Reactions of nitrogen oxides with heme models. Characterization of NO and NO2 dissociation from Fe(TPP)(NO2)(NO) by flash photolysis and rapid dilution techniques: Fe(TPP)(NO2) as an unstable intermediate

Article abstract of DOI:10.1021/ja020653a

Described are studies directed toward elucidating the controversial chemistry relating to the solution phase reactions of nitric oxide with the iron(II) porphyrin complex Fe(TPP)(NO) (1, TPP = mesotetraphenylporphinato^2-). The only reaction observable with clean NO is the formation of the diamagnetic dinitrosyl species Fe(TPP)(NO)₂ (2), and this is seen only at low temperatures (K₁ < 3 M^-1 at ambient temperature). However, 1 does readily react reversibly with N₂O₃ in the presence of excess NO to give the nitro nitrosyl complex Fe(TPP)(NO₂)(NO) (3), suggesting that previous claims that 1 promotes NO disproportionation to give 3 may have been compromised by traces of air in the nitric oxide sources. It is also noted that 3 undergoes reversible loss of NO to give the elusive nitro species Fe(TPP)(NO₂) (4), which has been implicated as a powerful oxygen atom transfer agent in reactions with various substrates. Furthermore, in the presence of excess NO₂, the latter undergoes oxidation to the stable nitrato analogue Fe(TPP)(NO₃) (5). Owing to such reactivity of Fe(TPP)(NO₂), flash photolysis and stopped-flow kinetics rather than static techniques were necessary for the accurate measurement of dissociation equilibria characteristic of Fe(TPP)(NO₂)(NO) in 298 K toluene solution. Flash photolysis of 3 resulted in competitive NO₂ and NO dissociation to give Fe(TPP)(NO) and Fe(TPP)(NO₂), respectively. The rate constant for the reaction of 1 with N₂O₃to generate Fe(TPP)(NO₂)(NO) was determined to be 1.8 × 10⁶ M^-1 s^-1, and that for the NO reaction with 4 was similarly determined to be 4.2 × 10⁵ M^-1 s^-1. Stopped-flow rapid dilution techniques were used to determine the rate constant for NO dissociation from 3 as 2.6 s^-1. The rapid dilution experiments also demonstrated that Fe(TPP)(NO₂) readily undergoes further oxidation to give Fe-(TPP)(NO₃). The mechanistic implications of these observations are discussed, and it is suggested that NO₂ liberated spontaneously from Fe(P)(NO₂) may play a role in an important oxidative process involving this elusive species.

Full text of DOI:10.1021/ja020653a

Published on Web 07/27/2002
Reactions of Nitrogen Oxides with Heme Models.
Characterization of NO and NO2 Dissociation from
Fe(TPP)(NO2)(NO) by Flash Photolysis and Rapid Dilution
Techniques: Fe(TPP)(NO2) as an Unstable Intermediate
Mark D. Lim,† Ivan M. Lorkovic,*,† Katrin Wedeking,† Andrew W. Zanella,‡
Carmen F. Works,† Steve M. Massick,† and Peter C. Ford*,†
Contribution from the Department of Chemistry and Biochemistry, UniVersity of California,
Santa Barbara, California 93106, and Joint Science Department, Claremont Colleges,
Claremont, California 91711-5916
Received May 6, 2002
Abstract: Described are studies directed toward elucidating the controversial chemistry relating to the
solution phase reactions of nitric oxide with the iron(II) porphyrin complex Fe(TPP)(NO) (1, TPP ) meso-
tetraphenylporphinato2-). The only reaction observable with clean NO is the formation of the diamagnetic
dinitrosyl species Fe(TPP)(NO)2 (2), and this is seen only at low temperatures (K1 < 3 M-1 at ambient
temperature). However, 1 does readily react reversibly with N2O3 in the presence of excess NO to give the
nitro nitrosyl complex Fe(TPP)(NO2)(NO) (3), suggesting that previous claims that 1 promotes NO
disproportionation to give 3 may have been compromised by traces of air in the nitric oxide sources. It is
also noted that 3 undergoes reversible loss of NO to give the elusive nitro species Fe(TPP)(NO2) (4),
which has been implicated as a powerful oxygen atom transfer agent in reactions with various substrates.
Furthermore, in the presence of excess NO2, the latter undergoes oxidation to the stable nitrato analogue
Fe(TPP)(NO3) (5). Owing to such reactivity of Fe(TPP)(NO2), flash photolysis and stopped-flow kinetics
rather than static techniques were necessary for the accurate measurement of dissociation equilibria
characteristic of Fe(TPP)(NO2)(NO) in 298 K toluene solution. Flash photolysis of 3 resulted in competitive
NO2 and NO dissociation to give Fe(TPP)(NO) and Fe(TPP)(NO2), respectively. The rate constant for the
reaction of 1 with N2O3 to generate Fe(TPP)(NO2)(NO) was determined to be 1.8 × 106 M-1 s-1, and that
for the NO reaction with 4 was similarly determined to be 4.2 × 105 M-1 s-1. Stopped-flow rapid dilution
techniques were used to determine the rate constant for NO dissociation from 3 as 2.6 s-1. The rapid
dilution experiments also demonstrated that Fe(TPP)(NO2) readily undergoes further oxidation to give Fe-
(TPP)(NO3). The mechanistic implications of these observations are discussed, and it is suggested that
NO2 liberated spontaneously from Fe(P)(NO2) may play a role in an important oxidative process involving
this elusive species.
