Activation barriers to meso-aryl group rotation in titanyl tetraaryltetrapyrroles. An investigation of the out-of-plane flexibility of hydroporphyrins

Article abstract of DOI:10.1021/ic010936o

The free-base and titanyl (Ti^IVO) complexes of meso-tetratolyl- and meso-tetra(3,5-xylyl)hydroporphyrins were synthesized and characterized. Metalation of the hydroporphyrins with titanium was achieved by reaction of the lithium salts of the hydroporphyrin with TiCl₄. Other methods used to metalate porphyrins with titanium required harsher reaction conditions and led to substantial oxidative dehydrogenation of the macrocycle when applied to hydroporphyrins. The titanyl group differentiates the two faces of the macrocycle and consequently the two sides of the meso-aryl groups, which are tilted nearly perpendicular to the macrocycle plane. The ¹H NMR signals for the nonequivalent ortho protons and nonequivalent meta protons averaged on the NMR time scale at elevated temperatures due to aryl group rotation. Activation barriers for aryl group rotation in the para-substituted and meta-disubstituted titanyl hydroporphyrin complexes and in related titanyl porphyrin complexes were determined from variable-temperature NMR spectra and ranged from 15.6 to 18.4 kcal/mol. In chlorin compounds, barriers for rotation of aryl groups located between a pyrrole and a pyrroline (reduced) ring are greater than those of aryl groups located between two pyrrole rings. Comparisons of barriers in complexes with different macrocycle saturation levels show that the increased barriers for aryl groups adjacent to pyrroline rings cannot be attributed solely to the increased steric bulk of the pyrroline β-CH₂ group relative to the pyrrole β-CH group. Variations in flexibility and electronic environments at meso carbons in the hydroporphyrins may also contribute. Rotation barriers for metadisubstituted aryl groups, which are higher than those for para-substituted aryl groups, increase with the size and mass of the substituent.

Full text of DOI:10.1021/ic010936o

Inorg. Chem. 2002, 41, 300−308
Activation Barriers to meso-Aryl Group Rotation in Titanyl
Tetraaryltetrapyrroles. An Investigation of the Out-of-Plane Flexibility of
Hydroporphyrins
Alan M. Stolzenberg* and G. Scott Haymond
Department of Chemistry, West Virginia UniVersity, P.O. Box 6045,
Morgantown, West Virginia 26506
Received August 29, 2001
IV
The free-base and titanyl (Ti O) complexes of meso-tetratolyl- and meso-tetra(3,5-xylyl)hydroporphyrins were
synthesized and characterized. Metalation of the hydroporphyrins with titanium was achieved by reaction of the
lithium salts of the hydroporphyrin with TiCl . Other methods used to metalate porphyrins with titanium required
4
harsher reaction conditions and led to substantial oxidative dehydrogenation of the macrocycle when applied to
hydroporphyrins. The titanyl group differentiates the two faces of the macrocycle and consequently the two sides
1
of the meso-aryl groups, which are tilted nearly perpendicular to the macrocycle plane. The H NMR signals for the
nonequivalent ortho protons and nonequivalent meta protons averaged on the NMR time scale at elevated
temperatures due to aryl group rotation. Activation barriers for aryl group rotation in the para-substituted and meta-
disubstituted titanyl hydroporphyrin complexes and in related titanyl porphyrin complexes were determined from
variable-temperature NMR spectra and ranged from 15.6 to 18.4 kcal/mol. In chlorin compounds, barriers for rotation
of aryl groups located between a pyrrole and a pyrroline (reduced) ring are greater than those of aryl groups
located between two pyrrole rings. Comparisons of barriers in complexes with different macrocycle saturation levels
show that the increased barriers for aryl groups adjacent to pyrroline rings cannot be attributed solely to the
increased steric bulk of the pyrroline â-CH group relative to the pyrrole â-CH group. Variations in flexibility and
2
electronic environments at meso carbons in the hydroporphyrins may also contribute. Rotation barriers for meta-
disubstituted aryl groups, which are higher than those for para-substituted aryl groups, increase with the size and
mass of the substituent.
Introduction
Chart 1
Hydroporphyrins are compounds in which one or more
double bonds of a porphyrin have been saturated by the
formal addition of hydrogen atoms or alkyl groups across a
1
double bond. Structures of di- and tetrahydroporphyrin
compounds are compared with that of the parent porphyrin
in Chart 1.
Metal complexes of hydroporphyrins and other non-
porphyrin tetrapyrroles play central roles as prosthetic groups
in the biochemical pathways of the carbon, nitrogen, and
sulfur cycles and in the metabolism of many anaerobic
bacteria. Examples include chlorophylls and pheophytins, the
magnesium and free-base chlorin and bacteriochlorin pig-
2
ments of photosynthesis; siroheme, the iron isobacterio-
*
Author to whom correspondence should be addressed. E-mail:
astolzen@wvu.edu.
(
1) Scheer, H. In The Porphyrins; Dolphin, D., Ed.; Academic Press: New
York, 1978; Vol. 2, pp 1-44.
chlorin prosthetic group of assimilatory (biosynthetic) nitrite
and sulfite reductases; and F430, a nickel hydrocorphinoid
3-5
(2) Scheer, H., Ed. Chlorophylls; CRC Press: Boca Raton, FL, 1991.
300 Inorganic Chemistry, Vol. 41, No. 2, 2002
10.1021/ic010936o CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/04/2002

meso-Aryl Rotation in Titanyl Tetraaryltetrapyrroles
6
prosthetic group involved in methanogenesis. The absence
demonstrated that the metal-centered reductions of the
octaethylisobacteriochlorin complexes Cu(OEiBC) and
Ni(OEiBC) result in large structural changes that involve
of enzymes that contain porphyrin prosthetic groups and are
competent to catalyze these processes is conspicuous and
raises the question of whether non-porphyrin tetrapyrroles
are specifically required. Thus, there is considerable interest
in delineating the effects of changes in the structure and
saturation level of a tetrapyrrole macrocycle on the chemistry
of its complexes and in particular on the chemistry of a
coordinated metal ion.
Several chemical differences between hydroporphyrins and
porphyrins have been observed. Hydroporphyrins have
intrinsically larger core sizes and exhibit both a greater
tendency to adopt nonplanar conformations and greater
displacements from planarity than the corresponding por-
2
2
in-plane flexibility. Indeed, the four Cu-N distances of
II
I
2.00 Å in the Cu complex increase to 2.06 Å in the Cu
II
complex, and the four Ni-N distances of 1.94 Å in the Ni
complex change to two Ni-N distances of 1.91 Å and two
I
of 2.07 Å in the Ni complex.
In an effort to probe the out-of-plane flexibility of
tetrapyrroles, we investigated the activation barriers to
rotation of meso-aryl groups in the previously unknown
titanyl (TidO) hydroporphyrin complexes. Earlier studies
of titanyl porphyrin complexes established that the coordi-
nated titanyl group differentiates the two faces of the
porphyrin, that the chemical shift differences in these
diamagnetic complexes are substantial between both the
nonequivalent ortho protons and the nonequivalent meta
protons of the meso-aryl group, which is tilted nearly
perpendicular to the porphyrin plane and has restricted
rotation, and that slow and fast exchange regimes for aryl
7,8
phyrin complexes that have similar peripheral substitution.
Standard reduction potentials of ligand-centered redox
processes generally decrease with increasing macrocycle
saturation.9 Thus, hydroporphyrin macrocycles are easier
to oxidize and more difficult to reduce than porphyrins. The
resistance of the macrocycle to reduction and the larger core
size are reasons that hydroporphyrins can stabilize metal ions
in less common, low-valent oxidation states such as Cu and
Ni , which are not readily accessible in porphyrins.
