NO disproportionation reactivity of Fe tropocoronand complexes

Article abstract of DOI:10.1021/ja991664f

The synthesis and characterization of divalent [Fe(TC-5,5)] (1) and trivalent [Fe(OTf)(TC-5,5)] (2) tropocoronand complexes are described. Compound 1 reacts with 1 equiv of NO to form the {FeNO}⁷ complex 3. A single-crystal X-ray structure determination of 3 reveals a trigonal bipyramidal geometry with a linearly coordinated nitrosyl (Fe-N-O = 174.3(4)°) having a short Fe-N distance of 1.670(4) A?. EPR and Mossbauer spectroscopy, SQUID susceptometry, and normal coordinate analysis indicate 3 to be a low-spin {Fe(III)(NO^-)}²⁺ species. In the presence of excess NO, 3 converts to a metastable nitrosyl-nitrito complex that decomposes by losing NO₂, which subsequently nitrates the aromatic tropolone rings of the ligand. The final products of the NO disproportionation reaction are N₂O and [Fe(NO)(TC-5,5-NO₂)] (4). The v(NO) stretching band of 4 is increased to 1716 cm^-1 from its value of 1692 cm^-1 in 3, owing to the electron- withdrawing nitro groups on the ligand, and the compound no longer promotes the disproportionation of NO. Mechanistic aspects of the reaction are discussed.

Full text of DOI:10.1021/ja991664f

10504
J. Am. Chem. Soc. 1999, 121, 10504-10512
NO Disproportionation Reactivity of Fe Tropocoronand Complexes
Katherine J. Franz and Stephen J. Lippard*
Contribution from the Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
ReceiVed May 19, 1999
Abstract: The synthesis and characterization of divalent [Fe(TC-5,5)] (1) and trivalent [Fe(OTf)(TC-5,5)] (2)
tropocoronand complexes are described. Compound 1 reacts with 1 equiv of NO to form the {FeNO}7 complex
3. A single-crystal X-ray structure determination of 3 reveals a trigonal bipyramidal geometry with a linearly
coordinated nitrosyl (Fe-N-O ) 174.3(4)°) having a short Fe-N distance of 1.670(4) Å. EPR and Mo¨ssbauer
spectroscopy, SQUID susceptometry, and normal coordinate analysis indicate 3 to be a low-spin {FeIII(NO-)}2+
species. In the presence of excess NO, 3 converts to a metastable nitrosyl-nitrito complex that decomposes
by losing NO2, which subsequently nitrates the aromatic tropolone rings of the ligand. The final products of
the NO disproportionation reaction are N2O and [Fe(NO)(TC-5,5-NO2)] (4). The νNO stretching band of 4 is
increased to 1716 cm-1 from its value of 1692 cm-1 in 3, owing to the electron-withdrawing nitro groups on
the ligand, and the compound no longer promotes the disproportionation of NO. Mechanistic aspects of the
reaction are discussed.
Introduction
The conversion of NO into species such as NO2, N2O3,
OONO-, and N2O, has important biological1 and environmental2
consequences. Some of these transformations occur by direct
oxidation, but transition metal ions are often required to activate
nitric oxide. The disproportionation of NO is effected by a
variety of metal complexes,3,4 including those of iron,5 ruthe-
nium,6,7 cobalt8,9 and copper.10-12
Figure 1. H2TC-5,5 ligand.
To explore the generality of this chemistry, we studied the
reaction of NO with the corresponding [Fe(TC-5,5)] complex.
As described here, a discrete iron nitrosyl complex forms which,
like its Mn analogue, also promotes NO disproportionation but
with some interesting and unexpected differences. The iron
complexes also provide the additional opportunity to apply
Mo¨ssbauer spectroscopy for characterizing formal oxidation
levels. Mechanistic details of the disproportionation reaction are
discussed that delineate the reaction pathways of these two
different metal complexes. The results provide a useful bench-
mark for evaluating the possible roles of Mn and Fe complexes
in processing NO in naturally occurring systems.
We recently reported novel NO disproportionation chemistry
promoted by a manganese tropocoronand (Figure 1) complex
[Mn(THF)(TC-5,5)].13 When the stoichiometry of NO addition
was carefully controlled, the nitrosyl complex [Mn(NO)(TC-
5,5)] could be isolated. This complex further reacts with NO to
yield N2O and the nitrito derivative [Mn(NO2)(TC-5,5)]. The
strongly donating nature of the tropocoronand ligand, coupled
with its geometric constraints that convey trigonal bipyramidal
geometry at the metal center,14 provide the framework for this
novel manganese nitrosyl reactivity, in which the coordinated
NO ligand is reductively activated.
(1) Marnett, L. J. Chem. Res. Toxicol. 1996, 9, 807-808.
(2) Trogler, W. C. J. Chem. Educ. 1995, 72, 973-976.
(3) Bottomley, F. Reactions of Coordinated Ligands; Plenum: New York,
1989; Vol. 2, pp 115-222.
Experimental Section
General Information. All reactions were carried out under nitrogen
in a glovebox or by using standard Schlenk techniques. Pentane,
tetrahydrofuran (THF), and 2-methyltetrahydrofuran (Me-THF) were
freshly distilled from sodium benzophenone ketyl under nitrogen.
Dichloromethane and dichloroethane were distilled from CaH2 under
nitrogen. The ligand H2(TC-5,5)15 and [Fe(OTf)2(CH3CN)2]16 were
synthesized as described in the literature. Nitric oxide (Matheson, 99%)
and 15NO (Aldrich, 99%) were purified of higher nitrogen oxides by
passage through a column of NaOH pellets and a mercury bubbler and
kept over mercury in gas storage bulbs. Analysis by GC of the NO
used in the experiments revealed no contaminants, such as NO2 or N2O,
at the limit of the thermal conductivity detector, about 30 nM. All other
reagents were obtained commercially and not further purified. UV-
(4) Richter-Addo, G. B.; Legzdins, P. Metal Nitrosyls; Oxford Univer-
sity: New York, 1992.
(5) Yoshimura, T. Inorg. Chim. Acta 1984, 83, 17-21.
(6) Lorkovic´, I. M.; Ford, P. C. Inorg. Chem. 1999, 38, 1467-1473.
(7) Miranda, K. M.; Bu, X.; Lorkovic´, I.; Ford, P. C. Inorg. Chem. 1997,
36, 4383-4848.
(8) Gans, P. J. Chem. Soc. A 1967, 943-946.
(9) Gwost, D.; Caulton, K. G. Inorg. Chem. 1974, 13, 414-417.
(10) Ruggiero, C. E.; Carrier, S. M.; Tolman, W. B. Angew. Chem., Int.
Ed. Engl. 1994, 33, 895-897.
(11) Paul, P. P.; Karlin, K. D. J. Am. Chem. Soc. 1991, 113, 6331-
6332.
(12) Schneider, J. L.; Carrier, S. M.; Ruggiero, C. E.; Young, V. G., Jr.;
Tolman, W. B. J. Am. Chem. Soc. 1998, 120, 11408-11418.
(13) Franz, K. J.; Lippard, S. J. J. Am. Chem. Soc. 1998, 120, 9034-
9040.
(14) Jaynes, B. S.; Ren, T.; Masschelein, A.; Lippard, S. J. J. Am. Chem.