via addition of NO.2 However, there have been conflicting
Introduction
reports2-7 describing products from the further reaction of 1
Given continuing elucidation of the biological functions of
nitric oxide (nitrogen monoxide),1 the interactions of NO and
other NOx species with relevant metal centers have become
increasingly important to model. Of particular interest is nitrogen
oxide speciation in the presence of ferri- and ferro-hemes in
aerobic aqueous and hydrophobic media given the roles played
by hemes in the biological synthesis and bioregulatory actions
of NO.
with excess NO in solution. A recent communication6a from
this laboratory demonstrated the equilibrium formation of the
diamagnetic dinitrosyl species Fe(TPP)(NO)2 (2) (eq 1), which
can be observed at low temperature using proton NMR
techniques. The interaction is relatively weak (K1 < 3 M-1 at
298 K), so there is little evidence for the formation of 2 in
ambient temperature solutions. EPR evidence consistent with
eq 1 was described in a much earlier report by Wayland and
Olson.2
The iron(II) heme model Fe(TPP) forms a stable nitrosyl
complex Fe(TPP)(NO) (1, TPP ) meso-tetraphenylporphinato2-)
(2) (a) Wayland, B. B.; Olson, L. W. J. Am. Chem. Soc. 1974, 96, 6037-
6041. (b) Scheidt, W. R.; Frisse, M. E. J. Am. Chem. Soc. 1975, 97, 17-
21.
(3) Lanc¸on, D.; Kadish, K. M. J. Am. Chem. Soc. 1983, 105, 5610-5617.
(4) Yoshimura, T. Inorg. Chim. Acta 1984, 83, 17-21.
(5) Ellison, M. K.; Schulz, C. E.; Scheidt, W. R. Inorg. Chem. 1999, 38, 100-
108.
(6) (a) Lorkovic, I. M.; Ford, P. C. J. Am. Chem. Soc. 2000, 122, 6516-6517.
(b) Lorkovic, I. M.; Ford, P. C. Inorg. Chem. 2000, 39, 632-633.
(7) Lin, R.; Farmer, P. J. J. Am. Chem. Soc. 2001, 123, 1143-1150.
* To whom correspondence should be addressed. E-mail: ford@
chem.ucsb.edu.
† University of California, Santa Barbara.
‡ Claremont Colleges.
(1) (a) Nitric Oxide, Biology, and Pathobiology; Ignarro, L. J., Ed.; Academic
Press: San Diego, 2000. (b) Feelisch, M., Stamler, J. S., Eds. Methods in
Nitric Oxide Research; Wiley: Chichester, U.K., 1996 and references
therein.
9
10.1021/ja020653a CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 9737-9743
9737

A R T I C L E S
Lim et al.
portionate to 1 plus 5.11c A similar nitro complex Fe(OEP)-
(NO2) was implicated by Castro13 as an intermediate in the
oxidation of alkenes (to form epoxides) by mixtures of Fe(OEP)-
Cl and nitrite salts dissolved in in N-methylpyrrolidinone
solution with 1% acetic acid. Presumably, this involves oxygen
atom transfer from N-coordinated nitro ligand to give the
oxidized substrate plus the stable Fe(II) nitrosyl complex Fe-
(OEP)(NO).
Clearly, the system that is generated from a solution of Fe-
(TPP)(NO) in equilibrium with gaseous NO plus small amounts
of air or other NOx as impurities represents a complex mixture
of reactive species and dynamic equilibria. The present work
is directed toward elucidating such interactions with the goals
of better understanding reactions of higher nitrogen oxides with
heme proteins and of reconciling seemingly conflicting observa-
tions regarding these deceptively complex and reactive systems.
In the interest of resolving the latter ambiguities, we describe
here quantitative investigations of the reaction equilibria and
kinetics that characterize the above mixture. The present report
will present additional documentation regarding reactions 1 and
2 as well as a kinetics study of the reactive species Fe(TPP)-
(NO2) formed from 3 by two methods, flash photolysis or by
stopped-flow rapid dilution.
K1
Fe(TPP)(NO)
Fe(TPP)(NO)
+ NO y\z
(1)
2
(1)
(2)
Other workers7 have subsequently disputed this observation
and argued that NO disproportionation is promoted by 1 to give
the nitro nitrosyl complex Fe(TPP)(NO2)(NO) (3). Indeed, this
reaction finds analogy in NO reactions with ruthenium(II) and
osmium(II) porphyrins8 as well as with certain non-heme iron
complexes.9 While we, in contrast, did not find NO dispropor-
tionation to result from mixing solutions of 1 with clean NO at
ambient temperature,6a we have demonstrated that Fe(TPP)(NO)
reacts readily with N2O3 in the presence of excess NO to give
the nitro nitrosyl complex Fe(TPP)(NO2)(NO) (3).6b
k2
Fe(TPP)(NO )(NO)
Fe(TPP)(NO)
+ N2O3 y z
+ NO (2)
2
k-2
(1)
(2)
These observations suggest that claims that 1 promotes NO
disproportionation in ambient temperature solutions may be
compromised by the presence of traces of air or other NOx in
the nitric oxide sources.
We have also reported6b that, upon removal of the NO
atmosphere from a solution containing 3, this species undergoes
further reaction (eq 3), NO dissociation to give the nitro (or
nitrito) complex Fe(TPP)(NO2) (4).5 However, as will be
described below, 4 is unstable and undergoes further oxidation
(eq 4) to the nitrato complex Fe(TPP)(NO3) (5), which has an
optical spectrum very similar to that of 4.10
Experimental Section
Toluene was distilled from CaH2. Spectrochemical grade toluene
was prepared by washing with sulfuric acid and neutralizing with NaOH
to remove residual thiophene, followed by distillation from sodium.
