Selective oxygen transfers with iron(III) porphyrin nitrite

Article abstract of DOI:10.1021/jo960546x

The reaction of octaethylporphyrin iron(III) chloride with potassium crown ether (18-crown-6) nitrite in N-methylpyrrolidone-1% acetic acid under argon generates the iron(III) nitrite salt (PFeNO₂). The latter is a unique and selective oxygen atom transfer reagent. The reaction of a broad range of substrates (S) proceeds quantitatively to yield the oxidized substrate and the iron(II) porphyrin-nitrosyl adduct: PFeNO₂ + S → PFeNO + SO. Diatomic molecules to which oxygen is directly transferred from PFeNO₂ are NO, CO, and O₂. The conversion NO to NO₂ is shown via ¹⁵NO₂^- labeling experiments to proceed exclusively by the O atom transfer process. The ozone, generated from dioxygen, was trapped with nitrite ion and the two olefins 2-methyl-2 butene and 2,3-dimethyl-2 butene. These substances are inert to PFeNO₂ under argon. However, in an oxygen-saturated reaction mixture, nitrite produced nitrate. The olefins, following reduction of the reaction mixture with Zn/HOAc, yielded 1 mol of acetone and acetaldehyde and 2 mol of acetone, respectively. Other simple O atom transfers under argon were observed with dimethyl sulfide and triphenylphosphine. The PFeNO₂ reagent shows a preference for O insertion into allylic, benzylic, and aldehydic C-H bonds. Thus, no olefin containing these moieties is epoxidized. However, styrene and cis-stilbene are converted to styrene oxide and cis-stilbene oxide, respectively. The double oxidation of allylbenzene to trans-cinnamaldehyde entails an allylic rearrangement that suggests radical character to the O insertion process. However, no kinetic evidence for this was obtained. The reaction is an overall third-order process, rate = k(PFe^III)(NO₂^-(S). There was no correlation of observed rates with relevant C-H bond dissociation energies of substrates. The fastest reacting substrate was nitric oxide (k₂₂° = 52 M^-2 s^-1) and the slowest was toluene (k₆₀° = 6.3 × 10^-4 M^-2 s^-1). The range and selectivity of these O atom transfers sets them apart from the catalytic oxidations brought about by reactions of iron(III) porphyrins with peroxides, iodosoaryls, hypochlorite, and other oxidants. The driving force for the relatively mild oxidations with PFeNO₂ resides in the thermodynamic stability of the heme-NO adduct. Given the broad presence of nitrite in the environment and the ubiquity of porphyrins in the biosphere, the activation of nitrite by iron porphyrins has both an environmental and biochemical significance.

Full text of DOI:10.1021/jo960546x

6388
J . Org. Chem. 1996, 61, 6388-6395
Selective Oxygen Tr a n sfer s w ith Ir on (III) P or p h yr in Nitr ite
Stephen K. O’Shea, Wen Wang, Ruth S. Wade, and Charles E. Castro*
The Environmental Toxicology Graduate Program, University of California, Riverside, California 92521,
and CEC Consulting, 1090 Madison Place, Laguna Beach, California 92651
Received March 22, 1996X
The reaction of octaethylporphyrin iron(III) chloride with potassium crown ether (18-crown-6) nitrite
in N-methylpyrrolidone-1% acetic acid under argon generates the iron(III) nitrite salt (PFeNO2).
The latter is a unique and selective oxygen atom transfer reagent. The reaction of a broad range
of substrates (S) proceeds quantitatively to yield the oxidized substrate and the iron(II) porphyrin-
nitrosyl adduct: PFeNO2 + S f PFeNO + SO. Diatomic molecules to which oxygen is directly
transferred from PFeNO2 are NO, CO, and O2. The conversion NO to NO2 is shown via 15NO2
-
labeling experiments to proceed exclusively by the O atom transfer process. The ozone, generated
from dioxygen, was trapped with nitrite ion and the two olefins 2-methyl-2 butene and 2,3-dimethyl-2
butene. These substances are inert to PFeNO2 under argon. However, in an oxygen-saturated
reaction mixture, nitrite produced nitrate. The olefins, following reduction of the reaction mixture
with Zn/HOAc, yielded 1 mol of acetone and acetaldehyde and 2 mol of acetone, respectively. Other
simple O atom transfers under argon were observed with dimethyl sulfide and triphenylphosphine.
The PFeNO2 reagent shows a preference for O insertion into allylic, benzylic, and aldehydic C-H
bonds. Thus, no olefin containing these moieties is epoxidized. However, styrene and cis-stilbene
are converted to styrene oxide and cis-stilbene oxide, respectively. The double oxidation of
allylbenzene to trans-cinnamaldehyde entails an allylic rearrangement that suggests radical
character to the O insertion process. However, no kinetic evidence for this was obtained. The
reaction is an overall third-order process, rate ) k(PFeIII)(NO2-)(S). There was no correlation of
observed rates with relevant C-H bond dissociation energies of substrates. The fastest reacting
substrate was nitric oxide (k22° ) 52 M-2 s-1) and the slowest was toluene (k50° ) 6.3 10-4 M-2
s-1). The range and selectivity of these O atom transfers sets them apart from the catalytic
oxidations brought about by reactions of iron(III) porphyrins with peroxides, iodosoaryls, hypochlo-
rite, and other oxidants. The driving force for the relatively mild oxidations with PFeNO2 resides
in the thermodynamic stability of the heme-NO adduct. Given the broad presence of nitrite in
the environment and the ubiquity of porphyrins in the biosphere, the activation of nitrite by iron
porphyrins has both an environmental and biochemical significance.
In tr od u ction
is catalytic in iron porphyrin. Systems with oxidation-
resistant porphyrins, notably 2,6-dichlorophenyl- and
pentafluorophenyl-TPP derivatives,9,10 in some cases
along with pentafluoroiodosobenzene, exhibit particularly
high turnovers. Moreover, a variety of other transition
metals have been employed successfully in these reac-
tions.1,11 In addition to epoxidation, the metalloporphy-
rin-oxidant reagents are capable of O atom insertion into
C-H bonds,12 and the conversion of sulfides to the
corresponding oxides.13 Mechanistic kinetic3,14 and spec-
tral15 studies point to metal-oxo and metal-oxo porphy-
The epoxidation of olefins by a combination of iron-
(III) porphyrin and various oxidants has been studied
extensively in the past two decades as a biomimetic
model for cytochrome P450.1 Following the initial report
by Groves and colleagues,2 with iron(III)TPP and io-
dosobenzene, corresponding research has expanded at an
astonishing rate, and this work has contributed to the
resurgence of interest in biological processes by organic
and inorganic chemists. Generally, the porphyrins em-
ployed have been TPP derivatives. The oxidants have
included principally iodosoaryls,3 perbenzoic acids,4 alkyl
and acyl hydroperoxides,5 amine oxides,6 persulfate,7 and
hypchlorite ion.8 These reactions have been conducted
with the oxidant in large excess such that the reaction
(5) (a) He, G-X.; Bruice, T. C. J . Am. Chem, Soc. 1991, 113, 2743.
