Facile N...N coupling of manganese(V) imido species

Article abstract of DOI:10.1021/ja066440t

(Salen)manganese(V) nitrido species are activated by electrophiles such as trifluoroacetic anhydride (TFAA) or trifluoroacetic acid (TFA) to produce N ₂. Mechanistic studies suggest that the manganese(V) nitrido species first react with TFAA or

Full text of DOI:10.1021/ja066440t

Published on Web 01/06/2007
Facile N‚‚‚N Coupling of Manganese(V) Imido Species
Shek-Man Yiu, William W. Y. Lam, Chi-Ming Ho, and Tai-Chu Lau*
Contribution from the Department of Biology and Chemistry, City UniVersity of Hong Kong,
Tat Chee AVenue, Kowloon Tong, Hong Kong, China
Received September 6, 2006; E-mail: bhtclau@cityu.edu.hk
Abstract: (Salen)manganese(V) nitrido species are activated by electrophiles such as trifluoroacetic
anhydride (TFAA) or trifluoroacetic acid (TFA) to produce N . Mechanistic studies suggest that the
2
manganese(V) nitrido species first react with TFAA or TFA to produce an imido species, which then
undergoes N‚‚‚N coupling. It is proposed that the resulting manganese(III) µ-diazene species decomposes
via internal redox to give N
2
and manganese(II). The manganese(II) species is then rapidly oxidized by
manganese(V) imide to give manganese(III) and CF
3
CONH (for TFAA) or NH (for TFA).
2
3
Introduction
transfer the imido group to electron rich alkenes. However, very
often a high concentration of the substrate relative to the active
manganese species is required for these amination reactions,
and it has been suggested that the reactive manganese(V) imido
species may be consumed in various side reactions.9d Indeed,
we observed that in the absence of organic substrates, these
N‚‚‚N coupling between metal nitrides (eq 1), which involves
a three-electron change between each metal and nitrogen, is of
fundamental interest.
(n-3)
(n-3)
2
MtN f M
-NtN-M
(1)
(salen)manganese(V) imido species undergo a clean N‚‚‚N
The factors that control the kinetics of this coupling reaction
would also be relevant to the reverse dinitrogen cleavage
process. Coupling between metal nitrides has been reported for
coupling reaction.
The following manganese nitrido complexes were investi-
gated:
that of osmium,1 ruthenium, and iron; a heterocoupling
-3
4
5
6
reaction between OstN and MotN has also been observed.
In contrast, the analogous coupling reaction for metal imides
(eq 2) has yet to be demonstrated.
(n-2)
(n-2)
2
MdNR f M
-NRdNR-M
(2)
In principle, imide coupling should also be a feasible process,
at least for R groups and ancillary ligands that are small enough
so that steric effects are not a problem. We report here kinetic
and mechanistic studies of N‚‚‚N coupling reaction of (salen)-
manganese(V) imido species, which are generated from reaction
of the corresponding nitrido species with electrophiles such as
trifluoroacetic anhydride (TFAA) or trifluoroacetic acid (TFA).
Manganese(V) nitrido complexes containing porphyrin and salen
ligands have received much attention in recent years because
of their use as amination reagents.7-12 In the presence of
electrophiles such as TFAA or TFA, these nitrido complexes
are converted to the corresponding imido species, which can
Experimental Section
Materials. All chemicals were of reagent grade unless otherwise
noted. (Salen)manganese(V) nitrido complexes were synthesized ac-
cording to a literature method.9a The Schiff base ligands were prepared
(
9) (a) Bois, J. D.; Hong, J.; Carreira, E. M.; Day, M. W. J. Am. Chem. Soc.
1
996, 118, 915-916. (b) Bois, J. D.; Tomooka, C. S.; Hong, J.; Carreira,
(
1) Ware, D. C.; Taube, H. Inorg. Chem. 1991, 30, 4605-4610.
2) (a) Che, C. M.; Lam, H. W.; Tong, W. F.; Lai, T. F.; Lau, T. C. J. Chem.
Soc., Chem. Commun. 1989, 1883-1884. (b) Lam, H. W.; Che, C. M.;
Wong, K. Y. J. Chem. Soc., Dalton Trans. 1992, 1411-1416.
E. M. J. Am. Chem. Soc. 1997, 119, 3179-3180. (c) Bois, J. D.; Tomooka,
C. S.; Hong, J.; Carreira, E. M.; Day, M. W. Angew. Chem., Int. Ed. Engl.
1997, 36, 1645-1647. (d) Bois, J. D.; Tomooka, C. S.; Hong, J.; Carreira,
E. M. Acc. Chem. Res. 1997, 30, 364-372.
(
(3) (a) Demadis, K. D.; Meyer, T. J.; White, P. S. Inorg. Chem. 1997, 36,
(10) (a) Minakata, S.; Ando, T.; Nishimura, M.; Ryu, I.; Komatsu, M. Angew.