Deuterated solvents for NMR studies (Cambridge Isotope Laboratories,
Inc.) were used without further purification. Nitric oxide (99%, Aire
Liquide) was purified by passage through a stainless steel column
containing Ascarite II (Thomas Scientific), attached via O-ring seal
(Viton) to a greaseless vacuum line.14 Solutions of known [NO] were
prepared manometrically by vacuum transfer techniques with flasks of
known volume. UV-vis measurements were conducted on an HP
8452A Diode Array spectrophotometer using custom-made cuvettes
fused to a flask equipped with a high-vacuum stopcock and coldfinger
for conducting anaerobic measurements. 1H NMR measurements were
conducted on a Varian 400 MHz instrument.
Fe(TPP)(NO2)(NO) solutions were prepared by adding measured
aliquots of air to toluene solutions of Fe(P)(NO) and NO. Sample
solutions for flash photolysis studies were prepared from 2 (10-20
µM) in CHCl3 or toluene under various NO pressures or under NO
with aliquots of air to prepare N2O3 solutions.
Flash Photolysis Experiments. The flash apparatus has been
described previously.15 This consisted of a frequency tripled Nd:YAG,
Continuum NY-61 excitation source (355 nm, 10 ns fwhm, 20 mJ/
pulse, 2 Hz), and the decays of transient absorbance or bleach at various
wavelengths were monitored with monochromatic probe light. The
probe beam was focused into the sample (at a right angle to the
excitation beam), through a double grating monochromator (SPEX
k3
Fe(TPP)(NO )(NO)
Fe(TPP)(NO ) + NO
y z
(3)
(4)
2
2
k-3
(3)
(4)
Fe(TPP)(NO )
Fe(TPP)(NO )
+ oxidant f
2
3
(4)
(5)
There have been multiple reports in the literature describing
the reactivity of ferric porphyrin complexes of the nitrite ion,
NO2-. For example, Scheidt and co-workers11c characterized
Fe(TpivPP)(NO2) as a high spin complex (â-pyrrole 1H NMR:
δ ) 75 ppm) intermediate formed during addition of [K(18-
crown-6)][NO2] to Fe(TpivPP)(H2O)(CF3SO3) (TpivPP )
R,R,R,R-tetrakis(2-pivalamidophenyl)porphinato2-). The same
species was isolated by Richter-Addo et al.12 as the aerobic
oxidation product of Fe(TpivPP)(NO) in the presence of
pyridine. Subsequently, Scheidt et al. used EPR to characterize
4 (formed by BF3 abstraction of NO2- from the stable anionic
complex Fe(TPP)(NO2)2-) and noted its propensity to dispro-
(8) (a) Miranda, K. M.; Bu, X.; Lorkovic, I. M.; Ford, P. C. Inorg. Chem.
1997, 36, 4838-4848. (b) Kadish, K. M.; Adamian, V. A.; Caemelbecke,
E. V.; Tan, Z.; Tagliatesta, P.; Bianco, P.; Boschi, T.; Yi, G.-B.; Khan, M.
A.; Richter-Addo, G. B. Inorg. Chem. 1996, 35, 1343-1348. (c) Bohle,
D. S.; Goodson, P. A.; Smith, B. D. Polyhedron 1996, 15, 3147-3150.
(d) Lorkovic, I. M.; Ford, P. C. Inorg. Chem. 1999, 38, 1467-1473. (e)
Lorkovic, I. M.; Leal, F. I.; Ford, P. C.; Richter-Addo, G. B., manuscript
in preparation.
(13) (a) O’Shea, S. K.; Wang, W.; Wade, R. S.; Castro, C. E. J. Org. Chem.
1996, 61, 6388-6395. (b) Castro, C. E. J. Am. Chem. Soc. 1996, 118,
3984-5398.
(14) For the preparation of the dinitrosyl complex 2, where traces of N2O3 were
especially damaging, more extensive procedures were adopted. NO was
purified by passing through an Ascarite scrubber, then through an activated
silica (80-100 mesh) packed column (5 ft, 3/8” stainless steel tubing coil)
at -78 °C. The latter procedure removed N2O and any residual NO2 and
N2O3. (Warning: At -78 °C, considerable NO absorbs in the column, and
precaution should be taken for the controlled NO release when the system
warms to room temperature.) Solutions were prepared and transferred to
cells with gastight syringes inside an inert atmosphere “drybox”. Materials
in contact with NO solutions (septa, stopcocks, and plungers) were deaerated
by storage in the drybox for at least 18 h prior to solution exposure.
(15) Lorkovic, I. M.; Miranda, K. M.; Lee, B.; Bernhard, S.; Schoonover, J. R.;
Ford, P. C. J. Am. Chem. Soc. 1998, 120, 11674-11683.
(9) Franz, K. J.; Lippard, S. J. J. Am. Chem. Soc. 1999, 121, 10504.
(10) The Soret band maxima in toluene solution for the various porphyrin
complexes discussed in this manuscript are: Fe(TPP)(NO) (1), 406 nm (ꢀ
) 1 × 105 M-1 cm-1); Fe(TPP)(NO2)(NO) (2), 432 nm (ꢀ ) 1.8 × 105
M-1 cm-1); Fe(TPP)(NO2) (3), 412 nm (ꢀ ) 1.1 × 105 M-1 cm-1); Fe-
(TPP)(NO3) (4), 412 nm (ꢀ ) 1.1 × 105 M-1 cm-1).
(11) (a) Nasri, H.; Goodwin, J. A.; Scheidt, W. R. Inorg. Chem. 1990, 29, 185-
191. (b) Finnegan, M. G.; Lappin, A. G.; Scheidt, W. R. Inorg. Chem.