(b) Beck, M. J .; Gopinath, E.; Bruice, T. C. J . Am. Chem. Soc. 1993,
115, 21. (c) Machii, K.; Watanabe, Y.; Morishima, I. J . Am. Chem. Soc.
1995, 117, 6691.
(6) Nee, M. W.; Bruice, T. C. J . Am. Chem. Soc. 1982, 104, 6123.
(7) Bernadou, J .; Fabiano, A.-S.; Robert, A.; Meunier, B. J . Am.
Chem. Soc. 1994, 116, 9375.
(8) (a) Collman, J . P.; Kodadek, T.; Raybuck S. A.; Meunier, B. Proc.
Natl. Acad. Sci. U.S.A. 1983, 80, 7039. (b) Meunier, B.; Guilmet, E.;
DeCarvalho, M. E.; Poilblanc, R. J . Am. Chem. Soc. 1984, 106, 6668-
6676.
(9) Traylor, P. S.; Dolphin, D.; Traylor T. G. J . Chem. Soc., Chem.
Commun. 1984, 279.
X Abstract published in Advance ACS Abstracts, August 15, 1996.
(1) McMurry, T. J .; Groves, J . T. Metalloporphyrin Models for
Cytochrome P-450. In Cytochrome P-450; Plenum Press: New York,
1986; Chapter 1, pp 1-28. (b) Groves, J . T.; Gross, Z. NATO ASI Ser.,
Ser. C 1995, 459 (Bioinorganic Chemistry) 39.
(2) Groves, J . T.; Nemo, T. E.; Myers, R. S. J . Am. Chem. Soc. 1979,
101, 1032-1033.
(3) (a) Groves, J . T.; Nemo, T. E. J . Am. Chem. Soc. 1983, 105, 5786-
5791. (b) Groves, J . T.; Myers, R. S. J . Am. Chem. Soc. 1983, 105,
5791-5796. (c) Kodadek, T.; Raybuck, S. H.; Collman, J . P.; Brauman,
J . I.; Papazian, L. 1985, 107, 4343-4345. (d) Traylor, T. G.; Marsters,
J . C., J r.; Nakano, T.; Dunlap, B. E. J . Am. Chem. Soc. 1985, 107,
5537-5539.
(4) (a) Groves, J . T.; Watanabe, Y. J . Am. Chem. Soc. 1986, 108,
507. (b) Yamaguchi, K.; Watanabe, Y.; Morishima, I. J . Am. Chem.
Soc. 1993, 115, 4058-4065. (c) Machili, K.; Watanabe, Y.; Morishim,
I. J . Am. Chem. Soc. 1995, 117, 6691-6697.
(10) Castellano, A. J .; Bruice, T. C. J . Am. Chem. Soc. 1988, 116,
158.
(11) (a) Arasasingham, R. D.; He, G.-X.; Bruice, T. C. J . Am. Chem.
Soc. 1993, 115, 7955. (b) Groves, J . T.; Ahn, K. H.; Quinn, R. J . Am.
Chem. Soc. 1988, 110, 4217. (c) Garrison, M. J .; Lee, R. W.; Bruice, T.
C. Inorg. Chem. 1990, 29, 2019. (d) Park, Y. C.; Na, H. G.; Kim, S. S.
J . Korean Chem. Soc. 1995, 39, 364. (e) Scharbert, B.; Zeisberger, E.;
Paulus, E. J . Organomet. Chem. 1995, 155, 59.
(12) (a) Traylor, T, G.; Hill, K. W.; Famm, W.-P. Tsuchuya, S.;
Dunlap, B. E. J . Am. Chem. Soc. 1992, 114, 1308. (b) Ohtake, H.;
Higuchi, T.; Hirobe, M. J . Am. Chem. Soc. 1992, 114, 10660.
S0022-3263(96)00546-4 CCC: $12.00 © 1996 American Chemical Society

Oxygen Transfers with Porphyrin Iron(III) Nitrite
J . Org. Chem., Vol. 61, No. 18, 1996 6389
Ta ble 1. Oxid a tion s w ith
Ch lor oir on (III)octa eth ylp or p h yr in a n d P ota ssiu m
(18-Cr ow n -6) Nitr ite in NMP -1%HOAc u n d er Ar gon a t 22
°C
rin cation radicals as intermediates in these processes.
However, recent work with efficient cation-radical traps,16
with zinc porphyrins,17 and water exchange studies
challenge this formulation.
As a part of our studies of the redox reactions of iron
porphyrins18 and hemeproteins19 with N-O-bonded spe-
cies, we have reported a stoichiometric oxygen atom
transfer from porphyrin iron-NO2 adducts to a range of
substrates under argon.20 In this work we expand upon
these preliminary findings and attempt to place this new
reaction in the general framework of porphyrin wrought
oxidation chemistry.
Resu lts a n d Discu ssion
Stoich iom etr y. The general chemistry described in
this work proceeds according to the stoichiometry shown
in eqn 1 (P ) porphyrin, S ) substrate). Starting
S
K} PFeNO2
8 PFeNO + SO (1)
-
PFeIII + NO2
{
krate
reactants are chloroiron(III) octaethylporphyrin, potas-
sium crown ether (18-crown-6) nitrite, and substrate. The
oxidized substrate and the porphyrin iron(II) porphyrin-
nitrosyl adduct are generated in each case. Reactions
were conducted under argon in the solvent N-methyl-
pyrrolidone-1% acetic acid. An exception is the reaction
with oxygen. This was conducted in an oxygen-saturated
vessel. Products and yields are summarized in Table 1.
In addition to the substances listed in Table 1, the
following were found to be inert under these conditions
(no reaction in 36 h): 2-methyl-2-butene, 2,3-dimethyl-
2-butene, vinyl chloride, cis- and trans-1,2-dichloro-
ethene, trans-stilbene, and nitrite ion.
The reactions feature (i) direct oxygen atom transfers
to the diatomic gases NO, O2, and CO, as well as to
triphenylphosphine and dimethyl sulfide (lines 1-5); (ii)
a preference for allylic, benzylic, and aldehydic C-H
insertion (lines 6-14); and (iii) the epoxidation of olefins
not containing an allylic or aldehydic C-H bond (lines
15-16). Some aspects of each of these reactions require
further comment.
a
b
In percent. 100(moles of product/moles of PFeIII). See text.
c Of TMS ester. 100(moles of product/2 moles of PFeIII).
d
ometry indicated in eq 1. In this case the reaction is (eq
2):
Th e Dia tom ic Ga ses. These oxygen transfers set the
PFeNO2 + NO f PFeNO + NO2
(2)
reactions of the PFeIII/NO2 system apart from those of
-
the iron porphyrin-catalyzed peroxide oxidations noted
in the Introduction. As far as we are aware, the latter
reagents do not bring about these transformations. The
CO-CO2 conversion is straightforward. However, the
NO and O2 oxidations need further characterization.
