O atom transfer from nitric oxide catalyzed by Fe(TPP)

Article abstract of DOI:10.1021/ja993643r

The reaction of NO-Fe(TPP) with low pressures of NO gas proceeds through three distinct transformations, the first of which we suggest is the formation of an N-N-coupled, (NO)₂ adduct intermediate. The subsequent formation of NO(NO₂)Fe(TPP), which under these conditions readily loses NO, suggests that it is formed by addition of free NO₂ to the starting nitrosyl. A mechanism is proposed which implies that the addition of a competitive O atom acceptor would lead to catalytic production of N₂O. In agreement with the proposed mechanism, the formation of N₂O is decoupled from the formation of the nitrite by using PPh₃ as the competitive acceptor. The mechanism of O atom transfer was examined by cross-labeling experiments, which show that both O atoms in the intermediate are equivalent, even under catalytic conditions, The formation of an intermediate was confirmed by IR spectroscopy of the heterogeneous reaction of an NO-Fe(TPP) film with gaseous NO, in which transient, isotope-sensitive v(NO) bands are seen prior to NO(NO₂)Fe(TPP) formation. Mixed ¹⁴N/¹⁵N label experiments demonstrate coupling between the two bound nitrosyls in the transient species.

Full text of DOI:10.1021/ja993643r

J. Am. Chem. Soc. 2001, 123, 1143-1150
1143
O Atom Transfer from Nitric Oxide Catalyzed by Fe(TPP)
Rong Lin and Patrick J. Farmer*
Contribution from the Department of Chemistry, UniVersity of California, IrVine, California, 92697-2025
ReceiVed October 11, 1999. ReVised Manuscript ReceiVed September 22, 2000
Abstract: The reaction of NO-Fe(TPP) with low pressures of NO gas proceeds through three distinct
transformations, the first of which we suggest is the formation of an N-N-coupled, (NO)2 adduct intermediate.
The subsequent formation of NO(NO2)Fe(TPP), which under these conditions readily loses NO, suggests that
it is formed by addition of free NO2 to the starting nitrosyl. A mechanism is proposed which implies that the
addition of a competitive O atom acceptor would lead to catalytic production of N2O. In agreement with the
proposed mechanism, the formation of N2O is decoupled from the formation of the nitrite by using PPh3 as
the competitive acceptor. The mechanism of O atom transfer was examined by cross-labeling experiments,
which show that both O atoms in the intermediate are equivalent, even under catalytic conditions. The formation
of an intermediate was confirmed by IR spectroscopy of the heterogeneous reaction of an NO-Fe(TPP) film
with gaseous NO, in which transient, isotope-sensitive ν(ΝÃ) bands are seen prior to NO(NO2)Fe(TPP)
formation. Mixed 14N/15N label experiments demonstrate coupling between the two bound nitrosyls in the
transient species.
3NO + M8 N2O + M-NO2
(1)
Introduction
The identification of the physiological activity of nitric oxide
has generated much interest in the chemistry of metal nitrosyls,
both as models of metalloprotein reactivity and as possible
pharmaceutical delivery systems.1,2 Though it has been studied
for many years, the chemistry of metal-coordinated nitrosyls is
complex and still not fully understood.3-6 For example, many
metal-nitrosyl complexes undergo further reaction with excess
NO, ultimately forming N2O and a metal-nitrite complex as
shown in eq 1. Examples of such reactivity are known for
nitrosyl complexes of Co,7,8 Fe,9-12 Ru,13,14 Mn,15 Cu,16,17 Rh,18
and Ir.19
Several distinct steps must occur during this metal-assisted
disproportionation reaction. First, an N-N coupling must occur
to form the precursor to N2O. Such reactivity mimics that of
the Cu-based nitrous oxide reductases (NoR) which transform
NO into N2O with the concurrent formation of water, eq 2.20
2NO Cu-NoR8 N2O + H2O
(2)
Various Cu complex models for NoR active sites reductively
disproportionate NO by eq 1.16,17 Work by Tolman and
co-workers with Cu tris(pyrazolyl)borate complexes has pointed
to a mononuclear coupling of nitrosyls, postulating either a
dinitrosyl (I) or a hyponitrous form (II) as the N2O-forming
intermediate.21 We have postulated an Fe-bound hyponitrous
intermediate similar to II in our investigations of the electro-
catalytic reduction of NO by myoglobin.22 Others have sug-
gested similar N-N-coupled intermediates for heme-based NoR
enzymes.23,24
(1) Butler, A. R.; Williams, D. L. H. Chem. Soc. ReV. 1993, 233-241.
(2) Ford, P. C.; Bourassa, J.; Miranda, K.; Lee, B.; Lorkovic, I.; Boggs,
S.; Kudo, S.; Laverman, L. Coord. Chem. ReV. 1998, 171, 185-202.
(3) Richter-Addo, G. B.; Legzdins, P. Metal Nitrosyls; Oxford University
Press: New York, 1992.
(4) Eisenberg, R.; Meyer, C. D. Acc. Chem. Res. 1975, 8, 26-34.
(5) McCleverty, J. A. Chem. ReV. 1979, 79, 53-76.
(6) Bottomley, F. Reactions of Coordinated Ligands; Plenum: New York,
1989; Vol. 2.
(7) Gans, P. J. Chem. Soc. (A) 1967, 943-946.
(8) Gwost, D.; Caulton, K. G. Inorg. Chem. 1974, 13, 414-417.
(9) Yoshimura, T. Inorg. Chim. Acta 1984, 83, 17-21.
(10) Ellison, M. K.; Schulz, C. E.; Scheidt, W. R. Inorg. Chem. 1999,
38, 100-108.