Soc. 1993, 115, 5589-5599.
(15) Davis, W. M.; Roberts, M. M.; Zask, A.; Nakanishi, K.; Nozoe, T.;
Lippard, S. J. J. Am. Chem. Soc. 1985, 107, 3864-3870.
(16) Bryan, P. S.; Dabrowiak, J. C. Inorg. Chem. 1975, 14, 296-299.
10.1021/ja991664f CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/21/1999

ReactiVity of NO in Fe Tropocoronand Complexes
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10505
visible spectra were recorded on a Cary 1 E spectrophotometer. Mass
spectra were determined in 3-nitrobenzyl alcohol matrix on a Finnegan
4000 mass spectrometer with 70-ev impact ionization.
Magnetic Susceptibility. Variable temperature magnetic suscepti-
bilities were measured with a Quantum Design MPMS SQUID
susceptometer. The samples were loaded in a drybox in gel capsules
and suspended in plastic straws. Field dependence studies at 5 and 150
K showed a linear correlation of magnetization vs field strength from
0 to 10 000 G for all compounds. The data were collected at 5000 G
in the temperature range from 5 to 300 K. The infrared spectrum of 2
following the measurement was unchanged, confirming that the sample
had not oxidized during the procedures required to obtain the data.
The susceptibilities of the straw and gel capsule were measured at the
same field and temperatures for accurate correction of their contribution
to the total measured susceptibility. The diamagnetic contributions from
the ligands were calculated from Pascal’s constants17 and applied. The
effective magnetic moments (µeff) were calculated from the expression,
Synthetic Procedures. [Fe(TC-5,5)] (1). A portion of H2TC-5,5 (502
mg, 1.33 mmol) was doubly deprotonated in 10 mL of THF with n-BuLi
(1.75 mL, 2.80 mmol of 1.6 M solution in hexanes) to give an orange
solution. The solution immediately turned burgundy red upon addition
of [Fe(OTf)2(CH3CN)2] (638 mg, 1.46 mmol). After the solution stirred
for 2.5 h, THF was removed in vacuo, and the resulting powder was
twice triturated with pentane and then extracted into CH2Cl2 and filtered
through Celite. The solvent was again removed in vacuo, and the
product was triturated twice with 3 mL of pentane. Dark burgundy
prisms of X-ray quality crystals were grown by pentane diffusion into
a THF solution of the complex. Yield: 340 mg, 60%. FTIR (KBr):
1586 (s), 1503 (s), 1466, 1451, 1407 (s), 1387, 1363, 1330, 1265 (s),
1228 (s), 1193, 1130, 1048 (s), 990, 936, 885, 721 (s), 484 cm-1. UV-
vis (THF) nm (M-1 cm-1): 277 (48000), 388 (37000), 425 (21000).
µeff ) 5.19 µB (300 K). Mo¨ssbauer δ ) 0.66(1) mm/s, ∆EQ ) 1.18(1)
mm/s. Anal. Calcd for FeC24H30N4: C, 66.98; H, 7.03; N, 13.02.
Found: C, 66.74; H, 7.11; N, 12.79.
[Fe(OTf)(TC-5,5)] (2). An equivalent of AgOTf (93 mg, 0.361
mmol) was added to a 5-mL THF solution of 1 (154 mg, 0.358 mmol);
the color changed from red-orange to darker red. Solid Ag metal was
removed by filtration after 0.5 h, and the solvent was evaporated. X-ray
quality crystals were grown from pentane diffusion into a THF solution
of the complex. Yield: 161 mg, 78%. FTIR (KBr): 1584, 1501 (s),
1433, 1410, 1384, 1339, 1314, 1264, 1225, 1200, 1166, 1044, 1023
(s), 885, 731, 632 (s) cm-1. UV-vis (THF) nm (M-1 cm-1) 274
(50000), 354 (19400), 388 (19700), 467 (13400), 500 (14200), 805
(7500). µeff ) 5.95 µB (300 K). EPR: g′ ) 8.69, 5.27, 4.25. Mo¨ssbauer
µeff ) 2.828(øT)1/2
.
Electrochemistry. Cyclic voltammetry was performed in a nitrogen-
filled glovebox with an EG&G model 263 potentiostat. A standard
three-electrode setup was used, consisting of a Ag/AgNO3 reference
electrode (0.01 M in acetonitrile), a platinum wire auxiliary electrode,
and a 1.75 mm2 platinum disk working electrode. The solute samples
were generally 2 mM in freshly distilled CH2Cl2 and 0.2 M in
supporting electrolyte (Bu4N)(PF6), triply recrystallized from acetone.
All spectra were externally referenced to ferrocene under identical
conditions, for which E1/2 ) 180 mV vs Ag/AgNO3 in CH2Cl2.
57Fe Mo1ssbauer Spectroscopy. Samples for 57Fe Mo¨ssbauer spec-
troscopy were finely ground, intimately mixed in Apiezon-N grease,
and then kept in a cryostat at 77 K during measurement. A 57Co/Rh
source was moved at constant acceleration at room temperature against
the absorber samples. All isomer shift (δ) and quadrupole splitting
(∆EQ) values are reported with respect to 57Fe-enriched metallic iron
foil that was used for velocity calibration. The displayed spectra were
folded to enhance the signal-to-noise ratio. Best fits for the data were
calculated by using the MBF Mo¨ssbauer plot and fit program.18
EPR Spectroscopy. EPR spectra were recorded as frozen Me-THF
glasses on a Bruker model 300 ESP X-band spectrometer operating at
9.47 GHz and running WinEPR software. Low temperatures were
maintained with a liquid He Oxford Instruments EPR 900 cryostat.
The following temperatures and powers were used for particular
compounds: 2, 26 K, 10 mW; 3, 4 K, 0.1 mW; 4, 10 K, 1.0 mW.
IR Spectroscopy. Standard IR spectra were recorded on a Bio Rad
FTS-135 instrument with Win-IR software. Solid samples were pressed
into KBr pellets; solution samples were prepared in an airtight Graseby-
Specac solution cell with CaF2 or KBr windows. In situ IR sample
monitoring was performed with a ReactIR 1000 from ASI Applied
Systems equipped with a 1-in. diameter, 30-reflection silicon ATR
(SiComp) probe optimized for maximum sensitivity. In a typical
reaction, the IR probe was inserted through a nylon adapter and O-ring
seal into an oven-dried, custom-made cylindrical flask equipped with
a gas-inlet adapter, a septum-sealed joint and a stir bar. The reaction
vessel was deaerated by three vacuum-fill cycles with Ar. Under Ar,
the flask was charged with 5 mL of the appropriate solvent, usually
THF, and background spectra were recorded at room temperature and/
or at -78 °C by placing the flask over a dry ice/acetone slush bath.
The solvent was removed in vacuo, and a 5-mL solution of sample
(10-20 mM) was injected under Ar and cooled to -78 °C. The NO
purification train, which extended from the gas tank regulator through
a mercury bubbler and solid NaOH column into the reaction flask
through a needle inserted into a rubber septum, was pre-purged with
Ar. The reaction was monitored through ReactIR 2.0 software. After a
few initial background spectra were recorded, the Ar flow was
terminated and replaced by a constant flow of NO into the headspace
of the reaction at a sufficient rate to ensure an immediate excess of
NO atmosphere.
