‚H2O (0.750 g, 3.76
mmol) in CH3OH. The reaction was monitored by absorption spec-
troscopy and silica TLC (eluent 50% CH2Cl2, 50% hexanes). Total
conversion to the desired product was observed after 10 min. The
solution was cooled to room temperature, poured over a silica gel plug,
and eluted with 1 L of CHCl3. After evaporation of the solvent, 985
mg (1.17 mmol) of a red solid was obtained (91% yield). UV-vis
[λmax (nm)]: CH2Cl2, 416, 540, 574.
2-Nitro-meso-tetramesitylporphyrinatocopper(II) (CuTMP-â-
NO2). This procedure was based upon a literature protocol.26 The copper
porphyrin (CuTMP) (147 mg, 0.174 mmol) was dissolved in 30 mL of
CH2Cl2 in a 250 mL round-bottom flask equipped with a stir bar and
sealed with a rubber septum. NO2(g) (8.8 mL, 0.36 mmol; calculated
using the ideal gas law) was added in two portions using a conventional
plastic syringe. The reaction was monitored with silica TLC (eluent
60% hexanes, 40% CH2Cl2). After the mixture was stirred for 30 min,
with the appearance of dinitrated products, the reaction was terminated
by flushing with nitrogen. Chromatographic purification was ac-
complished by preadsorbing the dissolved product on a minimal amount
of silica and loading this dried powder on a silica column (22 × 5 cm;
l × d) packed with hexanes. The material was eluted with a gradient
of CH2Cl2 (2-50%; 2% per 200 mL). The starting material (CuTMP)
eluted first (25-30% CH2Cl2) followed by the desired mononitrated
product (38-44% CH2Cl2). This procedure yielded 151 mg (0.17 mmol)
of the desired product (98% yield). UV-vis [λmax (nm)]: CH2Cl2, 420,
546, 584. MALDI MS: obsd m/z 891 (M + H)+, calcd for C56H51N5O2-
Cu 890.
2-Amino-meso-tetramesitylporphyrinatocopper(II) (CuTMP-â-
NH2). This procedure was based upon a literature protocol.27 Rigorously
anaerobic techniques were employed, and all glassware was oven dried
and cooled under a nitrogen flow. A round-bottom side arm flask was
charged with 127 mg (0.143 mmol) of CuTMP-â-NO2. The material
was subsequently dissolved in 30 mL of dry CH2Cl2 and 10 mL of dry
CH3OH. Palladium (10% on carbon) catalyst (120 mg) was added and
the solution purged with N2 for 5 min. Sodium borohydride (130 mg,
3.43 mmol) was added to the solution in portions over 10 min. The
reaction mixture was tested by silica TLC (eluent 60% hexanes, 40%
CH2Cl2) until all starting material (CuTMP-â-NO2) was consumed. The
amino product appears as a brown/red spot that turns green over time
on the plate, while CuTMP-â-NO2 is a stable red spot that moves
slightly faster than the amino product. Although the reaction appeared
complete after 45 min, it was stirred overnight under N2 in the dark.
The mixture was then cannula filtered into a dry round-bottom flask.
The solution was evaporated to dryness and the solid residue extracted
with dry CH2Cl2 until the solution was clear. The extraction was
performed using two cannulas, one to deliver the CH2Cl2 and one to
filter off the solution, leaving the insoluble material behind. The solvent
was then evaporated, and the air-sensitive product was carried on to
the next reaction without further purification.
turned green after approximately 5 min. The reaction was monitored
by silica TLC (eluent 50% CH2Cl2, 50% hexanes). A green spot due
to the product appeared that moved slower than CuTMP-â-NH2. The
solution was allowed to stir for 2 days in the window sill and purged
twice daily with dry O2. When no further reaction was observable by
TLC, the solvent was evaporated and the solid residue purified by flash
column chromatography. A silica column (22 × 5 cm; l × d) was
packed with a solution of 75% hexanes and 25% CH2Cl2. The column
was then eluted with a gradient of CH2Cl2 (25-70%; 5% per 200 mL).
This purification resulted in 69 mg (0.079 mmol) of the desired product
(55% yield based on CuTMP-â-NO2 used). UV-vis [λmax (nm)]: CH2-
Cl2, 408, 490. FAB MS: obsd m/z 875 (M + H)+, 1006 (M + Cs)+,
calcd for C56H50N4O2Cu 874.
2,9-Dimethyl-5-nitro-1,10-phenanthroline (NO2nc). This procedure
was adapted from a previously published preparation of 5-nitro-1,10-
phenanthroline.29 Neocuproine (1.000 g, 4.419 mmol) was dissolved
in 5 mL of oleum (100 mmol), and concentrated HNO3 (2.67 mL, 43
mmol) was added dropwise so that the solution temperature never
exceeded 170 °C. The solution was placed in an oil bath at 160 °C for
30 min. The reaction mixture was cooled to room temperature, poured
over 70 g of ice, and neutralized with 0.6 M aqueous NaOH (400 mL).
Enough concentrated HNO3 was then added to make the solution
slightly acidic (pH ≈ 5-6). Precipitation occurred overnight, after
which the solid was filtered. The filtrate was then further acidified with
concentrated HNO3 (pH ≈ 4), allowed to sit overnight, and filtered.