(11) Franz, K. J.; Lippard, S. J. J. Am. Chem. Soc. 1999, 121, 10504-
10512.
(12) Ju, T. D.; Woods, A. S.; Cotter, R. J.; Moenne-Loccoz, P.; Karlin,
K. D. Inorg. Chim. Acta 2000, 297, 362-372.
A second requirement in eq 1 is the loss of an O atom from
the N-N-coupled intermediate, ultimately yielding gaseous N2O
(13) Kadish, K. M.; Adamian, V. A.; Van Caemelbecke, E.; Tan, Z.;
Tagliatesta, P.; Bianco, P.; Boschi, T.; Yi, G.; Khan, M. A.; Richter-Addo,
G. B. Inorg. Chem. 1996, 35, 1343-1348.
(14) Miranda, K. M.; Bu, X.; Lorkovic, I. M.; Ford, P. C. Inorg. Chem.
1997, 36, 4838-4848.
(19) Johnson, B. F. G.; Bhaduri, S. J. Chem. Soc., Chem. Commun. 1973,
650
(20) Averill, B. A. Chem. ReV. 1996, 96, 2951-2964 and references
therein.
(15) Franz, K. J.; Lippard, S. J. J. Am. Chem. Soc. 1998, 120, 9034-
9040.
(16) Paul, P. P.; Karlin, K. D. J. Am. Chem. Soc. 1991, 113, 6331-
6332.
(17) (a) Ruggiero, C. E.; Carrier, S. M.; Tolman, W. B. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 895-897. (b) Schneider, J. L.; Halfen, J. A.; Young,
V. G.; Tolman, W. B. New J. Chem. 1998, 22, 459. (c) Schneider, J. L.;
Young, V. G.; Tolman, W. B. Inorg. Chem. 1996, 35, 5410.
(18) Hughes, W. B. Chem. Commun. 1969, 1126.
(21) Schneider, J. L.; Carrier, S. M.; Ruggiero, C. E.; Young, V. G., Jr.;
Tolman, W. B. J. Am. Chem. Soc. 1998, 120, 11408-11418.
(22) Bayachou, M.; Lin, R.; Farmer, P. J. J. Am. Chem. Soc. 1998, 120,
9888-9893.
(23) MacNeil, J. H.; Berseth, P. A.; Bruner, E. L.; Perkins, T. L.; Wadia,
Y.; Westwood, G.; Trogler, W. C. J. Am. Chem. Soc. 1997, 119, 1668-
1675.
10.1021/ja993643r CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/23/2001

1144 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Lin and Farmer
freeze-pump-thaw method. Purified nitric oxide was transferred into
the flask, and the solution was stirred vigorously overnight before
evaporation under vacuum. The resulting solid was washed with
degassed methanol several times.
(ii) Hydroxylamine Reduction.29 Fe(TPP)Cl was dissolved in
toluene in a round-bottom flask and degassed. After hydroxylamine
hydrochloride was dissolved in methanol, sodium methoxide in
methanol was added. The white precipitate formed was discarded after
the product solution was passed through the filter paper. The resulting
solution was added to the Fe(TPP)Cl solution and stirred overnight. A
large portion of methanol was added to precipitate the red-brown solid.
This solid was washed with methanol several times and dried under
vacuum.
and a metal nitrite. Several mechanisms have been suggested,
such as O atom transfer from a coordinated N2O2 to a second
metal nitrosyl.5,8 In a series of papers, Ford and co-workers
studied the NO disproportionation reaction catalyzed by Ru-
(TPP).25 Kinetic evidence identified an intermediate, suggested
as a dinitrosyl adduct, which subsequently reacted with two
additional NO molecules to yield a nitrosyl/nitrito adduct and
N2O. Unexpectedly, recent reports from the same laboratory
suggest that Fe(TPP) does not undergo the same reaction, but
does form a trans-dinitrosyl species at low temperature.26,27 In
our hands, the disproportionation reaction proceeds as previously
reported by Yoshimura and others.9,10
(iii) Zinc Amalgam Reduction.3 Fe(TPP)Cl was dissolved in toluene
and extensively degassed by the freeze-pump-thaw method. The
toluene used was commercial grade and not dried. After addition of
zinc amalgam powder, the solution was frozen in liquid nitrogen and
evacuated. Nitric oxide was introduced via a vacuum adapter. The
solution was then warmed to room temperature and stirred overnight.
The powder was filtered anaerobically, and the resulting solution could
be used directly or evaporated under vacuum and redissolved in dry
solvents if necessary.
15NO-Fe(TPP) was made from Fe(TPP)Cl and sodium nitrite-15N in
the presence of PPh3.30 Fe(TPP)Cl (79.3 mg, 0.11 mmol) and PPh3
(29.2 mg, 0.11 mmol) were dissolved in chloroform (5 mL). Sodium
nitrite-15N (11.4 mg, 0.16 mmol) in methanol (5 mL) was added, and
the solution was stirred overnight. After removal of the solvent, the
red-brown solid was redissolved in chloroform and passed down a
neutral alumina column (Fisher, grade I).
Reaction of NO-Fe(TPP) with NO. In a typical kinetic run, a 3-mL
sample of 0.1 mM NO-Fe(TPP) in toluene or chloroform was placed
in a homemade airtight quartz cuvette with a vacuum stopcock. After
the cell was evacuated, nitric oxide was introduced via a vacuum line,
and the pressure was measured with a mercury manometer. The cell
was pressurized with nitrogen, and the reaction was followed by the
UV-vis spectra.
