Stereoelectronic Aspects of Inter-Metal Nitrogen Atom Transfer Reactions between Nitridomanganese(V) and Chromium(III) Porphyrins

Article abstract of DOI:10.1021/ic960919c

Reactions of nitridomanganese(V) porphyrins with chromium(III) porphyrins resulted in the irreversible formation of nitridochromium(V) porphyrins and manganese(III) porphyrins. The progress of these reactions has been followed spectrophotometrically, electrochemically, and spectroscopically by EPR. Kinetic analysis of the spectrophotometric data obtained during these reactions for a variety of substituted porphyrins showed the reactions to be first order in each of the reactants. Rate constants were dependent upon the electronic and steric effects of the porphyrin substituent, upon the identity of the anion bound to the chromium(III) reactant, and upon the solvent dielectric constant. We propose that the mechanism of these nitrogen atom transfer reactions involves the nucleophilic attack of the nitridomanganese porphyrin donor on the cationic chromium(III) porphyrin acceptor facilitating a net, two-electron redox process mediated by a heterobimetallic μ-nitrido intermediate. This report represents the first systematic study of the stereoelectronic effects involved in the complete, inter-metal nitrogen atom transfer between two metalloporphyrins.

Full text of DOI:10.1021/ic960919c

Inorg. Chem. 1997, 36, 5435-5439
5435
Stereoelectronic Aspects of Inter-Metal Nitrogen Atom Transfer Reactions between
Nitridomanganese(V) and Chromium(III) Porphyrins
Lawrence A. Bottomley* and Frank L. Neely
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332-0400
ReceiVed August 1, 1996X
Reactions of nitridomanganese(V) porphyrins with chromium(III) porphyrins resulted in the irreversible formation
of nitridochromium(V) porphyrins and manganese(III) porphyrins. The progress of these reactions has been
followed spectrophotometrically, electrochemically, and spectroscopically by EPR. Kinetic analysis of the
spectrophotometric data obtained during these reactions for a variety of substituted porphyrins showed the reactions
to be first order in each of the reactants. Rate constants were dependent upon the electronic and steric effects of
the porphyrin substituent, upon the identity of the anion bound to the chromium(III) reactant, and upon the solvent
dielectric constant. We propose that the mechanism of these nitrogen atom transfer reactions involves the
nucleophilic attack of the nitridomanganese porphyrin donor on the cationic chromium(III) porphyrin acceptor
facilitating a net, two-electron redox process mediated by a heterobimetallic µ-nitrido intermediate. This report
represents the first systematic study of the stereoelectronic effects involved in the complete, inter-metal nitrogen
atom transfer between two metalloporphyrins.
Introduction
demonstrated that a two-electron nitrogen atom exchange is not
limited to the transfer of the nitride from manganese(V) to
Atom transfer reactions involve the transfer of a single atom
and at least one electron from a donor to an acceptor molecule.1
Metalloporphyrins participate in complete or incomplete atom
transfer reactions as both donors and acceptors.2 Inter-metal
oxygen atom transfer reactions are typically incomplete for this
class of compounds. For example, homo- and heterobimetallic
µ-oxo complexes are readily prepared by reaction of an
oxochromium(IV) porphyrin oxygen atom donor with metal-
(II) macrocyclic oxygen atom acceptor complexes.3 In contrast,
inter-metal nitrogen atom transfer reactions are generally
complete processes. Takahashi discovered the first complete
inter-metal nitrogen atom transfer reaction between nitridoman-
ganese(V) and chromium(II) porphyrins involving a three-
electron exchange.4 Woo has observed a similar reaction
between nitridomanganese(V) and manganese(II) porphyrins.5
In a preliminary communication,6 we demonstrated that
reaction of NMnV(OEP) with ClCrIII(TPP) in benzene proceeded
rapidly, irreversibly, and quantitatively to the manganese(III)
and chromium(V) products, i.e. NMn(OEP) + ClCr(TPP) f
ClMn(OEP) + NCr(TPP).7 This was the first example of
complete inter-metal nitrogen atom transfer between two
metalloporphyrins which involved a net, two-electron exchange.