The solids were combined, yielding 544 mg (2.15 mmol) of pure
1
product (49% yield). H NMR (CD3OD, 300 MHz) [δ (ppm)]: 2.90
(3H, s, CH3), 2.91 (3H, s, CH3), 7.73 (1H, d, 8-H, J ) 8.0 Hz), 7.76
(1H, d, 3-H, J ) 8.8 Hz), 8.46 (1H, d, 7-H, J ) 8.0 Hz), 8.71 (1H, s,
6-H), 8.84 (1H, d, 4-H, J ) 8.8 Hz). FAB MS: obsd m/z 254 (M +
H)+ and 276 (M + Na)+, calcd for C14H11N3O2 253.
6-Amino-2,9-dimethyl-5-nitro-1,10-phenanthroline (NO2NH2nc).
This procedure was based upon a literature protocol.30 Hydroxylamine
hydrochloride (1.458 g, 20.98 mmol) was added to a suspension of
NO2nc (0.759 g, 3.00 mmol) in 16.8 mL of absolute ethanol. The
mixture was brought to reflux, and 17 mL of an ethanolic KOH (1.66
M) solution was added dropwise over 25 min. After 40 min, the reaction
mixture was cooled to room temperature and poured into 20 mL of ice
cold water. The mixture was filtered (Whatman 50) and washed with
water, ethanol, and CHCl3, affording 227 mg of pure product.
Recrystallization of the material in the filtrate from ethanol and brine
resulted in an additional 89 mg of product. The final yield of this
1
reaction was 39% (316 mg, 1.18 mmol). H NMR (d6-DMSO, 300
MHz) [δ (ppm)]: 2.66 (3H, s, CH3), 2.77 (3H, s, CH3), 7.52 (1H, d,
8-H, J ) 8.4 Hz), 7.70 (1H, d, 3-H, J ) 8.7 Hz), 8.57 (2H, br s, NH2),
8.66 (1H, d, 7-H, J ) 8.4 Hz), 8.91 (1H, d, 4-H, J ) 8.7 Hz).
5,6-Diamino-2,9-dimethyl-1,10-phenanthroline (danc). This pro-
cedure was based upon a literature protocol.30 Under a nitrogen
atmosphere with oven-dried glassware, neat hydrazine monohydrate
(743.6 µL, 15.33 mmol) was added dropwise to a refluxing ethanol
(176 mL) solution of NO2NH2nc (196 mg, 0.731 mmol) and Pd on
activated carbon (5%, 198 mg). After 1 h of heating at reflux, the
solution was filtered hot (Whatman 50). The volume of the filtrate was
reduced by 50%, an equal volume of acetone added, and the solution
stored at 6 °C overnight. Filtration of this solution resulted in 51 mg
(0.214 mmol, 29% yield) of the desired product. 1H NMR (d6-DMSO,
300 MHz) [δ (ppm)]: 2.68 (6H, s, CH3), 5.04 (4H, br s, NH2), 7.44
(2H, d, 3,8-H, J ) 8.5 Hz), 8.33 (2H, d, 4,7-H, J ) 8.5 Hz).
(2′,9′-Dimethylpyrazo[5′,6′-e]-1′,10′-phenanthroline)[b]-meso-tet-
ramesitylporphyrinatocopper(II) (Cu1). This procedure was based
upon a literature protocol.31 Under a nitrogen atmosphere with oven-
dried glassware, a catalytic amount of trifluoroacetic acid (50 µL, 650
µmol) was added to a solution of danc (62 mg, 0.26 mmol) and
CuTMCO2 (69 mg, 79 µmol) in 150 mL of dry CH2Cl2 at reflux. Silica
TLC (eluent 5% MeOH in CH2Cl2) analysis after 10 min showed that
2,3-Dioxo-meso-tetramesitylchlorinatocopper(II) (CuTMCO2).
This procedure was based upon a literature protocol.28 The amino
derivative (CuTMP-â-NH2) was dissolved in 20 mL of dry CH2Cl2,
and the solution was purged with dry O2 for 2 min. The flask was
sealed with a septum and irradiated with ambient sunlight. The solution
(26) Catalano, M. M.; Crossley, M. J.; Harding, M. M.; King, L. G. J.
Chem. Soc., Chem. Commun. 1984, 1535-1536.
(27) Baldwin, J. E.; Crossley, M. J.; DeBernardis, J. Tetrahedron 1982,
38, 685-692.
(28) Crossley, M. J.; King, L. G. J. Chem. Soc., Chem. Commun. 1984,
920-922.
(29) Smith, G. F.; Cagle, F. W., Jr. J. Org. Chem. 1947, 12, 781-784.
(30) Bolger, J.; Gourdon, A.; Ishow, E.; Launay, J.-P. Inorg. Chem. 1996,
35, 2937-2944.
(31) Kozyrev, A. N.; Alderfer, J. L.; Srikrishnan, T.; Pandey, R. K. J. Chem.
Soc., Perkin Trans. 1 1998, 837-838.