A series of experiments were run in the presence of PPh3 and
monitored by 31P NMR of the reaction mix in NMR tubes with Schlenk
adapters. The amount of OPPh3 produced was calculated from the
integrals for PPh3 (-3 ppm) and OPPh3 (30 ppm).
In this report we give evidence of an N-N-coupled inter-
mediate, generated from reaction of NO-Fe(TPP) with NO, that
is capable of O atom transfer reactions which are uncoupled
from nitrite formation. This reactivity is demonstrated in
catalytic production of N2O in the presence of suitable O atom
acceptors. Isotopic labeling experiments under catalytic condi-
tions were undertaken to probe the nature of the O atom transfer
reaction, and rapid infrared spectroscopic studies are used to
characterize a possible N-N-coupled intermediate.
Experimental Section
Abbreviations. TPP, 5,10,15,20-Tetraphenyl-21H,23H-porphine; Fe-
(TPP)Cl, 5,10,15,20-tetraphenyl-21H,23H-porphine iron(III) chloride;
NO-Fe(TPP), nitrosyl iron tetraphenyl porphine; PPh3, triphenyl phos-
phine; OPPh3, triphenyl phosphine oxide.
General Procedures. All experiments were conducted under strict
anaerobic conditions by using standard Schlenk techniques. Toluene
was distilled from calcium oxide and degassed by freeze-pump-thaw
methods. Fe(TPP)Cl was purchased from Porphyrin Products. PPh3 and
OPPh3 were obtained from Aldrich. Sodium nitrite-15N was purchased
from Cambridge Isotope Laboratories. Nitric oxide obtained from Air
Gas was used for most experiments, but further purification was
necessary. To trap out NOx impurities, the gas was bubbled through a
1 M NaOH solution; to trap out N2O and H2O contaminants in the NO
reactant gas, the gas was passed through a U-tube immersed in acetone/
dry ice or liquid N2 before being exposed to the reaction mix. Trace
amounts of N2O remained and were measured at the start of each
experiment quantifying N2O production. [15N]Nitric oxide was generated
from L-ascorbic acid and sodium nitrite-15N and purified by the same
procedure. All other chemicals were reagent grade and used without
further purification.
To correlate the formation of N2O with changes in the solution-
based Fe species, the solution-phase UV-vis spectra and the gas-phase
IR spectra of the headgas were measured concurrently as the reaction
proceeded. A special apparatus was used, consisting of a quartz UV-
vis cell connected via a Schlenk adapter to an airtight CaF2 IR cell;
the details of this hybrid cell have been described in the literature.23
The absorbance spectra of solutions of NO-Fe(TPP) were unchanged
after several weeks of storage in this cell. The relative increase in
concentration of N2O in the headgas during a reaction was determined
The electronic spectra were recorded on a Hewlett-Packard 8453
spectrophotometer. Infrared spectra quantifying gaseous N2O production
were measured on a Nicolet 410 Impact spectrophotometer; solid-state
spectra were obtained with a ReactIR (Applied System) surface probe,
as will be described. The head gas samples from isotopic labeling
experiments were separated with a PLOT GC column (fused silica 50
m × 0.32 mm, Al2O3, KCl coating) and analyzed by a Micromass
Autospec mass spectrometer. Data were acquired in EI mode, and
fragmentation was reported as the percentage of N2O (44) peak.
The reaction of NO-Fe(TPP) with NO is inhibited by solvent
contamination (e.g., water, methanol) in the starting material or reactant
gas, and care was taken in the purification and thorough drying of the
complexes. Several methods were used to synthesize the starting NO-
Fe(TPP):
by integrating the IR absorbance from 2120 to 2280 cm-1 31
.
IR Characterization of Reaction Intermediates. An in situ two-
reflection ATR probe was used to obtain sequential IR spectra of solid
NO-Fe(TPP) films exposed to NO gas. A homemade glass adapter with
a rubber O-ring screw top, two gas inlets, and one septum can be fit
on the probe to form a small gastight chamber. The chamber was
evacuated for 10 min before each set of experiments and pressurized
with pure nitrogen gas. A 20-µL solution of NO-Fe(TPP) in toluene or
THF was transferred to the probe surface (diamond or silicon) via a
gastight syringe. A stream of nitrogen was passed over the probe until
the NO-Fe(TPP) solution dried to a film.
(i) Reductive Nitrosylation.28 Fe(TPP)Cl was dissolved in toluene
(20% v/v methanol) in a round-bottom flask and degassed by the
Traces of protic solvents affected the obtained spectra and inhibited
further reactivity of the NO-Fe(TPP). These observations were verified
by use of wet solvents during the deposition. To remove trace water in
solution, after evaporation of a thin Fe(TPP)(NO) film on the IR probe,
(24) (a) Shiro, Y.; Fujii, M.; Iizuka, T.; Adachi, S. I.; Tsukamoto, K.;
Nakahara, K.; Shoun, H. J. Biol. Chem. 1995, 270, 1617. (b) Obayashi, E.;
Takahashi, S.; Shiro, Y. J. Am. Chem. Soc. 1998, 120, 12964-12965.
(25) (a) Lorkovic, I. M.; Miranda, K. M.; Lee, B.; Bernhard, S.;
Schoonover, J. R.; Ford, P. C. J. Am. Chem. Soc. 1998, 120, 11674-11683.
(b) Lorkovic, I. M.; Ford, P. C. Chem. Commun. 1999, 1225-1226. (c)
Lorkovic, I. M.; Ford, P. C. Inorg. Chem. 1999, 38, 1467-1473.
(26) Lorkovic, I. M.; Ford, P. C. Inorg. Chem. 2000, 39, 632-633.
(27) Lorkovic, I. M.; Ford, P. C. J. Am. Chem. Soc. 2000, 122, 6516-
6517.