Other notable differences between hydroporphyrins or por-
-19
ring rotation are both accessible at experimentally convenient
I
23-26
temperatures.
In addition, the titanyl group is chemically
I
15,17,18
inert, and its porphyrin complexes are not subject to axial
ligand binding or exchange reactions that could lead to
2
0,21
24
phyrins have been reported or summarized elsewhere.
complications. The restricted rotation of the meso-aryl
We proposed that hydroporphyrins have shallower con-
formational energy surfaces than porphyrins and that this
could cause significant differences in the chemistries of the
complexes of these tetrapyrroles. In other words, hydro-
porphyrins are more “flexible” than porphyrins. The differ-
ence in flexibility could affect both the ease of changing the
tetrapyrrole hole size (in-plane flexibility) and the ease of
deforming the tetrapyrrole from planarity (out-of-plane
flexibility). EXAFS and resonance Raman studies have
groups is a consequence of steric interactions between the
aryl ortho protons and the porphyrin â-pyrrole protons that
occur when the aryl group and porphyrin are nearly coplanar.
Although evidence shows that electronic effects from
interaction of the aryl group and porphyrin π-systems
17
2
5,27
contribute to the rotation barriers,
the ability of the
porphyrin macrocycle to deform and permit the ortho and
â-pyrrole protons to avoid each other is clearly important.
As such, changes in macrocycle flexibility in hydroporphy-
rins could affect the rotation barriers of aryl groups situated
adjacent to pyrroline (saturated) rings. Direct comparisons
may be complicated by two factors, though. First, the steric
environments of the meso-aryl groups in hydroporphyrins
are not identical to each other or to those in porphyrins. Meso
positions adjacent to zero, one, and two pyrroline rings are
labeled A, B, and C, respectively, in Chart 1. The two
additional â-protons present in a pyrroline ring could increase
steric interactions. Second, the symmetry and electronic
inequivalence of the hydroporphyrin meso positions could
result in different electronic contributions to the rotation
barriers. The increased rate of electrophilic reactions at meso
positions adjacent to pyrroline rings has been taken as an
indication of increased electron density at these sites.2
(
(
3) Murphy, M. J.; Siegel, L. M. J. Biol. Chem. 1973, 248, 6911.
4) Vega, J. M.; Garrett, R. H.; Siegel, L. M. J. Biol. Chem. 1975, 250,
7980.
(
5) Vega, J. M.; Kamin, H. J. Biol. Chem. 1977, 252, 896.
6) Ellefson, W. L.; Whitman, W. B.; Wolfe, R. S. Proc. Natl. Acad. Sci.
U.S.A. 1982, 79, 3707-3710.
7) Scheidt, W. R.; Lee, Y. J. Struct. Bonding (Berlin) 1987, 64, 1.
8) Stolzenberg, A. M.; Schussel, L. J.; Summers, J. S.; Foxman, B. M.;
Petersen, J. L. Inorg. Chem. 1992, 31, 1678-1686.
9) Richardson, P. F.; Chang, C. K.; Spaulding, L. D.; Fajer, J. J. Am.
Chem. Soc. 1979, 101, 7736.
(
(
(
(
(
10) Stolzenberg, A. M.; Spreer, L. O.; Holm, R. H. J. Am. Chem. Soc.
1980, 102, 364-370.
(11) Chang, C. K.; Fajer, J. J. Am. Chem. Soc. 1980, 102, 848.
(
12) Stolzenberg, A. M.; Strauss, S. H.; Holm, R. H. J. Am. Chem. Soc.
1981, 103, 4763-4778.
(
(
(
(
(
(
13) Chang, C. K.; Hanson, L. K.; Richardson, P. F.; Young, R.; Fajer, J.
Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 2652.
8-31
14) Stolzenberg, A. M.; Stershic, M. T. Inorg. Chem. 1987, 26, 1970-
1977.
15) Stolzenberg, A. M.; Stershic, M. T. Inorg. Chem. 1987, 26, 3082-
083.
16) Stolzenberg, A. M.; Stershic, M. T. Inorg. Chem. 1988, 27, 1614-
620.
17) Stolzenberg, A. M.; Stershic, M. T. J. Am. Chem. Soc. 1988, 110,
391-6402.
18) Stolzenberg, A. M.; Schussel, L. J. Inorg. Chem. 1991, 30, 3205-
213.
(22) Procyk, A. D.; Stolzenberg, A. M.; Bocian, D. F. Inorg. Chem. 1993,
32, 627-633.
3
(23) Eaton, S. S.; Eaton, G. R. J. Am. Chem. Soc. 1975, 97, 3660-3666.
(24) Fournari, P.; Guilard, R.; Fontesse, M.; Latour, J.-M.; Marchon, J.-C.
J. Organomet. Chem. 1976, 110, 205-217.
1
6
(25) Eaton, S. S.; Eaton, G. R. J. Am. Chem. Soc. 1977, 99, 6594-6599.
(26) Shroyer, A. L. W.; Lorberau, C.; Eaton, S. S.; Eaton, G. R. J. Org.
Chem. 1980, 45, 4296-4302.
3
(
19) Choi, I.-K.; Ryan, M. D. New J. Chem. 1992, 16, 591.
(27) Eaton, S. S.; Fishwild, D. M.; Eaton, G. R. Inorg. Chem. 1978, 17,
1542-1545.
(20) Summers, J. S.; Stolzenberg, A. M. J. Am. Chem. Soc. 1993, 115,
1
0559-10567.
21) Stolzenberg, A. M.; Summers, J. S. Inorg. Chem. 2000, 39, 1518-
524.
(28) Bonnett, R.; Stephenson, G. F. J. Org. Chem. 1965, 30, 2791-2798.
(29) Bonnett, R.; Gale, I. A. D.; Stephenson, G. F. J. Chem. Soc. C 1966,
1600-1604.
(
1
Inorganic Chemistry, Vol. 41, No. 2, 2002 301

Stolzenberg and Haymond
(TXBC). A mixture of 1.0 g (1.4 mmol) of H TXP, 1.66 g
2
8.9 mmol) of p-toluenesulfonhydrazide, 4.0 g of anhydrous K -
Also, calculated spin densities at the meso positions of
hydroporphyrin cation and anion radicals differ within a
H
2
2
(
3
2,33
CO , and 250 mL of dry pyridine was reacted at 95 °C for 36 h.
macrocycle and between the different macrocycles.
We
3
After an initial heating period of 2 h, an additional aliquot of 1.0
g of p-toluenesulfonhydrazide was added every 4 h (with the
report in this paper the syntheses of titanyl hydroporphyrin
complexes and compare the activation barriers for rotation
of meso-aryl groups in these complexes with those in titanyl
porphyrin complexes.
exception of one addition during an 8 h overnight period). The
3
7
reaction was worked up following the procedure for H
The crude product was purified by preparative TLC on silica plates
to afford 50 mg (68 µmol, 5% yield). UV-vis (CH Cl ): λmax, nm
(rel abs), 355 (0.95), 365 (0.91), 376 (1.00), 522 (0.38), 739 (0.74).
(TXiBC). The compound was prepared from H (TXP) and
purified by the procedure reported for H
(TTiBC).37 When reacted
at 100 °C for 24 h, 1.0 g (1.4 mmol) of H (TXP) afforded 90 mg
(0.12 mmol, 9% yield) of isolated product after flash chromatog-
raphy. UV-vis (CH Cl ): λmax, nm (rel abs), 370 (sh, 0.66), 391
2
(TTC).
Experimental Section
2
2
All reactions, chromatography, recrystallizations, and sample
manipulations involving hydroporphyrins were carried out under
subdued lights and under a nitrogen atmosphere using standard
Schlenk techniques or in a Vacuum Atmospheres Co. drybox, unless
otherwise noted. Reagents and solvents used in this study were
HPLC or reagent grade. Toluene and THF were distilled from
sodium benzophenone ketyl. Methylene chloride and chloroform
H
2
2
2
2
2
2
(0.91), 414 (1.00), 516 (0.10), 553 (0.14), 598.7 (0.20).