1990, 29, 181-185. (c) Munro, O. Q.; Scheidt, W. R. Inorg. Chem. 1998,
37, 2308-2316.
(12) Cheng, L.; Powell, D. R.; Khan, M. A.; Richter-Addo, G. B. Chem.
Commun. 2000, 23, 2301.
9
9738 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

Reactions of Nitrogen Oxides with Heme Models
A R T I C L E S
Model 1680), and onto a photomultiplier tube (RCA IP28 for 400-
600 nm, RCA 8852/230 for 600-800 nm). The temporal response (50
shot averages) was recorded on a digital oscilloscope (Tektronix TDS
540) linked to a desktop computer. Plots of intensity versus time were
converted to absorbance, and the curves were fit to single or double
exponentials using Igor (Wavemetrics) software.
Stopped-Flow Experiments. These were performed on an Applied
Photophysics SX-17MV stopped-flow spectrometer using standard
asymmetric sequential mixing modes. The sample (A) and flush (F)
drive syringes were changed from the stock 2.5 mL syringe to 250 µL
(Hamilton Salt Line #1725 AD-SL) syringes. Solutions were loaded
into drive syringes from the solution flask via gastight syringes
(Hamilton Gastight #1010 SL) equipped with removable needles and
exit valves, permitting evacuation of the drive syringes and transfer
volume prior to loading. Rapid dilution of Fe(TPP)(NO2)(NO) solutions
was accomplished by asymmetric mixing of a solution of Fe(TPP)-
(NO2)(NO), NO, and N2O3 (1.5, 3-12, and 1-6 mM, respectively) in
toluene from a 250 µL syringe, with toluene from a 2.5 mL (11-fold
dilution) syringe into an aging loop. After 25 ms, the solution was
driven by toluene in a 250 µL syringe together with a 2.5 mL syringe
into the second mixing chamber (overall 121-fold dilution), observation
cell, and stop syringe.16 Dilution reproducibility was found to be within
5% using an unreactive absorbing sample of Fe(TPP)Cl (Midcentury
Chemicals) and dilution in toluene.
Preparation of 4. An authentic sample of Fe(TPP)(NO3) [1H
NMR: C6D6, 78.4 ppm (â-pyrrole)] was prepared by treatment of the
µ-oxo dimer, [Fe(TPP)]2O (Soret, λmax ) 410 nm; Q-band, λmax ) 572
nm in CH2Cl2), with dilute aqueous HNO3.17
Figure 1. Top: Optical spectra changes upon cooling a solution of Fe-
(TPP)(NO) in the presence of 8 mM NO in CHCl3 to 173 K, followed by
warming to ∼273 K. Bottom: Analogous experiment probing IR spectra
changes upon cooling a methylcyclohexane solution of Fe(TmTP)(NO) plus
NO (8 mM) to 173 K, followed by warming to ∼273 K. (The TmPP
complex was used in this case to improve solubility in methylcyclohexane,
TmTP ) tetra-(m-toluyl)porphinato2-.)
Results and Discussion
Reaction of Fe(TPP)(NO) with NO and with N2O3. As we
described in a preliminary communication,6a the reaction of
clean NO with 1 in toluene solution gave the dinitrosyl complex
Fe(TPP)(NO)2 (eq 1) at low temperature. This is further
demonstrated by the reversible changes in the optical and IR
spectra of the equilibrium mixture of 1 and 2 under NO upon
lowering T (Figure 1). The proton NMR spectra show 2 to be
diamagnetic, and variable T studies of NMR line broadening
and shifts allowed for the calculation of the equilibrium constant
K1 as 23 M-1 at 253 K and 3.1 × 103 M-1 at 179 K (∆H° )
-6.7 ( 0.8 kcal mol-1 and ∆S° ) -21 ( 4 cal mol-1 K-1) for
eq 1.6a Accordingly, K1 < 3 M-1 at ambient temperature. The
IR spectrum of 2 in 173 K methylcyclohexane solution prepared
under a 1:1 mixture of 15NO and 14NO gave three νNO peaks
with a 1:2:1 intensity ratio. This pattern is interpreted as
corresponding to the unlabeled, singly labeled, and doubly
labeled forms of Fe(P)(NO)2 suggesting a trans, centrosymmetric
structure for 2.
Figure 2. Equilibrium constant measurement for the formation of 3 from
1 and N2O3 in toluene at 22 ( 1 °C (eq 2), obtained by addition of air
aliquots (5, 10, 15, 20, 30, 40, 50, 100 µL) to a 117 mL cell containing 1
(8 µM) in toluene (25 mL) and NO (500 Torr, 8 mM). The equilibrium
constant shown was obtained by fitting the plot of ∆A-1 (432 nm) vs
[N2O3]-1 to a line (weighted by ∆A). The N2O3 generated by NO
autoxidation is assumed to be entirely in the solution. The slope and intercept
of this plot are equal to K2-1γ-1[NO], and γ-1, respectively, where γ is (ꢀ2
- ꢀ1)C0, and C0 is the total concentration of Fe(TPP) species. The intercept
divided by the slope gives K2[NO]-1, and multiplication by [NO] gives K2.