Chem., Int. Ed. Engl. 1998, 37, 3392-3394. (b) Minakata, S.; Nishimura,
M.; Takahashi, T.; Oderaotoshi, Y.; Komatsu, M. Tetrahedron Lett. 2001,
42, 9019-9022. (c) Nishimura, M.; Minakata, S.; Takahashi, T.; Oderao-
toshi, Y.; Komatsu, M. J. Org. Chem. 2002, 67, 2101-2110.
(11) Svenstrup, N.; Bøgevig, A.; Hazell, R. G.; Jørgensen, K. A. J. Chem. Soc.,
Perkin Trans. 1 1999, 11,1559-1566.
5678-5679. (b) Demadis, K. D.; El-Samanody, E.-S.; Coia, G. M.; Meyer,
T. J. J. Am. Chem. Soc. 1999, 121, 535-544.
(
4) Man, W. L.; Tang, T. M.; Wong, T. W.; Lau, T. C.; Peng, S. M.; Wong,
W. T. J. Am. Chem. Soc. 2004, 126, 478-479.
(5) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 6252-6254.
(6) Seymore, S. B.; Brown, S. N. Inorg. Chem. 2002, 41, 462-469.
(7) Groves, J. T.; Takahashi, T. J. Am. Chem. Soc. 1983, 105, 2073-2074.
(8) Bottomley, L. A.; Neely, F. L. J. Am. Chem. Soc. 1988, 110, 6748-6752.
(12) Ho, C. M.; Lau, T. C.; Kwong, H. L.; Wong, W. T. J. Chem. Soc., Dalton
Trans. 1999, 2411-2413.
10.1021/ja066440t CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 803-809
9
803

A R T I C L E S
Yiu et al.
by reacting the diamine with two equivalents of the corresponding
salicylaldehyde in ethanol. Dichloromethane, 1,2-dichloroethane, aceto-
nitrile, acetone, tetrabutylammonium hexafluorophosphate, and trans-
â-methylstyrene were of reagent grade and were further purified
Table 1. TFAA-Activated N‚‚‚N Coupling of (Salen)manganese(V)
Nitrido Complexesa
compound
% yield of N2b
reaction time (min)
1
2
3
4
73 ( 2
72 ( 2
53 ( 2
0
35
5
45
-
1
3
according to standard methods. Trifluoroacetic acid, trifluoroacetic
anhydride, 1,2-diaminoethane, 1,2-diaminocyclohexane, 3,5-tert-bu-
tylsalicylaldehyde, salicylaldehyde, and manganese(II) acetate tetrahy-
drate were obtained from Aldrich and were used as received. 2,3-
Diamino-2,3-dimethylbutane was synthesized according to a literature
a
V
Reaction conditions: Mn (N), 0.25 mmol; TFAA, 1.25 mmol; 1,2-
b
dichloroethane, 10 mL; T ) 0 °C. Percent yield of N2 was calculated by
assuming 2 mol of Mn (N) produce 1 mol of N2.
1
4
procedure.
V
Instrumentation. Gas chromatographic analyses were performed
on a HP 6890 gas chromatograph equipped with a HP-5MS (30 m ×
k
obs, was obtained by nonlinear least-squares fits of A
t
to time t according
and A are
0
.25 mm i.d.), a HP-FFAP (25 m × 0.2 mm i.d.), a molsieve 5A (25
to the equation A ) A + (A - A ) exp(-kobst), where A
t
∞
0
∞
0
∞
m × 0.32 mm i.d.) or a Carboplot (25 m × 0.53 mm i.d.) column.
GC-MS measurements were carried out on a HP 6890 gas chromato-
graph interfaced to a HP 5973 mass selective detector. Kinetic
experiments were performed by using either a Hewlett-Packard 8452A
diode array spectrophotometer or an Applied Photophysics DX-17MV
stopped-flow spectrophotometer. Electrospray ionization mass spec-
trometry (ESI/MS) was performed on a PE SCIEX API 365 mass
spectrometer. The samples were continuously infused with a syringe
the initial and final absorbances, respectively. The decay of the
intermediate was also monitored at 370 nm, however it appears to be
biphasic. The method of initial rate was applied to determine the order
2
of the reaction according to the equation A ) a
0
+ a
1
t + a
2
t + ... +
9 15
a
9
t .
The kinetics of the aziridination of trans-â-methylstyrene by
V
t
[
4
Mn (N)( Bu salen)] in the presence of TFAA were studied by using
an Applied Photophysics DX-17MV stopped-flow spectrophotometer.
-
1
pump at a constant flow rate of 5 µL min into the pneumatically
assisted electrospray probe with nitrogen as the nebulizing gas. H NMR
The concentration of trans-â-methylstyrene (0.1 to 0.8 M) was in large
1
V
t
-5
excess of [Mn (N)( Bu
4
salen)] (8 × 10 M). The reaction progress
spectra were recorded on a Varian (300 MHz) FT NMR spectrometer.