Lithium Salts of Tetrapyrroles. The free-base porphyrin or
hydroporphyrin was dried overnight in a vacuum oven at 60 °C. A
50 mg sample of the free-base compound was placed in an oven-
dried Schlenk flask and dissolved in 5 mL of toluene. The resulting
solution was freeze-thaw degassed. A 2.05 equiv sample of
butyllithium (2.5 M in hexane) was added by syringe. The mixture
was stirred for 5-10 min at room temperature, during which time
the solution changed color. The solvent was removed under vacuum
to afford a solid that was redissolved in an appropriate solvent for
spectroscopy.
2
were distilled from CaH . Pyridine was distilled from BaO. NMR
solvents were treated to remove traces of water and acid im-
mediately before use by passage down a dry column of grade I
basic alumina. The initial runnings were discarded.
The meso-tetraarylporphyrins H
(TpClPP), H (TpCF PP), H (TpMeOPP), H
(T3,5MeOPP) were prepared from pyrrole and the appropriate
2
(TPP), H
2
(TTP), H
2
(TXP),
H
H
2
2
3
2
2
(T3,5FPP), and
2
34
substituted benzaldehyde by either the Adler-Longo or Lindsey
method.3
5,36
H
2
(T3,5BrPP) and H
2
(T3,5tBuPP) were purchased from
Strem. H
2
(TTC) and H (TTiBC) were prepared and purified by
2
literature methods.3
7
(a) Li (TTP). UV-vis (CH Cl ): λ , nm (rel abs), 416 (sh,
2
2
2
max
1
Absorption spectra were recorded on a Perkin-Elmer Lambda 6
0.39), 434 (1.00), 576 (0.06), 620 (0.07). H NMR (C D ): δ 2.56
7
8
1
UV-vis spectrophotometer. H NMR spectra were recorded on a
(s, 12 H, Ph-CH ), 7.48 (d, 8 H, meta Ph), 8.28 (br d, 8 H, ortho
3
JEOL Eclipse 270 spectrometer (270.17 MHz) equipped with a
Ph), 8.92 (s, 8 H, py).
variable-temperature control system.
(
b) Li
60, 602, 626.
c) Li (TTiBC). UV-vis (CH
.53), 409 (sh, 0.86), 420 (1.00), 503 (0.09), 615 (0.19), 714 (0.01).
d) Li (TTBC). A 6.4 equiv sample of butyllithium was required.
The solution color changed from red to purple. UV-vis (C ):
max, nm (rel abs), 365 (0.96), 390 (1.0), 574 (0.36), 755 (0.77).
2
(TTC). UV-vis (CH
2
Cl
2
): λmax, nm, 407 (sh), 429, 521,
H
2
(TTBC). A mixture of 2.1 g (3.1 mmol) of H
32.2 mmol) of p-toluenesulfonhydrazide, 8.0 g (58 mmol) of
anhydrous K CO , and 250 mL of dry pyridine was placed in a
00 mL three-neck flask that was equipped with a gas inlet atop a
2
(TTP), 6.0 g
5
0
(
(
2
2
Cl
2
): λmax, nm (rel abs), 390 (sh,
2
3
5
(
2
reflux condensor, a septum, a stir bar, and a thermocouple probe
connected to a J-Kem model 210T temperature controller. The
contents of the flask were degassed and placed under nitrogen. The
reaction was heated and stirred at 95 °C for 96 h. The warm reaction
mixture was filtered anaerobically, and the filtrate was evaporated.
The resulting solid was suspended in dry acetone by stirring and
then collected on a Schlenk frit. The blue solid was washed with
methanol (100 mL) followed by acetone (200 mL) and dried in
7
H
8
λ
1
H NMR (C
7 8
D ): δ 2.42 (s, 12 H, Ph-CH
3 2
), 3.98 (s, 8 H, CH -
CH ), 7.33 (d, 8 H, metal Ph), 7.85 (br d, 8 H, ortho Ph), 8.31 (s,
H, py).
TiO(TTP). A 2.53 g (3.77 mmol) sample of dried H (TPP) and
2
00 mL of dry toluene were placed in an oven-dried 250 mL
Schlenk flask that was stoppered with a septum. The resulting
solution was freeze-thaw degassed. The lithium salt was prepared
by adding 3.1 mL (7.73 mmol) of butyllithium (2.5 M in hexanes)
by syringe. The solution color changed from red to bright forest
green while the solution was stirred at room temperature. When
the color change was complete after 10 min, 15.0 mL of 1.0 M
2
8
1
2
vacuo to afford 700 mg (1.0 mmol, 32% yield) of H (TTBC) that
contained roughly 10% of an overreduced impurity presumed to
be a hexahydroporphyrin.
H
2
(TXC). The compound was prepared from H
purified by the procedure reported for H
(TTC).37 When reacted at
0 °C for 12 h, 1.0 g (1.4 mmol) of H (TXP) afforded 0.22 g (22%
yield) of isolated product after flash chromatography. UV-vis
2
(TXP) and
2
9
2
4
TiCl in toluene solution (15.0 mmol) was added by syringe. The
solution was heated to 50 °C and stirred for 2 h, during which
time the color changed to dark green. The solution was cooled to
room temperature, then exposed to air, and stirred until the color
turned bright crimson red due to conversion of the titanium
dichloride complex to a titanyl complex. The solution was applied
to the top of a column of dry alumina (activity grade I) in a flash
chromatography column. Elution with toluene (under gravity flow
until the alumina was fully saturated with liquid and then with head
-
3
-1
-1
(
CH
2
Cl
2 M
): λmax, nm (10 ꢀ , M cm ), 420 (64.6), 518 (5.4),
547 (4.2), 597 (2.3), 653 (10.3).
(
(
(
(
(
(
(
30) Bonnett, R.; Gale, I. A. D.; Stephenson, G. F. J. Chem. Soc. C 1967,
168-1172.
31) Stolzenberg, A. M.; Laliberte, M. A. J. Org. Chem. 1987, 52, 1022-
027.
1
1
32) Fajer, J.; Davis, M. S. In The Porphyrins; Dolphin, D., Ed.; Academic
Press: New York, 1979; Vol. IV, pp 197-256.
2
pressure applied) gave a band of unreacted H (TPP). Chloroform
was then used to elute TiO(TTP). The chloroform was removed
on a rotary evaporator and the residue recrystallized to afford 2.04
33) Fujita, E.; Chang, C. K.; Fajer, J. J. Am. Chem. Soc. 1985, 107, 7665-
7669.
34) Adler, A. D.; Longo, F. R.; Finarelli, J. D.; Goldmacher, J.; Assour,
J.; Korsakoff, L. J. Org. Chem. 1967, 32, 476.
35) Lindsey, J. S.; Schreiman, I. C.; Hsu, H. C.; Kearney, P. C.;
Marguerettaz, A. M. J. Org. Chem. 1987, 52, 827-836.
36) Lindsey, J. S.; Wagner, R. W. J. Org. Chem. 1989, 54, 828-836.
(37) Stolzenberg, A. M.; Simerly, S. W.; Steffey, B. D.; Haymond, G. S.