(28) (a) Spaulding, L. D.; Chang, C. C.; Yu, N.; Felton, R. H. J. Am.
Chem. Soc. 1975, 97, 2517-2525. (b) Wayland, B. B.; Olson, L. W. J.
Am. Chem. Soc. 1974, 96, 6037-6041.
(29) Choi, I. K.; Liu, Y. M.; Wei, Z. C.; Ryan, M. D. Inorg. Chem.
1997, 36, 3113-3118.
(30) Castro, C. E.; O’Shea, S. K. J. Org. Chem. 1995, 60, 1922-1923.
(31) Trogler, W. C. Coord. Chem. ReV. 1999, 187, 303-327.

Fe(TPP)-Catalyzed O Atom Transfer from NO
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1145
Figure 1. Sequential changes in the absorbance spectra of a 0.07 mM solution of NO-Fe(TPP) under 60 mmHg NO: (I) over 2 min, (II) from 2
to 5 min, and (III ) from 5 to 30 min.
pure acetone was cast on the top and re-evaporated with a stream of
nitrogen gas. The drying can be followed by IR spectra, as the NO
stretch frequency shifts from 1677 to 1697 cm-1. After drying, the
system was evacuated, purified nitric oxide was introduced via the
Schlenk line, and the pressure was controlled by a manometer. The
system was pressurized with nitrogen while the IR spectra were
obtained. For mixed-labeling experiments, 15NO gas (ca. 95%) was used,
as described above.
Mass Spectroscopy of Labeled N2O. In a typical labeling experi-
ment, 15NO-Fe(TPP) (30.7 mg, 0.044 mmol) and PPh3 (0.6128 g, 2.3
mmol) were dissolved with 5 mL of chloroform in a 250-mL Schlenk
flask. Natural abundance nitric oxide was introduced via the vacuum
line. The solution was cooled with liquid nitrogen for 10 min and
warmed to room temperature. Afterward, the solution was pressurized
with He and stirred for 3 h. Next, 0.2-mL aliquots of head gas were
analyzed by GC/MS. The absolute abundance of 15NO (m/z 31) was
Figure 2. Changes in the absorbance at 545 nm of a 0.07 mM solution
of NO-Fe(TPP) under 60 mmHg NO, as in Figure 1. Under these
corrected by subtracting the contribution from the natural abundance
of NO (m/z 30), which is about 0.38%. The ratio between 15NO (m/z
conditions, the apparent rate constant of k2 is 2.7 × 10-3 s-1 and k3 is
31) and 14,15N2O (m/z 45) thus obtained was used to determine which
1.0 × 10-4 s-1
.
nitrogen atom was labeled, since only 14N15NO isomer could produce
a
15NO fragment.32
Results and Discussion
Reaction of NO with NO-Fe(TPP) Like many metal-
nitrosyl complexes, NO-Fe(TPP) is not stable in the presence
of excess NO gas, forming nitrite adducts.9,10 Using low NO
and NO-Fe(TPP) concentrations, three distinguishable trans-
formations are seen over time in the absorbance spectra, Figure
1. The sequential spectra show clear isosbestic points, suggesting
a clean conversion between the four species: the initial complex,
NO-Fe(TPP) (A), and species B, C, and D. The spectra of C
and D match those of the known compounds, trans-Fe(TPP)-
(NO2)(NO) and Fe(TPP)(NO2), respectively.9,10,33-34 Figure 2
shows the time course of the reaction at 545 nm which illustrates
the rates of the three transformations.
The transformation of C to D is reversible and is demonstrated
in Figure 3. At higher NO concentrations (>100 mmHg),
compound A is transformed directly to C, without buildup of
intermediate B (Supporting Information). Compound C thus
formed may be converted to D upon evacuating NO from the
headgas. Likewise, solutions of D will convert to C upon
exposure to NO gas. Under most conditions used in this work,
Figure 3. Formation of D from C upon evacuation of NO headgas.
Initial spectra of C were generated from a solution 0.15 mM NO-Fe-
(TPP) exposed to 10 psi of NO for 30 min. Absorbance spectra were
taken at 30 s intervals.
the endpoint of the reaction is a mixture of products C and D
depending on the [NO]-to-[NO-Fe(TPP)] ratio.
The initial transformation of A to B is reversible as well, as
illustrated in Figure 4; the spectral changes initiated by exposure
of NO-Fe(TPP) to NO can be reversed by evacuating the NO
from the reaction vessel within the first few minutes. The species
B, which precedes the formation of the nitrite adducts, is likely
the precursor to the formation of N2O. To our knowledge, this
is the first clear identification of an intermediate that precedes
(32) Bonner, F. T.; Kada, J.; Phelan, K. G. Inorg. Chem. 1983, 22, 1389-
1391.
(33) Finnegan, M. G.; Lappin, A. G.; Scheidt, W. R. Inorg. Chem. 1990,
29, 181-185.
(34) Fernandes, J.; Feng, D.; Chang, A.; Keyser, A.; Ryan, M. D. Inorg.
Chem. 1986, 25, 2606-2610.

1146 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Lin and Farmer
Table 1. Energies for Electronic Absorbance Maxima in the
Q-Band Region of Fe(TPP) Ligand Adductsa
species
Q-band maxima (nm)
(NO)Fe(TPP), A
537, 610
511, 575(sh), 653(br), 685(br)
545, 577(sh)
(N2O2)Fe(TPP), B
NO(NO2)Fe(TPP),b C
(NO2)Fe(TPP),c D
508, 570(sh), 687(br)
a Spectra were obtained in toluene, under conditions of turnover as
in Figure 1. (sh) shoulder, (br) broad. b Reference 9. c Reference 33.
