4378 J. Am. Chem. Soc., Vol. 118, No. 18, 1996
Kim et al.
Chart 1
hydroxide species to form the alcohol product in a rebound step
analogous to that proposed for heme-catalyzed hydroxylations.15
In the heme case, this rebound step19 is known to be faster than
the rate for trapping alkyl radicals with O2 (∼109 s-1).20
However, the fact that the syringe pump experiment for 3/20
equiv of tBuOOH when carried out in air (Table 2, run no. 3-5)
afforded both alcohol and ketone with a ratio close to 1 showed
that alkylperoxy radicals derived from substrate and O2 were
involved in that particular reaction. Thus the so-called “re-
bound” step in this nonheme system appears to be slower than
the trapping of O2 by the nascent alkyl radicals.
of the starting catalysts persist. These observations suggest that
the ligand exchange equilibrium has shifted significantly to the
left. The leftward shift can be rationalized by the fact that the
negatively charged halide (X) would be more difficult to displace
than the neutral aqua ligands on 3 and the displacement reaction
would result in the formation of a strong acid HX,14a i.e.
The notion of a slow “rebound step” is reminiscent of the
chemistry of bleomycin, another nonheme iron oxidation
catalyst. Bleomycin is an anti-tumor drug that has been shown
to oxidize its DNA target via dioxygen-dependent and dioxygen-
independent pathways.21 The active form of the drug is
“activated bleomycin”, which has recently been demonstrated
to be a hydroperoxoiron(III) species.22 “Activated bleomycin”
attacks the DNA ribose ring to generate a C-4′ radical. In the
absence of O2, the radical is converted to the C-4′-OH
derivative which ring opens to the 4′-keto-1′-aldehyde. In the
presence of O2, the C-4′ peroxy radical is formed and the ribose
ring is cleaved to afford the phosphoglycolate and base propenal
products of bleomycin-catalyzed DNA degradation. Thus O2
can compete with the metal center to trap the substrate radical
in bleomycin chemistry as well.
[Fe(TPA)X2]+ (1 or 2) + ROOH h
[Fe(TPA)X(OOR)]2+ + HX (11)
When formed, the alkylperoxoiron(III) intermediate reacts with
substrate to form haloalkane product. As reported earlier,5b the
addition of 1 equiv of BuOOH affords haloalkane as the sole
t
alkane-derived product in 70-80% yield, an excellent conver-
sion efficiency. The kH/kD values associated with the formation
of haloalkane by 1 and 2 (7 and 8, respectively) differ
significantly from the value for BuO• abstraction (4), in
t
agreement with the proposed metal-based oxidation. Further-
more, Me2S has been shown to intercept the formation of
haloalkane producing a stoichiometric amount of Me2SO instead.
This oxidative ligand transfer reaction is thus equivalent to the
metal-based alkane hydroxylation associated with 3 at low
ROOH concentrations.
In summary, we have formulated a unifying scheme that can
explain the many observations associated with Fe(TPA)-
catalyzed alkane hydroxylations.5,6 At the heart of this scheme
is a metal-based oxidant, probably an alkylperoxoiron(III)
species, which is a two-electron oxidant (and thus can be
intercepted by dialkyl sulfides) and whose oxidizing power can
be modulated by the nature of the tripodal ligand. We have
demonstrated that this metal-based oxidant is responsible for
alkane hydroxylation at low ROOH concentration. With
increasing ROOH concentration, ROOH competes with substrate
for the metal-based oxidant and thereby initiates a radical chain
process generating alkoxy radical and O2. Substrate alkyl
radicals generated by either the metal-based oxidant or the
alkoxy radical can then be trapped by O2 to afford the alcohol
and ketone products observed at high ROOH concentrations.
The complexity of this chemistry demonstrates the problems
associated with developing biomimetic versions of enzyme-
catalyzed oxidations, in particular, the competition of substrate
and peroxide for the transient metal oxidant. To circumvent
this problem, the enzyme active site enforces strict metal:
substrate:oxidant stoichiometry. In our biomimetic system, we
have resorted to using dilute solutions of alkyl hydroperoxide
to minimize this competition and obtain efficient conversion
of oxidant to hydroxylated product. However, new strategies
will have to be developed to make such catalytic hydroxylation
systems more practical.
t
As the amount of BuOOH added is increased, the yield of
haloalkane increases to a maximum value that is stoichiometric
with catalyst.5b Alcohol and ketone are also produced. As in
the case of 3/150 equiv of ROOH, the kH/kD values for alcohol
and ketone formation considered separately and together by 1
and 2 (Table 3) indicate the participation of both alkoxy radical
and a metal-based oxidant in the hydrogen abstraction step.
Thus, as with 3, the excess ROOH present can compete with
substrate for the oxidant and a radical chain autoxidation process
is initiated. The involvement of these competing reactions
reconciles the apparently conflicting observations reported by
Arends et al.5 and our group.6
The metal-based alkane hydroxylation we observe presumably
results from the interaction of substrate with the alkylperoxoiron-
(III) intermediate or a high-valent iron-oxo species derived
therefrom. Direct evidence for an alkylperoxo intermediate has
been obtained when 3 and tBuOOH are reacted in the presence
of alcohol substrate at -40 °C.17a In this study, the first-order
decay of the visible absorption band assigned to the intermediate
slowed down significantly when benzyl alcohol-d7 was used in
place of benzyl alcohol (kH/kD ) 5). We proposed a mechanism
involving the coordination of alcohol to the alkylperoxoiron-
(III) intermediate, formation of an attractive six-membered ring
transition state, and its subsequent decomposition via contem-
poraneous O-O and C-H bond cleavage to afford the product
aldehyde and tBuOH (Chart 1). However, the same mechanism
cannot apply to alkane oxidation, since alkanes cannot coordi-
nate to the iron(III) center. We are presently carrying out studies
to determine whether the alkylperoxoiron(III) intermediate
directly abstracts hydrogen from alkane as illustrated in Scheme
2 or first converts to a high-valent iron-oxo species, which
then cleaves the substrate C-H bond.
(19) (a) Inchley, P.; Lindsay Smith, J. R.; Lower, R. J. New. J. Chem.
1989, 13, 669-676. (b) Bowry, V. W.; Ingold, K. U. J. Am. Chem. Soc.
1991, 113, 5699-5707. (c) Harrison, R. G. Unpublished results.
(20) Porter, N. A.; Mills, K. A.; Caldwell, S. E.; Dubay, G. R. J. Am.
Chem. Soc. 1994, 116, 6697-6705.
(21) (a) Absalon, M. J.; Wu, W.; Kozarich, J. W.; Stubbe, J. Biochemistry
1995, 34, 2076-2086. (b) Stubbe, J.; Kozarich, J. W. Chem. ReV. 1987,
87, 1107-1136 and references therein.
(22) (a) Sam, J. W.; Tang, X.-J.; Peisach, J. J. Am. Chem. Soc. 1994,
116, 5250-5256. (b) Sam, J. W.; Tang, X.-J.; Magliozzo, R. S.; Peisach,
J. J. Am. Chem. Soc. 1995, 117, 1012-1018. (c) Westre, T. E.; Loeb, K.
E.; Zaleski, J. M.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am.
Chem. Soc. 1995, 117, 1309-1313. (d) Burger, R. M.; Kent, T. A.; Horwitz,
S. B.; Mu¨nck, E.; Peisach, J. J. Biol. Chem. 1983, 258, 1559-1564.
Once formed from the reaction of substrate with metal
oxidant, the substrate alkyl radical may be trapped by a metal