Under ambient temperature conditions (or lower), the spectra
of toluene solutions of 1 (Soret band, λmax 406 nm) under an
atmosphere of clean NO gave no indication of the nitro nitrosyl
complex Fe(TPP)(NO2)(NO), a species that is structurally5 and
spectrally18 well characterized. However, as we have reported,6b
introduction of small quantities of air leads to the rapid
appearance of 3 (λmax 432 nm) as the result of NO autoxidation
to NO2 (principally N2O3 under excess NO). The equilibrium
constant K2 was evaluated from the spectral changes as
microliter aliquots of air were introduced into a dilute toluene
solution of 1 (8 µM) and NO (8 mM) (Figure 1). Stepwise
addition of O2 led to decreased absorbance at 406 nm and
increases at 432 nm (λmax for 3) and increased absorbance at
306 nm due to the buildup of N2O3 (ꢀ ) 2 × 103 M-1 cm-1),
and these spectral changes values were used to determine K2 )
160 ( 20 by standard methods19 (Figure 2). Analogous studies
here have investigated the equilibrium constant for reaction of
N2O3 with other Fe(P)(NO) complexes (P ) various porphyrins)
and have noted K5 to range from 1.0 for P ) tetrakis-
(16) It should be noted that, upon dilution, N2O3 is expected to dissociate to
NO and NO2 (>104 s-1) and that, therefore, [NO] after dilution should be
regarded as a lower limit.
(17) (a) Phillippi, M. A.; Baenziger, N.; Goff, H. M. Inorg. Chem. 1981, 20,
3904-3911. (b) Notably, the species formed when 4 was dissolved in
CDCl3 was Fe(TPP)Cl, while in toluene the sample remained 4. (c) Reed,
C. A.; Guiset, F. J. Am. Chem. Soc. 1996, 118, 3281-3282.
(18) Settin, M. F.; Fanning, J. C. Inorg. Chem. 1988, 27, 1431-1435.
(19) Hoshino, M.; Ozawa, K.; Seki, H.; Ford, P. C. J. Am. Chem. Soc. 1993,
115, 9568-9575.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9739

A R T I C L E S
Lim et al.
Figure 3. Attempted steady-state equilibrium (K3) measurement for eq 3,
performed by sequential controlled diminution of [NO] in a solution
originally containing 3 and excess NO. Inset: Plot of the change in
absorbance vs [NO], monitored at 432 nm. The dashed line demonstrates
the expected Lineweaver-Burke behavior for a K2 of 5 × 10-4 M. The
poor agreement to the fit suggests that the model does not describe the
system adequately under these conditions.
(pentafluorophenyl)porphine to 400 for P ) OEP.20 The
consistent trend is for NO2 binding to be more favorable for
the more electron-donating porphyrins.
Figure 4. (a) Difference spectra obtained by subtraction of the spectrum
of 3 from the spectrum of Fe(P)(NOx) (obtained upon removal of NO from
a solution of 3). (b) Transient difference spectrum obtained using a CCD
detector in the 355 nm laser flash photolysis of a CHCl3 solution of 3.
K5
Fe(P)(NO) + NO3 y
\
z Fe(P)(NO2)(NO) + NO
(5)
While Fe(TPP)(NO2)(NO) proved to be relatively stable under
high [NO] (>4 mM) and well defined [N2O3], it decomposed
when [NO] was lowered. For example, 1H NMR analysis
showed the final product in CHCl3 to be Fe(TPP)Cl, (82.2 ppm)
while in toluene it was Fe(TPP)(NO3)(δ 78.46 ppm) (5). The
identities of these products were corroborated by comparing the
δ values of the paramagnetically shifted â-pyrrole protons to
authentic samples of Fe(TPP)Cl and of 5.17
In another experiment, a concentrated toluene solution of Fe-
(TPP)(NO2)(NO) was prepared under a NO/N2O3 atmosphere;
then the system was rapidly evacuated so that the solvent and
all volatile NOx species were removed. The solid residue was
then redissolved in benzene-d6 under inert atmosphere and
transferred to an airtight NMR tube for observation. The
spectrum showed a resonance of 76.55 ppm, which was
concluded to be the â-pyrrole of 4. Over the course of an hour,
this peak disappeared and gave way to one at 78.46 ppm,
matching the spectrum of a sample of 5 prepared by literature
methods.17
Steady-State Experiments To Measure K3. Initial attempts
to determine the equilibrium constant for NO dissociation from
Fe(TPP)(NO2)(NO) to give Fe(TPP)(NO2) (eq 3) employed
steady-state techniques, that is, controlled removal of NO from
the headspace gas over a solution of 3. The resulting changes
in the optical spectra are shown in Figure 3. Although isosbestic
points were apparent, the spectral changes were not completely
reversible upon reintroduction of NO to original pressures.
Furthermore, the spectral changes did not give a good fit to the
[NO] dependence expected for simple equilibrium between 3
and 4 (represented by the dashed line in the inset of Figure 3,
which is a fit to an apparent dissociation constant of 5 × 10-4
M). Although the K3 thus estimated was close to that reported
for the analogous reaction of Fe(TpivPP)(NO2)(NO) (K ) 8 ×
10-4 M),21 the poor fit to Lineweaver-Burke behavior19 and
the irreversibility of the reaction indicated that there may be a
secondary process which converts 4 to a spectroscopically
similar species which binds NO differently.
Flash Photolysis of 3. Flash excitation of metalloporphyrins
under excess NO results in photolabilization of the nitrosyl
complex.22 The subsequent exponential relaxation gives kobs
values equal to kon[NO] + koff, and the respective rate constants
can be extracted from a plot of kobs versus [NO]. The equilibrium
constant for NO dissociation can then be calculated from the
ratio koff/kon.
Flash photolysis of 3 in both toluene and CHCl3 (298 K) led
to transient difference spectra (recorded with a CCD detector)
such as Figure 4. These difference spectra are nearly identical
to the difference spectrum between 3 and the product (or
products) formed when excess NO and other volatiles were
quickly removed from a solution of the former.23 The temporal
absorbance at 680 nm (characteristic of the NO loss photoprod-
uct, presumably 4) decayed exponentially to give kobs values
that proved to be a linear function of [NO] (Figure 5). The slope
of this plot was taken to be the second-order rate constant, k-3
) (4.2 ( 0.5) × 105 M-1 s-1. Notably, the kobs values measured
at 680 nm were independent of the concentration of N2O3,
consistent with the view that the process observed at this
wavelength is indeed the reaction of 4 with NO to give 3.