Elemental analyses were done on an Elementar Vario EL Analyzer.
Infrared spectra were recorded as KBr pellets on a Nicolet Avatar 360
FTIR spectrophotometer. Ion chromatographic analyses were performed
on a Dionex LC 20 ion chromatograph equipped with a GP 40 gradient
pump, a Dionex IonPac CS12A analytical column (4 mm × 250 mm)
and a Dionex ED 40 electrochemical detector. Suppression of the eluent
was achieved with a Dionex cation CSRS ULTRA II self-regenerating
suppressor (4 mm).
was monitored by observing absorbance changes at 370 nm.
Results and Discussion
1. TFAA-Activated N‚‚‚N Coupling of (Salen)manganese-
(V) Nitrido Complexes. 1.1. Products and Stoichiometry.
V
[Mn (N)(salen)] (1) is stable for weeks in common organic
solvents at room temperature. However, when 5 equiv of TFAA
were added to a suspension of 1 in 1,2-dichloroethane under
argon at room temperature, the green solid gradually dissolved
to give a dark brown solution and gas bubbles could be seen.
The gas evolved was determined to be N2 by GC and GC/MS.
Procedure for N‚‚‚N Coupling Reactions. Isolation and Analysis
of Products. All experiments were carried out under argon, unless
otherwise specified. TFAA/TFA was added using an airtight syringe
to a green solution of (salen)manganese(V) nitrido complex dissolved
in a solvent. The reaction temperature was maintained by a water bath.
The volume of the gas produced was measured using a gas burette.
The gas was analyzed by GC and GC/MS.
1
5
V 15
When the 98% N-labeled complex [Mn ( N)(salen)] was
14
15
used, the gas evolved was found to be a mixture of N N and
N2 in the ratio of 1:20, indicating that both N atoms come
1
5
2
+
+
V
Mn and NH
4
were analyzed by ion chromatography after
from Mn tN. These observations are consistent with a N‚‚‚N
coupling reaction of 1 upon activation by TFAA. This coupling
reaction also occurs with other (salen)manganese(V) nitrido
species. The results in Table 1 show that introducing tert-butyl
groups on 3- and 5- positions of salen does not affect the
coupling reaction. The longer reaction time for 1 than 2 is
probably due mainly to its low solubility in ClCH2CH2Cl. On
the other hand, the cyclohexylene-bridged salen complex 3 gives
a lower yield and a slower rate. Moreover, addition of TFAA
extraction with water. Trifluoroacetamide and aziridines were analyzed
by GC and GC/MS.
V
t
The manganese product after N‚‚‚N coupling reaction of [Mn (N)( -
Bu salen)] (2) was isolated by the following procedure. The reaction
4
mixture was filtered and the filtrate was evaporated to dryness under
vacuum at room temperature. The residue was purified by column
chromatography using silica gel (CH
from methanol/water to give [Mn ( Bu
2
Cl
2
/acetone) and recrystallized
salen)CF CO ].CH OH as a
brown crystalline solid. Anal. Calcd. for C35 Mn: C, 60.86;
H, 7.29; N, 4.06. Found: C, 60.55; H, 7.25; N, 4.19.
III
t
4
3
2
3
50 2 5 3
H N O F
V
to a dark-green solution of [Mn (N)(saltmen)] (4) did not result
in any color change nor formation of gas bubbles, indicating
that N‚‚‚N coupling does not occur readily for this nitrido
species. The dark-green solution eventually turned brown after
several days. Simple molecular modeling (Chem3D) shows that
there are little steric effects to N‚‚‚N coupling by tert-butyl
groups in the aromatic rings but there can be substantial effects
from substituents in the ethylene bridge, which are closer to
the nitrido moiety.
Among the complexes 1-4, [Mn (N)( Bu4salen)] (2) was
used to study the coupling reaction in detail because of its good
solubility in various organic solvents. Because the only gas
detected by GC and GC/MS is N2 with a yield of 72%, attempts
The amount of carbon monoxide produced after N‚‚‚N coupling
reaction was detected by the following procedure. TFAA (0.75 mmol)
was added into a green solution of 2 (0.15 mmol) in 1,2-dichloroethane
(3 mL) in a 10 mL reaction vessel equipped with a septum at 22 °C in
air. After 5 min, 50 µL of gas was withdrawn from the head space of
the reaction vessel using a gastight syringe and then analyzed by GC/
MS and GC/TCD.
Kinetics. The kinetics of the coupling reaction was monitored by
using a HP 8452A diode array spectrophotometer. The formation of
the intermediate was monitored at 370 nm under pseudo-first-order
conditions with the concentration of TFAA in at least 10-fold excess
V
t
n
of that of 2. Ionic strength was maintained with Bu
4 6
NPF . The reaction
follows first-order kinetics, and the pseudo-first-order rate constant,
(
13) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory Chemicals,
rd ed.; Pergamon: New York, 1988.