Figure 4. Formation of A from B upon evacuation of NO headgas.
Initial spectra of B were generated from a solution 0.05 mM NO-Fe-
(TPP) under 60 mmHg of NO for 2 min. Absorbance spectra were
taken at 30 s intervals.
nitrite formation. A similar intermediate was previously seen
to be formed from reaction of A with low pressures of NO but
was reported to be in equilibrium with C, which at that time
was thought to be a trans-dinitrosyl.35 Physical characterization
of this species identifies it as the mononitrite D, which has
absorbance spectra very similar to those of B, Table 1.
Figure 5. (A) 31P NMR spectrum of 0.01 M PPh3 in chloroform after
exposure to 2 psi NO gas for 3 h. (B) Same conditions with 1 mM
NO-Fe(TPP).
The short-lived species B irreversibly converts to C within
minutes. Thus, species C must be the initial nitrite adduct
formed, as it is unstable with regard to loss of NO under the
conditions in which it is formed. From these observations, a
tentative mechanism can be proposed for these transformations.
The starting nitrosyl A reacts with free NO to form an N-N-
coupled intermediate, corresponding to B, eq 3. The N2O2-Fe
adduct then transfers an O atom to free NO in solution, releasing
the gases N2O and NO2, eq 4. The unligated Fe(TPP) thus
formed may then react with NO in solution to re-form the
starting nitrosyl complex A, eq 5, while the NO2 released can
react with A directly to give the initially formed nitrite complex
C, eq 6. Under these conditions, loss of NO from C gives the
mononitrite adduct D, eq 7. In support of this mechanism, both
N2O and NO2 gases were recently detected during metal-assisted
NO disproportionation catalyzed by a binuclear Fe complex.1
Scheme 1
We tested this hypothesis by using PPh3 as a competitive O
atom acceptor, eq 8. Addition of a 100-fold excess of PPh3 to
a 0.1 mM solution of NO-Fe(TPP) under 2 psi NO for 3 h
resulted in the formation of over 60% of the PPh3 to OPPh3 by
31P NMR, Figure 5. The headgas was analyzed by GC/MS, and
a strong signal was seen at m/z 44, corresponding to N2O. Under
identical conditions in the absence of NO-Fe(TPP), less than
1% OPPh3 was produced, as compared with experiments done
in the presence of the metal complex.
NO-Fe(TPP) + NO a N2O2-Fe(TPP)
(3)
N2O2-Fe(TPP) + NO f Fe(TPP) + N2O + NO2 (4)
PPh3 NO-Fe(TPP)8 OdPPh3
(8)
excess NO
Fe(TPP) + NO f NO-Fe(TPP)
(5)
(6)
(7)
To correlate the transformations of the Fe complexes with
the formation of N2O, we utilized a cell designed for concurrent
determination of UV/vis spectra and IR analysis of the headgas
above the reaction.23 Integration of the absorbance peaks of N2O
in the IR between 2260 and 2160 cm-1 was used to quantify
its increase in the headgas and plotted on the same time course
as the changes in Soret absorbances of NO-Fe(TPP) (407 nm)
and Fe(TPP)(NO2)(NO) (433 nm), Figure 6. The results show
that NO-Fe(TPP) is the major species in solution during catalytic
N2O formation and that the catalysis precedes the formation of
the nitrite complexes C and D. In fact, the addition of PPh3 to
the reaction mixture simply delays the formation of C; the lag
time before the conversion is dependent on the relative
concentrations of PPh3 and NO. For example, 10-fold excess
of PPh3 under 40 mmHg delays the conversion by ca. 3 h, and
a 100-fold excess results in a lag time of over 16 h before a
NO-Fe(TPP) + NO2 f (NO)(NO2)Fe(TPP)
(NO)(NO2)Fe(TPP) a (NO2)Fe(TPP) + NO
Catalytic O Atom Transfers. In the proposed mechanism,
free NO acts as an O atom acceptor, forming NO2 which
subsequently reacts with the starting NO-Fe(TPP). If another
O atom acceptor were available, the reactions described by eqs
6 and 7 would be suppressed, and the formation of N2O would
become catalytic, Scheme 1. The catalytic efficiency would
depend on the relative rates of reaction of B with free NO and
with the O atom acceptor.
(35) (a) Olson, L. W.; Schaeper, D.; Lancon, S. D.; Kadish, K. M. J.
Am. Chem. Soc. 1982, 104, 2042-2044. (b) Lancon, D.; Kadish, K. M. J.
Am. Chem. Soc. 1983, 105, 5610-5617.

Fe(TPP)-Catalyzed O Atom Transfer from NO
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1147
Figure 8. ATR-IR spectra of conversion of the ν(ΝÃ) band of an
NO-Fe(TPP) film from 1677 (*) to 1697 cm-1(9), (a) as first observed
from wet solvent and (b) and (c) taken as the film was dried under an
N2 stream over 1 h.
PPh3; under the conditions given, the reaction of C is negligible,
and the reaction of D given by eq 9 is ca. 2 orders of magnitude
slower than those described by eqs 3-8; sample data are given
in the Supporting Information.
(NO2)Fe(TPP) + PPh3 f (NO)Fe(TPP) + OPPh3 (9)
Figure 6. Concurrent measurement of N2O in the headgas and
absorbance changes in a solution of 0.05 mM NO-Fe(TPP) and 1 mM
PPh3 during reaction under 40 mmHg NO. Upper graph, the IR
absorbance of N2O in headgas (area); lower graph, absorbance changes
at 407 nm (solid line) corresponding to the Soret band of A, NO-Fe-
(TPP), and at 433 nm (dotted line) corresponding to the Soret band of
C, (NO2)NO-Fe(TPP).