We proved that the reaction rate was first order in each of the
reactants and that the reaction rate can be significantly dimin-
ished either by the placement of steric groups on the porphyrin
periphery or by addition of nitrogenous bases capable of axial
ligation with the chromium(III) reactant. It has since been
chromium(III) porphyrins. Woo has observed nitrogen atom
transfer from manganese(V) to manganese(III) porphyrins.8 We
have extended our studies to include the nitrogen atom transfer
from chromium(V) to chromium(III) porphyrins9 and have
generalized the manganese(V) to chromium(III) reaction to other
macrocyclic complexes.10
In this report, we present the first systematic study of the
stereoelectronic effects involved in the complete, inter-metal
nitrogen atom transfer between two metalloporphyrins. We
have found that the nitrogen atom transfer rate is dependent
upon the electronic and steric effects of the porphyrin substitu-
ent, upon the identity of the anion bound to the chromium(III)
reactant, and upon the solvent dielectric constant and donor
strength. A mechanism for this novel atom transfer reaction is
proposed.
Experimental Section
Materials. All manganese and chromium porphyrins used in this
study were prepared by the method of Adler.11 The free-base form of
Baldwin’s capped porphyrin, H2(C2Cap), was synthesized and purified
by the method of Almog.12 Metal ion insertion into this porphyrin
was accomplished as previously described by Ellis.13 Nitridomanganese
and nitridochromium porphyrins utilized in this study were prepared
according to the procedure of Buchler14 from the corresponding
hydroxomanganese(III) or hydroxochromium(III) porphyrins. Product
purity was verified by UV-vis, NMR, and mass spectral measurements.
(8) Woo, K. L.; Czapla, D. J.; Goll, J. G. Inorg. Chem. 1990, 20, 3915-
3916.
(9) (a) Neely, F. L. Ph.D. Dissertation, Georgia Institute of Technology,
1989. (b) Neely, F. L.; Bottomley, L. A. Inorg. Chem. 1997, 36, 5432-
5434.
X Abstract published in AdVance ACS Abstracts, October 1, 1997.
(1) Holm, R. H. Chem. ReV. 1987, 87, 1401-1449.
(2) Woo, L. K. Chem. ReV. 1993, 93, 1125-1136.
(3) (a) Elliott, R. L.; Nichols, P. J.; West, B. O. Aust. J. Chem. 1986, 39,
975. (b) Liston, D. J.; West, B. O. Inorg. Chem. 1985, 24, 1568-
1576.
(10) Neely, F. L.; Bottomley, L. A. Inorg. Chim. Acta 1992, 192, 147-
149.
(4) Takahashi, T. Ph.D. Dissertation, University of Michigan, 1985.
(5) Woo, K. L.; Goll, J. G. J. Am. Chem. Soc. 1989, 111, 3755-3756.
(6) Bottomley, L. A.; Neely, F. L. J. Am. Chem. Soc. 1989, 111, 5955-
5957.
(11) Adler, A. D.; Longo, F. R.; Finarelli, J. D.; Goldmacher, J.; Assour,
J.; Korsakoff, L. J. Org. Chem. 1967, 32, 476.
(12) Almog, J.; Baldwin, J. E.; Crossley, M. J.; DeBernardis, J. F.; Dyer,
R. L.; Huff, J. R.; Peters, M. K. Tetrahedron 1988, 37, 3587.
(13) Ellis, P. E.; Linard, J. E.; Szymanski, T.; Jones, R. D.; Budge, J. R.;
Basolo, F. J. Am. Chem. Soc. 1980, 102, 1889-1896.
(14) Buchler, J. W.; Dreher, C.; Lay, K.-L.; Raap, A.; Gersonde, K. Inorg.
Chem. 1983, 22, 879-884.