14) Sayre, R. J. Am. Chem. Soc. 1955, 77, 6689-6690.
(15) (a) Lee, D. G.; Lee, E. J.; Chandler, W. D. J. Org. Chem. 1985, 50, 4306-
4309. (b) Chandler, W. D.; Lee, E. J.; Lee, D. G. J. Chem. Ed. 1987, 64,
878-881.
3
(
8
04 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007
9

Facile N‚‚‚N Coupling of Manganese(V) Imido Species
A R T I C L E S
Figure 2. Plot of kobs vs [TFAA] for the formation of the intermediate (A)
-
4
between 2 (2 × 10 M) and TFAA in ClCH2CH2Cl at 298.0 K and I )
Figure 1. Spectral changes at 288-s intervals for the reaction between 2
-
5
0.1 M. (Insert) Corresponding plot of 1/k vs 1/[TFAA] (slope ) (6.32 (
.17) × 10 ; y-intercept ) (6.46 ( 3.48) × 10 ; r ) 0.998).
(
0
4 × 10 M) and TFAA (0.1 M) in ClCH2CH2Cl at 298.0 K and I )
obs
-
3
-2
0
.1 M. (Dotted line is the spectrum before adding TFAA.) (Insert) Kinetic
trace at 370 nm.
were made to search for the remaining 28% nitrogen atoms in
the solution. Analysis of the organic products in the reaction
mixture by GC and GC/MS showed the presence of (25 ( 2)%
of CF3CONH2, thus the mass balance for nitrogen is 97 ( 2%.
Also, it was found that (67 ( 2)% of TFAA was consumed in
the reaction. The yields are based on the number of moles of 2
used. There is no difference in the yields between reactions
performed under argon or in air. From the reaction mixture,
III t
[
Mn ( Bu4salen)(CF3CO2)] was isolated in 93 ( 2% yield. The
net result as observed experimentally is represented by eq 3,
which is not balanced.
V t
III
[
Mn ( Bu salen)(N)] + 0.67 (CF CO) O f [Mn
4
3
2
t
V
2
(
Bu salen)(CF CO )] + 0.36 N + 0.25 CF CONH (3)
Figure 3. Plot of initial rate vs [Mn (N)] for the reaction between 2 and
TFAA (0.10 M) in ClCH2CH2Cl at 298.0 K and I ) 0.1 M (slope )
4
3
2
2
3
2
4
-4
(
5.46 ( 0.15) × 10 ; y-intercept ) -(0.95 ( 1.66) × 10 ; r ) 0.998).
1
.2 Kinetic Studies. The kinetics of the coupling reaction
were investigated by UV/vis spectrophotometry. Repetitive
scanning of the UV/vis spectrum of a solution of 2 and TFAA
in ClCH2CH2Cl at 298.0 K revealed the rapid formation of an
intermediate (A) with λmax at 334 and 370 nm, which then slowly
decayed (Figure 1). The maximum absorbance for A is the same
for [TFAA] ranging from 0.025 to 0.1 M, suggesting that the
formation of A is quantitative at these TFAA concentrations.
The formation of A was monitored at 370 nm, and it was
found to follow pseudo-first-order kinetics. The pseudo-first-
order rate constant (kobs) is independent of ionic strength (0.01-
On the other hand, the decay of the intermediate A appears
to be biphasic; spectral changes show an isosbestic point at 312
nm for approximately the first half-life of the decay, which shifts
to 313 and 338 nm on further reaction. The final spectrum is
III t
similar to that of [Mn ( Bu4salen)(CF3CO2)]. The first phase
of the decay does not obey first-order kinetics, and the method
15
of initial rates was used to determine the order of the reaction.
V
2
V
-5
A plot of initial rate against [Mn (N)] ([Mn (N)] ) 5 × 10
-4
-
5 × 10 M) gives a straight line (Figure 3), indicating a
second-order rate law:
0.1 M) but increases with [TFAA], and kinetic saturation occurs
at [TFAA] > 0.1 M (Figure 2). The plot of 1/kobs vs 1/[TFAA]
is linear, this is consistent with a preassociation step prior to
the formation of the intermediate A as shown in eq 4 and eq 5;
the rate law is shown in eq 6. A nonlinear least-squares fit of
d[A]
dt
V
2
2
-
) 2k [Mn (N)] ) 2k [A]
(7)
2
2
V
The [Mn (N)] is taken as equal to [A] because the formation
of A from Mn (N) is quantitative. At 298.0 K, [TFAA] ) 0.10
V
-
1
the data to eq 6 gives k1 ) (6.4 ( 0.3) s and Kf ) (36 ( 5)
4
-1 -1
-1
M, I ) 0.1 M, and k is found to be (2.7 ( 0.2) × 10 M s .
M
at 298.0 K.