Neocuproine-Extended Porphyrin Complexes
Inorganic Chemistry, Vol. 39, No. 2, 2000 343
the reaction was approximately 50% complete. The mixture was
shielded from light and heated at reflux overnight. After being cooled
to room temperature, the mixture was stirred over an equal volume of
4% aqueous sodium bicarbonate, washed with water (2 × 150 mL)
and brine (75 mL), and dried over Na2SO4. The solution was then loaded
onto a silica column (12 × 6 cm; l × d) packed in CH2Cl2. The desired
product was eluted with a gradient of CH3OH (1/2% per 50 mL) up to
4%, at which point the product eluted off the column with 400 mL of
4% MeOH. This method afforded 80 mg (74 µmol) of pure material
as a purple/brown solid (94% yield). UV-vis [λmax (nm) (log(ꢀ), M-1
cm-1)]: MeCN, 404 (4.97), 442 (5.09), 524, 558 (4.27), 598 (3.92).
ESI MS: obsd m/z 1076.5 (MH)+, calcd for C70H60N8Cu 1075.5.
(2′,9′-Dimethylpyrazo[5′,6′-e]-1′,10′-phenanthroline)[b]-meso-tet-
ramesitylporphyrin (H21). This procedure was based upon a literature
protocol.32 Concentrated sulfuric acid (62 µL, 1.12 mmol) was added
slowly to a rapidly stirring solution of Cu1 (15 mg, 0.014 mmol) in 5
mL of CH2Cl2. The flask was shielded from light and the solution stirred
for 5 min, at which time all the porphyrin material had dissolved in
the acid layer. The mixture was then stirred over an equal volume of
aqueous EDTA (0.01 M H4EDTA, 0.1 M acetate buffer, pH 4.7). The
organic layer was separated, and the aqueous layer was further washed
with CH2Cl2 until the organic washes were clear. The combined CH2-
Cl2 solutions were washed with equal volumes of 4% aqueous sodium
bicarbonate, water, and brine. The solution was dried with Na2SO4 and
the solvent evaporated. The reaction proceeded quantitatively, yielding
14 mg (0.014 mmol) of pure H21. UV-vis [λmax (nm) (log(ꢀ), M-1
cm-1)]: CH2Cl2, 438 (5.04), 526 (4.00), 562 (sh), 598 (3.70), 654. 1H
NMR (CDCl3, 400 MHz) [δ (ppm)]: -2.40 (2H, br s, NH), 1.81 (12H,
s, o-aryl-C-H3), 1.89 (12H, s, o-aryl-CH3), 2.63 (6H, s, p-aryl-CH3),
2.87 (6H, s, p-aryl-CH3), 3.04 (6H 2,9-phenanthroline-CH3), 7.28 (4H,
s, m-aryl-H), 7.45 (4H, s, m-aryl-H), 7.64 (2H, d, 3,8-phenanthroline-
H, J ) 8.4 Hz), 8.55 (2H, s, pyrrole-â-H), 8.78 (2H, d, pyrrole-â-H,
J ) 5.1 Hz), 8.87 (2H, d, pyrrole-â-H, J ) 5.1 Hz), 8.92 (2H, d, 4,7-
phenanthroline-H, J ) 8.4 Hz). FAB MS: obsd m/z 1015 (MH)+ and
1147 (M + Cs)+, calcd for C70H62N8 1014.
(2′,9′-Dimethylpyrazo[5′,6′-e]-1′,10′-phenanthroline)[b]-meso-tet-
ramesitylporphyrinatozinc(II) (Zn1). A saturated methanolic solution
of Zn(CH3CO2)2‚2H2O (1 mL) was added to a solution of H21 (14 mg,
14 µmol) in 25 mL of CHCl3 at reflux. The reaction was heated
overnight, after which UV/vis analysis of the Q bands showed complete
conversion to the metalated porphyrin. The cooled solution was then
stirred with an equal volume of aqueous EDTA (0.01 M H4EDTA, 0.1
M acetate buffer, pH 4.7) for 24 h. The organic layer was separated
and washed with 50 mL each of 4% aqueous sodium bicarbonate, water,
and brine. The material was then purified using silica chromatography
with CHCl3 as the eluent (6.5 × 6.5 cm; l × d). This method yielded
10.6 mg (9.8 µmol, 71%) of the desired product in high purity. UV-
vis [λmax (nm) (log(ꢀ), M-1 cm-1)]: MeCN, 446 (4.7), 538 (3.8), 576
(4.3), 616 (3.8). 1H NMR (CDCl3, 400 MHz) [δ (ppm)]: 1.81 (12H, s,
o-aryl-C-H3), 1.88 (12H, s, o-aryl-CH3), 2.62 (6H, s, p-aryl-CH3), 2.87
(6H, s, p-aryl-CH3), 3.02 (6H 2,9-phenanthroline-CH3), 7.28 (4H, s,
m-aryl-H), 7.45 (4H, s, m-aryl-H), 7.65 (2H, d, 3,8-phenanthroline-H,
J ) 8.0 Hz), 8.65 (2H, s, pyrrole-â-H), 8.76 (2H, d, pyrrole-â-H, J )
4.8 Hz), 8.82 (2H, d, pyrrole-â-H, J ) 4.8 Hz), 8.96 (2H, d,
4,7-phenanthroline-H, J ) 8.0 Hz). ESI MS: obsd m/z 1077 (MH)+,
calcd for C70H60N8Zn 1076.