(7) Abbreviations used: monovalent anion ) X-; 1,2-dichloroethane )
DCE; porphyrin dianion ) POR; meso-tetraphenylporphinato dianion
) TPP; substituted meso-tetraphenylporphinato dianion ) T-R-PP;
octaethylporphinato dianion ) OEP.
S0020-1669(96)00919-6 CCC: $14.00 © 1997 American Chemical Society

5436 Inorganic Chemistry, Vol. 36, No. 24, 1997
Bottomley and Neely
Table 1. Effect of Solvent Dielectric Constant on the Rate
Constant for the Reaction between NMn(OEP) and ClCr(TPP)
dielectric rate const dielectric rate const
solvent
const
(M-1 s-1
48 ( 6 CH2Cl2
)
solvent
const
9.08
(M-1 s-1
)
C6H6
CHCl3
C6H5Cl
2.27
4.79
5.70
350 ( 20
66 ( 5 CH2ClCH2Cl 10.35 1170 ( 140
200 ( 17
The kinetics of the manganese(V) to chromium(III) nitrogen
atom transfer reaction were determined from the changes in the
electronic spectrum of the reaction mixture as a function of time.
Data analysis was performed at the absorption maxima of each
of the products and reactants. A typical analysis of the data is
depicted in the inset of Figure 1. This reaction is first order in
each of the reactants; at 23 °C, a second-order rate constant of
48 ( 6 M-1 s-1 was obtained.
Figure 1. Electronic spectra acquired as a function of time after mixing
equal volumes of 62 µM solutions of NMn(OEP) and ClCr(T-4-Me-
PP) dissolved in benzene. The inset figure depicts an analysis of the
absorbance data for ClCr(T-4-Me-PP) as a function of time verifying
that the reaction is first order in this reactant and second order overall.
The ordinate has been normalized to the initial concentration of reactant
for purposes of display. Identical traces were obtained for the decrease
in concentration of NMn(OEP) and the increase in concentration of
the products NCr(T-4-Me-PP) and ClMn(OEP).
In an effort to identify long-lived intermediates, this reaction
was monitored with EPR spectrometry and electrochemistry.
Upon reaction of an equimolar aliquot of NMn(OEP) with ClCr-
(TPP) in benzene under strictly anaerobic conditions, an EPR
signal centered at g ) 1.985 increased in intensity with time.
No signals were observed for either reactants or the Mn product
at ambient temperatures. The spectrum is identical to that
previously reported for NCr(TTP) and NCr(OEP). The intense
11-line pattern and the symmetrical hyperfine structure14,16 are
consistent with the four pyrrole N atoms and the axial nitrido
group being magnetically equivalent. The isotropic signal at
g0 ) 1.985 is invariant with changes in the porphyrin structure.
Similarly, when the reaction was carried out in an electrochemi-
cal cell, the cyclic voltammetric fingerprints17 of the reactants
were replaced by those of the products. No new or transient
redox processes were observed within the accessible solvent/
supporting electrolyte potential range of 0.1 M TBAP in DCE.
Solvent Effects. Solvent donor strength and dielectric
constant have a marked influence on the rate of the reactions
under study. Table 1 lists the rate constants measured for the
nitrogen atom transfer between NMn(OEP) and ClCr(TPP) as
a function of solvent. Over the limited solvent series listed in
this table, the log of the rate constant increased linearly with
dielectric constant. However, in pyridine, no nitrogen atom
transfer between NMn(OEP) and ClCr(TPP) was observed. In
benzene solution containing the reactants and pyridine in a 1:1:2
molar ratio, the observed rate constant was one-eighth that
observed in the absence of pyridine. As the relative concentra-
tion of pyridine was increased, the rate of the reaction decreased.
No reaction was observed at relative pyridine concentrations
greater than 10. Solvents with dielectric constants higher than
those listed in Table 1 possess appreciable solvent donor strength
(e.g. benzonitrile, dimethyl sulfoxide, pyridine, etc.). With these
solvents, greatly diminished reaction rates were observed,
suggesting that the donor ability of the solvent plays a more
important role in inhibiting the reaction than the solvent
dielectric in facilitating the formation of a key reaction
intermediate (vide infra).