NO-NO2. As detailed in the Experimental Section,
the reaction proceeds according to the general stoichi-
The stoichiometry alone, however, does not establish
that an oxygen atom transfer from the nitrite adduct has
occurred. NO is the only substrate studied that is both
a substrate and product. Thus, in this case, assuming
there is some iron(II) character to the iron(III)-nitrite
+
adduct (PFeIII , NO2- T PFeII, NO2), the reaction (eq 2)
could represent a simple ligand exchange. In order to
address this and other potential schemes that may
explain these results, the NO oxidation was conducted
with the 15N-labeled nitrite salt.21 Thus, an oxygen atom
transfer (eq 3) must result in 15NO but no 15NO2.
(13) (a) Tomofumi, S.; Naoki, M.; Masaki, H. Chem. Pharm Bull.
1984, 32, 1252. (b) Ando, W.; Tajima, R.; Takata, T. Tetrahedreon Lett.
1982, 23, 1685.
(14) (a) Groves, J . T.; Watanabe, Y. J . Am Chem. Soc. 1986, 108,
507. (b) Traylor, T. G.; Miksztal, A. R. J . Am. Chem. Soc. 1987, 109,
2770. (c) Inchley, P.; Smith, J . R. L. J . Chem. Soc., Perkin Trans. 2
1995, 1579.
(15) (a) Groves, J . T.; Gross, Z.; Stern, M. K. Inorg. Chem. 1994, 33,
5065. (b) Yamaguchi, K.; Watanabe Y.; Morishima, I. J . Am. Chem.
Soc. 1992, 114, 1721. (c) Balch, A. L.; Cornman, C. R.; Latos-
Gratzynski, L.; Renner, M. W. J . Am. Chem. Soc. 1992, 114, 2230.
(16) (a) Mirafzal, G. A.; Kim, T.; Liu. J .; Bauld, M. L. J . Am. Chem.
Soc. 1992, 114, 10968. (b) Kim, T., Mirafszal, G. A.; Liu. J .; Bauld, N.
L. J . Am. Chem. Soc. 1993, 115, 7653.
+
PFeIII
+
15NO2- f PFe15NO2 NO8 PFe15NO + NO2
(3)
Analysis of the product gases by mass spectroscopy and
by 15N-NMR of an aqueous solution into which they were
passed agreed exactly with this occurrence. The mass
spectra of a sample stored in an evacuated tube and one
(17) (a) Nam, W.; Valentine, J . S. J . Am. Chem. Soc. 1993, 115, 1772.
(b) The Zn references.
(18) Castro, C. E.; Bartnicki, E. W. J . Org. Chem. 1994, 59, 4051.
(19) Wade, R. S.; Castro, C. E. Chem. Res. Toxicol. 1990, 3, 289.
(20) (a) Castro, C. E.; O’Shea, S. K. J . Org. Chem. 1995, 60, 1922.
(b) Castro, C. E. J . Am. Chem. Soc. 1996, 118, 3984.
(21) A nitrogen rather than oxygen label was selected because the
former can give an all or nothing result, and it is subject to both MS
and NMR analysis.

6390 J . Org. Chem., Vol. 61, No. 18, 1996
O’Shea et al.
F igu r e 1. Mass spectra for the reaction gases from PFeIII
,
15NO2-, and NO: (a) stored gas sample, (b) sample taken
directly above liquid in reaction and analyzed immediately,
(c) sample of NO.
F igu r e 2. 15N-NMR of (a) product gases bubbled through D2O;
(b) standard 0.02 M D2O solution of K (crown ether) 15NO2.
Ta ble 2. Rela tive Ma ss Sp ectr a l In ten sities for NO a n d
The 15N-NMR spectra of this aqueous solution is
compared with that of a 0.02 M solution of the potassium-
(18-crown-6) 15N-nitrite, used in these reactions, in Figure
2. The aqueous solution from the reaction shows only
the resonances for the 15N-benzamide (nondecoupled)
added as a reference. Thus, in agreement with the mass
spectral findings no 15NO2 is generated in these reactions.
In sum, these results establish that an O atom transfer
from PFeNO2 to NO does occur.
NO2 fr om th e 15NO2 Exch a n ge Exp er im en ts
-
sample
NO(30)
15NO(31)
NO2(46)
4a
52
37
31
31
60
32
13
12
12
1
20
13
10
10
5b
6b,c
7b,c
8d
a
b
Reaction gas from stored sample. Reaction gas, just above
liquid, directly analyzed. c Intensities from single ion monitoring
for masses 30, 31, 46, and 47. The NO reference.
d
O2-O3. Three reactants were chosen to demonstrate
the generation of ozone in these reactions. They were
nitrite ion, and the two olefins 2-methyl-2-butene and 2,3-
dimethyl-2-butene. A preliminary account of this work
has recently appeared.20b None of these substances are
taken directly from a reaction mixture via gas tight
syringe and immediately injected into the mass spec-
trometer are shown in Figure 1. In both cases masses
for 14NO and 15NO (30 and 31) are apparent along with
a mass for 14NO2 (46). No mass at 47 was observed in
any sample. A reference spectrum of pure NO, obtained
in similar fashion, establishes the validity of the experi-
mental technique (no oxygen oxidation of NO). The
relative intensities of the relevant ions from four runs
and the NO standard are shown in Table 2.
substrates for the PFeIII/NO2 reaction under argon.
-
However, nitrite23 and the olefins24 are efficient scaven-
gers of ozone. Derived products are well established and,
with the olefins, are uniquely characteristic of ozone. The
splitting of olefins by the sequence ozonolysis followed
by reduction with zinc and acetic acid is the classical
procedure for locating the position of double bonds in
unknown structures. Reactions were conducted in the
general manner outlined above (Table 1) except a thor-
oughly oxygen-purged and saturated system under slight
oxygen pressure (10 mm) was employed. With nitrite as
the ozone trap, following reaction, the mixture was
concentrated and nitrite and nitrate were analyzed in an
aqueous extract. The total recovery of nitrite and nitrate
was 73%, and the nitrate/nitrite ratio was 0.5 ( 0.1
(theoretical, 0.6). The presence of nitrate was confirmed
by infrared analysis (1350 cm-1). With the olefin traps,
a large excess of olefin was employed to compete with
the nitrite oxidation. Following reaction, the mixture
was purged with argon and stirred with powdered zinc
and additional acetic acid for 3 h. The mix was then
distilled at low temperature (<60 °C), and the distillate
was analyzed by GC, GC-MS, and IR. The yield of
There is consistency to these analyses. 15NO is gener-
ated. 15NO2 is not. The ability to see the 15NO in the
gas phase is the result of the equilibrium between iron-
bound NO in solution and the excess nonlabeled NO in
the gas and solution phases (eq 4).