2
k2 is independent of ionic strength (I ) 0 - 0.1 M). The
Kf
q
q
V
V
activation parameters ∆H and ∆S , obtained from the plot of
Mn (N) + TFAA y
\
z Mn (N).TFAA
(4)
(5)
-
1
ln (k2/T) vs 1/T, are found to be 6.9 ( 0.4 kcal mol and -(15
k1
-1
-1
V
( 1) cal K mol respectively at [TFAA] ) 0.10 M.
The effect of [TFAA] on the initial rate has also been
investigated (Figure 4). There is a slight increase in initial rate
Mn (N).TFAA
98 A
k K [TFAA]
1 f
V
d[Mn (N)]
V
V
-
) kobs[Mn (N)] )
[Mn (N)]
-2
on increasing the concentration of TFAA from 2 × 10 to
dt
1 + K [TFAA]
f
-
1
(6)
3 × 10 M; this is probably due to medium effects.
J. AM. CHEM. SOC.
9
VOL. 129, NO. 4, 2007 805

A R T I C L E S
Yiu et al.
1.3. N‚‚‚N Coupling vs Aziridination. To gain more insight
into the nature of the intermediate A, the kinetics of the coupling
reaction were carried out in the presence of trans-â-methylsty-
rene (0.1-0.8 M). The same intermediate A was observed and
the rate constant for the appearance of this intermediate is the
same as in the absence of the alkene. However, the decay of
the intermediate A is first order in A when an excess of the
olefin is used (Figure 5). The pseudo-first-order rate constant
depends linearly on [trans-â-methylstyrene], and the second-
-
1
-1
order rate constant is (7.74 ( 0.03) × 10 M
s
at 298.0 K
(
Figure 5). The product was also analyzed by GC and GC/MS
and it was found that the corresponding N-trifluoroacetyl
aziridine was formed in 80% yield (based on 2). These results
strongly suggest that A is an imido species, which is capable
of transferring the imido group to an alkene. At [trans-â-
methylstyrene] < 0.1 M, however, clean pseudo-first-order
kinetics were not observed, suggesting that N‚‚‚N coupling is
competing with aziridination, as illustrated in Scheme 11.4.
Mechanism of N‚‚‚N Coupling. On the basis of the above
experimental results, a proposed mechanism for the coupling
reaction is shown in eqs 8-11; the overall reaction is represented
by eq 12.
-5
Figure 4. Plot of initial rate vs [TFAA]. [2] ) (9.84 × 10 M) in ClCH2-
CH2Cl at 298.0 K and I ) 0.1 M.
Figure 5. Absorbance change vs time trace at 370 nm during the reaction
-
5
of 2 (8 × 10 M) with trans-â-methylstyrene (0.1 M) in the presence of
TFAA (0.1 M) at 298.0 K in ClCH2CH2Cl. (Insert) Plot of kobs vs [trans-
â-methylstyrene].
coupling, this pathway can be neglected. UV/vis spectral studies
suggest that formation of the imido species is quantitative at
the TFAA concentrations used in our studies. Moreover the rate
of the coupling reaction is insensitive to [TFAA] (0.02-0.3 M),
which is consistent with an imido-imido coupling process. For
nitrido-imido coupling a bell-shape relationship between rate
and [TFAA] is expected, since the amount of TFAA governs
the relative amounts of nitride and imide in the solution.
Imido coupling should produce initially a manganese(III)
µ-diazene species, however, attempts to detect such a species
by various techniques including mass spectrometry were unsuc-
cessful. Because N2 is produced immediately after addition of
V
TFAA to Mn tN without an induction period, it is reasonable
to assume that the manganese(III) µ-diazene species rapidly
decomposes via intramolecular electron transfer according to
eq 10. A two-electron oxidation of CF3CONdNCOCF3 would
+
produce N2 and 2 mol of CF3CO ; the latter species would be
highly unstable. Thus, it is proposed that the reaction is initiated
-
by nucleophilic attack of CF3CO2 on coordinated CF3CONd
The first step (eq 8) is the reaction of 2 with TFAA to produce
an imido species (A) Mn dNCOCF3; this is supported by its
V
3 2
NCOCF , which would produce N and 2 mol of TFAA.
Production of TFAA would also satisfy the experimental
stoichiometry of the reaction. Apart from internal redox, prior
dissociation of coordinated CF3CONdNCOCF3 to give free
diazene followed by intermolecular electron transfer is also a
possibility. Free CF3CONdNCOCF3 is known to be unstable
ability to react with an alkene to produce an aziridine.
The next step (eq 9) is coupling of the imido species to
produce a manganese(III) µ-diazene species. This is supported
by a rate law that is second order in the imido species. Although
a second-order rate law is also consistent with nitrido-imido
806 J. AM. CHEM. SOC.
9
VOL. 129, NO. 4, 2007

Facile N‚‚‚N Coupling of Manganese(V) Imido Species
A R T I C L E S
Scheme 1
II t
Table 2. TFA-Activated N‚‚‚N Coupling of (Salen)manganese(V)
reaction vessel instead of by Mn(V). [Mn ( Bu4salen)] is known
to be very air-sensitive, and this side reaction would lead to
more N2 and less CF3CONH2 production.