All reagents were obtained from Aldrich Chemical Co. and purified in
the manner previously described.15
Instrumentation and Methods. Visible spectral measurements
were obtained with a diode array rapid-scanning spectrometer system
composed of a Tracor Northern Model 6050 spectrometer containing
a crossed Czerny-Turner spectrograph in conjunction with a Tracor
Northern Model 1710 multichannel analyzer. Spectra were acquired
by irradiation of the sample with polychromatic light from a xenon
arc lamp and the subsequent spatial dispersion of the transmitted
radiation onto a 512 diode array detector by a grating with a 300 groove/
mm rule and a blaze of 500 nm. Wavelength calibration was achieved
with either a holmium oxide or an NIST standard filter. All measure-
ments reported herein represent the ensemble average of at least 56
spectral acquisitions. For display purposes, a five-point Savitsky-
Golay smoothing algorithm was applied to each spectrum depicted.
EPR experiments were performed on a Varian E3 spectrometer. All
experiments were carried out at ambient temperature (23 ( 1 °C).
For kinetic studies, equimolar solutions approximately 60 µM in
XCr(POR) or NMn(POR) were prepared in distilled benzene and their
electronic spectra were obtained. Equal volumes of each solution were
mixed, and the solutions were rapidly injected into a cuvette with an
optical path length of 0.71 mm. Electronic spectra were acquired over
time with the rapid-scanning spectrometer system. The concentrations
of reactants and products were calculated by solving four simultaneous
equations at the absorbance maximum for the Soret band of each
reactant and product in solution. Analysis of the concentration data
as a function of time indicated that the data were best-fit by a second-
order rate law. The rate constants given in each of the tables is the
mean value from at least three separate experiments plus or minus the
standard deviation from replicated determinations.
Results
The reaction of NMn(OEP) with ClCr(TPP) resulted in the
rapid, quantitative, and irreversible production of ClMn(OEP)
and NCr(TPP). Electronic absorption spectra acquired during
the course of this reaction are depicted in Figure 1. Well-defined
isosbestic points are observed at 410, 435, and 464 nm. The
final spectrum is that of a mixture of NCr(TPP) and ClMn-
(OEP) in the exact concentration ratios as that of the reactants.
Spectrophotometric analysis of mixtures of solutions containing
NCr(TPP) and ClMn(OEP) in concentration ratios up to 10:1
showed no evidence of any significant back-reaction even after
maintaining these solutions at 40 °C for 72 h.
The rate constants were invariant with changes in solution
ionic strength. The ionic strength of benzene was systematically
varied by addition of n-Bu4NClO4 to the reagent solutions. The
observed rate constants determined at a salt concentration of
(16) Groves, J. T.; Takahashi, T.; Butler, W. M. Inorg. Chem. 1983, 22,
884-887.
(17) (a) Bottomley, L. A.; Neely, F. L.; Gorce, J.-N. Inorg. Chem. 1988,
27, 1300-1303. (b) Bottomley, L. A.; Kadish, K. M. J. Chem. Soc.,
Chem. Commun. 1981, 1212-1214. (c) Bottomley, L. A.; Kadish, K.
M. Inorg. Chem. 1983, 22, 342-349. (d) Kadish, K. M.; Kelly, S. L.
Inorg. Chem. 1979, 18, 2968. (e) Kelly, S. L.; Kadish, K. M. Inorg.
Chem. 1982, 21, 3631. (f) Bottomley, L. A.; Neely, F. L. Inorg. Chem.
1990, 29, 1860-1865.
(15) Bottomley, L. A.; Deakin, M. R.; Gorce, J.-N. Inorg. Chem. 1984,
23, 3563.