PFe15NO + NO ) PFeNO + 15NO
(4)
For NMR analyses, following completion of a 15NO2
/
-
NO reaction, the mixture was purged with argon, and
effluent gases were passed through D2O. The volume of
the latter was chosen such that the concentration of
nitrite and nitrate from the rapid hydrolysis22 of NO2
(N2O4, eq 5) would be 0.02 M.
2NO2 ) N2O4 H O8 NO2- + NO3- + 2 H+
(5)
2
(23) Kontrakis, P.; Wolfson, J . M.; Bunyaviroch, A.; Fruehlich, S.
E.; Hirono, K.; Mulik, J . D. Anal. Chem. 1993, 65, 209.
(24) March, J . B. Advanced Organic Chemistry, 3rd ed.; J ohn Wiley
and Sons: New York, 1985; pp 1066 ff.
(22) Schwartz, E.; White, W. H. Trace Atmospheric Constituents,
Properties, Transformation and Fates; J ohn Wiley and Sons: New
York, 1983; pp 1-117.

Oxygen Transfers with Porphyrin Iron(III) Nitrite
J . Org. Chem., Vol. 61, No. 18, 1996 6391
acetone from 2,3-dimethyl-2-butene was 100 ( 5% (2 mol
of acetone/mol PFeIII). A corresponding reaction sequence
with 2-methyl-2-butene yielded 1 mol of acetone and 1
mol of acetaldehyde. The latter were confirmed by mass
spectral and infrared analysis. Thus, ozone is trapped
from these reactions in an oxygen atmosphere. These
results are summarized in eqs 6 and 7.
We presume reaction (6) is reversible. Though stable
in air for hours, the iron(II)-NO adduct is not stable in
an oxygen saturated environment.
Th e C-H In ser tion s. Reactions with allyl chloride,
allylbenzene, and cyclohexene are striking in that they
show no products derived from olefin epoxidation. Rather,
the products result from allylic C-H insertion exclu-
sively. With allyl chloride, the first formed halohydrin
loses HCl to yield acrolein directly (eq 8). Note, epichlo-
rohydrin, the product of epoxidation, could not be de-
tected in the reaction mixture, and it is not transformed
to acrolein under these conditions.
F igu r e 3. Repeat spectral scans of NO reaction, (PFeIII)o )
3.7 10-4 M, (NO)o ) 3.7 10-3 M, (NO2-)o ) 7.4 10-3 M:
(- - -), PFeIII, (‚‚‚) at 51 s, final PFeNO lowest spectrum.
phenylacetaldehyde is straightforward and has also been
reported with the PFeIII/ArIO system.1 The successful
reaction with cis-stilbene, but the lack of reaction of the
trans isomer, suggests some steric constraint in ap-
proaching the iron porphyrin -bound NO2. The single
product here is in contrast to the variety of substances
observed with PFeIII/ArIO.10
PFeIIINO2
-
CH2dCHCH2-Cl
8
Kin etics. The general features of the visible spectrum
associated with the conversion of iron(III) porphyrin to
iron(II) nitrosyl adduct by triphenylphosphine have been
previously presented.20 We show in this work only the
diminution of the iron(III) band at 619 nm attendant
upon the addition of NO to the nitrite-iron(III) solution
(Figure 3). A pseudo-first-order plot of the data is shown
in Figure 4. These reactions are overall third order (eq
10). This is in agreement with the formulation given in
(CH2dCHCH(OH)-Cl) f CH2dCHCHO + HCl (8)
The trans-cinnamaldehyde resulting from allylbenzene
is surprising. There are several explanations for this.
We believe the simplest is that this results from a second
O insertion into the rearranged allylic alcohol. The result
suggests an allyl radical intermediate and some radical
character to the O insertion process25 (eq 9).
CH2dCHCH2Ph f HOCH2CHdCHPh f
trans-PhCHdCHCHO (9)
-d(PFeIII)/dt ) d(PFeNO)/dt ) k(PFeIII)(NO2-)(S)
(10)
The reactions with cyclohexenol and cylclohexene again
express the selectivity of the reagent for allylic C-H
insertion rather than epoxidation, and the results with
cumene and toluene are in accord with oxygen insertion
into activated C-H bonds. The olefin 1,5-hexadiene
yields a product that was not fully characterized. How-
ever, the mass spectrum of it is consistent with the
terminal allylic alcohol (1-hydroxy-2,5-hexadiene). The
results with the olefins differ markedly from the general
pattern of epoxidation observed with the peroxide-iron
porphyrin reagents.
eq 1, wherein k ) krateKeq. NO is the fastest reacting
substrate, and the overall third-order rate constant for
it at 22 °C is 52 L2/mol2/s. A typical plot for cumene, one
of the slower reacting hydrocarbons, is shown in Figure
5.
The rate constants for selected substrates are given
in Table 3, along with the relevant bond dissociation
energies. For the relatively slow C-H insertions, kinet-
ics were monitored at 50 °C. The reproducibility of rate
constants evaluated in this manner was (20% run to run.
While all of the C-H oxygen insertions are much
slower than the direct oxygen atom transfers to NO or
triphenylphosphine, there is clearly no correlation of
these rates with available C-H bond dissociation ener-
gies. However, it is unknown to what extent the various
substrates may have influenced the equilibrium between
Ep oxid a tion s. The epoxidation of styrene and sub-
sequent rearrangement of the oxide (upon standing in
the reaction mixture and upon gas chromatography) to
(25) An allylic C-H abstraction followed by HO return to the
terminal carbon.

6392 J . Org. Chem., Vol. 61, No. 18, 1996
F igu r e 4. Pseudo-first-order plot of the data in Figure 3.
F igu r e 5. Pseudo-first-order plot for cumene.
O’Shea et al.
Ta ble 3. Ra te Con sta n ts for Rea ction of Va r iou s
Su bstr a tes w ith Ch lor oir on (III) Octa r th ylp or p h yr in w ith
P ota ssiu m Cr ow n Eth er (18-Cr ow n -6) Nitr ite in
NMP a -1%HOAc a t 50 °C
substrate
k (M-2 s-1
)
103
DC-H (kcal/mol)b
NO
52 000c
1300c,20
190
triphenylphosphine
allyl chloride
propionaldehyde
cumene
89
87
84
87
45
45
benzaldehyde
allyl benzene
cyclohexene
toluene
5.8
3.8
3.5
0.63
88
a
b
NMP ) N-methylpyrrolidone. At 25 °C.37 c At 22 °C.
ride26 indicate it is completely dissociated under the
conditions employed for kinetic analysis. Given that
nitrite is in the rate law, we infer the initial equilibrium
(eq 1) is between porphyrin-iron-solvate and nitrite ion.