[Zn(Zn1)2](CH3CO2)2. A methanolic saturated solution of Zn(CH3-
CO2)2‚2H2O (0.5 mL) was added to a nitrogen-purged solution of Zn1
(5.3 mg, 4.9 µmol) in 10 mL of CHCl3. The mixture was shielded
from light and stirred for 5 min under a nitrogen atmosphere. The
solvent was evaporated, and the solid residue was taken up in a minimal
amount of CH2Cl2 and filtered through Celite 521. The reaction proceeds
quantitatively. UV-vis [λmax (nm)]: CH2Cl2, 418, 456, 538, 576, 614.
1H NMR (CDCl3, 300 MHz) [δ (ppm)]: 1.78 (12H, s, o-aryl-C-H3),
1.87 (12H, s, o-aryl-CH3), 2.01 (6H, CO2CH3), 2.62 (6H, s, p-aryl-
CH3), 2.85 (6H, s, p-aryl-CH3), 3.26 (6H 2,9-phenanthroline-CH3), 7.27
(4H, s, m-aryl-H), 7.43 (4H, s, m-aryl-H), 7.85 (2H, d, 3,8-phenan-
throline-H, J ) 8.4 Hz), 8.66 (2H, s, pyrrole-â-H), 8.76 (2H, d, pyrrole-
Figure 2. Synthetic scheme for the preparation of the molecules
investigated. Ar ) 2,4,6-trimethylphenyl. Conditions: (i) 2 equiv of
NO2 in dry CH2Cl2; (ii) Pd/C, NaBH4 in CH2Cl2/MeOH; (iii) O2/hν in
dry CH2Cl2; (iv) 3.3 equiv of danc, 0.033% TFA in CH2Cl2; (v) CH2-
Cl2/H2SO4; (vi) Zn(CH3CO2)2 in CHCl3/EDTA; (vii) Zn(CH3CO2)2 in
CHCl3.
â-H, J ) 4.8 Hz), 8.83 (2H, d, pyrrole-â-H, J ) 4.8 Hz), 9.13 (2H, d,
4,7-phenanthroline-H, J ) 8.4 Hz). MALDI MS: obsd m/z 2217(M +
e-)+, calcd for C140H120N16Zn3 2217.
Results and Discussion
(I) Syntheses. Figure 2 presents the synthetic strategy used
to obtain compounds H21, Cu1, and Zn1. Reaction of the copper-
(II) porphyrin (CuTMP) with NO2 is essential for a clean
reaction to the nitrated complex (CuTMP-â-NO2) because the
regiospecificity of the nitration is directed by the electronega-
tivity of the central metal ion.26 If one nitrates a zinc or free-
base tetraarylporphyrin, oxidation occurs at the meso positions
and the porphyrin ring is ruptured.26 Reaction of CuTMP with
NO2 usually produced a small amount of dinitrated products
which were easily removed by column chromatography. Care
must be taken to not add too much NO2 and to not let the
reaction proceed for too long, since such conditions favor the
formation of polynitrated species. Silica TLC analysis was found
to be an effective means by which this reaction could be
monitored.
Reduction of CuTMP-â-NO2 proceeds smoothly, but must
be performed under rigorously maintained inert atmosphere
conditions. The amino derivative (CuTMP-â-NH2) is extremely
(32) Beavington, R.; Rees, P.; Burn, P. J. Chem. Soc., Perkin Trans. 1
1998, 2847-2851.

344 Inorganic Chemistry, Vol. 39, No. 2, 2000
Vannelli and Karpishin
Table 1. Absorption Maxima and Band Assignmentsa
B(0,0) Q(2,0) Q(1,0)
Cu1 406 (By), 442 (Bx) 526 560
Q(0,0)
600
H21 438 526 (Qy), 598 (Qx) 562 (Qy), 654 (Qx)
Zn1 412 (By), 446 (Bx) 532 568 608
a All values are reported in nanometers, and the spectra were
performed in CH2Cl2 at room temperature.
Figure 3. Absorption spectra in CH2Cl2 at room temperature of H21
(A), Zn1 (B), and Cu1 (C).
reactive to hydrolysis and must be handled with care.27 Once
CuTMP-â-NH2 is oxidized to the dioxochlorin, CuTMCO2,
these precautions are no longer necessary since CuTMCO2 is
air stable. The dioxochlorin is then subjected to a condensation
with 5,6-diaminoneocuproine (danc, Figure 2). The synthesis
of danc was carried out in a manner similar to that reported for
the preparation of 5,6-diamino-1,10-phenanthroline.30 However,
the yield (29%) in the final step of the synthesis of danc fell
well below the reported yield for the analogous reaction
performed with 1,10-phenanthroline (83%).30
Figure 4. Representation of the electron densities in the two HOMOs
(bottom) and the two LUMOs (top) for the D4h porphyrin ring.23
zinc porphyrin, Zn1, being a regular porphyrin, displays a
normal type spectrum.33 The Q-band envelope of both Cu1 and
Zn1 contains the Q(0,0) transition as well as the two vibronic
overtones, Q(1,0) and Q(2,0) (Table 1). These are typical
characteristics common to all metalloporphyrins.
The conditions used for the condensation of CuTMCO2 with
danc were based upon the reported work of Pandey et al.31 These
conditions were found to be far superior to conditions reported
by Crossley et al. for a similar condensation.12 The condensation
reaction produces the desired product (Cu1) in high yield (94%).