Nitrido Complexesa
b
compound
% yield of N
2
reaction time (min)
1
2
3
4
93 ( 2
72 ( 2
58 ( 2
35 ( 2
30
3
40
90
2. TFA-Activated N‚‚‚N Coupling of (Salen)manganese-
V) Nitrido Complexes. 2.1. Products and Stoichiometry. N‚
‚N coupling also occurs readily when TFA (instead of TFAA)
is added to (salen)manganese(V) nitrido complexes. Addition
of 5 equiv of TFA to a solution of 1 (0.25 mmol) in
,2-dichloroethane (10 mL) under argon at room temperature
(
‚
a
V
Reaction conditions: Mn (N), 0.25 mmol; TFA, 1.25 mmol; 1,2-
dichloroethane, 10 mL; T ) 22 °C. Percent yield of N2 was calculated by
assuming 2 mol of Mn (N) produce 1 mol of N2.
b
1
V
resulted in the evolution of 93% of N2 after 30 min (assuming
V
15
2
mol of Mn tN produce 1 mole of N2). Using N-labeling,
and decomposes to give C2F6, CO, and N2.16 By using GC/MS
and GC/TCD, 0.4 ( 0.1 mol % of CO was detected in the head
space of the reaction vessel (the yield was calculated based on
the amount of Mn tN). Although this is a very small amount,
it is reproducible and is well above the detection limit of the
V
it is shown that both nitrogen atoms in N2 come from Mn tN.
As in activation by TFAA, the yield of N2 and the reaction rate
depend on the structure of the salen ligand (Table 2). The longer
reaction time for 1 than 2 is probably due to its low solubility,
whereas the low yields and rates for 3 and 4 are probably due
to steric effects; as expected, these effects are less than in TFAA-
activated reactions because of a less-bulky imido intermediate.
The reactions of 1 and 2 have been studied in more detail.
V
GC used (∼0.05 mol %). No CO was detected in the absence
V
of Mn tN or TFAA. This result suggests the existence of a
trace amount of free diazene; it also provides some support for
the proposed formation of a manganese(III) µ-diazene species.
Decomposition of the manganese(III) µ-diazene species would
result in the formation of Mn(II), however [Mn ( Bu4salen)-
CF3CO2)] is produced quantitatively. We propose that Mn(II)
is oxidized rapidly by Mn dNCOCF3 according to eq 11. This
is supported by the detection of CF3CONH2 in the reaction
mixture, which may be produced by the reduction of Mn d
NCOCF3 to Mn -NCOCF3 (the CF3CON ligand has a -1
charge in both Mn oxidation states) followed by protonation
with TFA (presumably formed by hydrolysis of TFAA by trace
water in the solvent).18
The overall stoichiometry of the proposed mechanism (eq
2) can be compared with the experimental results (eq 3). The
observed yield of N2 (72 ( 2) % is in reasonable agreement
with the yield of 67% according to eq 12. Moreover, the
observed amount of TFAA consumed (69%) is in good
agreement with the predicted value (67%), assuming that the
two moles of TFA consumed come from hydrolysis of one mole
of TFAA by trace water in the solvent. On the other hand, the
observed yield of 25% for CF3CONH2 is lower than the
predicted yield of 33%. The slightly higher yield of N2 and the
lower yield of CF3CONH2 than expected can be accounted for
by the oxidation of some Mn(II) to Mn(III) by trace O2 in the
Analysis of products in the solution by ion chromatography
+
(
after extraction with water) showed the presence of NH4 and
III t
2
+
Mn (aq). In a separate experiment, the solution after reaction
was evaporated to dryness, and recrystallization of the residue
(
V
17
III
t
produced [Mn (L)(CF3CO2)] (L ) salen or Bu4salen). The
yields of the various products for compound 1 and 2 are
summarized in Scheme 2. For both compounds, there is excellent
mass balance for manganese and nitrogen.
V
III
-
2
.2. Kinetic Studies. As in the case of TFAA-activated
reaction, the spectral changes for a solution containing 2 and
TFA in ClCH2CH2Cl at 298.0 K reveal the rapid formation of
an intermediate (B) with λmax at 330 and 372 nm, which then
slowly decays (Figure 6). The UV/vis spectra of the intermedi-
ates B and A are rather similar (see also Figure 1). The
maximum absorbance for B is the same for [TFA] ranging from
1
0
.025 to 0.1 M; this suggests that formation of B is quantitative
at these TFA concentrations.
The formation of the B, however, is too fast to be followed
even by stopped-flow techniques. The decay of B, monitored
at 372 nm, did not obey first-order kinetics, and the method of
V
initial rates was applied. A plot of initial rate against [Mn -
2
(
N)] gives a straight line (Figure 7), indicating a second-order
rate law. The slope of the line gives the second-order rate
4
-1 -1
constant k2′, which is equal to (1.4 ( 0.1) × 10 M
s
at
(
16) Young, J. A.; Durrell, W. S.; Dresdner, R. D. J. Am. Chem. Soc. 1962, 84,
105-2109.