Thus, the kinetics are consistent with a rate-limiting step
that entails all three reactants, but we have no evidence,
for the C-H insertion reactions, to suggest there is
significant C-H bond cleavage in the transition state.
Th er m od yn a m ics. At this time, it is not possible to
assess accurately the thermodynamics of these reactions.
The true stability of the free heme-NO adduct, though
high, is unknown. It is considered to be formed “ir-
reversibly”, and the rate of its dissociation has been
approximated as 0.27d The nonavailability of other
relevant thermodynamic factors in this solvent is also
complicating. However, any crude approximation of free
energies for reaction of the diatomic gases, based on the
data at hand (and an assumed dissociation rate of 10-6
s-1 for heme-NO), would indicate the NO and CO
oxidations are thermodynamically favorable, but the O2
to O3 reaction is not. Thus, we assume the equilibrium
in eq 6 may lie far to the left. This would be in keeping
with the fast reaction of O3 with NO in the gas phase.28
Presumably then this reaction occurs because the prod-
ucts are removed by further reaction.
In our view, the porphyrin iron(III)-nitrite salt exhib-
its a broader spectrum of reactivity than the correspond-
ing metalloporphyrin-peroxide systems, but it appears
to be less reactive with hydrocarbons and more selective.
With olefins, it exhibits a preference for allylic and
benzylic C-H oxygen atom insertion over epoxidation.
It shows none of the properties of a peroxide or “HO+”
type of reagent.
The prevalence of nitrite in the environment, and the
ubiquitousness of iron porphyrins in the biosphere,
imparts an environmental relevance to this work. As an
example, tissue or other structural damage could result
in vivo via the ozone reaction. Moreover, the oxidation
and hydrolysis of nitric oxide to nitrite22,29 may well occur
at sites near its generation. That is, NO synthase is a
P450-like hemeprotein,30 and oxygen is a requisite for
reaction. Thus, other oxidative processes proceeding via
PFeIIINO2 are possible. In addition, nitrogen dioxide is
PFeIII and PFeNO2. The starting iron(III) spectrum in
Figure 2 exhibits a λmax at 619 nm. This is shifted from
the 627 nm observed for the chloroiron(III) porphyrin in
this solvent. The relatively high concentrations of por-
(26) Belser, N. O.; Castro, C. E. J . Agric. Food Chem. 1971, 12, 22.
(27) (a) Sharma, V. S.; Traylor, T. G.; Gardiner, G. R.; Mizuka, H.
Biochemistry 1987, 26, 3837. (b) Gibson, Q, H. et al. J . Biol. Chem.
1991, 226, 13097. (c) Petrich, J . W. et al. Biochemistry 1991, 30, 2975.
(d) Hishino, M. et al. J . Am. Chem. Soc. 1993, 115, 9568.
(28) Atkinson, R. et al. Supplement 4. J . Phys. Chem. Ref. Data
1992, 21, 1125.
phyrin (3.7
10-4 M) and nitrite salt (7.4
10-3 M)
suggest that this spectral shift reflects the generation of
the iron(III)-nitrite or it represents a significant dis-
sociation of chloride from the inner coordination sphere
of iron (a salt effect). Potentiometric analyses for chlo-
(29) Wink, D. A.; Darbyshire, J . F.; Nims, R. W.; Saavedva, J . E.;
Ford, P. C. Chem. Res. Toxicol. 1993, 6, 23.
(30) White, K. A.; Marletta, M. A. Biochemistry 1992, 31, 6627.

Oxygen Transfers with Porphyrin Iron(III) Nitrite
J . Org. Chem., Vol. 61, No. 18, 1996 6393
from 1,2-dichloroethane-1,2-13C as previously described.34 All
substances were pure as judged by GC and NMR analysis.
a well-known air pollutant, and it is acutely toxic.31 It
is an interesting question whether or not an entity
similar to the iron-nitrite generated herein could result
from the reaction of NO2 (N2O4) with iron(II) porphyrin
(eq 11):
Nitrite and nitrate analyses were conducted by the proce-
dure of Chow and J ohnstone.35 A Hewlett-Packard 5989A
mass spectrometer fitted with a 5980 gas chromatograph and
data station was employed for GC-MS analyses. For inline
GC-FT-IR-MS, a Hewlett-Packard 5965B FT-IR-unit was
interposed between the GC exit and MS inlet or separate run
PFeIII + NO2- f PFeNO2 7?
9
PFeII + NO2 (11)
1
spectrums for GC-IR and GC-MS were performed. 31P, H,
and 13C NMR analyses were acquired with a General Electric
QE-300 spectrometer. Routine IR and UV-vis spectra were
obtained with a Mattson 4020 Galaxy FT-IR and a Cary 118
°C UV-vis spectrophotometer, respectively.
The X-ray structures32 of sterically encumbered iron
porphyrin-nitrites establish the bonding of the nitro
ligand to iron is through nitrogen:
Potassium crown ether (18-crown-6) 15N-nitrite was pre-
pared from K 15NO2- (Cambridge Isotopes). The original salt
(1.0 g) contained 5-10% of KNO3. It was purified at great
loss by dissolving in ice-water and concentrating the solution
on a rotary evaporator for 3.5 min in a 50 °C bath. The white
crystals were vacuum filtered and dried for 6 h in the rotary
evaporator at 50 °C. The IR (KBr) showed a dominant intense
nitrite band at 1240 cm-1. A trace of nitrate (1352 cm-1) was
apparent, but it did not show in the 15N-NMR (Figure 2).
Reaction with 1,4,7,10,13,16-hexaoxacycloctdecane in metha-
nol yielded the crown ether salt. Final recrystalization was
from benzene. The 15N-NMR of this salt is shown in Figure
2.
PFe
N
O
O
It is not unreasonable that oxygen transfers may
proceed from such an adduct. The driving force for these
reactions, we believe, derives from the enormous stability
of the PFeIINO product.27 These reactions also differ from
those of other transition metal-nitro adducts33 in that
they require no metal or ligand activator (BF3-Et2O,
PdII).
Non r ea ctive Su bstr a tes. The general procedure is il-
lustrated with 2,3-dimethyl-2-butene. A 25 mL three-neck
flask equipped with magnetic stirrer, argon inlet and outlet
stopcocks, and a splayed set of two serum capped stopcocks
fitted to the center neck was thoroughly purged with argon.