δmajor (>92%) ) 0.27(1) mm/s, ∆EQ ) 2.19(1) mm/s. δminor
)
-0.01(3) mm/s, ∆EQ 0.56(3) mm/s. Anal. Calcd for
)
FeC25H30N4O3F3S: C, 51.82; H, 5.22; N, 9.67. Found: C, 51.60; H,
5.13; N, 9.37.
[Fe(NO)(TC-5,5)] (3). A portion of 1 (110 mg, 0.256 mmol) was
dissolved in 10 mL of THF in a 25-mL, thick-walled, round-bottom
flask equipped with a screw-top Teflon stopcock and a sidearm for
attachment to a high-vacuum manifold. This red solution was subjected
to three freeze-pump-thaw cycles before NO gas (0.307 mmol) was
transferred under reduced pressure to the solution. Submerging the
reaction flask in a liquid N2 bath facilitated the complete transfer of
NO. The color remained deep red upon exposure to NO. The reaction
was allowed to stir for 2 h as the solution warmed to room temperature.
The solvent was removed in vacuo, and X-ray quality crystals were
grown in the drybox by pentane diffusion into a THF solution of the
complex. Yield: 81 mg, 67%. FTIR (KBr): 1692 (s, νNO), 1582 (s),
1503 (s), 1412, 1342, 1264, 1231, 1089, 1044, 1018, 978, 936, 886,
870, 799, 727, 582, 555, 532, 522, 515, 475, 432 cm-1. FTIR (THF)
1710 (s, νNO). UV-vis (THF) nm (M-1 cm-1): 277 (25100), 420
(14700), 515 (5200). µeff ) 2.40 µB (300 K). EPR: g ) 2.11, 2.03,
2.01. Mo¨ssbauer δ ) 0.06(1) mm/s, ∆EQ ) 1.39(1) mm/s. Anal. Calcd
for FeC24H30N5O: C, 62.61; H, 6.57; N, 15.21. Found: C, 62.17; H,
6.88; N, 14.66.
[Fe(NO)(TC-5,5-NO2)] (4). A portion of 1 (65 mg, 0.151 mmol)
was dissolved in 10.0 mL of THF in a 25-mL septum-sealed Schlenk
flask. The solution was cooled to -78 °C over a dry ice/acetone slush
bath, and purified NO was purged through the solution for 5 min. After
1 h at -78 °C, the cold bath was allowed to warm slowly to room
temperature; the reaction was left sealed and stirring for 12 h. The
solvent was removed in vacuo, and the black residue was extracted
into dichloroethane in the glovebox to give a deep purple solution after
some insoluble black material was filtered away. Blackish-purple
needles suitable for X-ray analysis were grown from pentane diffusion
into a dichloroethane solution. Yield: 25 mg, 30%. FTIR (KBr): 1716
(s, νNO), 1588, 1504, 1278, 1242, 1063, 1043, 873, 822, 737, 530, 474
cm-1. FTIR (THF) 1727 (s, νNO). UV-vis (THF) nm (M-1 cm-1): 267
(26000), 432 (20000), 522 (25000). Mo¨ssbauer δ ) -0.02(1) mm/s,
∆EQ ) 1.26(1) mm/s. EPR: g ) 2.14, 2.03, 2.01. Anal. Calcd for
FeC24H29N6O3: C, 57.04; H, 5.76; N, 16.63. Calcd for FeC24H28N7O5:
C, 52.27; H, 5.31; N, 17.78. Found: C, 52.21, 52.82, 51.46; H, 5.18,
4.99, 5.11; N, 15.67, 15.78, 15.95.
Vibrational Analysis. Normal-mode analysis was performed on a
Silicon Graphics Indigo workstation with the Svib program.19 Cartesian
coordinates for the Fe-N-O fragment were obtained from the X-ray
(17) Carlin, R. L. Magnetochemistry; Springer-Verlag: New York, 1986.
(18) Nagel, S. MBF Mo¨ssbauer Plot and Fit Program. 890330/c; Philipps
Universita¨t; Marburg, Germany.
(19) Mukherjee, A.; Spiro, T. G. Program 656, Bulletin Quantum
Chemistry Program Exchange. Indiana University; 1995.

10506 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Table 1. Summary of X-ray Crystallographic Data
Franz and Lippard
compound
formula
formula wt (g mol-1
crystal system
space group
a (Å)
[Fe(TC-5,5)] 1
FeC24H30N4
430.37
hexagonal
P6522
11.0367(4)
[Fe(OTf)(TC-5,5)] 2
FeC25H30N4O3F3S
579.44
monoclinic
P21/n
15.4688(2)
9.4959(2)
18.5944(2)
[Fe(NO)(TC-5,5)] 3
FeC24H30N5O
460.38
monoclinic
P21/c
9.210(3)
10.096(5)
23.334(8)
)
b (Å)
c (Å)
30.423(1)
R (deg)
â (deg)
113.551(1)
94.58(3)
γ (deg)
V (Å3)
3209.3(2)
6
1.336
0.722
188(2)
2503.83(7)
4
1.537
0.744
188(2)
2163.0 (1)
4
1.414
0.724
188(2)
Z
F
calcd (g cm-1
)
absorption coeff (mm-1)
temp (K)
radiation λ (Å)
reflections collected
data/parameter
Ra
0.71073
20224
2602/133
0.0796
0.71073
14175
5715/454
0.0623
0.71073
12772
4919/280
0.0608
wR2 b
0.1491
0.1434
0.1768
crystal size (mm)
0.66 × 0.24 × 0.14
0.40 × 0.25 × 0.15
0.34 × 0.14 × 0.03
max/min peak (e/Å3)
0.267/-0.453
0.889/-0.505
0.859/-0.422
a R ) Σ||Fo| - |Fc||/Σ|Fo|. wR2 ) {Σ[w(Fo - Fc2)2]/Σ[w(Fo )2]}1/2
.
b
2
2
structures of 3 and 4 and input to the program to generate potential
energy matrixes. Incorporating force constants, the program constructs
and solves the vibrational secular equation.20 The three principal and
three interaction force constants associated with the MNO oscillator
were refined by least-squares fitting of the experimental frequencies.
For 3, these values were 1692, 582, 515 cm-1 for 14NO and 1660, 579,
and 509 cm-1 for 15NO. For 4 these values were 1716, 737, 530 for
14NO and 1686, 721, 522 for 15NO.
GC Headspace Analysis. Procedures for the detection and quan-
titation of N2O have been described previously.13
X-ray Crystallography. Single crystals were mounted on quartz
fibers with Paratone N (Exxon) and transferred rapidly to the -85 °C
cold stream of a Bruker (formerly Siemens) CCD X-ray diffraction
system controlled by a Pentium-based PC running the SMART software
package.21 The general procedures for data collection and reduction
followed those reported elsewhere.22 Empirical absorption corrections
were calculated and applied from the SADABS program.23 Structures
were solved by the direct methods program XS, and refinements were
carried out with XL, both part of the SHELXTL program package.24
Non-hydrogen atoms were refined by a series of least-squares cycles.