Demetalation of Cu1 with H2SO4 generates H21, which is then
metalated with zinc(II) acetate to provide Zn1. Although Cu1
is indefinitely stable, the free base (H21) and zinc(II) complex
(Zn1) are susceptible to photooxidation if exposed to light in
solution for extended periods of time. We have also observed
that the copper(I) and zinc(II) complexes of H21, [Cu(H21)2]+,
and [Zn(H21)2]2+ slowly undergo transmetalation, whereby the
externally coordinated metal is transferred to the porphyrin core.
This process occurs overnight in solution and over a period of
a couple of months in the solid state. Transmetalation is not a
problem with the metal complexes of the metallo derivatives
(Cu1 and Zn1).
(b) Theory. Gouterman’s early research resulted in the widely
accepted four-orbital porphyrin model which is a combination
of simple Hu¨ckel calculations with the free-electron and cyclic
polyene models.33,34 This theory proposes that all porphyrin
systems contain two nearly degenerate top-filled π orbitals and
two lowest unoccupied molecular orbitals (LUMOs) that are
degenerate. In the case of perfect D4h symmetry, the top-filled
orbitals differ in their nodal and density structure, resulting in
an orbital of a2u symmetry and one having a1u symmetry (Figure
4). On the other hand, the two LUMOs both have eg symmetry.
Since the porphyrin molecules discussed in this paper have C2V
symmetry rather than D4h symmetry, we will refer to the a2u
and a1u orbitals as b1 and b2, respectively, and the two eg orbitals
as c1 and c2 (Figure 4). Gouterman’s model adequately explains
the electronic absorption spectra of most porphyrinic compounds
and thus is the foundation for many orbital-based explanations
of the electronic properties of porphyrin-containing complexes.
(II) Absorption Spectra. (a) General Features. The spectral
features of H21 have been previously reported.21 The absorption
spectra of H21, Cu1, and Zn1 are shown in Figure 3. They are
typical porphyrin spectra in that they contain a strong absorption
band in the UV region, designated as the B band, and weaker
visible absorption bands designated as the Q bands. The copper
porphyrin, Cu1, displays a hypso type spectrum which is in
accord with the fact that it is an irregular porphyrin, while the
A knowledge of the relative energy levels of these four
orbitals in the individual porphyrins within a multiporphyrin
system is relevant to understanding interporphyrin ground-state
and excited-state electronic communication. If one wishes to
(33) Gouterman, M. The Porphyrins; Academic Press: New York, 1978;
Vol. III.
(34) Gouterman, M. J. Mol. Spectrosc. 1961, 6, 138-163.

Neocuproine-Extended Porphyrin Complexes
Inorganic Chemistry, Vol. 39, No. 2, 2000 345
Table 2. A(Q(0,0))/A(Q(1,0)) Values for Selected Compounds
PFPa
TPPa
TMP
1
b
H2
0.13
0.26
0.47
0.43
0.13
0.095
0.65
0.39
0.11
0.093
0.42
0.30
0.43
0.45
Zn
Cu
Cu-â-NO2
a Taken from ref 22. b For the free-base compounds the ratio of the
sums of the x and y components of the Q transitions was used
[A(Qx(0,0)) + A(Qy(0,0))]/[A(Qx(1,0)) + A(Qy(1,0))].
link the chromophores via the meso position, then a porphyrin
with the b1 orbital as the HOMO is desirable to maximize
electronic coupling. On the other hand, if one wishes to join
the chromophores through the â-pyrrole positions, then a
porphyrin with a b2 HOMO is desirable. Inspection of the two
top-filled orbitals in Figure 4 shows that the b1 orbital is
stabilized either by electron-withdrawing groups at the meso
positions or by metal ion coordination by the pyrrole nitrogens.
The b2 orbital can be stabilized only by electron-withdrawing
groups on the â-pyrrole carbons.
Coupling between the four electronic transitions from the two
top-filled orbitals into the two LUMOs results in the high-energy
B state and the low-energy Q state observed in porphyrin
absorption and emission spectra.33 The two pairs of singlet
transitions in which these orbitals are involved are a2u f eg
and a1u f eg. According to the four-orbital model, when the
top-filled π orbitals are equal in energy and the two unoccupied
π* orbitals are also equal in energy, the B and Q states are
comprised of two degenerate transitions polarized in the x and
y directions.33 In this case the Q state is formally forbidden.33
If this degeneracy is lifted, then the Q(0,0) transition becomes
partially allowed.33,34 Therefore, the intensity of the Q(0,0)
absorption band is directly related to the energy spacing between
the two HOMOs. The degree of degeneracy and the energy
splitting of the two top-filled π orbitals can be determined
through analysis of the absorption spectra in the low-energy
region (the Q-band region).22,23 This method utilizes the intensity
ratio of the two visible transitions typical of porphyrin absorption
spectra. The two transitions used are Q(0,0) and its vibronic
overtone Q(1,0). The Q(1,0) transition arises from vibronic
mixing of the Q state with the B state, and its intensity varies
little among various porphyrins.22,33 The intensity of the Q(1,0)
transition can, thus, be used as a normalizing factor when
different porphyrin compounds are compared. This ratio is
related to the energy of the HOMO by the following equation:
Figure 5. Energy differences between the a1u and a2u orbitals in the
indicated porphyrins based on the Q-band ratios shown in Table 2.