Inter-Metal Nitrogen Atom Transfer Reactions
Inorganic Chemistry, Vol. 36, No. 24, 1997 5437
Table 2. Influence of the Chromium(III) Counterion on the Rate of
the Reaction of XCr(TPP) with NMn(OEP) Determined in Benzene
Soret
Soret
band rate const
band rate const
reactant
(nm) (M-1 s-1
)
reactant
(nm) (M-1 s-1
)
(ClO4)Cr(TPP) 457 >2000
ClCr(TPP)
650 ( 30 (N3)Cr(TPP) 448 40 ( 2
451 48 ( 6
(TFA)Cr(TPP) 455
(SCN)Cr(TPP) 453
340 ( 20 FCr(TPP)
437
5 ( 0.2
Table 3. Dependence of Rate Constants (M-1 s-1) on the Identity
of the Nitrogen Atom Donor and Acceptor Phenyl Ring Substituents
solvent
nitrogen atom donor
NMn(OEP)
nitrogen atom acceptor C6H6 CH2ClCH2Cl
ClCr(T-4-PhCH2O-PP) 33 ( 2 676 ( 20
Figure 2. Linear free energy relationships for the reaction of NMn-
(OEP) with a series of chlorochromium(III) para-substituted tetraphe-
nylporphyrin nitrogen atom acceptors (triangles) and the reaction of
nitridomanganese(V) para-substituted tetraphenylporphyrin nitrogen
atom donors with ClCr(OEP) (circles) in DCE.
ClCr(T-4-MeO-PP)
ClCr(T-4-Me-PP)
ClCr(TPP)
ClCr(T-4-F-PP)
ClCr(T-4-Cl-PP)
ClCr(T-4-F3C-PP)
36 ( 5 826 ( 58
41 ( 3 947 ( 104
48 ( 6 1170 ( 140
60 ( 3 1280 ( 160
1700 ( 200
110 ( 3
NMn(T-4-PhCH2O-PP) ClCr(OEP)
NMn(T-4-MeO-PP)
NMn(T-4-Me-PP)
NMn(TPP)
NMn(T-4-F-PP)
NMn(T-4-Cl-PP)
86 ( 10
energy relationship was constructed (Figure 2), and a F value
of 0.17 was obtained for the rate data acquired in each solvent
using σp as a measure of the electron-donating or -withdrawing
effect of the phenyl ring substitutent. This F value attests to a
significant inductive effect, especially when one considers the
distance of the substituent on the para position of the phenyl
ring to the nitrido reaction site.
50 ( 4
20 ( 4
15 ( 0.3
10 ( 0.3
9.0 ( 0.2
NMn(T-4-F3C-PP)
1.8 ( 0.2
10-5 M were identical to those at 10-3 M. However, the
observed rate constants were markedly dependent upon the
concentration of chloride (in the form of n-Bu4NCl) added to
solution. The observed rate decreased with increasing concen-
tration of n-Bu4NCl. In fact, no appreciable reactivity was
observed over a 24 h period when the [n-Bu4NCl]/[ClCr(TPP)]
ratio exceeded 5. Previous work has shown that ClCr(POR)
has little affinity for an anionic ligand.17c,18
Counterion Effects. The influence of the CrIII(POR) coun-
terion on the rate of nitrogen atom transfer was determined by
reaction of NMn(OEP) with XCr(TPP) where X ) F-, Cl-,
TFA-, N3-, SCN-, and ClO4-. The rate constants obtained are
listed in Table 2. The rate constants varied over 2 orders of
magnitude, illustrating the dramatic influence of counterion
identity on the reaction. Anions with coordinating abilities less
than that of TFA- (e.g., ClO4-) reacted too quickly to obtain
reliable rate data with our method. The rate constants were
found to be exponentially related both to the wavelength of the
Soret band maximum for XCr(TPP) and to Jorgensen’s f
parameter, an empirically based measure of ligand field strength
for anions.19 Such relationships demonstrated that the rate
constants were inversely related to the ligand field strength of
the chromium(III) counterion.