17) Attempts to study this reaction independently did not give any reproducible
2
q
2
98.0 K and [TFA] ) 0.10 M. The activation parameters ∆H
(
V
t
4
results. If [Mn ( Bu salen)(N)] was mixed with TFAA first before adding
V
II
t
Scheme 2
[
Mn ( Bu
4 3
salen)], N‚‚‚N coupling of Mn dNCOCF occurred rapidly. On
V t
the other hand, there was a direct reaction between [Mn ( Bu
4
salen)(N)]
18) Under our experimental conditions, the presence of around 0.01 wt % of
II
t
4
and [Mn ( Bu salen)] if they were mixed first.
(
2 3 2
H O in the solvent is enough to produce 25% of CF CONH . Attempts to
add water to the system to study the effects resulted in lower yields of all
products.
J. AM. CHEM. SOC.
9
VOL. 129, NO. 4, 2007 807

A R T I C L E S
Yiu et al.
q
-1
and ∆S are found to be 8.8 ( 0.7 kcal mol and -(10 ( 2)
-
1
-1
cal K mol , respectively, at [TFA] ) 0.10 M. These values
are similar to that for TFAA-activated reaction, although B
should be less bulky than A.
2
.3. Mechanism of N‚‚‚N Coupling. Based on the experi-
mental results a proposed mechanism for the TFA-activated
coupling reaction of 2, which is similar to the TFAA-activated
reaction, is shown in eqs 13-16, the overall reaction is
represented by eq 17.
Figure 6. Spectral changes at 144 s intervals for the reaction between 2 (4
-
5
×
10 M) and TFA (0.1 M) in ClCH2CH2Cl at 298.0 K and I ) 0.1 M.
Dotted line is the spectrum before adding TFA.) (Insert) Kinetic trace at
72 nm.
(
3
V
2
Figure 7. Plot of initial rate vs [Mn (N)] for the reaction of 2 with TFA
(
×
0.10 M) in ClCH2CH2Cl at 298.0 K and I ) 0.1 M (slope ) (2.84 ( 0.05)
4
-4
10 ; y-intercept ) (3.56 ( 0.07) × 10 ; r ) 0.999).
2
1
also produced. Because we were unable to detect any N2H4,
-
H2, or N3 from the reaction mixture, we conclude that free
V
The first step is the formation of the imido species Mn d
N2H2 was either not produced or was more readily oxidized by
Mn . NH3 is proposed to come from the reduction of
Mn (NH)(L)(CF3CO2)] by [Mn (L)] (L ) salen or Bu4salen)
in the presence of TFA (eq 16). According to eq 17, the yield
for N2 and NH3 should be 67 and 33%, respectively, which are
slightly different from the results for 2 but very different for
that of 1. Because Mn ions were detected, these discrepancies
can be readily accounted for by a demetalation reaction of
NH (eq 13). This step is expected to be very fast. This proposed
intermediate is supported by our previous studies, which show
that in the presence of TFA, (salen)manganese(V) nitrido
complexes can transfer an NH group to an alkene to give an
aziridine.12 The next step is coupling of the imido species to
produce a manganese(III) µ-diazene species (eq 14), which is
supported by a second-order rate law. An imido-nitrido
coupling mechanism can be ruled out because UV/vis spectral
studies suggest that formation of the imido species is quantitative
at the TFA concentrations used in our studies. Attempts to detect
the proposed manganese(III) µ-diazene species were unsuc-
cessful, although trans-[(µ-N2H2){CpMn(CO)2}2] and [(µ-
N2H2){Fe(“NHS4”)}2]20 have been reported. N2H2 is a strong
reducing agent; hence, it is not unreasonable to assume that, in
this case, the µ-diazene species undergoes rapid internal redox
to give N2 (eq 15). Free N2H2 is known to decompose to give
N2, N2H4, and a small amount of H2; in acid solution, HN3 is
III
V
II
t
[
2+
II
[
Mn (L)] by TFA according to eq 18.
II
[
Mn (L)] + 2CF CO H f Mn(CF CO ) + H L (18)
3
2
3
2 2
2
19
This is supported by the detection of Mn2+ ions using ion
2+
chromatography. Control experiments show that Mn does not
V
reduce Mn dNH, and reaction of 1 or 2 with TFA in the
presence of an equivalent amount of Mn(CF3CO2)2 did not result
in an increase in the yield of NH3. Demetalation would result
(
21) Holleman, A. F.; Wiberg, E. In Inorganic Chemistry; Academic Press: New
(
19) Sellmann, D. J. Organomet. Chem. 1972, 44, C46-C48.
20) Sellmann, D.; Soglowek, W.; Knoch, F.; Moll, M. Angew. Chem., Int. Ed.
Engl. 1989, 28, 1271-1272.