moiety onto the â-positions of the TMP core there is an
inversion in the nature of the HOMO. This assignment is based
on the observed trend in the ratio of the Q-band absorbance
values upon metalation. The orbital most affected by metalation
is the b1 orbital, which has considerable electron density at the
pyrrole nitrogens (Figure 4). The electron-withdrawing metal
ion has a stabilizing effect on this orbital. Therefore, when the
HOMO is the b1 orbital, metalation reduces the energy gap
between the b1 and b2 orbitals, resulting in a smaller value for
the ratio of the Q-band absorbance values when compared to
those of the free-base porphyrin. This trend is indeed observed
within the TMP and TPP compounds, which are known to have
the b1 orbital as the HOMO. Thus, the values of A[Q(0,0)]/
A[Q(1,0)]}1/2 are negative for those compounds with an a2u (b1)
HOMO, as can be seen by eq 1. If, however, the b2 orbital is
the HOMO, then metalation would increase the energy gap
between the two top-filled orbitals, resulting in an increase in
the value of the ratio of the Q-band absorbance values. The
PFP compounds, which are known to have the b2 orbital as the
HOMO, show this trend.11,22 Compounds H21, Zn1, and Cu1
have a b2 HOMO on the basis of the fact that the ratio of the
Q bands increases in the order H21 < Zn1 < Cu1 (Figure 5).
Determination of the HOMO structure is important if one wishes
to study energy- and electron-transfer involving such complexes.
Since we plan to study energy and electron transfer between
chromophores linked through the fused neocuproine unit, a b2
HOMO is desirable to maximize the rates of such processes.
(d) B-Band Analysis of H21, Cu1, and Zn1. The B-band
structure of H21, Cu1, and Zn1 is dramatically different from
those of the parent porphyrin, H2TMP, and its copper(II) and
zinc(II) complexes (see the Supporting Information). The
observed broadening of the B band has been attributed to the
extension of the π-system of the porphyrin into the neocuproine
portion of the molecule.21 The full width at half-maximum
(fwhm) values are 1808, 3730, and 3433 cm-1 for H21, Cu1,
and Zn1, respectively, while the fwhm values for H2TMP,
CuTMP, and ZnTMP are 687, 521, and 622 cm-1, respectively.
The B-band envelope of the metallo complexes, Cu1 and Zn1,
is considerably different from that of H21, in that it is split into
two relatively sharp transitions (Figure 3). Excitonic coupling
is precluded as a possible explanation since it has been
established that good electronic communication exists between
the porphyrin and neocuproine substituents.21,35 A more plausible
explanation would be to describe the two transitions as the Bx
and By transitions of the molecule. The x and y molecular dipole
axes lie along the two planes containing opposite pyrrole
A[Q(0,0)]/A[Q(1,0)] ) (const)[1E(a2u,eg) - 1E(a1u,eg)]2 (1)
where 1E(a2u,eg) - 1E(a1u,eg) is the energy gap between the two
pairs of singlet transitions.
(c) Nature of the HOMOs. Q-Band Analysis of H21, Cu1,
and Zn1. Here we compare the experimentally determined
Q-band ratios for our compounds with those of other benchmark
porphyrins such as meso-tetraphenylporphyrin, TPP, and its
tetra(pentafluorophenyl) analogue, PFP. Using the ratios in Table
2 we can then plot {A[Q(0,0)]/A[Q(1,0)]}1/2, which in turn can
be used to determine which of the two top-filled porphyrin
orbitals is the HOMO. This plot is shown in Figure 5.
The compounds H2TMP, CuTMP, CuTMP-â-NO2, and
ZnTMP are spectroscopically very similar to their TPP ana-
logues, which is reflected in the fact that they all have a HOMO
that is a2u (b1) in character. This is not surprising since the
mesityl groups of TMP are electronically very similar to the
phenyl groups of TPP. The most interesting piece of information
derived from Figure 5 is that upon fusing the neocuproine
(35) Kasha, M.; Rawls, H. R.; El-Bayoumi, M. A. Pure Appl. Chem. 1965,
11, 371-392.

346 Inorganic Chemistry, Vol. 39, No. 2, 2000
Vannelli and Karpishin
Figure 6. Overlay of absorption spectra in CH3CN at room temperature
for Cu1 with 0, 0.17, 0.25, 0.33, 0.41, 0.50, 0.83, and 1.2 equiv of
[Cu(CH3CN)4](PF6). Initial [Cu1] ) 6.36 × 10-6 M. All spectra were
corrected for changes in concentration due to the addition of copper-
(I).
nitrogens. For purposes of this discussion, we will assign the
y-axis as that containing the pyrrole fused to the neocuproine
subunit. If we invoke this formalism, then it becomes obvious
that the two transition dipoles are no longer equivalent in our
molecules. Furthermore, we can now assign the high-energy
band as the By transition and the lower energy band as the Bx
transition.33,34
Figure 7. Absorption spectra in CH2Cl2 at room temperature for (top)
H21 (thin line) and [Zn(H21)2](CH3CO2)2 (thick line) and (bottom) Zn1
(thin line) and [Zn(Zn1)2](CH3CO2)2 (thick line).