The exit stopcock was connected to a mercury trap. Into the
Exp er im en ta l Section
Gen er a l Meth od s. Argon (liquid carbonic) 99.998% was
used throughout. N-Methylpyrrolidone (NMP) was passed
through A-540 alumina to remove traces of peroxides, dried
over sodium sulfate, and distilled under argon (bp 80-81 °C/
10 mm). Chloroiron(III) octaethylporphyrin was prepared
from the porphyrin in the manner previously described.20
Potassium (18-crown-6) nitrite was obtained by refluxing
potassium nitrite with 1,4,7,10,13,16-hexaoxacyclooctadecane
(18-crown-6) in methanol. Following concentration and crys-
tallization, the white solid was recrystallized from benzene,
dried, and stored in an evacuated desiccator over calcium
sulfate. The white needles exhibited IR bands characteristic
flask were placed 0.0125 g (2.0
10-5 mol) of ClFe(III)OEP,
0.090 g (2.6
10-5 mol) of K (18-crown-6) nitrite, 2 mL of
NMP, and 0.04 mL of HOAc. The flask contents were stirred
and purged with argon for 15 min. A 5 mL reactivial, fitted
with serum cap, was charged with 2.2 mL of tetramethyleth-
ylene and purged vigorously with argon until the volume had
diminished to 2.0 mL (15 min.). The olefin was then cannu-
lated into the reaction vessel with hypodermic tubing.
A
separately purged 0.005 cm barrel-shaped spectrophotometric
cuvette equipped with serum-capped argon inlet and outlet
stopcocks was connected in line to the reaction mixture via
the center neck, and gentle purging was continued until a few
drops of sample were removed under argon pressure. (Note
the face of the narrow cell fills readily with a small amount of
liquid due to surface tension.) The visible spectrum cor-
responded to the starting iron(III) porphyrin at 2.5, 20, and
36 h. The sealed reaction mixture had been stirred and heated
for 4 h at 72-74 °C before taking the 36 h sample. At this
point an argon-purged solution of triphenylphosphine (0.05g,
-
of NO2 (1269 cm-1) and crown ether (CO at 1107 cm-1).
Carbon monoxide-13 (99% 13C, Cambridge isotopes) showed
only one 13C resonance in the NMR at δ 183 and was used
without purification. Nitric oxide (Matheson) was conve-
niently held for low-pressure storage by displacing water from
an inverted 1 L separatory funnel that was fitted above the
stopcock with a serum cap. This procedure removes any traces
of NO2(N2O4) that may be present in the NO. Analysis of the
gas, stored in this manner, by gas chromatography upon a 9
M-1/8 in. 8 Å molecular sieves column (30 °C, 30 mL/min
He, thermal conductivity detector) showed only a trace of argon
(syringe purge), 2.1 min, and NO, 5.7 min. The emergence
times of other relevant gases under these conditions are as
follows: N2, 3.2 min; NO2, 3.8 min; and N2O > 30 min.
Dimethyl sulfide, propionaldelhyde, styrene, cyclohexene,
and cyclohexenol were freshly distilled under argon. Tri-
phenylphosphine (Aldrich) was used without purification. It
showed only one 31P resonance in the NMR at δ 8.0 corre-
sponding to an authentic standard. Benzaldehyde, allyl
chloride, cumene, tolueme, cyclohexen-3-ol, 1,5-hexadiene,
2-methyl-2-butene, 2,3-dimethyl-2-butene, cis- and trans-1,2-
dichloroethene, cis-stilbene, epichlorohydrin, and phenyl vinyl
ketone were distilled under argon, and trans-stilbene was
recrystallized before use. Vinyl chloride-1,2-13C was prepared
1.8
10-4 mol) in 0.6 mL of NMP was transferred into the
reaction. The spectrum of a subsequent sample indicated the
beginning generation of the PFeNO adduct. Alternatively,
unreactive substrates (No PFeNO in 36 h at room tempera-
ture) were allowed to react in the presence of the phosphine.
In all cases the spectrum of the iron(II)-NO adduct fully
developed overnight. Unreactive substrates established in this
fashion are given in the text following Table 1.
Ozon e Tr a p p in g Exp er im en ts. (a ) With 2,3-Dim eth yl-
2-bu ten e (Tetr a m eth yleth ylen e). A reaction exactly like
that described above under nonreactive substrates was con-
ducted in an atmosphere of pure oxygen. The iron(II)-NO
adduct is not stable under these conditions, and it was not
observed. At 16 h, the reaction flask was flushed with argon,
and 0.4 g of powdered zinc and 0.5 mL of acetic acid were
added. Stirring was continued for 3 h, whereupon the now
red iron(II) porphyrin solution was opened to air and the flask
(31) Elsayed, N. M., Ed. Effects of Exposure to Nitrogen Dioxide.
Toxicology 1994, 89, 161.
(32) (a) Nasri, H.; Goodwin, J . A.; Scheidt, W. R. Inorg. Chem. 1990,
29, 185. (b) Nasri, H.; Haller, K. J .; Wang, Y.; Huynh, B. H.; Scheidy,
W. R. Inorg. Chem. 1992, 31, 3459.
(33) (a) Solar, J . P.; Mares, F.; Diamond, S. E. Catal. Rev. 1985, 27,
1. (b) Andrews, M. A.; Chang, T.-C-T.; Cheng, C. W.-F. Organometallics
1985, 4, 268. (c) Leising, R. A.; Takeuchi, K. J . Am. Chem. Soc, 1988,
110, 4079. (d) Ercolani, C.; Pavletti, A.; Pennesi, G.; Rossi, G. J . Chem.
Soc., Dalton Trans. 1991, 1317.
(34) Castro, C. E.; Wade, R. S.; Riebeth, D. M.; Bartnicki, E. W.;
Belser, N. O. Environ. Toxicol. Chem. 1992, 11, 757.
(35) Chow, T. J .; J ohnstone, M. S. Anal. Chim. Acta 1962, 27, 441.
(36) Castro, C. E.; Robertson, C.; Davis, H. Biorg. Chem. 1974, 3,
343.
(37) CRC Handbook of Chemistry and Physics; Lide, D. R., Ed.;
CRC: Boston, 1990; Vol. 9, p 96.

6394 J . Org. Chem., Vol. 61, No. 18, 1996
O’Shea et al.
contents were distilled through a short, small, minilab (vacuum)
distillation head into an ice-cooled receiver fitted with a drying
tube. The distillate was analyzed by gas chromatography on
gently purged with argon into 2 mL of argon prepurged water.