Protic Inhibition. The reaction of A with NO is inhibited
by protic solvents such as water or methanol. Under the same
conditions, there is no reaction of A in wet chloroform with
exogenous NO, but the addition of trace O2 rapidly produces
spectral changes characteristic of mixtures of ferric porphyrins.
interaction of water and A is evidenced by subtle changes in
the electronic absorbance spectra; the band at 536 nm sharpens
and shifts to 539 nm upon addition of approximately stoichio-
metric H2O to a dry chloroform solution (shown in Supporting
Information). The effect is more apparent in the IR spectra of
a film of NO-Fe(TPP), shown in Figure 8, where the ν(ΝÃ)
stretch increases from 1677 to 1697 cm-1 during exhaustive
evacuation. We, and others, have also observed the protic
contaminant shifting of ν(ΝÃ) of NO-Fe(TPP) in chloroform
solution and in KBr pellets, where mild heating under a vacuum
returns the peak frequency to the “dry” value.36
The source of the spectral shift is most likely hydrogen-
bonding between the nitrosyl and the protic impurity. A recent
crystallographic determination of nitrosyl myoglobin shows a
strong hydrogen-bonding interaction between the nitrosyl N
atom and the distal histidine.37 We have recently shown that
the one-electron-reduced form of nitrosyl myoglobin is proto-
nated at the nitrogen, forming an unusual HNO adduct.38 Much
evidence suggests that the reactivity of a metal nitrosyl with
NO, eq 1, depends on the anionic character of the coordinated
nitrosyl ligand. For example, the NO disproportionation reactiv-
ity of an Fe(topocorand) complex is shut down by nitration of
the ligand, which significantly decreases the electron density
at the Fe center.11
Figure 7. In a continuation of the experiment described in Figure 6
after 16 h, the absorbance at 407 nm (solid line) corresponding to the
Soret band of A, NO-Fe(TPP), and at 433 nm (dotted line) correspond-
ing to the Soret band of C, (NO2)NO-Fe(TPP).
comparatively rapid formation of the nitrite adducts occurs,
Figure 7. Interpreted in terms of our proposed mechanism, the
competitive O atom acceptor PPh3 decouples N2O formation
from that of the nitrite adducts.
Nitrite adducts of FeIII porphyrinates are well known to be
unstable to nitrite disproportionation, generating ferrous nitrosyls
and apparently nitrate in a stoichiometric reaction.33 Likewise,
it is known that D can act as an O atom donor to PPh3 under
certain conditions, eq 9, again in a stoichiometric reaction.30
The nitrite adducts, C and D, were tested for their reaction with
Mass Spectroscopy of Mixed-Labeled N2O. No buildup of
B occurs under catalytic conditions; thus, the following irrevers-
ible reaction must be fast enough to limit its concentration. This
is consistent with the supposition that species B is itself the O
(36) Scheidt, W. R.; Wyllie, G. R. A. Personal communication.
(37) Bruker, E. A.; Olson, J. S.; Ikeda-Saito, M.; Phillips, G. N., Jr.
Proteins: Struct., Funct. Genet. 1998, 30, 352-356.
(38) Lin, R.; Farmer, P. J. J. Am. Chem. Soc. 2000, 122, 2393-2394.

1148 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Lin and Farmer
Scheme 2
Figure 10. Sequential ATR-IR spectra of a film of NO-Fe(TPP) film
taken before exposure and 5, 30, and 60 s after exposure to a stream
of purified NO. Absorbances marked are due to the ν(ΝÃ) of A,
NO-Fe(TPP) (b); intermediate B, assigned as (NO)2-Fe(TPP) (*); and
C, (NO2)NO-Fe(TPP) (9). Experimental details are as given in the
text.
Figure 9. Mass spectrum of N2O fraction of headgas above the reaction
of 15NO-Fe(TPP) with natural abundance NO. Peaks at 45 (14,15N2O)
and 31 (15NO) were used to calculate the abundance of 14N15NO.
atom donor preceding irreversible N2O formation. To look more
closely at this O atom transfer mechanism, labeled 15NO-Fe-
(TPP) (ca. 98%) was reacted with excess natural abundance
14NO under catalytic conditions, and the N2O produced was
characterized by GC/MS, Scheme 2.
by eq 8, or the two nitrosyls are indistinguishable in the
precursor complex. Ready exchange of exogenous NO with
metal nitrosyls has been previously seen for non-porphyrin Mn
and Cu nitrosyl adducts15,17 and is apparent in the ATR-IR
experiments described below.
Solid-State IR Characterization. Proof of the formation and
structure of an N-N-coupled precursor may be obtained from
the IR spectra, from the number and stretching frequencies of
bands assignable to N-O bonds of a transient species. Several
examples of possible N-N-bound species, such as the neutral
(NO)2 dimer and the radical anion N2O2-, have recently been
characterized by IR in matrix isolation.39,40
When ionized in a mass spectrometer, N2O fragments to yield
+
both N2 and NO+ fragments in addition to the mother ion
N2O+. The ratio of the corresponding ions can be used to
identify the position of 15N label in mixed-label N2O.32 For
+
example, if the terminal N in N2O is labeled, then only the N2
fragment will have the label, as depicted in eq 10.
*NNO ionization8 *NNO+ + *NN+ + NO+
(10)
In our hands, the progression of the reaction was difficult to
observe by using standard solution IR cells because of the low
solubility of NO-Fe(TPP) and masking absorbances of solvents
in the region of interest. As an alternative, a solution of NO-
Fe(TPP) was evaporated onto the surface of a reflectance IR
probe, and the reaction was initiated by exposing the dry film
to purified NO gas. The reaction between the dry NO-Fe(TPP)
film and gaseous NO occurs rapidly, and conditions were
manipulated in order to obtain multiple sequential spectra within
the first minute of exposure.