J. Am. Chem. Soc. 1997, 119, 11843-11854.
302 Inorganic Chemistry, Vol. 41, No. 2, 2002

meso-Aryl Rotation in Titanyl Tetraaryltetrapyrroles
g of TiO(TTP) (2.78 mmol, 74% yield). UV-vis (CH
2
Cl
2
): λmax
,
Table 1. Abbreviations for Meso-Substituted Porphyrins
nm (rel abs), 426 (1.00), 548 (0.06), 586 (0.01).
porphyrin
aryl group
porphyrin
aryl group
TiO(TXP). TiO(TXP) (and the titanyl complexes of other
substituted tetraphenylporphyrins) was prepared similarly in about
TPP
TTP
TXP
TpClPP
phenyl
4-methylphenyl (tolyl)
3,5-dimethylphenyl (xylyl) T3,5BrPP
4-chlorophenyl
TpMeOPP
T3,5FPP
4-methoxyphenyl
3,5-difluorophenyl
3,5-dibromophenyl
8
(
0% yield. UV-vis (CH
0.08), 590 (0.04).
TiO(TTC). The procedure for TiO(TTP) was modified by
2 2
Cl ): λmax, nm (rel abs), 425 (1.00), 551
T3,5MeOPP 3,5-dimethoxyphenyl
TpCF PP 4-(trifluoromethyl)phenyl T3,5tBuPP 3,5-di-tert-butylphenyl
3
Results
decreasing the reaction time at 50 °C to 1.5 h and not exposing the
reaction solution to air. Conversion of the titanium dichloride
solution to a titanyl complex and initial purification were achieved
by applying the reaction solution to the column of dry alumina
under a nitrogen atmosphere. The column was run as above. The
green chloroform solution of TiO(TTC) was concentrated and
brought into a nitrogen-filled glovebag for further purification by
preparative TLC on silica. The plates were developed with
chloroform. Development was performed in the dark by covering
the tank with aluminum foil. Solvent was evaporated from the plates
using a heavy stream of nitrogen. The band of silica containing
the purified complex was scraped off the plate and the compound
recovered by washing the silica with chloroform. A total of 45 mg
Synthesis. The porphyrin and hydroporphyrin complexes
used in this study were para-substituted or meta-disubstituted
tetraphenylporphyrin complexes. The abbreviations used to
specify a particular tetrapyrrole consist of a T (for tetra)
followed by a designation for the meso-aryl group and then
P, C, BC, or iBC to, respectively, indicate porphyrin, chlorin,
bacteriochlorin, or isobacteriochlorin. Abbreviations for
porphyrin compounds are listed in Table 1 to illustrate the
designations for the meso-aryl groups.
Replacement of the phenyl para H atom or meta H atoms
with groups that do not spin couple with the remaining aryl
protons simplifies the latters’ resonances from multiplets to
doublets or singlets. Given that hydroporphyrins have several
inequivalent meso-aryl group environments, the greater ease
of resolution and assignment of the NMR spectra of phenyl-
substituted complexes is a distinct advantage.
of TiO(TTC) (61 µmol) was obtained from 70 mg of H
µmol) using corresponding quantities of solvent and reagents, a
yield of 59%. UV-vis (CH Cl ): λmax, nm (rel abs), 402 (sh, 0.34),
22 (1.00), 521 (0.15), 593.4 (0.15), 630.8 (0.24).
TiO(TTiBC). The procedure for TiO(TTC) was followed with
2
(TTC) (104
2
2
4
the exception that the reaction time at 50 °C was 30 min. A total
of 70 mg of TiO(TTiBC) (95 µmol) was obtained from 95 mg of
Syntheses of the titanyl hydroporphyrin complexes requires
two transformations of a parent free-base porphyrin com-
pound: reduction of one or more â-â double bonds in the
porphyrin π-system and metalation of a free-base compound
with titanium. In principle, either reduction followed by
metalation or metalation followed by reduction could afford
the desired complexes. However, the tendency of hydropor-
phyrin compounds to undergo oxidative dehydrogenation to
porphyrins, especially during metalation reactions, and
the potential differences in the ease of purification of free-
base hydroporphyrin vs metallohydroporphyrin compounds
made the most efficient approach to pure samples of the
desired complexes less than obvious. Thus, we investigated
both approaches.
H
2
(TTiBC) (129 µmol), a yield of 74%. UV-vis (CH
nm (rel abs), 385 (0.60), 405 (0.76), 419 (1.00), 567 (0.29), 615
0.44).
TiO(TTBC). The procedure for TiO(TTC) was modified by
2 2
Cl ): λmax,
(
using 6.4 equiv of butyllithium and heating at 50 °C for 2.5 h. The
solution of the titanium dichloride complex was filtered through a
small bed of Celite, which served to convert it to the titanyl
complex. The complex was used with no additional purification.
UV-vis (CH Cl ): λmax, nm, 359, 402, 556, 790.
2 2
TiO(TXC). The procedure for TiO(TTC) was followed. A total
of 50 mg of TiO(TXC) (63 µmol) was obtained from 85 mg of
10
39
H
2
(TXC) (116 µmol), a yield of 54%. UV-vis (CH
nm (rel abs), 402 (sh, 0.23), 423 (1.00), 522 (0.04), 551 (0.05),
94 (0.05), 633 (0.15).
Variable-Temperature NMR Spectra. H NMR spectra were
obtained using freshly prepared 1,1,2,2-tetrachloroethane-d solu-
2 2
Cl ): λmax,
Several methods have been used for metalation of free-
base porphyrins with titanium. The earliest methods reported,
5
1
4
0
reaction with (C
and reaction with TiO(acac)
5 5 2
H )TiCl in refluxing diethylene glycol
2
41
2
in refluxing phenol, required
tions of titanyl complexes. The NMR tubes containing the solutions
were sealed off with a glassblowing torch under a partial nitrogen
harsh conditions and afforded titanyl porphyrin complexes
in less than 60% yield. Better yields (90%) of the titanyl
2
atmosphere. Chemical shifts in tetrachloroethane-d are only slightly
complexes were obtained by reaction with TiCl
4
in refluxing
anhydrous toluene. Mild heating of lithium porphyrin salts
Li (THF) P with TiCl (THF) in toluene affords titanium
different (generally within 0.1 ppm) from those reported in Table
24
-
3
3, which are in CDCl
3
. Solutions were typically (5-10) × 10
M
in complex. Spectra were recorded at appropriate temperatures over
the range from the slow exchange to fast exchange limits. Samples
were given adequate time to equilibrate at each temperature before
the spectrum was recorded. The reproducibility of the spectra of a
particular complex at a given temperature was good, both for
repeated measurements of an individual sample cycled over the
temperature range and for measurements of independent samples
of the complex. Eaton and Eaton have shown that the exchange
process is independent of the sample concentration.23 The probe
temperature was calibrated using methanol and ethylene glycol
standards and the temperature-dependent shifts of Van Geet, which
have errors (rms) of 0.6 and 0.3 °C, respectively.38 The temperature
achieved in the probe was reproducible to (0.2 °C.
2
4
4
2
4
2
porphyrin dihalide complexes. The latter are readily
converted to titanyl complexes upon exposure to moisture.
TiO(TTP) and TiO(TXP) were accessible by any of the
above methods. Reduction of either complex with diimide,
(38) Van Geet, A. L. Anal. Chem. 1968, 40, 2227-2229.
(
(
(
(
39) Lahiri, G. K.; Summers, J. S.; Stolzenberg, A. M. Inorg. Chem. 1991,
30, 5049-5052.
40) Fuhrhop, J.-H.; Kadish, K. M.; Davis, D. G. J. Am. Chem. Soc. 1973,
95, 5140-5147.
41) Buchler, J. W.; Eikelmann, G.; Puppe, L.; Rohbock, K.; Schneehage,
H. H.; Weck, D. Liebigs Ann. Chem. 1971, 745, 135-151.
42) Berreau, L. M.; Hays, J. A.; Young, V. G.; Woo, L. K. Inorg. Chem.
1994, 33, 105-108.