Analysis of this aqueous solution for nitrite and nitrate11c
indicated both were produced, but the yield of nitrate never
exceeded 2.4 10-6 mol (52%). Note: NO2 (N2O4) hydrolyzes
a 2-ft
1/8 in. Porapak P column at 85 °C/30 mL/min He. A
-
thermal conductivity detector was employed. Only two peaks
were observed. They corresponded exactly to acetone (1.8 min)
and tetramethylethylene TME (2.6 min). The yield of acetone
assessed with an authentic standard of acetone in TME was
in water to NO3 and NO2-3. NO does not react with water,
-6
but in the presence of oxygen yields only NO2
. (b) More
quantitative runs were accomplished using thermal conductiv-
ity GC analysis on the column described above. For these
reactions, a 20 mL reaction solution containing 2.0 10-5 mol
95 ( 5% (4.0
10-5 mol). Infrared spectra of the reaction
distillate and standard run against TME were identical and
showed a sharp carbonyl absorbtion at 1722 cm-1. (b) With
2-m eth yl-2-bu ten e (Tr im eth yleth ylen e). The distillate of
a reaction identical to the one above, except with 2-methyl
of PFeIIICl and 2.6
10-5 mol of K (crown ether) nitrite in
NMP 1% HOAc was treated with 2 mL of NO (8 10-5 mol).
Two mL of the reaction mixture was transferred under argon
via hypodermic tubing to a 2.5 mL barrel shaped cell with an
0.1 cm light path length. The cell was injected with 0.2 mL of
NO. At 55 min, the reaction in the cell was 50% complete.
GC analysis (above) of the cell gas phase showed argon NO2
and NO. After 1 day, the reaction mixture was analyzed by
GC. The spectrum was that of the PFeIINO adduct. For this
purpose, a 250 µL gas-tight syringe was repeatedly flushed
with argon, and following insertion into the gas phase of the
reaction vessel was repeatedly filled and emptied into the gas
phase. GC analysis showed both unreacted NO and NO2 in
the ratio of (NO/NO2) 3.44. The ratio calculated from the
2-butene, was analyzed by GC on a 5 ft
1/8 in. Porapak T
column at 120 °C/40 mL/min He. Peaks corresponding to
acetaldehyde (4.0 min), trimethylethylene (6-12 min, off
scale), and acetone (14 min) were observed. Comparison with
an authentic standard of acetaldehyde and acetone in 2-meth-
yl-2-butene indicated the yields of acetone and acetaldehyde
were 95 ( 5% and 90 ( 10%, respectively (2.0 10-5 mol each).
Infrared spectra of the reaction distillate matched that of the
standard. Though broadened, the two carbonyls were not
resolved in these spectra (CdO at 1718 cm-1, CHO, 2743 cm-1).
The combined yield of acetone and acetaldehyde estimated
from carbonyl intensities was 90 ( 10%. GC-MS analyses of
these reaction distillates was accomplished on a low load DB-5
column at -25 °C. For runs with 2-methyl-2-butene acetal-
dehyde was cleanly resolved (P, 44, P-H, 43, P-CHO, 15 and
29), but acetone tailed into the trimethylethylene peak.
Nonetheless, it was clearly observable (P, 58, P-CH3, 43). (c)
With Nitr ite. A stirred reaction mixture composed of 0.0125
g (2.0 10-5 mol) of ClFe(III)OEP, 0.018 g (5.2 10-5 mol) of
K (18-crown-6) NO2 in 4 mL of NMP, and 0.04 mL of HOAc
was purged with oxygen for 3 min and sealed. After 16 h the
solution was concentrated to dryness on the rotary evaporator
at 45 °C initially at 20 mm and finally at 1 mm. The dried
black concentrate was triturated with 4.0 mL of water and
vacuum filtered. Two mL of this clear solution was analyzed
for nitrite and nitrate by the procedure of Chow and
J ohnstone.35 The recovery of nitrate and nitrite was 73%, and
the ratio nitrate/nitrite was 0.5 ( 0.1 (theoretical 0.6). The
remaining 2 mL were concentrated to dryness on the rotary
evaporator and analyzed by IR (KBr). Strong bands for nitrate
amount of material charged is 6.2/1.8 ) 3.48. That is, moles
18
of NO2 produced ) mol of PFeIII
)
/ (2 10-5) ) 1.8 10-5
.
.
20
Moles of NO remaining is 8.0 10-5-1.8 10-5 ) 6.2 10-5
CO to CO2. A reaction mixture like that generally de-
scribed above was flushed with 10 mL of 13CO. The flask was
shaken on a rotary shaker. After 1 day, the PFeIINO spectrum
was observed and the reaction vessel was purged with argon
into 2 mL of an argon purged 0.1 N NaOH solution. 13C NMR
analysis of the aqueous solution showed only a small amount
of NMP and HOAc along with a large HCO3- signal at δ 167.
The yield was quantitated by comparison with an NaH13CO3
standard spectrum obtained under identical conditions.
A
blank NMP-1% HOAc solution charged with 13CO, but
without porphyrin or nitrite salt treated in exactly the same
-
manner, exhibited no H13CO3 in the NMR analysis.
Direct gas chromatographic analyses of the reactions of
liquid substrates using flame ionization detection was ac-
complished with the columns indicated below. For GC-MS
and GC-FT-IR-MS, the reaction mixture was first partially
distilled at 20 mm into a cold trap (2-propanol-CO2) to
separate volatiles from NMP. The distillate was analyzed. For
propionic acid, the distillate was treated with trimethylsilyl
chloride and hexamethyldisilazane (Pierce-Trisil Conc.) 0.7
mL, and after the mixture was allowed to sit overnight, 50 µL
of methanol was added before GC-MS analyses. For all
analyses authentic standards were employed for quantitation
and mass spectral analyses. Mass spectra and IR spectra of
products were identical to authentic samples. Yields of
products are reported in Table 1.
30 M GSQ capillary column, at 165°/40 mL/min He: di-
methyl sulfide (1.2 min), dimethyl sulfoxide (20 min); at 100°/
40 mL/min He followed by a temperature ramp at 8 min of
30°/min to 190°: cyclohexene (8.0 min), cyclohexenol (15 min),
cyclohexenone (16.5 min).
30 M, RSL-150 capillary column, at 60°/4 mL/min He:
propionaldyde (1.0 min), acetic acid (2.5 min), propionic acid
(4.0 min); at 20°/4 mL/min He 2°/min temperature ramp
trimethylsilyl propionate (15 min); at 60°/5 mL/min He:
styrene (1.5 min), phenylacetaldehyde (5.0 min), styrene oxide
(5.7 min). For GC-FT-IR-MS, a 30 M DB 5 column was
employed at 60°/5 mL/min followed by a temperature ramp to
230 °C. Styrene oxide and phenylacetaldehyde emerged at
26.1 and 24 min, respectively.
(1350 cm-1), nitrite (1275 cm-1),. and crown ether (1109 cm-1
)
were present. A blank run without the porphyrin conducted
and worked up in the same fashion generated no nitrate. In
other blanks, under the conditions of a and b above, it was
demonstrated the olefins are not oxidized by oxygen alone or
by oxygen in the presence of iron porphyrin (no nitrite). Taken
together with the lack of reactivity of these substrates under
argon, it is established that porphyrin, oxygen, and nitrite are
all essential for these reactions to occur.