All hydrogen atoms in 1 and 3 were assigned idealized positions and
given a thermal parameter 1.2 times that of the carbon atom to which
each was attached. The hydrogen atoms of compound 2 were located
in difference Fourier maps and refined isotropically. The structure of
1 was refined in the chiral space group P6522 and assigned the correct
absolute configuration. The Flack25 absolute structure parameter was
0.09(6), compared to expected values of 0.0 for the correct and +1.0
for the inverted structure. Crystallographic information is provided in
Table 1, and Figures S1-S3 (Supporting Information) display the
structures with complete atom labeling schemes.
Figure 2. ORTEP diagram of [Fe(TC-5,5)] 1 showing selected atom
labels and 50% probability ellipsoids for all non-hydrogen atoms.
previously described for the manganese analogue.13 The
iron(II) complex 1 was synthesized in good yield by reaction
of [Fe(OTf)2(CH3CN)2] with the dilithium salt of TC-5,5 in
THF. The distorted tetrahedral stereochemistry of this complex,
which has crystallographically required C2 symmetry, is depicted
as an ORTEP plot in Figure 2. Selected bond distance and angle
information is contained in Table 2. Previously we demonstrated
that profound changes in the stereochemistry of transition metal
tropocoronand complexes attend alterations in the number of
polymethylene linker chains of length n and m in the TC-n,m
ligand (Figure 1).15,26-28 To relieve the strain imposed on the
pentamethylene chains of the TC-5,5 ligand upon metal che-
lation, the two five-membered chelate ring planes must twist
away from planarity. For CoII, NiII, and CuII TC-5,5 complexes,
the dihedral angle Θ, which measures the degree of twist, is
69.9,28 70.1,15 and 61.3°,27 respectively. Compound 1 has a Θ
value of 66.0°, within that range. This ligand-imposed distorted
tetrahedral geometry in 1 differs from the square-planar
geometry encountered in most Fe(II) complexes of other
tetraazamacrocyclic ligands. Porphyrins,29,30 phthalocyanins,31
Results and Discussion
Synthesis and Characterization. An investigation of iron
tropocoronand complexes and their reaction with nitric oxide
was undertaken, and the results were compared with those
(20) Wilson, E. B.; Decius, J. C.; Cross, P. C. Molecular Vibrations;
McGraw-Hill: New York, 1955.
(21) SMART. 4.0; Bruker Analytical X-ray Systems; Madison, WI, 1996.
(22) Feig, A. L.; Bautista, M. T.; Lippard, S. J. Inorg. Chem. 1996, 35,
6892-6898.
(23) Program written by Professor George Sheldrick at the University
of Go¨ttingen which corrects data collected on Bruker CCD and multiwire
detectors for absorption and decay.
(24) SHELXTL: Structure Analysis Program. 5.1; Bruker Analytical
X-ray Systems; Madison, WI, 1997.
(25) Flack, H. D. Acta Crystallogr. 1983, A44, 499-506.
(26) Doerrer, L. H.; Lippard, S. J. Inorg. Chem. 1997, 36, 2554-2563.
(27) Davis, W. M.; Zask, A.; Nakanishi, K.; Lippard, S. J. Inorg. Chem.
1985, 24, 3737-3743.
(28) Jaynes, B. S.; Doerrer, L. H.; Liu, S.; Lippard, S. J. Inorg. Chem.
1995, 34, 5735-5744.
(29) Collman, J. P.; Hoard, J. L.; Kim, N.; Lang, G.; Reed, C. A. J. Am.
Chem. Soc. 1975, 97, 2676-2681.

ReactiVity of NO in Fe Tropocoronand Complexes
Table 2. Selected Bond Distances and Anglesa
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10507
cmpd
distances
(Å)
angles
(deg)
1
Fe-N(TC)avg
2.020(3)
N1-Fe-N1A
N1A-Fe-N2
N2-Fe-N2A
N1-Fe-N2
N3-Fe-N1
N2-Fe-O1
N4-Fe-O1
N4-Fe-N2
N3-Fe-O1
N1-Fe-O1
N3-Fe-N1
N2-Fe-N5
N5-Fe-N4
N4-Fe-N2
Fe-N5-O1
N3-Fe-N5
N1-Fe-N5
80.2(3)
114.8(2)
79.6(3)
139.0(2)
168.0(2)
120.3(2)
129.0(2)
110.0(2)
89.9(1)
2
3
Fe-N(TC)avg
Fe-O1
2.009(6)
2.117(3)
85.1(1)
Fe-N(TC)avg
Fe-NO
N5-O1
1.96(2)
1.670(4)
1.176(5)
172.0(2)
129.9(2)
130.7(2)
99.4(2)
174.3(4)
92.3(2)
95.6(2)
a Numbers in parentheses are estimated standard deviations of the
last significant figure. Atoms are labeled as indicated in Figures 2, 5,
and 7.
Figure 4. Mo¨ssbauer spectra of (a) [Fe(TC-5,5)] 1, δ ) 0.66(1) mm/
s, ∆EQ ) 1.18(1) mm/s and (b) [Fe(NO)(TC-5,5)] 3, δ ) 0.06(1) mm/
s, ∆EQ ) 1.39(1) mm/s. Both spectra were recorded as solid samples
at 77 K.
Figure 3. Effective magnetic moments from 5 to 300 K for O [Fe-
(TC-5,5)] 1, 2 [Fe(OTf)(TC-5,5)] 2, and b [Fe(NO)(TC-5,5)] 3.
and others32,33 afford square-planar complexes that exhibit
unusual intermediate-spin (S ) 1) ground states and short Fe-N
bond distances, <2.00 Å.29 In contrast, the average Fe-N bond
distance in 1 is 2.020(3) Å, and the effective magnetic moment
µeff, as shown in Figure 3, is 5.19 µB at 300 K, consistent with
an S ) 2 ground state. The Mo¨ssbauer spectrum, displayed in
Figure 4a, reveals an isomer shift δ of 0.66(1) mm/s, a value in
the range of high-spin ferrous complexes.34
Compound 1 can be oxidized with AgOTf in THF to afford
[Fe(OTf)(TC-5,5)] (2) in high yield. In a manner similar to that
observed for 4-coordinate [M(TC-n,m)] complexes, which twist
from square-planar to tetrahedral geometry as n and m increase,
5-coordinate [M(X)(TC-n,m)] complexes vary from square
pyramidal to trigonal bipyramidal geometry.35 The ORTEP
diagram in Figure 5 and metrical data in Table 2 reveal that 2
assumes the expected trigonal bipyramidal geometry of 5-co-
ordinate TC-5,5 complexes, with the triflate anion being coor-
dinated in the equatorial plane. Figure 3 displays the temperature
Figure 5. ORTEP diagram of [Fe(OTf)(TC-5,5)] 2 showing selected
atom labels and 50% probability ellipsoids for all non-hydrogen atoms.
dependence of the effective magnetic moment, µeff, which ranges
from 5.49 to 5.95 µB for 30 < T < 300 K. The latter value is
in excellent agreement with that expected for an S ) 5/2 spin-
only ion. The rapid decrease in µeff below 30 K is consistent
with zero-field splitting effects.17 The EPR spectrum of 2 (Figure
6) displays a sharp signal at effective g′ ) 8.69 together with
a multicomponent resonance centered at g′ ) 4.3. The spectrum
is consistent with high-spin Fe(III) in a noncubic coordination
environment, in which the magnetic levels in the absence of an
external magnetic field comprise three Kramers doublet states,
each of which may be described by a set of three effective g′
values.36 The signal at g′ ≈ 9 arises from the lowest doublet in
one principal direction, and the multicomponent signal at g′ ≈
4.3 arises from the middle doublet in more than one principal
direction.37 The highest doublet is often not observed.37
(30) Strauss, S. H.; Silver, M. E.; Long, K. M.; Thompson, R. G.;
Hudgens, R. A.; Spartalian, K.; Ibers, J. A. J. Am. Chem. Soc. 1985, 107,
4207-4215.