In contrast, the temporal absorbance changes monitored at
432 nm showed a strong bleach in absorbance followed by a
(21) Frangione, M.; Port, J.; Baldiwala, M.; Judd, A.; Galley, J.; DeVega, M.;
Linna, K.; Caron, L.; Anderson, E.; Goodwin, J. A. Inorg. Chem. 1997,
36, 1904-1911.
(22) Laverman, L. E.; Ford, P. C. J. Am. Chem. Soc. 2001, 123, 11614-11622.
(23) The spectral changes in Figure 1 are consistent with the formation of any
high spin Fe(TPP)X, where X may be NO2-, NO3-, or halides, all of which
have nearly identical electronic spectra (ref 9).
(20) Lorkovic, I. M.; Lim, M. D.; Wedeking, K.; Ford, P. C., manuscript in
preparation.
9
9740 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

Reactions of Nitrogen Oxides with Heme Models
A R T I C L E S
Scheme 1. Reactions of Species Generated by Flash Photolysis
of Fe(TPP)(NO2)(NO)
Scheme 2. Processes Observed Subsequent to Stopped-Flow
Rapid Dilution of a Solution Containing 3, NO, and N2O3
Figure 5. Flash photolysis of solutions of Fe(TPP)(NO2)(NO) in toluene
solution, under varying concentrations of NO and monitored at 680 nm.
Top: Transient absorbance trace at 680 nm upon flash photolysis of 3 (25
µM). [NO] ) 2.5 mM, [N2O3] ) 100 µM; the fit to an exponential function
(kobs ) 1.45 × 103 s-1) is shown by the dashed line. Bottom: Plot of kobs
vs [NO] for spectral changes at 680 nm, representing reformation of 3 from
4 (slope ) kNO ) (4.2 ( 0.5) × 105 M-1 s-1).
K2 ) k2/k-2, use of the previously measured K2(160)2 allowed
for the calculation of k-2 as 1.1 × 104 M-1 s-1
.
Stopped-Flow Experiments. Although, in principle, the [NO]
) 0 intercept of Figure 5 (bottom) gives the rate constant for
NO dissociation from 3 (k3), the intercept is too small to measure
reliably in this manner. Instead, stopped-flow experiments were
attempted. Toluene solutions of 3 were prepared in situ by
stopped-flow mixing of solutions of 1 with dilute solutions of
NO containing small concentrations of N2O3. Under these
conditions, measured values of k3 were extremely variable.
Qualitative observations suggested that k3 increased with
increasing [N2O3] at constant [NO], offering the possibility that
NO is abstracted from 3 by NO2, although some other
mechanism may be responsible. Therefore, it appeared to be
necessary to minimize both [NO] and [N2O3] to characterize
unimolecular NO dissociation rates. Because both NO and N2O3
are required for the long-term stability of 3, rapid high dilution
techniques were pursued.
Concentrated toluene solutions of 3, NO, and N2O3 were
prepared and rapidly mixed in two sequential 11-fold dilutions
into pure toluene (121-fold overall). The rapid dilution resulted
in a nonequilibrium concentration of 3, which then decayed via
NO dissociation to form 4 and subsequently to form 5 (Scheme
2).
The results of these experiments are shown in Figure 7. The
curves in Figure 7 (top) show the absorbance growth at 412
nm resulting from the rapid dilution of a solution containing 3
(∼1.5 mM), NO (4.6 mM), and N2O3 (∼3 mM). The absorbance
changes were bimodal; there was a rapid absorbance increase
complete within the first 0.5 s and then a slower increase
typically complete within 2 min. The rapid step can be attributed
to the rapid re-equilibration between 3 and 4 at the new lower
[NO], while the slower process represents subsequent oxidation
of 4 to the spectrally similar 5.23 The magnitude of the
Figure 6. Flash photolysis of Fe(TPP)(NO2)(NO) in toluene solution, under
varying concentrations of NO and monitored at 432 nm. Top: Transient
absorbance decay after 355 nm photolysis of 3 (10 µM, [NO] ) 8 mM,
[N2O3] ) 100 µM), showing evidence of two processes leading to
reformation of 3. The dotted line shows the fit to a double exponential
decay with two first-order rate constants, kfast (7.7 × 103 s-1) and kslow (3.5
× 102 s-1). Note: The rate constant for the faster process (kfast) was the
same as the kobs determined under these conditions when monitored at 680
nm (see Figure 3). Bottom: Plot of kslow vs [N2O3]. This was assigned to
the reformation of 3 from photogenerated 1 (slope ) kN O ) (1.8 ( 0.2)
2
3
× 106 M-1 s-1).
biexponential return to the parent compound (Figure 6). The
rate constant of the faster process (kfast) was the same as the
kobs for the single exponential found when monitoring at 680
nm under the same conditions. The slower process gave rate
constants kslow that proved to be linearly dependent on the
concentration of N2O3 (Figure 6 (bottom)). Thus, this process
was assigned to reaction of 1 with N2O3 to re-form 3. The slope
of the kobs versus [N2O3] plot gave the second-order rate
constant, k2 ) (1.8 ( 0.2) × 106 M-1 s-1
.
Thus, the flash photolysis of 3 leads to the competitive
dissociation of both NO and NO2, and the spectroscopic
properties of the products (1 and 4) are sufficiently distinct to
study the respective relaxation processes (Scheme 1). Given that
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9741

A R T I C L E S
Lim et al.