Substituent Effects. The influence of the steric and inductive
effects of porphyrin ring substituents on the rate of the nitrogen
atom transfer reaction was determined for both the donor and
the acceptor metalloporphyrins. The inductive effect of phenyl
ring substituents on the chlorochromium(III) porphyrin acceptor
was determined by monitoring the kinetics of the reaction of
selected para-substituted chromium tetraphenylporphyrins with
NMn(OEP). Table 3 contains a listing of the rate constants
obtained in benzene and in DCE. Note that the span in kobsd
ranges from 33 to 110 M-1 s-1 in benzene and from 676 to
1700 M-1 s-1 in DCE. In both solvents, electron-withdrawing
substituents facilitate the reaction. A Hammett-Taft linear free
The inductive effect of phenyl ring substituents on the NMn-
(POR) donor was determined by monitoring the kinetics of the
reaction of selected para-substituted nitridomanganese tetraphe-
nylporphyrins with ClCr(OEP). The rate constants for this series
are also listed in Table 3. Note that the span in rate constants
determined in DCE ranges from 1.8 to 86 M-1 s-1, significantly
lower than when the donor and acceptor porphyrins were
reversed (vide supra). This decrease in rate reflects the increased
basicity of OEP relative to that of the substituted tetraphe-
nylporphyrins and is consistent with the trend observed with
the substituted tetraphenylporphyrin nitrogen atom acceptor
study. A Hammett-Taft linear free energy relationship (Figure
2) gave a F value of -0.40. The sign of F indicates that the
reaction rate increases with increasing nucleophilicity of the
nitridomanganese reaction center. The magnitude indicates that
the donor porphyrin is twice as sensitive to the effect of
induction by phenyl ring substituents as the acceptor porphyrin.
Steric effects on the nitrogen atom transfer reaction were
probed using ortho-substituted tetraphenylporphyrins. Tetraphe-
nylporphyrins possessing one ortho substituent on each of the
phenyl rings displayed reduced rates of reactivity when such
substitution was on either the nitride donor or the acceptor
complex. For example, the rate of the reaction of NMn(OEP)
with ClCr(T-2-MeO-PP) was about 30% of the observed rate
of the reaction between NMn(OEP) and ClCr(T-4-MeO-PP).
Similarly, a comparable diminution in rate was observed for
the reaction between NMn(T-2-MeO-PP) and ClCr(OEP) com-
pared with the reaction between NMn(T-4-MeO-PP) and ClCr-
(OEP). Ortho-substituted metallotetraphenylporphyrins possess
six interconvertable atropisomers. The diminished rates ob-
served for the ortho-substituted metallotetraphenylporphyrins
imply markedly different rates for each atropisomer. Tetraphe-
nylporphyrins possessing two ortho substituents on each of the
phenyl rings of either the nitride donor or the acceptor were
unreactive. Specifically, no reaction was observed between
NMn(OEP) and the two potential nitrogen atom acceptors ClCr-
[T-2,6-(MeO)2-PP] and ClCr[T-2,4,6-(MeO)3-PP] or between
NCr(OEP) and ClMn[T-2,6-(MeO)2-PP] or ClMn[T-2,4,6-
(MeO)3-PP]. From this finding, we infer that the all-syn
atropisomer is unreactive. A comparable decrease in rate was
(18) (a) Summerville, D. A.; Jones, R. D.; Hoffman, B. M.; Basolo, F. J.
Am. Chem. Soc. 1977, 99, 8195. (b) Basolo, F.; Jones, R. D.;
Summerville, D. A. Acta Chem. Scand., Ser. A 1978, A32, 771.
(19) Jorgensen, C. K. Absorption Spectra and Chemical Bonding in
Complexes; Pergamon Press: Elmsford, N.Y., 1962.