(31) Kirner, J. F.; Dow, W.; Scheidt, W. R. Inorg. Chem. 1976, 15,
1685-1690.
(32) Goedken, V. L.; Pluth, J. J.; Peng, S.-M.; Bursten, B. J. Am. Chem.
Soc. 1976, 98, 8014-8021.
(33) Little, R.; Ibers, J. A.; Baldwin, J. E. J. Am. Chem. Soc. 97, 7049-
7053.
(34) Dickson, D. P. E.; Berry, F. J. Mossbauer Spectroscopy; Cambridge
University: New York, 1986.
(35) Jaynes, B. S.; Ren, T.; Liu, S.; Lippard, S. J. J. Am. Chem. Soc.
1992, 114, 9670-9671.
(36) Pilbrow, J. R. Transition Ion Electron Paramagnetic Resonance;
Oxford University: New York, 1990.

10508 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Franz and Lippard
Figure 6. EPR spectrum of [Fe(OTf)(TC-5,5)] 2 in Me-THF glass at
26 K, 10 mW and 9.4 GHz.
Figure 8. EPR spectrum of [Fe(NO)(TC-5,5)] 3 in Me-THF glass at
4 K, 0.1 mW, 9.4 GHz.
impose trigonal bipyramidal geometry on five-coordinate com-
plexes, where the exogenous ligand lies in the equatorial
plane.13,27,28,35 The linearly coordinated nitrosyl ligand of 3 is
in the equatorial plane of a trigonal bipyramid, as evident in
Figure 7 and Table 2, thereby reconciling its linear coordination
with the low NO stretching frequency. The [Fe(NO)(TMC)]2+
complex included in Table 3 also has distorted trigonal
bipyramidal geometry and a linearly coordinated nitrosyl.39 The
Fe-NNO bond distance in 3 is slightly shorter than those in the
comparable compounds, implying a rather strong Fe-N bond.
In the literature of {FeNO}7 compounds, there are examples
3
40-43
1 44-47
of both S ) /2
and S ) /2
ground states, as well as
several compounds that exhibit spin-crossover behavior.39,48,49
Various physical methods were employed to probe the electronic
and magnetic properties of 3, from which we conclude that it
is an S ) 1/2 species having no spin-crossover behavior. Figure
3 reveals that µeff remains nearly constant from 5 to 300 K,
ranging from 1.89 to 2.40 µB. These values are only slightly
higher than the 1.73 µB spin-only value expected for an S ) 1/2
compound. The EPR spectrum of 3, shown in Figure 8, displays
g values at 2.11, 2.03, and 2.01, closely resembling those of
other S ) /2 {FeNO}7 compounds.44,46 An S ) /2 value can
arise either from a low-spin Fe(II) with the unpaired electron
on the nitrosyl ligand, as occurs in [Fe(NO)(TPP)],45 or from a
formally low-spin Fe(III) (S ) 1/2) coupled antiferromagnetically
Figure 7. ORTEP diagram of [Fe(NO)(TC-5,5)] (3) showing selected
atom labels and 50% probability ellipsoids for all non-hydrogen atoms.
1
Table 3. Comparison of Selected S ) /2 {FeNO}7 Complexes
νNO
,
M-N(NO), N-O, M-N-O,
deg
cm-1
Å
Å
S
ref
[Mn(NO)(TC-5,5)] 1662 1.6993(3) 1.179(3) 174.1(3)
2
/2
13
a
58
48
50
39
1
1
1
1
1
1
[Fe(NO)(TC-5,5)] 1692 1.670(4)
1.176(5) 174.3(4)
1.12 (1) 149.2(6)
[Fe(NO)(TPP)]b
[Fe(NO)(salen)]c
[Fe(NO)L]d
1670 1.717(7)
1630 1.80(15)
/
/
/
2
1.15
132.(5)
144.1(9)
2
2
1636 1.716(11) 1.17(2)
[FeNO(TMC)]2+ e 1840 1.737(6)
1.137(6) 177.5(5)
e
a This work. b TPP ) tetraphenylporphyrin. c Data from -175 °C,
(38) Hoffmann, R.; Chen, M. M. L.; Elian, M.; Rossi, A. R.; Mingos,
D. M. P. Inorg. Chem. 1974, 13, 2666-2675. The {MNO}n notation denotes
the number of electrons (n) in the MNO unit as the sum of the d electrons
of the metal plus the electrons in π* orbitals of NO [Enemark, J. H.; Feltham,
R. D. Coord. Chem. ReV. 1974, 13, 339-406.
3
at rt this compound is S ) /2. d L ) tetramethyldibenzotetraazacy-
clotetradecine. e TMC ) tetramethylcyclam; this complex has a S )
3
1/2 to /2 spin-equilibrium.
(39) Hodges, K. D.; Wollmann, R. G.; Kessel, S. L.; Hendrickson, D.
N.; Van Derveer, D. G.; Barefield, E. K. J. Am. Chem. Soc. 1979, 101,
906-917.
When 1 reacts with one equivalent of NO in solution, the
mononitrosyl compound 3 forms, the structure of which is
shown in Figure 7 (see also Table 2). Solutions of this nitrosyl
complex are stable to vigorous Ar purging, vacuum and
refluxing in toluene, but rapidly oxidize in air. The Fe-N-O
metrical parameters are listed in Table 3, where they are
compared with the values from [Mn(NO)(TC-5,5)] as well as
other 5-coordinate {FeNO}7 species. Of the tabulated iron
compounds with νNO between 1630 and 1690 cm-1, the
tropocoronand complex is unique in having a nearly linear Fe-
N-O linkage. This difference is related to the geometry at the
metal center. Molecular orbital studies for {MNO}7 complexes
predict a bent Fe-N-O angle if the NO is coordinated in the
apical position of a square pyramid.38 Trigonal bipyramidal
geometry, on the other hand, can accommodate a linearly
coordinated nitrosyl if the nitrosyl lies in the equatorial plane.38
As mentioned previously, the constraints of the TC-5,5 ligand
(40) Brown, C. A.; Pavlosky, M. A.; Westre, T. E.; Zhang, Y.; Hedman,
B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1995, 117, 715-
732.
(41) Zhang, Y.; Pavlosky, M. A.; Brown, C. A.; Westre, T. E.; Hedman,
B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1992, 114, 9189-
9191.
(42) Shepherd, R. E.; Sweetland, M. A.; Junker, D. E. J. Inorg. Biochem.
1997, 1-14.
(43) Ray, M.; Golombek, A. P.; Hendrich, M. P.; Yap, G. P. A.; Liable-
Sands, L. M.; Rheingold, A. L.; Borovik, A. S. Inorg. Chem. 1999, 38,
3110-3115.