Scheme 3. Possible Scenario for the Back Reaction of NO with 4
decreases. The initial rate of change gave the best measurement
of k3, as follows. If ê(t) is taken to represent the time-dependent
extent of the reaction of 3, and rapid approach to K3 is observed,
then the following relationship holds
K3 + [NO]
[4]
êint
)
)
-1 ) k3(k3 + k-3[NO])-1 (6)
K3
[3] + [4]
where êint is defined as the extent of reaction at the conclusion
of the rapid process, before evolution of the slower process.
The approach to this equilibrium may also be written in terms
of ê and the time derivative to give
d
dt
[ê(t) ) êint{1 - e-(k +k [NO])t}] ) k3e-(k +k [NO])t w k3
3
-3
3
-3
Figure 7. Temporal absorbance changes upon 121-fold rapid dilution of
toluene solutions of Fe(TPP)(NO2)(NO), NO (3-12 mM), and N2O3 (∼3
mM). Top: Absorbance at 412 nm showing initial rapid rise representing
rapid equilibration between 3 and 4 (inset), followed by a longer term
absorbance rise representing complete conversion of 3 and 4 to 5. Middle:
Time dependence of extent of reaction (ê) of K3 using averaged spectral
change data at 432 nm (fall) and 412 nm (rise). Extent of reaction was
obtained by normalizing the rapid absorbance change to the total absorbance
change over the longer term. Linear fit of the first 25 ms of the normalized
data gives an [NO] independent initial rate of koff ) 2.6 ( 4 s-1. Bottom:
Rate of approach to initial equilibrium as a function of [NO]; a linear fit
for
t , (k3 + k-3[NO])-1
(7)
In other words, the initial rate of change in extent of reaction
is simply k3, for any given [NO]. This is illustrated by Figure
7(middle), which gives k3 ) 2.6 ( 0.4 s-1. The full course of
this initial equilibration monitored at 412 and 432 nm may also
be fit to an exponential to obtain kobs. Values of kobs plotted
versus [NO] are linear (Figure 7 (bottom)) in accord with a
second-order reaction of 4 with NO to regenerate 3. Accord-
gives a slope corresponding to k-3 ) (2.2 ( 5) × 105 M-1 s-1
.
absorbance change during the fast stage was much less than
would be expected for complete conversion of 3 to 4 (∆Ai )
0.10 vs ∆Atot ≈ 0.3 expected at 412 nm for [3] ) 12 µM).
However, the two processes together did account for complete
conversion of 3. The spectral characteristics of the products from
the fast and slow processes were similar and corresponded to
the spectral change seen in Figure 4. On this basis, the ratio of
initial absorbance change (∆Ai) to total absorbance change
(∆Atot) for the overall process may be used to quantify K3 ≈
2.4 × 10-5 M (see Supporting Figure S-1). Notably, this is about
20-fold smaller than that estimated from the Lineweaver-Burke
plot described in Figure 3.
In this context, a plot of ki versus [NO] should be linear
according to the relationship ki ) k3 + k-3[NO] for a system
approaching equilibrium. To approach the y-intercept (k3) as
closely as possible, it was desirable to prepare solutions of 3
with minimal [NO], although solutions having [NO] < 4 mM
underwent slow decomposition. However, the relatiVe contribu-
tions of the fast and slow components, that is, ∆Ai/∆Atot, may
still be used to represent the extent of reaction for the fast
process, even for cases when partial conversion of 3 to 5 has
occurred prior to dilution and observation.
ingly, the slope is k-3 ) (2.2 ( 0.5) × 105 M-1 s-1 24
. Notably,
this value is an upper limit because the [NO] values used did
not account for NO released owing to N2O3 dissociation,
although the correction would be <25%. In principle, the
intercept gives k3, but the initial rate method (Figure 7 (middle))
is much more accurate in this case.
Comparison of the k-3 values measured by the flash pho-
tolysis and rapid dilution techniques for the “on” reaction of
NO with 3 shows that these are not in exact agreement. The
disagreement may have its origin in the different techniques
employed. Under the high [NO] (∼5 mM) used in flash
photolysis experiments, the kobs values fall in the range 103-
104 s-1 ()k-3[NO], Figure 6b), much higher than the kobs’s (∼10
s-1) observed in the rapid dilution studies where [NO] is very
low (∼50 µM). One might speculate that on the longer time
scale of the stopped-flow experiments, there is equilibration
between linkage isomers of 3 (Scheme 3). If the nitro isomer
reacts more rapidly with NO, the depleted concentration of that
species would be reflected in lower apparent values of k-3
.
Figure 7 (middle) shows that as [NO] is increased, kobs for
the rapid process increases, while the initial extent of reaction
(ê ) ∆Ai/∆Atot) due to the equilibration between 3 and 4
(24) Notably, this value is an upper limit because the [NO] values used did not
account for NO released owing to N2O3 dissociation, although the correction
would be <25.
9
9742 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

Reactions of Nitrogen Oxides with Heme Models
A R T I C L E S
Table 1. Summary of Equilibrium and Rate Constants (298 K) Described in This Worka
reaction
k
f
kb
K
i
1 + NO h 2
<3 M-1 b
1 + N2O3 h 3 + NO
3 h 4 + NO
1.8 × 106 M-1 s-1 c
1.1 × 104 M-1 s-1 c
4.2 × 105 M-1 s-1 c
2.2 × 105 M-1 s-1 d
160b
2.6 s-1 d
1.2 × 10-5
6.2 × 10-6
M
M
4 + NO2 h 5 + NO
∼1 × 103 M-1 s-1 d
a 1 ) Fe(TPP)(NO), 2 ) Fe(TPP)(NO)2, 3 ) Fe(TPP)(NO2)(NO), 4 ) Fe(TPP)(NO2), 5 ) Fe(TPP)(NO3); kf is the forward rate constant, kb is the back
rate constant, and Ki is the equilibrium constant (kf/kb) determined for the reaction shown. Error limits are described in the text. b From static spectroscopic
measurements. c From flash photolysis experiments. d From stopped-flow rapid dilution experiments.