A similar splitting in the B band is observed in the reduced
zinc porphyrin molecule, zinc hexahydroporphyrin, in which
three pyrrole rings are reduced at the â positions.33 This
molecule has C2V symmetry, like the metallo compounds Cu1
and Zn1. Therefore, the molecular symmetry of these porphyrins
may contribute to the splitting of the B bands.
stirring period before changes were observable in the absorption
spectra. When lithium(I) was titrated into an acetonitrile solution
of Cu1, there were no spectral changes even upon addition of
up to 100 equiv of Li+. However, when the titration was
performed in methylene chloride, lithium behaved much like
the zinc and copper ions. These results suggest that the binding
constant for lithium and the neocuproine moiety is significantly
lower in acetonitrile than in methylene chloride.
(III) “External” Metal Binding. (a) Titrations in Aceto-
nitrile. Various metal salts have been titrated into acetonitrile
solutions of Cu1 to determine if there is a spectral dependence
on the nature of the metal ion coordinated “externally” to the
neocuproine moiety. The results of a typical spectrophotometric
titration can be seen in Figure 6, where [Cu(CH3CN)4](PF6) was
the metal ion source added. Other metal salts used include
LiBF4, Zn(BF4)2, Ni(BF4)2, and Cu(BF4)2. All of these metal
ions form bis(chelate) complexes with neocuproine.36 All of the
spectrophotometric titrations gave the same general results, with
only small variations in the extent to which the By and Bx bands
shifted. The general spectroscopic changes observed are a red
shift of all peaks, a reduction in the intensity of the Bx band, an
increase in the intensity of the By band, and a general broadening
of the B band (Figure 6). It is clear from these results that the
B state in Cu1 is extremely sensitive to external metal ion
coordination. Also, the absorption spectrum of Cu1 is much
more sensitive to external metal binding than that of the
previously studied free-base porphyrin, H21.21
The spectral changes that occur upon binding of metal ions
to the neocuproine moiety of Cu1 are believed to be electrostatic
in nature, since excitonic theory predicts that the Bx band should
blue shift, while the By band is predicted to shift to the red.3,35
Also, excitonic theory does not predict the observed dramatic
changes in intensities of the Bx and By bands.3,35
Although the shifts and intensity changes in the spectra were
insensitive to the nature of the metal ion titrated, the rates at
which spectroscopic changes were observable did vary slightly.
Zinc(II) and both copper(I) and copper(II) ions resulted in
instantaneous changes in the spectra even upon addition of 0.1
equiv of the metal ion. Nickel(II), however, required a longer
An additional experiment was performed to test the stability
of the metal-neocuproine complexes in acetonitrile. A solution
containing 4 equiv of Zn2+/equiv of Cu1 was diluted, and spectra
were taken over time. It was observed that the spectra slowly
reverted to an “intermediate spectrum”, similar to one in which
fewer equivalents of metal ion were added. Further dilution
resulted in a spectrum very similar to that of Cu1. Thus, the
concentration dependence of the spectra demonstrate that
dissociation is likely occurring, and that, at low concentrations,
there is likely a mixture of Cu1, [Zn(Cu1)]2+, and [Zn(Cu1)2]2+
in acetonitrile.
(b) Examination of the Zinc Bis(porphyrin) Complex, [Zn-
(Zn1)2]2+. Since we were interested in directly probing the
spectral differences between the metallo derivatives of 1 and
the bis(porphyrin) [M(M1)2]n+ complexes, we examined the
spectra of the complex [Zn(Zn1)2]2+ in methylene chloride.
Mass spectrometry and 1H NMR data confirm that, in methylene
chloride, this complex is stable as the bis(chelate), without
dissociation to [Zn(Zn1)]2+ 37
. In addition, the spectrum of [Zn-
(Zn1)2]2+ is invariant over the concentration range of 1 × 10-5
to 1 × 10-6 M, allowing us to compare both the absorption
and emission properties with those of the previously reported
complex [Zn(H21)2]2+ 21
. Figure 7 shows the absorption spectra
of Zn1 and H21 and their Zn2+ complexes. The changes in the
spectrum of Zn1 induced by external metal binding are
practically identical to those observed in the titrations of Cu1
with the different metal ions (Figure 6). This is not surprising
since the Q-band ratios for Cu1 and Zn1 (Figure 5) are nearly
identical. Zn1 is spectroscopically much more sensitive to
external metal binding than H21, not only because of the
observed dramatic changes in the B band, but also because the
(36) Gantzel, P. K.; Vannelli, T. A.; Joshi, H. S.; Guidolin, M.; Karpishin,
T. B. Manuscript in preparation.

Neocuproine-Extended Porphyrin Complexes
Inorganic Chemistry, Vol. 39, No. 2, 2000 347
intensities vary little between Zn1 and its zinc(II) complex, 60%
and 64%, respectively (ZnTMP ) 100%). This increase in
intensity upon complex formation with zinc(II) is also observed
in the free-base system, but to a greater extent; the emission
spectrum of the zinc(II) complex of H21 is 25% more intense
than that of H21. It can also be seen in Figure 8 that the emission
bands of [Zn(Zn1)2]2+ are broader than those of Zn1, and that,
relative to ZnTMP, for both Zn1 and [Zn(Zn1)2]2+ the Q(0,0)
emission band increases in intensity relative to the Q(0,1) band.