(44) Nasri, H.; Ellison, M. K.; Chen, S.; Huynh, B. H.; Scheidt, W. R.
J. Am. Chem. Soc. 1997, 119, 6274-6283.
(45) Wayland, B. B.; Olson, L. W. J. Am. Chem. Soc. 1974, 96, 6037-
6041.
(46) Chen, Y.; Sweetland, M. A.; Shepherd, R. E. Inorg. Chim. Acta
1997, 260, 163-172.
(47) Hauser, C.; Wieghardt, K. Thesis, Ruhr-University Bochum, 1998.
(48) Haller, K. J.; Johnson, P. L.; Feltham, R. D.; Enemark, J. H. Inorg.
Chim. Acta 1979, 33, 119-130.
(37) Peisach, J.; Blumberg, W. E.; Lode, E. T.; Coon, M. J. J. Biol.
Chem. 1971, 246, 5877-5881.
(49) Wells, F. V.; McCann, S. W.; Wickman, H. H.; Kessel, S. L.;
Hendrickson, D. N.; Feltham, R. D. Inorg. Chem. 1982, 21, 2306-2311.

ReactiVity of NO in Fe Tropocoronand Complexes
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10509
Figure 10. In situ IR reaction spectra of [Fe(TC-5,5)] 1 in THF at
-78 °C as NO is introduced into the headspace of the reaction flask.
The label “FeNO” signifies the νNO of [Fe(NO)(TC-5,5)] 3 at 1710
cm-1, whereas “Fe(NO)(NO2)” represents the νNO of the nitrosyl-nitrito
Figure 9. Cyclic voltammograms of (a) [Fe(NO)(TC-5,5-NO2)] 4, with
E1/2 ) -1.24 V and -0.046 V and (b) [Fe(NO)(TC-5,5)] 3, with E1/2
) -1.81 V and -0.24 V, referenced to Cp2Fe/Cp2Fe+.
species at 1850 cm-1. N2O appears at 2223 cm-1
.
with NO- (S ) 1), as was favored for another tetraazamacro-
cyclic FeNO complex.50
To probe further the nature of the iron center in 3, a
Mo¨ssbauer spectrum was measured at 77 K. The results are
shown in Figure 4b. As a general trend, a low isomer shift
reflects increased formal charge on iron. Moreover, low-spin
compounds tend to have smaller δ values than their high-spin
counterparts.34 The low δ of 0.06(1) mm/s of 3 places it well
below the 0.66(1) mm/s of the high-spin Fe(II) compound 1,
as well as the 0.27(1) mm/s value of the high-spin Fe(III)
compound 2 (Figure S4). The Mo¨ssbauer data suggest that iron
has been oxidized to Fe(III) in compound 3. The shorter average
Fe-NTC bond distance of 1.96(2) Å in 3 compared to the values
of 2.020(3) Å in 1, an Fe(II) species, and 2.009(6) Å in high-
spin ferric 2, is also consistent with oxidation of iron to afford
low-spin Fe(III).
Figure 11. Profile traces of absorbance vs time of the relevant IR
bands from Figure 10.
Assignment of the iron center as Fe(III) implies that the
coordinated nitrosyl has been formally reduced to NO-. To
evaluate further such an assignment, a normal coordinate
analysis was performed to extract the force constants of the Fe-
N-O fragment. The nitrosyl stretching vibration for [Fe(NO)-
(TC-5,5)] appears at 1692 cm-1 and shifts to 1660 cm-1 upon
isotopic labeling with 15NO, in agreement with the 1662 cm-1
value computed by assuming a classical diatomic oscillator
model. By using six experimental frequencies (1692, 582, 515
cm-1 for 14NO and 1660, 579, and 509 cm-1 for 15NO), the
force constants were adjusted by least-squares refinement until
the calculated and experimental values matched one another.
The results indicate that, apart from the assignment of the high-
frequency mode to the N-O stretch, the band at 579 cm-1
derives from the Fe-N-O bend and that at 509 cm-1 from the
Fe-N stretch. The force constant derived for the N-O stretch,
11.41(4) mdyne/Å, reflects NO- character, the force constants
of NO-, NO, and NO+ being 8.4, 15.5, and 21.3 mdyne/Å,
respectively.40 Combined with the physical data presented above,
the results support the formal assignment of 3 as low-spin
consistent with the formulation [Fe(NO)(TC-5,5)(OTf)]. Reac-
tion of 1 equiv of NO with the triflate complex 2 provides an
alternative route to this {FeNO}6 species, which has an NO
stretch at 1707 cm-1 (ν15 ) 1675 cm-1) in the solid state,
ΝÃ
15 cm-1 higher in energy than that of the parent nitrosyl
complex 3.
Reactivity. We previously reported that [Mn(THF)(TC-5,5)]
converts to [Mn(NO)(TC-5,5)] in the presence of NO before
reacting further to disproportionate NO, releasing N2O and
forming [Mn(NO2)(TC-5,5)].13 The iron nitrosyl compound 3
is structurally and electronically very similar to its Mn analogue.
They have both been characterized as {MIII(NO-)}2+ species
with nearly the same N-O bond stretching force constants. It
was therefore of interest to determine whether 1 would also
promote the disproportionation of NO. Figure 10 displays the
results of in situ IR monitoring of the reaction of 1 in THF at
-78 °C under an NO atmosphere. The characteristic νNO of
compound 3 appears at 1710 cm-1, and after it reaches its
maximum intensity, two new bands emerge, one at 1850 cm-1
and one at 2223 cm-1. The latter band corresponds to the
formation of N2O. It takes about 30 min for 3 to be completely
consumed, during which time the two new bands reach their
maximum intensities and level off. This behavior is revealed
by the speciation profiles in Figure 11. The band at 1850 cm-1
corresponds to cis-[Fe(NO)(NO2)(TC-5,5)]. The infrared spec-
trum was inconclusive in distinguishing the NO2 ligand
coordination mode as either N-bound nitro or O-bound nitrito.
Although cis-nitrosyl-nitro iron complexes are known,51,52 we
favor the nitrosyl-nitrito formulation by analogy with the
{FeIII(NO-)}2+
.
Electrochemical investigation of compound 3 revealed two
reversible redox couples, as indicated by the cyclic voltammo-
gram in Figure 9b. A reversible reduction occurs at -1.81 V,
and a reversible oxidation appears at -0.24 V vs Cp2Fe/Cp2-
Fe+. Attempts to reduce 3 chemically to afford an {FeNO}8
species did not lead to any isolable product. Oxidation of 3 with
[Cp2Fe]OTf in THF, on the other hand, did afford a product
(50) Berno, P.; Floriani, C.; Chiesi-Villa, A.; Guastini, C. J. Chem. Soc.,
Dalton Trans. 1988, 1409-1412.

10510 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Franz and Lippard
Scheme 1
Figure 12. Continuation of the reaction shown in Figure 10 when the
solution warms from -78 °C to room temperature. The band at 1710
cm-1 corresponds to νNO of [Fe(NO)(TC-5,5)] 3, which reappears as
the reaction warms, but shifts to 1725 cm-1 as 3 is nitrated to form 4.
The band at 1850 cm-1 corresponds to the nitrosyl-nitrito species.