To determine which O atom is lost in the formation of
N2O, the ratio of 15NO+ (m/z 31) to 14,15N2O (m/z 45) from the
mixed label experiments were compared to the ratio of NO+
(m/z 30) to N2O (m/z 44) for natural abundance N2O. Be-
cause of the ongoing catalytic turnover, only the label-containing
peaks at m/z 31 and 45 would correspond to the N2O formed
during an initial turnover from the labeled 15NO-Fe(TPP)
precursor.
Three limiting scenarios are possible: (i) equal ratios if the
O atom lost is from 14NO; (ii) the ratio of 31/45 should approach
zero if the O atom lost is from 15NO; and (iii) if both O atoms
are lost equally, or if the label is exchanged with the exogenous
NO, the ratio of 31/45 from mixed label N2O will be about
one-half of that of 30/44 from natural abundance N2O.
The results of a series of experiments are most consistent
with the last scenario; i.e., that about one-half of the NO+
fragments from the mixed label N2O intermediate are labeled.
A typical run is given in Figure 9 (further data are given in the
Supporting Information). Likewise, the height of the small peak
at m/z 46, corresponding to doubly labeled 15N2O, is close to
that statistically expected for complete label exchange with the
NO headgas.
A typical series of spectra of a NO-Fe(TPP) film before and
during the reaction with NO is shown in Figure 10. Upon
exposure to NO, the band at 1697 cm-1 assigned to the ν(ΝÃ)
of NO-Fe(TPP) decreases rapidly, and several new isotope-
sensitive bands grow in. Two of these bands, at 1873 and 1299
cm-1, are identical to those of independently synthesized and
characterized Fe(TPP)(NO2)(NO).9,10,32 Several other isotope-
sensitive, transient bands were seen in various experiments. But
of these, only the two high-energy bands (1898 and 1858 cm-1
)
were consistently formed in the first seconds of reaction during
the loss of A and subsequently decreased during the formation
of C, Figure 11.41
The use of 15NO gas during the reaction yields corresponding
peaks ca. 30 cm-1 lower in frequency, a shift somewhat smaller
These experiments show that both the initial Fe-bound NO
and exogenous NO are equally active as O atom donors in these
experiments. Either the exchange reaction of eq 3 predominates,
even under conditions that maximize the rate of O atom transfer
(39) Krim, L.; Lacome, N. J. Phys. Chem. A 1998, 102, 2289.
(40) Andrews, L.; Zhou, M. J. Chem. Phys. 1999, 111, 6036.

Fe(TPP)-Catalyzed O Atom Transfer from NO
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1149
Figure 12. ATR-IR difference spectra of films of 14NO-Fe(TPP) after
ca. 10 s exposure to (a) natural abundance NO (>99% 14NO), (b)
14NO/15NO (ca. 50/50), and (c) 15NO (ca. 95%). To emphasize the
transient peaks, a normalized product spectrum obtained after ca. 2
min has been subtracted. The band at 1820 cm-1 is present in initial
spectra.
Figure 11. Sequential ATR-IR spectra of the NO region of a film of
NO-Fe(TPP) film obtained within the first minute of exposure to a
stream of purified NO.
Table 2. Energies, in cm-1, for ν(NO) bands of Fe(TPP) Ligand
Adducts
Scheme 3
species
(NO)Fe(TPP), Aa
(NO)2Fe(TPP), Ba
14NO
1697
14NO/15NO
15NO
1669
1898, 1856
1888, 1846
plus others
1867, 1823
NO(NO2)Fe(TPP),a C
1872, 1299
1863, 1776
1836, 1280
1831, 1745
spin ferric porphyrin, with distinctive broad bands in the region
between 600 and 700 nm. We originally postulated an oxidation
of the metal center after N-N coupling, as theoretical calcula-
tions suggest that the (NO)2 dimer has an electron affinity on
the order of 30 kcal/mol.44 But the ν(ΝÃ) peak frequencies of
the intermediate are in the region of neutral or cationic NO metal
adducts, inconsistent with reduction of the ligand.
cis-(NO)2 in Arb
1850, 1757
a Spectra were obtained by heterogeneous reaction as described in
the text. b Reference 39.
than that observed in the ultimate product of the reaction, C,
Table 2.42 The exchange of the 14NO and 15NO in the starting
complex A is also apparent in these experiments, though the
rate of exchange is not noticeably faster than the production of
C. Assignment of the transient species to a dinitrosyl species is
supported by bands at intermediate energies in mixed 14NO/
15NO experiments, as compared to those of 14NO/14NO and
15NO/15NO species, Figure 12.
The presence of unique absorbances for the mixed-label
species is indicative of coupling between two nitrosyls, though
not conclusive proof of N-N coupling. The bands are broad
and overlap those of the product. Also, self-ordering of the
Fe(TPP) on the surface of the probe may affect the intensities
and positions of observed IR modes by orienting the bound
nitrosyls relative to the reflectance, as has been seen for (NO)2
dimers adsorbed on Ag(1,1,1).43 Further spectroscopic experi-
ments are underway to better characterize the structure of this
precursor.