5438 Inorganic Chemistry, Vol. 36, No. 24, 1997
Bottomley and Neely
previously observed in the conversion of nitridomanganese
porphyrins into (acylimido)manganese porphyrin complexes.20,21
Reaction of NMn(C2Cap) with ClCr(OEP) proceeded at a rate
comparable to that of NMn(TPP) whereas reaction of NMn-
(OEP) and ClCr(C2Cap) proceeded at a rate approximately 25
times slower than that of NMn(TPP).
nitrogen atom acceptor. Alternatively, the reaction may be
inhibited by the formation of the six-coordinate, anionic
chromium(III) complex {(Cl)2Cr(POR)}+. However, there is
no literature precedent for such a complex and our efforts to
provide spectroscopic or electrochemical evidence for its
existence (at 1 e [Cl-]/[ClCr(TPP)] e 100) have been in vain.
We propose that the second step involves nucleophilic attack
by the nitrido lone pair on the cationic chromium(III) porphyrin,
forming a positively charged heterobinuclear µ-nitrido inter-
mediate. We have yet to obtain spectroscopic or electrochemical
evidence of its existence and thus conclude that this intermediate
is present in very low concentration. Support for this proposed
step arises from the results of several experiments. First, when
the reaction of NMn(OEP) with ClCr(TPP) was carried out in
benzene/pyridine solution, the rate constants were inversely
proportional to [pyridine]. It has been previously shown that
pyridine readily binds to the chromium(III) center but does not
bind to the vacant axial position of NMn(POR).18 Thus, as the
pyridine concentration is increased, the equilibrium between the
five-coordinate ClCr(POR) and four-coordinate {Cr(POR)}+
complexes is shifted to either of the six-coordinate Cl(py)Cr-
(POR) or {(py)2Cr(POR)}+Cl- complexes. The formation of
either six-coordinate complex renders the chromium center
inaccessible to the nitrido lone pair. Second, the nonreactivity
of the porphyrins possessing substituents at the 2,6-positions
of the phenyl rings infers the formation of a binuclear
intermediate, since such substitution has been shown to inhibit
formation of µ-oxo-bridged23 and µ-nitrido-bridged24 iron por-
phyrin dimers. Third, the linear free energy trends demonstrate
that the reaction rate is significantly enhanced by increasing
the nucleophilic character of the donor and the electrophilic
character of the acceptor, which is consistent with the notion
of nucleophilic attack. Fourth, we have recently communicated
the synthesis and spectral characterization of an analogous
cationic heterobimetallic µ-nitrido complex, {(POR)RetNfCr-
(POR)}+, formed by reaction of NRe(POR) with ClCr(POR).25
The cationic heterobinuclear µ-nitrido dimer then undergoes
an intramolecular, bridging nitrogen atom mediated, two-electron
transfer as shown in eq 3. The product of this electron transfer
reaction dissociates to form NCr(POR) and {Mn(POR)}+. Kelly
and Kadish17e have concluded from conductivity measurements
that the extent of ionization of XMn(POR) is significantly less
than that of XCr(POR). We suggest that the affinity of the
manganese(III) center for anionic axial ligands provides a
plausible driving force for the dissociation of the heterobinuclear
µ-nitrido complex.
Discussion
In our preliminary communication,6 we suggested that the
initial step in the nitrogen atom transfer sequence involved back-
side nucleophilic attack on the five-coordinate chromium(III)
center. This suggestion is at odds with the observed dependence
of reaction rate on [n-Bu4NCl]/[ClCr(TPP)] and reaction of
NMn(OEP) with ClCr(C2Cap); Vide infra. We now propose
the following mechanism for the inter-metal nitrogen atom
transfer reaction between NMn(POR) and XCr(POR):
XCrIII(POR2) { k1 } {CrIII(POR2)}+X-
(1)
k-1
:NtMnV(POR1) + CrIII(POR2)+ { k2
}
k-2
{(POR1)MnVtNfCrIII(POR2)}+ (2)
{(POR1)MnVtNfCrIII(POR2)}+ { k3
}
k-3
{(POR1)MnIIIrNtCrV(POR2)}+ (3)
{(POR1)MnIIIrNtCrV(POR2)}+ { k4
}
k-4
{MnIII(POR1)}+ + :NtCrV(POR2) (4)
{MnIII(POR1)}+X- { k5 } XMnIII(POR1)
(5)
k-5
The initial step is the ionization of the chromium(III) complex
via loss of the anion. Kelly and Kadish22 have published
conductivity data which infer that the extent of ionization of
chromium(III) porphyrins differs markedly with anion ligand
field strength. Their evidence indicates that ∼40% of
(ClO4)Cr(TPP) is ion-paired in CH2Cl2 or DCE compared to
only 2% for ClCr(TPP). This fact and the marked dependence
of the rate of reaction between NMn(OEP) and ClCr(TPP) in
the presence of added chloride suggest that the dissociated, four-
coordinate {Cr(POR)}+ is the nitrogen atom acceptor. The
dependence of the reaction rate on both solvent dielectric
constant (Table 1) and anion ligand field strength (Table 2)
provides additional support of this hypothesis. Ionization of
the coordinately and covalently bound anion should be favored
in solvents of higher dielectric strength, since this property
measures the inherent ability of the solvent to separate charges.