The residual peak at 1857 cm-1 is assumed to be some nitrosyl-nitrito
species in which the ligand rings have been nitrated.
in Figures 10 and 12. Generation of N2O coincides with
formation of the nitrosyl-nitrito species (νNO ) 1850 cm-1),
which is stable at -78 °C, but loses NO2 at higher temperature.
Maximizing the formation of the nitrosyl-nitrito species
therefore ensures that 1 equiv each of N2O and NO2 are
generated. Loss of the NO2 ligand upon warming regenerates
the nitrosyl compound 3, which can then further react with NO
to continue the cycle and generate more N2O and NO2, as
portrayed in Scheme 1. Under these reaction conditions, the
catalytic cycle is quenched by the nitration of 3 to form 4, which
does not react further with NO (step e). When the reaction is
executed at room temperature, NO2 is released much more
rapidly, forming 4 before each iron center has had a chance to
form the nitrosyl-nitrito intermediate, and thereby limiting the
generation of N2O.
The observation that 4 is incapable of promoting the
disproportionation of NO is significant for understanding the
factors that affect this reactivity. Compounds 4 and 3 are
structurally similar; both are 5-coordinate, mononitrosyl com-
plexes. The electron-withdrawing nitro groups on the TC ligand
in 4, however, decrease the electron density at the iron center,
which in turn affects the electronic distribution in the {FeNO}7
unit. The cyclic voltammogram of 4, displayed in Figure 9a,
reflects this electronic modulation. Compound 4 is easier to
reduce and harder to oxidize than the parent nitrosyl compound
3. Predictably, the Mo¨ssbauer isomer shift (Figure S5) decreases
from the value in 3, consistent with decreased electron density
at the metal center.34 Like its relative 3, compound 4 is also a
low-spin Fe(III) complex, as evidenced by EPR spectroscopy
(Figure S5). These subtle changes at the iron center affect the
nitrosyl adduct by increasing the energy of the NO stretch. The
NO force constant, derived from normal coordinate analysis, is
12.30(2) mdyne/Å, almost a whole unit increase from the value
found for 3. Given the structural similarities of 3 and 4, we
conclude that the decreased electron density at the metal in 4
deactivates the coordinated nitrosyl with respect to electrophilic
attack by NO, thereby inhibiting NO disproportionation reactiv-
ity. This conclusion corroborates an observation made in a recent
study of the NO disproportionation reactivity of various copper
tris(pyrazolyl)borate (Tp) complexes. A significant decrease in
the rate of NO disproportionation occurred when an electron-
withdrawing trifluoromethyl substituted Tp ligand was used.12
corresponding [Mn(NO)(NO2)(TC-5,5)] complex, for which a
crystal structure revealed the O-linked nitrito ligand.13
Up to the formation of this intermediate species, the reaction
pathways of the Mn and Fe systems appear to be identical. A
major difference occurs, however, when the solution warms to
room temperature. As shown in Figure 12, a band at 1710 cm-1
reappears upon warming that shifts to 1725 cm-1. Concomi-
tantly, the band at 1850 cm-1 diminishes in intensity and shifts
to 1857 cm-1. After 14 h, a weak band persists in solution at
1857 cm-1, although it decreases upon purging with Ar or
evacuation, whereas the strong nitrosyl stretch at 1725 cm-1
persists.
The final product from the reaction of 1 with excess NO was
determined to be [Fe(NO)(TC-5,5-NO2)], 4, an {FeNO}7 species
in which one or both rings of the tropocoronand ligand have
been nitrated. Whereas the intermediate [Mn(NO)(NO2)(TC-
5,5)] loses NO to form [Mn(NO2)(TC-5,5)],13 the analogous [Fe-
(NO)(NO2)(TC-5,5)] loses NO2. The assignment of the nitrating
species as gaseous NO2 was verified by reaction of 3 with NO2,
which afforded 4 as evidenced by IR spectroscopy. X-ray
crystallographic studies of 4 provided data that were sufficient
to reveal the connectivity of the structure.53 Disorder of the nitro
groups on the 4-position of the ligand rings indicated that some
of the compound might be nitrated on both rings. This finding
was confirmed by a FAB(+)MS of the organic material that
was extracted after acidic hydrolysis of 4, which confirmed the
presence of both singly and doubly nitrated H2TC-5,5-NO2. A
mixture of products is also evident in the C, H, and N analytical
determinations of three distinct preparations.
Since the amount of NO2 present in the products must match
the amount of N2O generated in the reaction on a molar basis,
the headspace gas was sampled by gas chromatography to
quantitate the corresponding N2O. The amounts varied consider-
ably, depending on the reaction conditions, ranging from 0.3
to 1.4 equiv of N2O per iron. The highest yields were obtained
by keeping CH2Cl2 solutions of 1 cooled to -78 °C for the
first 2 h after introducing NO and then sampling the headspace
after 48 h at room temperature in a sealed vessel. Eliminating
the cooling stage resulted in decreased N2O production.
The differing product distributions based on reaction tem-
perature can be explained by the in situ IR experiments shown
(51) Ileperuma, O. A.; Feltham, R. D. Inorg. Chem. 1977, 16, 1876-
(53) Crystals were grown by vapor diffusion of pentane into a dichlo-
roethane solution in a glovebox. The structure was determined in space
group Pbcn with. a ) 19.6694(4) Å, b ) 12.7979(1) Å, c ) 18.5090(3) Å,
V ) 4659.2(1) Å3, Z ) 8, T ) 188(2) K, R ) 9.12%, wR2 ) 23.65%.
1883.
(52) Ileperuma, O. A.; Feltham, R. D. J. Am. Chem. Soc. 1976, 98, 6039-
6040.

ReactiVity of NO in Fe Tropocoronand Complexes
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10511
The reduction of NO to N2O by iron compounds is not
-
without precedent. FeSO4 is reported to reduce NO2 and NO
to N2O in weakly alkaline solutions.54,55 In bacterial nitrite
reductases, heme centers carry out this transformation. Some
iron porphyrin complexes also display similar reactivity. Early
reports that a dinitrosyl complex forms when [Fe(NO)(TPP)]
is exposed to excess NO45 were later reexamined to disclose a
[Fe(NO)(NO2)(TPP)] species that is stable only in the presence
of excess NO.5 N2O was also detected in the reaction mixture.5
Some ruthenium porphyrins also promote NO disproportionation
and, in the process, form trans-[Ru(P)(NO)(NO2)] complexes.6,7
14NO/15NO Exchange. In an attempt to gain some insight
into the mechanism of the disproportionation reaction, in situ
IR spectroscopy was used to monitor the reaction of labeled
[Fe(15NO)(TC-5,5)] in the presence of excess of 14NO at 1 atm.