Inorganic Chemistry, Vol. 41, No. 2, 2002 303

Stolzenberg and Haymond
N
2
H
2
, generated in situ from p-toluenesulfonhydrazide af-
reduced, hexahydroporphyrin products produced at extended
reaction times is decreased to less than 10% by lowering
the temperature of the reduction reaction.
forded gross mixtures of the corresponding titanyl porphyrin,
chlorin, and isobacteriochlorin complexes that effectively
could not be separated. Little, if any, bacteriochlorin was
present because reduction of metalloporphyrins results in
selective formation of the isobacteriochlorin isomer in the
second reduction step.43 Thus, metalation followed by
reduction does not provide ready access to pure samples of
all four compounds.
The presence of additional methyl substituents in tetra-
xylylporphyrin, H
difficult than that of H
complete consumption of H
sion of the tetrahydroporphyrins H
2
(TXP), made its reduction even more
(TTP). Conditions that forced
(TXP) led to substantial conver-
(TXBC) and H (TXiBC)
2
2
2
2
to overreduced compounds. Thus, yields of the tetrahydro-
porphyrins were quite low, and we did not pursue the
syntheses of the titanyl complexes of these compounds.
The utility of the various metalation methods (above) when
applied to hydroporphyrins was investigated using 80-90%
We reported previously the syntheses of the tetratolyl-
3
7
hydroporphyrins H
These were obtained by diimide reduction of tetratolylpor-
phyrin, H (TTP). Consistent with our experience with the
diimide reduction of H
(TPP),17 the materials obtained
2 2 2
(TTC), H (TTBC), and H (TTiBC).
2
2
2
pure free-base compounds. Reaction of H (TTC) with either
directly from the reduction were mixtures of the various
compounds that contain at best 80-90% of the target
hydroporphyrin. Purities were somewhat lower for the
tetratolyl compounds because the rates of reduction are
slowed by the electron-donating methyl group. In particular,
the reduction step from porphyrin to chlorin is slowed to a
greater extent than subsequent steps. Thus, it is harder to
both effect complete reduction of the porphyrin and avoid
the formation of overreduced compounds. The selective
quinone reoxidation and phosphoric acid extration procedures
that were reported to afford pure individual tetraphenyl-
hydroporphyrin compounds43 did not work in our hands.
Purified tetratolylhydroporphyrin compounds can be obtained
by flash chromatography but only on a 10-100 mg scale.
These difficulties were not an issue in the context of the
(C )TiCl in diethylene glycol or TiO(acac)
5
H
5
2
2
in phenol
resulted in quantitative conversion to titanyl complex.
However, the harsh conditions also led to quantitative
conversion of the chlorin to porphyrin. Reactions of hydro-
porphyrins with 12 equiv of TiCl in refluxing toluene for
4
30 h resulted in near-quantitative conversion to metal
complexes and partial oxidation of the hydroporphyrin
2 2
macrocycle. For H (TTC) and H (TTiBC), the crude reaction
product contained roughly 60% of the respective titanyl
complex, which could be substantially purified by repeated
chromatography. In contrast, the crude product obtained from
17
H
2
(TTBC) contained only about 10% TiO(TTBC), which
did not survive chromatography. Reaction of lithium hydro-
porphyrin salts (see below) with 4 equiv of TiCl (or TiCl
(THF) ) in toluene at 50 °C for a few hours resulted in near-
4
4
-
2
37
synthesis of N-alkyl-substituted hydroporphyrin compounds.
complete metalation accompanied by minimal oxidation of
the hydroporphyrin macrocycle. The titanium dichloride
complexes of TTC and TTiBC formed in this reaction were
converted to titanyl complexes upon contact with the alumina
column packing, which served to separate the desired titanyl
complex from traces of unreacted free-base and excess
It was most efficient to use the 80-90% pure mixtures in
the alkylation reactions. Chromatographic separation of the
N-methylporphyrin and -hydroporphyrin products from each
other and from unreacted starting materials was substantially
easier and could be conducted on a larger scale than
separation of the individual free-base hydroporphyrins. In
contrast, chromatographic purification of titanyl hydropor-
phyrin complexes is relatively difficult. Thus, more highly
purified samples of free-base hydroporphyrins had to be
employed in the metalation reactions used to prepare
complexes for the dynamic NMR experiments reported here.
The titanyl complexes generally were further purified by
preparative TLC. Unfortunately, TiO(TTBC) is not suf-
ficiently stable to permit its chromatographic separation from
TiO(TTC) and TiO(TPP), whose presence interferes with the
titanium compounds. A similar conversion of Ti(TTBC)Cl
2
to TiO(TTBC) was effected by rapid filtration through Celite,
which, unlike alumina or silica, did not oxidize the bacte-
riochlorin.
Application of the last metalation method required us to
prepare the previously unknown lithium hydroporphyrin salts.
The reported procedure for isolation of lithium porphyrin
salts involves reaction of free-base porphyrin with 2 equiv
of LiN(SiMe )
3 2
for 8 h in refluxing THF or DME.4
4,45
Spectroscopic monitoring of the reaction showed that hy-
droporphyrins were incompletely deprotonated and partially
oxidized under these conditions. Hence, we chose to rapidly
generate the lithium salts in toluene at room temperature with
the stronger base butyllithium. Slightly more than 2 equiv
dynamic NMR experiments. Thus, samples of H
used in metalation reactions must be free of H (TTC) and
(TTP). In addition, the metalation reaction must not form
TiO(TTC) and TiO(TPP) by oxidation of the bacteriochlorin.
The procedure that we reported for H (TTBC) is inadequate
for our current purposes. Chromatographic purification of
(TTBC) leads to great loss of material and does not
completely remove H (TTC) and H (TTP). We report here
2
(TTBC)
2
H
2
2
of butyllithium was required to deprotonate H
2
(TTP), H
2
-
(TTC), or H (TTiBC). Complete titration of H
2
2
(TTBC)
H
2
required roughly 6 equiv. Presumably, some of the protic
solvents used to wash the solid during isolation were retained
and consumed butyllithium. Organolithium reagents have
been reported to add to the meso positions and â-positions
2
2
a modified procedure that eliminates these impurities by
extending the reaction time of the diimide reduction and
simplifying the workup. In addition, the amount of over-
(
44) Arnold, J. J. Chem. Soc., Chem. Commun. 1990, 976-978.
(
43) Whitlock, J., H. W.; Hanauer, R.; Oester, M. Y.; Bower, B. K. J. Am.
(45) Arnold, J.; Dawson, D. Y.; Hoffman, C. G. J. Am. Chem. Soc. 1993,
115, 2707-2713.
Chem. Soc. 1969, 91, 7485-7489.