The slower process observed in the stopped-flow dilution
experiments is the conversion of the 3 plus 4 mixture to the
nitrato complex 5. The rate of this transformation is dependent
on [N2O3]. Although concentrated (>100 µM) solutions of 3
have been reported to disproportionate (to 1 plus 5),10 this
process occurred over the course of hours. Thus, it is likely
that the 3 f 5 transformation seen here involves oxidation by
low concentrations of NO2/N2O3 or mostly NO2 in dilute
solution (eq 8). Despite significant experimental uncertainty
under these conditions (e.g., kslow ) 0.02 ( 0.01 for [NO2/N2O3]
) 20 µM), the rate constant k4 (Scheme 2) can be estimated as
that 3 is not formed by the direct reaction of 1 with NO and
that reports to the contrary may have been compromised by
trace impurities of higher NOx species. We have not detected
the presence of an ambient temperature intermediate attributed7
to NO addition to the nitrosyl nitrogen of 1. While it has been
argued7 that our conditions are sufficiently different to prevent
such an observation, we doubt that the difference lies in the
proposal7 that these systems are subject to inhibition by protonic
impurities. Kutikyan et al. have found the IR frequency shifts
claimed as evidence of this interaction in thin films are more
likely the result of annealing the films.25b
When N2O3 (NO2) is introduced to the stable solutions of 1
under clean NO, the nitro nitrosyl complex Fe(TPP)(NO2)(NO)
(3) is the result. This is relatively stable in the presence of excess
NO and N2O3, but when the concentrations are lowered by
stopped-flow dilution, NO dissociates from 3 to form Fe(TPP)-
(NO2), which undergoes future reaction to give the nitrate
complex 5. The similarities of the optical spectra of 4 and 5
may have misled others with regard to the relative stability of
the former. In the present studies, it is clear that 4 undergoes
fairly rapid oxidation by NO2/N2O3 to 5. Because this reaction
also occurs in the solid state,25 we suggest that NO2 dissociation
followed by oxygen atom transfer may be one source of such
oxidations.
Table 1 summarizes the rate constants elucidating the present
system. While the chemical transformations characterized here
are in media far from biological conditions, such processes are
likely to be general to the reactivity of heme enzyme models
with nitric oxides in a hydrophobic environment. The quantita-
tive studies reported provide a baseline of reactivity patterns to
be expected in the absence of specific effects induced by the
protein. Furthermore, these studies emphasize the importance
of characterizing not only the reactions of NO with biologically
relevant targets but also the need to elucidate the reactivities of
other free and coordinated NOx species.
∼103 M-1 s-1
.
Fe(TPP)(NO )
Fe(TPP)(NO )
+ NO2 f
+ NO (8)
2
3
(4)
(5)
These observations are in agreement with IR spectral studies
by Kurtikyan et al.25 of the reactions between films of sublimed
Fe(TPP) and NOx gases in a high-vacuum system. Reaction with
NO2 leads to the nitrato complex 5 (νNO 1526, 1271 cm-1) as
3
the stable product, presumably via oxidation of the intermediate
4 by NO2. On the other hand, when Fe(TPP) films were exposed
to a mixture of NO and NO2 (i.e., N2O3), the IR spectrum
characteristic of 3 (νNO ) 1877 cm-1; νNO ) 1464, 1300 cm-1
)
2
was immediately apparent.
Summary
The present study was initiated in part to elucidate the
disproportionation of NO reported to be promoted by reaction
with Fe(TPP)(NO) (1) and which parallels studies in this
laboratory with analogous ruthenium and osmium complexes.
However, such disproportionation does not occur readily with
1 under NO at ambient temperature. There are no obvious
spectral changes upon exposing solutions of 1 to clean NO, and
analogous studies elsewhere with films of solid 1 gave a similar
result.25 In lower temperature solutions, the optical, IR, and 1H
NMR spectral properties all point to the reversible formation
of a dinitrosyl complex Fe(TPP)(NO)2 (2) in agreement with
an earlier interpretation of ESR spectral changes for a similar
solution.2a On the other hand, 1 does react readily with N2O3
(generated from traces of O2 in excess NO), so we conclude
Acknowledgment. The authors thank the National Science
Foundation (C.F.W.), the ACS Petroleum Research Fund
(M.D.L., I.M.L., A.A.Z.), and the U.S. Department of Energy
(S.M.M.) for financial support of this work. We thank Dr. T.
S. Kurtikyan of the Armenian Research Institute of Applied
Chemistry for bringing ref 25 to our attention.
(25) (a) Kurtikyan, T. S.; Stepanyan, T. G.; Akopyan, M. E. Russ. J. Coord.
Chem. 1999, 25, 721-725. (b) These workers (ref 25c) find νNO changes
Supporting Information Available: Plot of [NO] versus 1/ê
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
in these films associated with a phase change from amorphous (1677 cm-1
)
to ordered film (1697 cm-1), induced by annealing the film at ∼80 °C,
and then recooling. Similar changes have been attributed to H2O binding,7
but in these high-vacuum experiments the water activity should be very
low. (c) Personal communication from T. S. Kurtikyan.
JA020653A
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9743