Figure 13 shows the IR spectra and Figure 14 the time-
dependence profiles of the relevant bands. When 14NO is
Figure 13. In situ IR reaction spectra of [Fe(15NO)(TC-5,5)] in THF
at -78 °C as 14NO is introduced into the headspace of the reaction
flask. The label “Fe14NO” signifies the ν14NO of [Fe(14NO)(TC-5,5)] 3
at 1710 cm-1, whereas “Fe15NO” appears at 1670 cm-1. The nitrosyl-
nitrito ν14NO appears at 1850 cm-1, and the corresponding ν15NO, which
can be seen as a small shoulder alongside the 1850 band, appears at
1816 cm-1. The inset shows the magnified region containing the
bands from mixed-labeled N2O. The assignments were made as
introduced to the reaction, the ν15
band at 1670 cm-1
NO
decreases in intensity and is replaced by the ν14NO band at 1710
cm-1. An increase in bands at 1850 and 2223 cm-1 corresponds
to the growing concentrations of [Fe(14NO)(NO2)(TC-5,5)] and
14N14NO, respectively. A small shoulder can be seen at 1816
cm-1, corresponding to [Fe(15NO)(NO2)(TC-5,5)]. It is apparent
from these results that most of the labeled 15NO is exchanged
prior to further reactions. The formation of mixed-label N2O,
as shown in the inset of Figure 13, is therefore the result of this
exchange process. The 14NO/15NO exchange reaction, which
was also detected in the Mn study,13 occurs faster than the
subsequent reactions that lead to NO disproportionation.
The exchange can be envisioned to occur by either of two
different mechanisms, shown in eqs 1-2. The first process is
follows: 14N14NO: 2223 cm-1,14N15NO: 2200 cm-1 15N14NO: 2176
,
cm-1, and 15N15NO: 2150 cm-1. The presence of an apparent feature
at 2200 cm-1 at the beginning of the reaction is an artifact. As the
reaction proceeds, the overlapping signal from 14N15NO appears well
above noise level in this region.
associative and involves a dinitrosyl intermediate, analogous
to that previously proposed for [Mn(NO)(TC-5,5)].13 Compound
3, like [Mn(NO)(TC-5,5)], does not lose NO upon purging with
N2 or by heating it in the solid state or in solution in vacuo.
The stability of these nitrosyl compounds favors eq 1 over eq
2, although an equilibrium between free and dissociated NO
such as in eq 2 cannot be ruled out.
Figure 14. Profile traces of absorbance vs time of the relevant IR
From the mixed-labeling studies it appears that the NO
exchange is a pre-equilibrium process that occurs prior to the
NO disproportionation reaction. The critical step leading to an
N-N bond formation has not been spectroscopically observed.
The N-N bond can be envisioned to form in several different
ways. A dinitrosyl intermediate could be induced, perhaps by
an incoming NO, to collapse and give rise to a hyponitrito
M{N(O)NO} species. Alternatively, direct attack of NO on the
coordinated nitrosyl would lead to a similar intermediate. Both
scenarios have been proposed previously but not definitively
proved.5,8,10,13,56 A dinitrosyl ruthenium porphyrin intermediate
was recently reported to form during NO disproportionation.6
Because of the configurational constraints of the porphyrin
ligand, this species would be a trans-dinitrosyl and therefore
not the direct precursor to N-N bond formation. Although the
results of the mixed-labeling studies favor the presence of a
cis-dinitrosyl intermediate for the NO exchange pathway in [Fe-
bands from Figure 13.
(NO)(TC-5,5)], there is no evidence that such an intermediate
leads directly to N-N bond formation. The difference in
reactivity between 3 and the electron-deficient 4 and the slowed
rates for the electron-deficient [Cu(NO)(Tp)] complex12 con-
vincingly argue that an electrophilic attack of NO on coordinated
NO is a critical aspect in these reactions. The attack could arise
either from free NO or from NO that is pre-coordinated to the
metal center, such as in a dinitrosyl complex.
The proposed reaction pathway is outlined in Scheme 1. The
first step (a) is the formation of the well-characterized nitrosyl
complex 3. Between steps (a) and (b) is the NO exchange
equilibrium, which is not shown explicitly in the scheme, but
is most likely by eq 1. Electrophilic attack on the reductively
activated nitrosyl ligand of 3 is proposed, yielding an unidenti-
fied {Fe(N2O2)} intermediate in step (b). Concomitant with the
release of N2O in step (c) is the formation of the nitrosyl-
nitrito species that is observable at -78 °C. In step (d) this
(54) Pearsall, K. A.; Bonner, F. T. Inorg. Chem. 1982, 21, 1978-1985.
(55) Bonner, F. T.; Pearsall, K. A. Inorg. Chem. 1982, 21, 1973-1978.

10512 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Franz and Lippard
species reverts to the original nitrosyl complex 3 by releasing
NO2 upon warming to room temperature. The liberated NO2
eventually kills the cycle by nitrating the ligand to form the
unreactive nitrosyl complex 4 in step (e).
can shut down this process, as is seen for ligand-nitrated 4.
Production of Fe(NO)(NO2) species and N2O has been docu-
mented in more biologically relevant Fe porphyrins,5 although
the fate of the NO2 was not investigated. The present study
reveals that the decomposition of Fe(NO)(NO2) species can
release the reactive NO2 radical, a potent agent for cytotoxic
DNA- and protein-damaging events.1 This study also suggests,
however, that such reactivity can be controlled by electronic
modification of the ligand environment.
Summary and Conclusions
The iron compound 3 displays interesting reactivity with NO.
Geometric constraints of the TC-5,5 ligand lead to trigonal
bipyramidal metal nitrosyl complexes in which NO is bound in
the equatorial plane; as in the case of Mn,13 the Fe-N-O
linkage is nearly linear. The strongly donating environment of
the ligand leads to metal nitrosyl complexes that formally
contain oxidized metal centers and reduced coordinated nitrosyl
ligands. In the presence of excess NO, both the Mn compound
reported previously13 and the Fe compound reported here
promote the disproportionation of NO. During this reaction,
metal nitrosyl-nitrito species cis-[M(NO)(NO2)(TC-5,5)] form.
The manganese analogue [Mn(NO)(NO2)(TC-5,5)] loses NO
to form [Mn(NO2)(TC-5,5)], whereas iron retains the NO and
loses NO2, which then reacts to give the final ligand-nitrated
[Fe(NO)(TC-55-NO2)] 4. The mechanism of the overall reaction
involves an NO-exchange step that could go through an
equilibrium between bound and dissociated NO, but more likely
involves a dinitrosyl intermediate. The steps of the proposed
reaction pathway are summarized in Scheme 1. This study
strongly supports previous arguments that a coordinated NO-
ligand is required to activate NO reductively to produce N2O.57
Modulation of the reduced nature of the coordinated nitrosyl
Acknowledgment. This work was supported by a grant from
the National Science Foundation. We thank Dr. W. Reiff at
Northeastern University for the use of his Mo¨ssbauer equipment,
Mr. D. Lee for his help in running the Mo¨ssbauer samples, Dr.
B. Spingler for assistance with X-ray crystallography and Dr.
R. Davydov for assistance with EPR spectroscopy.
Supporting Information Available: Figures S1-S3 display
fully labeled ORTEP diagrams for each reported structure.
Figures S4 and S5 contain Mo¨ssbauer spectra of 2 and 4,
respectively, and Figure S6 exhibits the EPR spectrum of 4
(PDF). X-ray crystallographic files, in CIF format, are available
free of charge via the Internet at http://pubs.acs.org.
JA991664F
(56) 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.
(57) McCleverty, J. A. Chem. ReV. 1979, 79, 53-76.
(58) Scheidt, W. R.; Frisse, M. E. J. Am. Chem. Soc. 1975, 97, 17-21.