304 Inorganic Chemistry, Vol. 41, No. 2, 2002

meso-Aryl Rotation in Titanyl Tetraaryltetrapyrroles
Table 2. 1H NMR Data for Free-Base Compoundsa
compd
H2(TXP)
NH
Ph-CH3
CH2CH2
para H
ortho H
pyrrole H
8.87 (s, 8H)
8.21 (d, 5.4 Hz, 2H)
8.46 (s, 2H)
-2.80 (br s, 2H)
-1.48 (br s, 2H)
2.60 (s, 24H)
2.53 (s, 12H)
.56 (s, 12H)
7.40 (s, 4H)
7.29 (s, 2H)
7.34 (s, 2H)
7.83 (s, 8H)
7.49 (s, 4H)
7.76 (s, 4H)
H2(TXC)
4.18 (s, 4H)
2
8.60 (d, 5.4 Hz, 2H)
H2(TXBC)
H2(TXiBC)
-1.37 (br s, 2H)
0.82 (br s, 2H)
2.50 (s, 24H)
2.40 (s, 18H)b
4.00 (s, 8H)
3.28 (m, 8H)
7.23 (s, 4H)
7.39 (br s, 4H)
7.43 (s, 8H)
7.05 (s, 2H)
7.10 (s, 2H)
7.97 (s, 4H)
6.91 (d, 4.4 Hz, 2H)
7.42 (d, 4.4 Hz, 2H)
2.42 (s, 6H)
7
.18 (s, 4H)
c
d
H2(T3,5FPP)
-2.94 (br s, 2H)
-2.90 (br s, 2H)
-2.85 (br s, 2H)
-2.72 (br s, 2H)
7.32 (t, 4H)
7.79 (d, 8H)
8.29 (br, 8H)
7.39 (s, 8H)
8.07 (br, 8H)
8.93 (s, 8H)
8.88 (s, 8H)
8.92 (s, 8H)
8.89 (s, 8H)
H2(T3,5BrPP)
H2(T3,5MeOPP)
H2(T3,5tBuPP)
8.14 (br, 4H)
6.89 (s, 4H)
7.76 (br, 4H)
3.95 (s, 24H)
1.50 (s, 72H)
a
Parts per million relative to TMS in CDCl3 solution at 20 °C. Two peaks (of 6H and 12H) overlap. c Triplet with barely resolved doublet splitting, JFH
b
3
4
d 3
)
8.9 Hz, JHH ) 2.0 Hz.
JFH ) 6 Hz; peaks too broad to resolve splitting by para H.
of free-base porphyrins and metalloporphyrins.46-48 We did
not observe addition to the hydroporphyrins under the
conditions that we employed. Generally, the lithium salts
were generated in situ and used in metalation reactions
without isolation.
1
H NMR data for free-base tetratolylporphyrin and -hy-
droporphyrin compounds were reported previously.37 Data
for free-base tetraxylylporphyrin and -hydroporphyrin com-
pounds and other free-base meta-disubstituted porphyrin
compounds are reported in Table 2.
1
1
H NMR Spectra. The H NMR spectra of free-base, para-
Metalation of the free-base compounds with lithium or
titanium results in several changes to the NMR spectra. The
upfield NH proton peak disappears due to the replacement
of these protons by the metal. Metalation with lithium does
not remove the equivalency of the two faces of the tetra-
substituted tetraphenylporphyrins consists of a broad singlet
near -3 ppm for the NH protons, a singlet near 8.9 ppm for
the pyrrole â-protons, a pair of “doublets” (with unequal
intensities typical of a four-line AB pattern that has ∆ν .
J) between 8.1 and 7.4 ppm for the ortho and meta protons,
respectively, and appropriate peaks for the para substituent.
The coupling constants between adjacent ortho and meta
protons are typically 8 Hz. All other couplings in the phenyl
rings are small and unresolved. Spectra for free-base, meta-
disubstituted tetraphenylporphyrins are similar with the
exceptions that a singlet near 7.4 ppm is present for the para
proton, the meta proton peak is absent, and the ortho proton
peak is a singlet.
The lower symmetry of hydroporphyrins removes the
equivalency of the four meso-aryl groups. Chlorins have two
aryl environments. Two aryl groups are adjacent to two
pyrrole rings (A type), and two are adjacent to one reduced,
or pyrroline, ring (B type). Isobacteriochlorins have three
environments: one A-type aryl ring, two B-type aryl rings,
and one aryl ring between two pyrroline rings (C type).
Bacteriochlorins have four equivalent B-type aryl groups.
Although the increased saturation of the hydroporphyrins
decreases the ring current and in turn moves the NH protons
downfield and the aryl group protons upfield, the ranges of
chemical shifts of the different aryl groups still overlap each
other. The pyrrole â-protons of hydroporphyrins shift upfield
relative to their position in porphyrins and can fall in the
same chemical shift region as the aryl protons. The pyrrole
â-proton peaks are readily distinguished from aryl proton
peaks, though, by the integration of singlets or by the
characteristic 4.5 Hz coupling of doublets. Finally, new peaks
appear for the hydroporphyrin pyrroline â-protons between
1
pyrrole macrocycles. Hence, the H NMR spectra of the
lithium complexes have the same multiplicity of peaks as
the spectra of the free-base compounds. Small changes in
chemical shifts occur, though. In contrast, the titanyl group
removes the equivalency of the two faces. Multiplicities of
the peaks for pyrrole â-protons, which lie in the macrocycle
plane, and para phenyl substituents, which rotate freely, are
unchanged relative to those of the spectra of the free-base
compounds. However, the peaks for the hydroporphyrin
pyrroline â-protons have greater multiplicity due to the
additional coupling between the now inequivalent protons.
In addition, the spectra of titanyl porphyrin complexes at
room or lower temperature show a doubling of peaks for
ortho protons and meta protons or substituents. The in-
equivalence of the faces and the restricted rotation of the
meso-aryl groups about the meso carbon to aryl carbon bonds
cause the AB pattern of the free-base compounds to change
to an ABCD pattern in the titanyl complexes. The chemical
shift differences between the two doublets for the ortho
protons and for the meta protons are both significant. The
chemical shift difference between the doublets is 106 Hz
for the ortho protons and 26 Hz for the meta protons of TiO-
(
TTP) when the spectra are obtained on a 270 MHz
spectrometer. As the temperature is raised, the rate of rotation
of the aryl groups increases and the pairs of doublets for the
ortho and meta protons broaden and then average to an
apparent AB pattern similar to that observed in H (TTP).
2
H NMR data for metal complexes in the slow exchange
limit are reported in Table 3.
1
4.2 and 3.2 ppm.
1
The H NMR spectra of titanyl hydroporphyrins that have
(
(
(
46) Kalisch, W. W.; Senge, M. O. Angew. Chem., Int. Ed. Engl. 1998,
7, 1107-1109.
47) Senge, M. O.; Kalisch, W. W.; Bischoff, I. Chem.sEur. J. 2000, 6,
721-2738.
48) Krattinger, B.; Callot, H. J. Eur. J. Org. Chem. 1999, 1857-1867.