York, 2001; p 630.
(
(22) Mansuy, D.; Battioni, P.; Mahy, J. P. J. Am. Chem. Soc. 1982, 104, 4487-
4489.
808 J. AM. CHEM. SOC.
9
VOL. 129, NO. 4, 2007

Facile N‚‚‚N Coupling of Manganese(V) Imido Species
A R T I C L E S
V
in less Mn dNH being reduced to give NH3 but more N‚‚‚N
tetramethylpiperidyl) was reported to give Fe(TPP)(py)2 and
C9H18N-NdN-NC9H18 upon treatment with pyridine (py);22
this reaction may also occur by coupling of the iron nitrene
complex. On the other hand, stable (imido)manganese(V) corrole
coupling to produce more N2. Independent experiments using
Mn (L)] also indicate that this demetalation indeed occurs in
the presence of TFA, with that of [Mn (salen)] occurring more
readily than [Mn ( Bu4salen)]. By lower the reaction temperature
II
[
II
II t
23
complexes have recently been reported. These complexes
to 0 °C, the amount of N2 (69 ( 2)% and NH3 (30 ( 2)% from
contain bulky 2,4,6-trisubstituted phenylimido groups, so
N‚‚‚N coupling is not expected to occur.
2
are in good agreement with the predicted values, presumably
demetalation is slowed down.
. Concluding Remarks. We have presented a definitive
Our results show that facile N‚‚‚N coupling occurs for imido
species derived from the (salen)manganese(V) nitrido complexes
1 and 2. This probably explains why amination experiments
are usually performed with 4, which forms an imido species
3
example of N‚‚‚N coupling of imido species. Imido species
generated from (porphyrin)manganese(V) nitrido complexes
have been reported to slowly decompose with time to the
corresponding manganese(III) porphyrin complexes; it is pos-
9
that is much more stable with respect to N‚‚‚N coupling. In
some cases, the more bulky tosyl anhydride (Ts2O) is used as
8
sible that N‚‚‚N coupling also occurs for these species. An iron-
the activating reagent for alkene amination, which gives better
1
0,11
(II) nitrene (or iron(IV) hydrazido) complex Fe(TPP)(N-
yields than TFAA.
NC9H18) (TPP ) tetraphenylporphyrin, NC9H18 ) 2,2,6,6-
The reverse process of N‚‚‚N coupling, i.e., cleavage of a
µ-diazene species to produce imido species, is also of interest,
and there are a number of systems reported in the literature that
undergo NdN splitting via µ-diazene intermediates.24
(
23) Eikey, R. A.; Khan, S. I.; Abu-Omar, M. M. Angew. Chem., Int. Ed. 2002,
1, 3592-3595.
24) (a) Cotton, F. A.; Duraj, S. A.; Roth, W. J. J. Am. Chem. Soc. 1984, 106,
749-4751. (b) Hill, J. E.; Profilet, R. D.; Fanwick, P. E.; Rothwell, I. P.
4
(
4
Angew. Chem., Int. Ed. Engl. 1990, 29, 664-665. (c) Hill, J. E.; Fanwick,
P. E.; Rothwell, I. P. Inorg. Chem. 1991, 30, 1143-1144. (d) Cui, Q.;
Musaev, D. G.; Svensson, M.; Sieber, S.; Morokuma, K. J. Am. Chem.
Soc. 1995, 117, 12366-12367. (e) Aubart, M. A.; Bergman, R. G.
Organometallics 1999, 18, 811-813. (f) Peters, J. C.; Cherry, J.-P. F.;
Thomas, J. C.; Baraldo, L.; Mindiola, D. J; Davis, W. M.; Cummins, C. C.
J. Am. Chem. Soc. 1999, 121, 10053-10067. (g) Mindiola, D. J.; Meyer,
K.; Cherry, J.-P. F.; Baker, T. A.; Cummins, C. C. Organometallics 2000,
Acknowledgment. The work described in this paper was
supported by the Research Grants Council of Hong Kong (CityU
1
105/02P) and the City University of Hong Kong (7001799).
We thank Prof. Seth Brown for helpful discussions.
Supporting Information Available: Tables S1-5 and Figures
S1-5. This material is available free of charge via the Internet
at http://pubs.acs.org.
1
9, 1622-1624. (h) Guillemot, G.; Solari, E.; Scopelliti, R.; Floriani, C.
Organometallics 2001, 20, 2446-2448. (i) Greco, G. E.; Schrock, R. R.
Inorg. Chem. 2001, 40, 3861-3878. (j) Yandulov, D. V.; Schrock, R. R.
J. Am. Chem. Soc. 2002, 124, 6252-6253. (k) Figueroa, J. S.; Piro, N. A.;
Clough, C. R.; Cummins, C. C. J. Am. Chem. Soc. 2006, 128, 940-950.
JA066440T
J. AM. CHEM. SOC.
9
VOL. 129, NO. 4, 2007 809