The ν(NO) bands observed in the ATR-IR spectra of the
intermediate are distinct from those previously suggested to be
of trans-dinitrosyl adducts of Fe(II) porphyrins.27,28b We for-
mulate the intermediate as an upright, asymmetrically bound
cis-(NO)2 adduct, Scheme 3. The formation of the N-N bond
removes two ligand-based electrons from N-O antibonding
orbitals, resulting in the observed upfield shift in ν(ΝÃ). The
bands are at higher frequency than comparable absorbances in
the low-temperature spectra of cis-(NO)2, perhaps indicative of
σ-bonding interactions with the metal. Such an N2O2 adduct
may be susceptible to a fluxional shift between coordination
by the two N atoms, Scheme 3, or may undergo NO exchange
by eq 3. Both pathways are in accordance with equivalence of
N atom positions in the mass spectroscopic experiments and
the exchange evident over time in the ATR-IR spectra.
It is interesting to note the parallel between solution and
surface chemistry of NO, as similar intermediates have been
postulated as forming on metal surfaces. Weakly adsorbed (NO)2
dimers have been characterized at low temperatures on noble
metals (Ag, Cu) and metal oxide (Mo, Cr) surfaces under
conditions which result in the reductive formation of N2O from
NO.43,45,46 In all cases, the apparent role of metal coordination
is to provide a nucleation site for the catenation of NO.
The Nature of the N-N-Coupled Intermediate. The
absorbance spectrum of intermediate B resembles that of a low-
(41) Free NO, with a ν(ΝÃ) ca. 1850 cm-1, should not be seen in the
ATR experiment unless it is absorbed onto the surface. Several control
experiments under identical NO flow conditions but in the absence of NO-
Fe(TPP) show no equivalent peaks ca. 1898 and 1856 cm-1. For instance,
an Fe(TPP)Cl film was reacted under these conditions, and only a strong
ν(ΝÃ) of the known NO-Fe(TPP)Cl was observed at 1882 cm-1 (Supporting
Information). In the course of this work, we ran over a dozen experiments
with NO-Fe(TPP) films that did show these transient peaks under an NO
stream.
(43) (a) Brown, W. A.; Gardner, P.; Jigato, M. P.; King, D. A. J. Chem.
Phys. 1995, 102, 7277-7280. (b) Brown, W. A.; Gardner, P.; Jigato, M.
P.; King, D. A. J. Chem. Phys. 1995, 102, 7277-7280.
(44) (a) Snis, A.; Panas, I. Surf. Sci. 1998, 412, 477-488. (b) Snis, A.;
Panas, I.; Chem. Phys. 1997, 221, 1. (c) Snis, A.; Panas, I. J. Chem. Phys.
1999, 110, 10345.
(42) The peak assignments in Table 2 were determined by difference
spectra of the normalized product and intermediate spectras (Supporting
Information).

1150 J. Am. Chem. Soc., Vol. 123, No. 6, 2001
Lin and Farmer
Conclusions
bound nitrosyls in the transient species, which may be indicative
of N-N bond formation.
We have identified an intermediate in the reaction of NO-
Fe(TPP) with low concentrations of NO gas, which precedes
formation of the nitrite adducts of the Fe complex. We suggest
the intermediate to be due to N-N coupling between NO and
the metal-bound nitrosyl. The subsequent formation of NO-
(NO2)Fe(TPP) under such conditions that it readily loses NO
suggests that it is formed by addition of free NO2 to the starting
nitrosyl and that free NO acts as the O atom acceptor during
the formation of N2O. The addition of the competitive O atom
acceptor, PPh3, leads to catalytic production of N2O, which was
observed and quantified. In agreement with the proposed
mechanism, the formation of N2O is decoupled from the
formation of the nitrite adducts under these catalytic conditions.
The mechanism of O atom transfer was examined by cross-
labeling experiments, which show that both O atoms in the
intermediate are equivalent, even under catalytic conditions. The
formation of an intermediate was confirmed by IR spectroscopy
of the heterogeneous reaction of an NO-Fe(TPP) film with
gaseous NO. The disproportionation reaction is rapid under these
conditions, but transient isotope-sensitive NO bands are seen,
assignable to a precursor to the O atom transfer. Mixed 14N/
15N label experiments demonstrate coupling between the two
Acknowledgment. We thank Dr. John Greaves for valued
discussions and mass spectroscopic analysis, Professor Frank
Feher for use of a Fischer Porter tube, and Professors Larry
Overman and Bill Evans for use of their ReactIR spectrometers.
Special thanks are extended to Professor Bill Trogler for the
use of a hybrid solution UV-vis/gas-phase IR cell, used exten-
sively in this work. This research was supported by the National
Science Foundation (CHE-9702332) and the Petroleum Research
Fund, administered by the American Chemical Society (PRF-
31804-G3).
Supporting Information Available: Absorbance spectra of
the reaction of NO-Fe(TPP) under high [NO], the reactions of
(NO)(NO2)Fe(TPP) and (NO2)Fe(TPP) with PPh3 as followed
by absorbance, a UV-vis spectrum of wet and dry NO-Fe-
(TPP), a table of the ion abundances obtained from mixed
labeled N2O with explanation of the data analysis, ATR-IR
spectra of the reaction of NO(g) with an Fe(TPP)Cl film, and
difference ATR-IR spectra obtained from experiments run
under 14NO, 15NO, and 14NO/ 15NO mixtures (PDF). This
material is available free of charge via the Internet at http://
pubs.acs.org.
(45) (a) Kugler, E. L.; Kokes, R.; Gryder, J. W. J. Catal. 1975, 35, 142-
151. (b) Kugler, E. L. Gryder, J. W. J. Catal. 1975, 36, 152-158.
(46) (a) Queeney, K. T.; Pang, S.; Friend, C. M. J. Chem. Phys. 1998,
109, 8058-8061. (b) Queeney, K. T.; Friend, C. M. J. Phys. Chem. B 1998,
102, 9251-9257.
JA993643R