3
multiple aryl group environments are considerably more
complicated than that of TiO(TTP). The slow exchange
spectrum of TiO(TTC) has four doublets corresponding to
2
Inorganic Chemistry, Vol. 41, No. 2, 2002 305

Stolzenberg and Haymond
Table 3. 1H NMR Data for Titanyl Tetrapyrrole Complexesa,b
compd
TiO(TTP)
Ph-CH3
CH2CH2
para H
meta H
ortho H
pyrrole H
9.22 (s, 8H)
2.76 (s, 12H)
7.58 (d, 8.1 Hz, 4H)
8.06 (d, 8.1 Hz, 4H)
8.43 (d, 8.1 Hz, 4H)
7.58 (d, 8.1 Hz, 2H)
7.81 (d, 8.1 Hz, 2H)
7
.70 (d, 8.1 Hz, 4H)
c
d
TiO(TTC)
2.60 (s, 6H)
.64 (s, 6H)
4.15-4.40 (4H)
7.48 (8H)
8.11 (d, 4.4 Hz, 2H)
8.45 (s, 2H)
2
7
8
.88 (d, 8.1 Hz, 2H)
.13 (d, 8.1 Hz, 2H)
8.54 (d, 4.4 Hz, 2H)
c
TiO(TTBC)
TiO(TTiBC)
2.57 (s, 12H)
2.48 (s, 3H)
3.95-4.35 (8H)
7.42 (d, 7.4 Hz, 4H)
7.56 (d, 7.4 Hz, 4H)
7.84 (d, 7.4 Hz, 4H)
e
7.96 (s, 8H)
7.48 (d, 7.4 Hz, 4H)
c
3.40-3.60 (8H)
e
7.18 (d, 4.3 Hz, 2H)
7.70 (d, 4.3 Hz, 2H)
2
2
.50 (s, 6H)
.54 (s, 3H)
TiO(TXP)
TiO(TXC)
2.63 (s, 12H)
.71 (s, 12H)
2.49 (s, 6H)
7.50 (s, 4H)
7.17 (s, 4H)
7.78 (s, 4H)
7.29 (s, 2H)
7.54 (s, 2H)
7.66 (s, 2H)
7.87 (s, 2H)
9.25 (s, 8H)
2
c
4.15-4.40 (4H)
7.30 (s, 2H)
7.33 (s, 2H)
8.20 (d, 4.4 Hz, 2H)
8.52 (s, 2H)
8.62 (d, 4.4 Hz, 2H)
2
2
2
.52 (s, 6H)
.55 (s, 6H)
.58 (s, 6H)
TiO(TpClPP)
7.76 (d, 8.0 Hz, 4H)
.85 (d, 8.0 Hz, 4H)
7.26 (d, 8.0 Hz, 2H)
.37 (d, 8.0 Hz, 2H)
8.07 (d, 7.7 Hz, 4H)
.15 (d, 7.7 Hz, 4H)
8.06 (d, 8.0 Hz, 4H)
8.40 (d, 8.0 Hz, 4H)
8.04 (d, 8 Hz, 4H)
8.38 (d, 8 Hz, 4H)
8.26 (d, 7.7 Hz, 4H)
8.61 (d, 7.7 Hz, 4H)
9.14 (s, 8H)
9.15 (s, 8H)
9.11 (s, 8H)
9.19 (s, 8H)
9.21 (s, 8H)
9.23 (s, 8H)
9.21 (s, 8H)
7
TiO(TpMeOPP)
TiO(TpCF3PP)
TiO(T3,5FPP)
TiO(T3,5BrPP)
TiO(T3,5MeOPP)
TiO(T3,5tBuPP)
4.10 (s, 12H)
7
8
f
g
7.36 (dt, 4H)
7.67 (d, 4H)
g
8
.03 (d, 4H)
h
h
8.23 (s, 4H)
8.23 (s, 4H)
.60 (s, 4H)
8
3.95 (s, 12H)
.00 (s, 12H)
1.53 (s, 36H)
.57 (s, 36H)
6.96 (s, 4H)
7.85 (s, 4H)
7.33 (s, 4H)
7.66 (s, 4H)
8.05 (s, 4H)
8.34 (s, 4H)
4
1
a
Parts per million relative to TMS in CDCl3 solution. Data are at the slow exchange limit. Typically T ) 0-20 °C. c Multiplet. d Multiple unresolved
b
e
f
3
4
overlapping doublets. Broad, overlapping peaks between 7.30 and 7.55 ppm. Triplet with barely resolved doublet splitting, JFH ) 8.9 Hz, JHH ) 2.0 Hz.
g
3
JFH ) 7.9 Hz; peaks too broad to resolve splitting by para H. h Coincident at the slow exchange limit but resolved at higher temperature.
different ortho protons. The doublets for the ortho protons
of the A-type tolyl groups are centered at 7.81 and 8.13 ppm.
Those for the B-type tolyl groups are centered at 7.58 and
easier to resolve and interpret because ortho H peaks are
singlets when meta protons are not present to spin couple.
Unfortunately, the electronic effects of the additional methyl
substituents in the xylyl group (above) led to the tetra-
hydroporphyrin complexes being unavailable in the quantities
and purity necessary for study. Results are included for TiO-
(TXP) and TiO(TXC). The activation barriers to rotation of
the xylyl groups in these complexes were unexpectedly
higher than those for tolyl groups (see below). In an effort
to understand the reason for the increase, we expanded the
study to include the titanyl complexes of other meta-
disubstituted and para-substituted tetraphenylporphyrin com-
plexes that had substituents of varied size and electronic
properties.
7
.88 ppm. The chemical shifts of the four different meta
protons lie in a sufficiently narrow range around 7.45-7.60
ppm that well-resolved doublets are not observed, even at
low temperatures. The ortho H doublet at 7.58 overlaps the
meta H multiplet. In addition, the ortho H doublet at 8.13
ppm overlaps a doublet for pyrrole â-protons at 8.11 ppm.
As the temperature is increased, the ortho H multiplets for
the two tolyl groups broaden and shift position through each
other at different rates. Consequently, the coalescence
temperatures of the ortho protons of the two tolyl groups of
TiO(TTC) cannot be judged as precisely as those of TiO-
(
TTP). The situation is even worse for TiO(TTiBC), which
Analysis of Activation Parameters. Two approaches have
been reported to determine the activation parameters for
chemical exchange from variable-temperature spectral data.
The first approach utilizes the Gutowsky-Holm approxima-
has three different tolyl group environments. We were unable
to resolve or identify the expected six doublets for the ortho
protons that should be observed in the slow exchange limit,
even at -60 °C, and were unable to judge the temperatures
of coalescence of the related pairs of doublets. It is not clear
whether this is due simply to the number of overlapping
peaks or this is because exchange is fast at low temperatures.
The latter could result from smaller activation barriers and/
or smaller frequency differences between the exchanging
peaks for this complex.
4
9
q
tion for coalescence, and obtains ∆G Tc values from ∆ν
5
0
and T
c
c
, where ∆ν is the frequency difference and T is the
coalescence temperature of the exchanging peaks. Coales-
cence occurs when the valley between the separate, exchang-
ing peaks just disappears. This approach gives only the value
q
of ∆G at a single temperature and hence provides no
q
q
information about ∆H or ∆S . The second approach uses
full line shape analysis. Rate constants for exchange, k , are
The inaccessibility of activation parameters for all four
of the complexes in the tetratolylporphyrin series led us to
investigate the complexes in the tetraxylylporphyrin series.
We expected the spectra of the xylyl complexes would be
r
(
49) Pople, J. A.; Schneider, W. G.; Bernstein, H. J. High-Resolution
Nuclear Magnetic Resonance; McGraw-Hill: New York, 1959; p 223.
(50) Shanan-Atidi, H.; Bar-Eli, K. H. J. Phys. Chem. 1970, 74, 961-963.
306 Inorganic Chemistry, Vol. 41, No. 2, 2002

meso-Aryl Rotation in Titanyl Tetraaryltetrapyrroles
Table 4. Activation Energies for Aryl Group Rotationa
determined by comparisons at each temperature of experi-
mental and simulated NMR data for the region containing
the exchanging protons. Activation parameters are obtained
typeb
q
c
compd
TiO(TTP)
Tc, °C
∆G , kcal/mol
A
A
B
B
A
A
B
A
A
A
A
A
A
A
70
50
60
55
98
87
100
70
71
58
16.1 ( 0.1
15.6 ( 0.25
16.2 ( 0.25
15.9 ( 0.25
17.9 ( 0.1
17.5 ( 0.15
18.0 ( 0.25
16.5 ( 0.1
16.5 ( 0.1
16.0 ( 0.1
17.0 ( 0.1
18.4 ( 0.1
18.1 ( 0.1
18.4 ( 0.1
from weighted least-squares fits of k
/T) or Eyring (ln(hk /kT) vs 1/T) equations over the full
temperature range.
r
to Arrhenius (ln k
r
vs
TiO(TTC)
1
r
TiO(TTBC)
TiO(TXP)
TiO(TXC)
The best computer program for full line shape analysis of
the NMR spectra of exchanging systems that we were able
TiO(TpClPP)
51
to locate was DNMR5. The parameters required to simulate
the spectrum can be specified as fixed values or variables
whose values will be determined by the program. These
TiO(TpCF3PP)
TiO(TpMeOPP)
TiO(T3,5FPP)
TiO(T3,5BrPP)
TiO(T3,5MeOPP)
TiO(T3,5tBuPP)
82
108
105
110
2
include chemical shifts, coupling constants, T relaxation
times, relative populations of the exchanging configurations,
and the rate constants for exchange. The program uses
iterative, least-squares fits of a calculated spectrum to the
experimental spectrum to model the best values of the varied
parameters. Unfortunately, DNMR5 was unable to handle
the number of parameters that had to be specified to simulate
the hydroporphyrin spectra, which had multiple exchanging
spin systems and nonexchanging spin systems whose lines
overlapped in the region of interest. Thus, we determined
activation parameters by the approximate method.
a
In 1,1,2,2-tetrachloroethane-d2 solution. b A ) aryl adjacent to two
pyrrole rings, and B ) aryl adjacent to one pyrrole and one pyrroline ring.