Similarly, weakly coordinating anions should make better
leaving groups and promote the formation of the cationic
Woo and co-workers have proposed a similar mechanism
involving the reversible nitrogen atom transfer between man-
ganese(V) and manganese(III) porphyrins.26 Since chromium-
(V) and chromium(III) porphyrins also undergo reversible
nitrogen atom transfer, a common mechanism seems likely.9
All steps in our proposed mechanism have been written as
equilibria even though we have not observed the reverse process
with the nitrogen atom donor-acceptor pairs reported herein.
We have concluded that eq 2 is rate determining on the basis
of the lack of observed reversibility, our inability to detect
binuclear intermediates, and the known facility of inner-sphere
electron transfer reactions.27 Application of the steady-state
(23) Morgan, B.; Dolphin, D. In Structure and Bonding; Buchler, J. W.,
Ed.; Springer-Verlag: Berlin, 1987, Vol. 64, pp 115-204.
(24) Wagner, W. D.; Nakamoto, K. J. Am. Chem. Soc. 1988, 110, 4044-
4045.
(25) Tong, C.; Bottomley, L. A. Inorg. Chem. 1996, 35, 5108-5109.
(26) Woo, L. K.; Goll, J. G.; Czapla, D. J.; Hays, J. A. J. Am. Chem. Soc.
1991, 113, 8478.
(20) Bottomley, L. A.; Luck, J. F.; Neely, F. L.; Quick, J. S. In Redox
Chemistry and Interfacial BehaVior of Biological Molecules; Dryhurst,
G., Niki, K., Eds.; Plenum: New York, 1988; pp 77-86.
(21) Bottomley, L. A.; Neely, F. L. J. Am. Chem. Soc. 1988, 110, 6748-
6752.
(22) Kelly, S. L.; Kadish, K. M. Inorg. Chem. 1984, 23, 679.
(27) Haim, A. Prog. Inorg. Chem. 1983, 30, 273.

Inter-Metal Nitrogen Atom Transfer Reactions
Inorganic Chemistry, Vol. 36, No. 24, 1997 5439
approximation yields the following kinetic expression:
field strength and solvent dielectric constant supports this claim.
Present efforts are directed toward generalizing the scope of
this reaction with other metallomacrocyclic nitrogen atom
donor-acceptor pairs.
k1k2
k1k2
kf )
≈
(6)
k-1[X-] + k2[NCr(POR)] k-1[X-]
We submit that the absence of the back-reaction between NCr-
(POR) and XMn(POR) may reflect the differential stability of
NCr(POR) compared to NMn(POR), i.e. Keq(reacn 4) > Keq-
(reacn 2), as well as the extent of ionization of XCr(POR)
compared to XMn(POR), i.e. Keq(reacn 5) > 1/Keq(reacn 1).
The dramatic dependence of reaction rate on counterion ligand
Acknowledgment. We thank Profs. L. Keith Woo, John
Groves, and Laren Tolbert for helpful suggestions. Support of
this research by a grant from the DHHS National Institute of
Heart and Lung (No. HL-33734) is gratefully acknowledged.
IC960919C