Rea ction s. Typically, reactions were conducted in small
flasks (25 mL or less) equipped with argon inlet and outlet
stopcocks, a serum-capped stopcock, and a magnetic stirring
bar. The argon outlet was connected to a mercury trap.
Generally, for gaseous substrates, a solution of chloroiron(III)
octaethylporphyrin (9.6
ether) nitrite (3.84
10-6 mol) and potassium (crown-
10-5 mol) in 6 mL of NMP-1% HOAc
was stirred and purged with argon for 0.5 h. Gaseous
substrates were added via hypodermic syringe under argon.
For liquid substrates 2 mL of substrate were separately purged
before cannulating into 4 mL of NMP-1% HOAc containing
the same amounts of porphyrin and nitrite salt indicated
above. With the solid, triphenylphosphine, 2 mL of an 0.031
M solution in NMP was purged and transferred into the iron
porphyrin/nitrite solution to begin the reaction. Thus, for most
reactions (PFeIIICl)o ) 1.6 10-3 M and (K(crown ether)NO2)o
Separate GC analysis of pure styrene oxide always showed
some phenylacetaldehyde upon chromatography (5% or less).
When the solution of styrene oxide in NMP-1% HOAc was
allowed to stand for 2 days, it exhibited a 40% conversion to
phenylacetaldehyde. A reaction with cyclohexenol yielded
cyclohexenone quantitatively. This reaction is faster than the
conversion of cyclohexene to cyclohexenone.
) 6.4
All reactions were conducted at least three and usually five
times.
10-3 M.
NO to NO2. (a) A reaction that yielded only qualitative
results was initiated by the injection of 0.4 mL of NO gas into
the above reaction mixture. Reaction proceeded immediately.
A sample removed for spectral analysis exhibited the PFeII-
NO spectrum.20 The reaction mixture and product gases were
Triphenylphosphine and triphenylphosphine oxide were
analyzed by 31P NMR as indicated previously indicated.20a

Oxygen Transfers with Porphyrin Iron(III) Nitrite
J . Org. Chem., Vol. 61, No. 18, 1996 6395
Allyl Ch lor id e. The red PFeNO reaction mixture was
distilled after 36 h to bp 70 °C. Acrolein was quantitated in
the distillate on a 30 m GSQ capillary column, yield 1.06
10-6 (112%). The mass spectrum of the 4.7 min peak matched
exactly that of an authentic standard (P, 56, PH, 55, PCHO,
27). Upon standing under argon (without workup) this reac-
tion slowly returned to the original spectrum of PFeIIICl (see
kinetics). We presume this oxidation is the result of the slower
reaction of excess allyl chloride with PFe(II) derived from the
slow dissociating PFeNO adduct.33 This would allow for some
additional PFeNO2 generation and additional oxidation, thus
accounting for the consistently greater than 100% yields of
product in this case. Note: epichlorohydrin (not detected) was
not converted to acrolein under reaction conditions.
Allylben zen e. After 36 h, the reaction mixture was
distilled at 110 °C/20 mm. trans-Cinnamaldehyde was quan-
titated from the distillate on a 30 m RSL-150 column, yield
4.8 10-6 mol (100%). The 11.1 min peak showed a MS and
IR spectrum identical with authentic aldehyde. IR: CO,1710,
CdC, 1624, CHO, 2735, 2805 cm-1. MS: P, 132, PH (domi-
nant) 131, PCHO, 103.
°C/20 mm. Benzyl trimethylsilyl ether was quantitated on the
RSL-150 column. The 5 min peak (30 °C/10 mL/min He)
coincided with a sample of authentic ether, yield, 9.6
mol (100%).
10-6
cis-Stilben e. Two mL of a 1.87 M solution of olefin in
benzene was added under argon to 4 mL of NMP-1% HOAc
containing the porphyrin and nitrite salt. After 36 h, the
reaction was distilled. The fraction boiling at 90-110 °C/10
mm was analyzed on the RSL-150 column (70-220 °C/10 mL/
min He). The 15 min peak comerged with cis-stilbene oxide
and had a mass spectrum identical with an authentic standard
(P, 196, PO, 180). In particular, no trans-stilbene oxide (17
min) was detected. The trans-olefin is inert.
1,5-Hexa d ien e. After 36 h, the reaction mixture was
distilled at 130 °C. Analysis on the RSL-150 column showed
a product peak at 8.5 min, the mass spectrum of which
corresponded to cis- or trans-CH2dCHCH2CHdCHCH2OH (P,
PCH2OH, i.e., 98, 67). This structure is unconfirmed.
Kin etics. For kinetic analysis small-scale reactions were
conducted in 0.1 or 1.0 cm cuvettes fitted with a serum-capped
stopcock. NO, and the previously reported triphenylphosphine
reaction, were examined at room temperature. However, the
much slower reacting hydrocarbons and aldehydes were
studied at 50 °C. For this purpose, initial solutions of iron
porphyrin and nitrite salt were prepared and thoroughly
purged with argon at room temperature. Spectra were re-
corded before and after the addition of nitrite. A shift in the
λmax of the iron(III) band from 627 to 619 nm was attendant
upon the addition of nitrite, but there was no significant
change in the spectrum upon warming the cuvette to 50 °C.
Cells were thermostated at 50 °C, and reaction was initiated
by injection of substrate. For initial slope analysis, each
reactant was varied by a factor of at least 10. This established
the general rate law. Typical concentrations were in the range
Ben za ld eh yd e. After 36 h, the reaction mixture was
treated with 80 µL of trisil concentrate. Excess reagent was
decomposed with methanol, and the reaction mixture was
distilled at 50-80 °C/7 mm. Trimethylsilyl benzoate was
quantitated on the RSL-150 column, yield 9.6
10-6 mol
(100%). A sample of the entire reaction mixture was subjected
to GC-MS analysis prior to derivitization. The 19 min peak
corresponded exactly to benzoic acid (P, 122, POH, 105,
PCO2H, 77).
Cu m en e. After 36 h, the reaction mixture was distilled at
110 °C/20 mm cumyl alcohol was quantitated on the RSL-150
column, yield, 9.6 10-5 mol (100%). The mass spectrum of
the 30 min peak corresponded exactly to a standard, (P, 136,
PCH3, 121).
Tolu en e. For GC-MS analysis the reaction was distilled
at 36 h. The MS of the 31 min peak on the RSL-150 column
corresponded to authentic benzyl alcohol (P, 108, POH, 91,
PCH2OH, 77). For quantitation, a reaction at 36 h was treated
with 80 µL of trisil concentrate and, after 10 min, with 0.5
mL of methanol. The mixture was then distilled to bp 110
-
(PFeIII
)
12-4
10-4 M , (NO2
) 0.01-1 M, (substrate)o
o
o
0.01-1 M for pseudo-first-order runs. At least three runs were
made with each substrate. The reproducibility run to run was
( 20%.
J O960546X