These emission data, combined with the absorption data, provide
compelling evidence that the binding of a closed shell metal
ion (Zn2+) to the neocuproine unit of Zn1 elicits significant
changes in the electronic structure of the chromophore.
Conclusions
Notable differences in the B-band region are observed
between H21 and its metal complexes, Cu1 and Zn1. From
analysis of absorption and emission spectra, these metallopor-
phyrins also display an increased spectroscopic sensitivity to
external metal ion binding relative to H21. Interestingly, the
absorption spectra of Cu1 and Zn1 are insensitive to the nature
of the metal ion bound externally. Assignment of the b2 orbital
as the top-filled porphyrin π orbital (HOMO) in compounds
H21, Cu1, and Zn1 was achieved using a spectroscopic technique
based on Gouterman’s four-orbital model. Although the HOMO
is the same in all three compounds, the energy gap between
the two HOMOs is greater for Cu1 and Zn1 than it is for H21.
This difference in the HOMO energy levels likely contributes
to the observed spectroscopic differences between H21 and the
metallo derivatives, Cu1 and Zn1.
Since the molecules under study possess a b2 HOMO, which
has substantial electron density at the â-pyrrole positions (Figure
4), these compounds are excellent candidates as building blocks
for multiporphyrin systems with a high degree of electronic
communication between the chromophores. Thus, having now
characterized the HOMO of these novel chromophores, future
work is planned to examine bis(porphyrin) complexes with two
different porphyrins, utilizing methods recently developed for
making heteroleptic bis(phenathroline) complexes.38
Figure 8. Corrected emission spectra in CH2Cl2 at room temperature
for (top) H2TMP (λex ) 514 nm) (dashes), H21 (λex ) 528 nm) (thin
line), and [Zn(H21)2](CH3CO2)2 (λex ) 532 nm) (thick line) and (bottom)
ZnTMP (λex ) 550 nm) (dashes), Zn1 (λex ) 568 nm) (thin line), and
[Zn(Zn1)2](CH3CO2)2 (λex ) 576 nm) (thick line). All spectra were
normalized to identical OD values at the excitation wavelength.
bands shift to the red more than those in the free-base system.
The By and Bx bands of Zn1 shift by 348 and 492 cm-1 while
the B band of H21 only shifts by 104 cm-1 upon complex
formation with zinc(II). The Q(1,0) band in Zn1 shifts to the
red by 244 cm-1 in comparison to 21 and 11 cm-1, respectively,
for the Qy(1,0) and Qx(1,0) bands of H21. Another interesting
difference between the two complexes is that [Zn(Zn1)2]2+ has
a shoulder on the low-energy side of the B band, while the
analogous free-base compound lacks this feature (Figure 7).
(IV) Emission Spectra. Figure 8 shows the corrected Q-band
emission spectra of H21, Zn1, and the Zn2+ complexes, [Zn-
(H21)2]2+ and [Zn(Zn1)2]2+. The corrected spectra for H2TMP
and ZnTMP are provided for comparison. Note that, as a general
trend, the emission intensities for the zinc porphyrins are lower
than those of the free-base porphyrins. The higher energy
emission band is due to the Q(0,0) transition, i.e., from the
lowest vibrational mode of the excited state to the lowest
vibrational mode of the ground state. The lower energy band is
assigned as Q(0,1), which arises from relaxation of the excited
electron to the first vibrational mode of the ground state.33 The
general trends observed in the absorption spectra among
ZnTMP, Zn1, and [Zn(Zn1)2]2+ are also observable in the
emission spectra. There is a red shift of 497 cm-1 in the Q(0,0)
emission band between ZnTMP and Zn1 and 419 cm-1 between
Zn1 and [Zn(Zn1)2]2+, while the Q(0,1) transition shifts by 593
and 259 cm-1, respectively. When H21 is compared to its zinc-
(II) complex, these shifts are much smaller: 23 and 48 cm-1
for the Q(0,0) and Q(0,1) transitions, respectively. The integrated
Acknowledgment. We thank Professor Y. Tor for the use
of his emission spectrometer. This research was supported by
the NIH (training grant to T.A.V.) and by a Hellman Faculty
Fellowship (to T.B.K.).
Supporting Information Available: Absorption spectra of H2TMP,
CuTMP, and ZnTMP. This material is available free of charge via the
Internet at http://pubs.acs.org.
IC990914Q
(37) 1H NMR titrations of Zn(CH3COO)2‚2H2O into Zn1 were performed
in CDCl3. As various amounts of Zn2+ ion were added, all three of
the porphyrin species, Zn1, [Zn(Zn1)(CH3COO)2], and [Zn(Zn1)2]2+
,
could be observed by examination of the chemical shifts of the
phenanthroline methyl groups. This information allows for unambigu-
ous identification of [Zn(Zn1)2]2+ as the only species in CDCl3 when
the complex is prepared as described in the Experimental Section.
(38) Miller, M. T.; Gantzel, P. K.; Karpishin, T. B. J. Am. Chem. Soc.
1999, 121, 4292-4293.