Uncertainty is a relative rather than absolute error.
c
q
in ∆G Tc for hydroporphyrin complexes are somewhat larger
because the multiple exchanging spin systems made it more
difficult to judge coalescence. Values of ∆G Tc and its
q
associated error were estimated by taking the average and
half the difference of values calculated for the temperatures
at which an observer was certain coalescence had not yet
occurred and at which coalescence had definitely occurred.
q
Absolute errors in ∆G Tc values determined by the ap-
proximate method include contributions from experimental
errors and from systematic errors introduced as a result of
the approximations made to simplify the Bloch equation and
the deviations of the actual systems from the idealized system
modeled in the approximation. Several studies have shown
that the approximate method leads to errors of at worst 0.3
q
The ∆G Tc values reported in Table 4 were determined
using data for the ortho protons, which have the largest ∆ν
and therefore are closest to the idealized system modeled in
the approximation. The values that we report for TTP,
TpClPP, TpCF
3
PP, and TpMeOPP are in good agreement
q
52,53
with literature values that were determined at lower field
kcal/mol in ∆G Tc compared to full line shape analysis.
2
5
and referenced to 298 K, especially when the -∆T∆S
The systematic errors are expected to be of similar size and
magnitude for the compounds in this study, which have
q
correction term is included. (Values of ∆S for phenyl ring
rotation in titanyl tetraphenylporphyrin complexes are typi-
similar spin systems, populations, T values, and ∆ν values.
c
2
5
cally about -10 eu. ) In several of the complexes that we
We will consider only the relative errors due to experimental
measurement errors in making comparisons between com-
pounds.
studied the ortho protons and meta protons or substituents
provided for two independent determinations of ∆G . The
agreement of these values was reasonable. They will not be
q
q
The main sources of relative error in ∆G Tc are errors in
q
identical because the difference in T
c
affects ∆G .
c
measurement of ∆ν and T . Given the typical values of
activation energies, T , and ∆ν for the complexes in this
c
Discussion
study, the error in ∆ν would have to be about 15 Hz to
q
The results in Table 4 establish that reduction of the
pyrrole rings in hydroporphyrins affects the barriers for
rotation of meso-aryl groups. The effects of reduction cannot
be explained by simple interpretations. The barriers for
B-type aryl groups, which are next to one pyrroline ring,
are larger in TiO(TTC) and TiO(TXC) than the barriers for
A-type groups in the same compound. Moreover, they may
be larger than barriers for the A-type groups in their
respective porphyrin complexes. The increased barriers for
the B-type groups cannot be attributed solely to the increased
change ∆G Tc by (0.1 kcal/mol. The accuracy in ∆ν
measurements in this study was better than 0.5 Hz, which
q
introduces negligible error in ∆G Tc. Errors in T
c
result both
from errors in the actual probe temperature and from errors
in judging the temperature at which coalescence occurs. The
reproducibility of the probe temperature was about 0.2 °C
(although the absolute error is larger). The coalescence point
for porphyrin complexes could be judged to (1 °C, which
introduces an error of roughly (0.05 kcal/mol in the systems
studied. Thus, relative errors in the values of porphyrin
complexes are estimated as 0.1 kcal/mol. The relative errors
2
steric bulk of the pyrroline â-CH group relative to the
pyrrole â-CH group. If sterics were the only important factor,
then the A-type barriers in porphyrin and chlorin should be
identical and the B-type barriers in chlorin and bacterio-
chlorin should be identical. This is clearly not the case.
A-type barriers for chlorins are less than those for parent
porphyrins, and the B-type barrier in TiO(TTBC) is smaller
than that in TiO(TTC). These observations could be con-
(
51) LeMaster, C. B.; LeMaster, C. L.; True, N. S. DNMR5: IteratiVe
Nuclear Magnetic Resonance Program for Unsaturated Exchange-
Broadened Bandshapes; Quantum Chemistry Program Exchange;
Indiana University: Bloomington, IN.
(
52) Kost, D.; Carlson, E. H.; Raban, M. J. Chem. Soc., Chem. Commun.
1
971, 656-657.
53) Egan, W.; Tang, R.; Zon, G.; Mislow, K. J. Am. Chem. Soc. 1971,
3, 6205-6216.
(
9
Inorganic Chemistry, Vol. 41, No. 2, 2002 307

Stolzenberg and Haymond
sistent with more saturated tetrapyrroles having somewhat
increased out-of-plane flexibility. This would make deforma-
tion of the macrocycle more facile and in turn permit the
ortho H atoms and â-H atoms to avoid each other more
readily. However, the observations might also be conse-
quences of electronic effects, which are expected to make
variable contributions to the barriers in porphyrins and
hydroporphyrins. The aryl group π-system and the tetra-
pyrrole π-system are coplanar (or nearly so) in the transition
state for aryl group rotation. The orbital overlap between
these π-systems that occurs at the meso carbon-ipso carbon
bond will affect the energy of the transition state. The
magnitude of the overlap will differ both for different
tetrapyrroles and for the inequivalent aryl groups of chlorins
examine the effects of the size and electronic nature of the
meta substituent on barriers to rotation. The data in Table 4
establish that barriers increase from 17.0 kcal/mol for TiO-
(T3,5FPP) to 18.4 kcal/mol for TiO(T3,5tBuPP). Comparison
of the barriers for Br and F meta substituents shows that
stronger electron-withdrawing substituents do not have
increased barriers, unlike the case for para-substituted aryl
groups. A slowing of aryl group rotation for meta-disubsti-
tuted groups relative to para-substituted groups had been
reported previously and was attributed to the steric bulk of
the meta group “buttressing” the ortho hydrogens. Our data
lead us to question this suggestion. The van der Waals radius
of bromine is comparable to or slightly smaller than that of
a methyl group. Yet, the barrier for aryl group rotation in
TiO(T3,5BrPP) is larger than that of TiO(TXP) and com-
parable to that of TiO(T3,5tBuPP). Barriers appear to be
related to both the size and mass of the meta substituent.
The size effect may be due to a phenomenon analogous to
drag, the resistance to movement through a fluid medium.
Larger meta substituents, which are off-axis, will sweep out
larger volumes of solvent as the aryl group rotates. More
massive substituents, on the other hand, will increase the
moment of inertia of the aryl group.
26
2
5
and of bacteriochlorins. Data in Table 4 and the literature
clearly show that the electronic nature of the para substituent
in titanyl tetraarylporphyrins leads to a variation in barrier
that is comparable to or larger than the differences observed
here between barriers for the same type (A or B) aryl group
in different tetrapyrroles. Barriers are lower for electron-
donating substituents than for electron-withdrawing substit-
uents, but were not linearly related to the Hammett-Taft
parameters for the substituents.
An interesting observation is that barriers for xylyl group
rotation in TiO(TXP) and TiO(TXC) are roughly 1.5 kcal/
mol greater than the corresponding barriers for tolyl group
rotation in TiO(TTP) and TiO(TTC). These results led us to
Acknowledgment. This research was supported by the
National Institutes of Health (Grant GM 33882).
IC010936O
308 Inorganic Chemistry, Vol. 41, No. 2, 2002