Sarah A Cook1, Joseph W Ziller, A S Borovik. 1. Department of Chemistry, University of California-Irvine , 1102 Natural Sciences II, Irvine, California 92697, United States.
Abstract
High-valent iron species are known to act as powerful oxidants in both natural and synthetic systems. While biological enzymes have evolved to prevent self-oxidation by these highly reactive species, development of organic ligand frameworks that are capable of supporting a high-valent iron center remains a challenge in synthetic chemistry. We describe here the reactivity of an Fe(II) complex that is supported by a tripodal sulfonamide ligand with both dioxygen and an oxygen-atom transfer reagent, 4-methylmorpholine-N-oxide (NMO). An Fe(III)-hydroxide complex is obtained from reaction with dioxygen, while NMO gives an Fe(III)-alkoxide product resulting from activation of a C-H bond of the ligand. Inclusion of Ca(2+) ions in the reaction with NMO prevented this ligand activation and resulted in isolation of an Fe(III)-hydroxide complex in which the Ca(2+) ion is coordinated to the tripodal sulfonamide ligand and the hydroxo ligand. Modification of the ligand allowed the Fe(III)-hydroxide complex to be isolated from NMO in the absence of Ca(2+) ions, and a C-H bond of an external substrate could be activated during the reaction. This study highlights the importance of robust ligand design in the development of synthetic catalysts that utilize a high-valent iron center.
High-valent iron species are known to act as powerful oxidants in both natural and synthetic systems. While biological enzymes have evolved to prevent self-oxidation by these highly reactive species, development of organic ligand frameworks that are capable of supporting a high-valent iron center remains a challenge in synthetic chemistry. We describe here the reactivity of an Fe(II) complex that is supported by a tripodal sulfonamide ligand with both dioxygen and an oxygen-atom transfer reagent, 4-methylmorpholine-N-oxide (NMO). An Fe(III)-hydroxide complex is obtained from reaction with dioxygen, while NMO gives an Fe(III)-alkoxide product resulting from activation of a C-H bond of the ligand. Inclusion of Ca(2+) ions in the reaction with NMO prevented this ligand activation and resulted in isolation of an Fe(III)-hydroxide complex in which the Ca(2+) ion is coordinated to the tripodal sulfonamide ligand and the hydroxo ligand. Modification of the ligand allowed the Fe(III)-hydroxide complex to be isolated from NMO in the absence of Ca(2+) ions, and a C-H bond of an external substrate could be activated during the reaction. This study highlights the importance of robust ligand design in the development of synthetic catalysts that utilize a high-valent iron center.
High-valent iron species
are often reactive intermediates and are understood to be involved
in C–H bond functionalization of a variety of substrates.
For example, nonheme iron-containing monooxygenases utilize a mononuclear
iron(IV)–oxo unit as the active species to perform a diverse
set of reactions, including hydroxylation, halogenation, desaturation,
and epoxidation.[1] These diverse and important
reactions have made nonheme high-valent iron complexes a target for
synthetic chemists, both for understanding the functional aspects
of active sites in enzymes and for developing new synthetic oxidants
for chemical transformations such as C–H activation.[2,3]One challenge associated with preparing complexes that can
support oxidized Fe centers and harnessing their reactivity for substrate
activation is designing supporting ligands that can withstand the
highly reactive nature of these species. Indeed, several well-characterized
Fe(IV)–oxo[4] and Fe(IV)–imido[5] species have been shown to undergo self-decay
via reactivity with the supporting ligand. In this report, we describe
C–H activation of the tripodal ligand N,N′,N″-[2,2′,2″-nitrilotris(ethane-2,1-diyl)]tris(2,4,6-trimethylbenzenesulfonamido)
([MST]3– = mesityl sulfonamide tripod, Chart 1) upon oxidation of its Fe(II) complex with an oxygen-atom
(O-atom) transfer reagent to generate an Fe(III)–alkoxide species.
The ligand [MST]3– contains mesityl groups whose
methyl positions are susceptible to oxidation. Redesign of the ligand
to remove the methyl groups that are positioned closest to the metal
center (Chart 1, [TST]3– =
tolyl sulfonamide tripod) allowed us to observe an intermediate species,
possibly an Fe(IV)–oxo complex, which was capable of activating
C–H bonds on external substrates to give an Fe(III)–hydroxide
species.
Chart 1
Ligand Derivatives Described in This Report
Experimental Section
General
Methods
Syntheses of metal complexes were completed under
a nitrogen atmosphere in a VAC drybox. Solvents were sparged with
argon and dried over columns containing Q-5 and molecular sieves.
All reagents were purchased from commercial suppliers and used as
received unless otherwise noted. Sodium hydride as a 30% suspension
in mineral oil was filtered and washed five times each with Et2O and pentane and dried under vacuum. H3MST,[6a] NMe4[FeIIMST],[6b] and H3TST[7] were prepared according to literature procedures.
Complex Syntheses
and Reactivity Studies
NMe4[FeIIMST(OH2)]
A suspension of NMe4[FeIIMST] (0.100 g, 0.12 mmol) in 4 mL of THF at room temperature was
treated with 3 μL (0.17 mmol) of H2O via syringe,
causing the solution to become homogeneous. The reaction was stirred
vigorously for 5 min, after which the solvent was removed under vacuum.
The resulting residue was redissolved in dichloromethane (DCM) and
filtered through Celite to remove fine particulate solid. The product
was crystallized from the DCM filtrate via diffusion of pentane to
give 95 mg (93%) of NMe4[FeIIMST(OH2)] as colorless needle crystals. Anal. Calcd for NMe4[FeIIMST(OH2)], C37H59N5O7S3Fe: C, 53.03; H, 7.10; N, 8.36. Found:
C, 53.00; H, 7.03; N, 8.38. FTIR (KBr disc, cm–1, selected bands, strong (s), medium (m), weak (w)): 3255 (m), 2968
(m), 2852 (m), 1604 (w), 1491 (m), 1255 (s), 1127 (s), 974 (s), 812
(s), 654 (s).
NMe4[FeIII–O–MST]
A solution of NMe4[n class="Chemical">FeIIMST] (0.250 g, 0.30
mmol) in 15 mL of DCM at room temperature was treated with a solution
of NMO (72 mg, 6.1 mmol) in 5 mL of DCM, resulting in an immediate
color change to red. The reaction was stirred for 3 h and then filtered
through Celite. The product was recrystallized twice from DCM via
pentane diffusion to give 128 mg (50%) of orange crystals. Anal. Calcd
for NMe4[FeIII–O–MST]·0.5CH2Cl2, C37.5H57N5O7S3Cl2Fe: C, 51.34; H, 6.55; N,
7.98. Found: C, 51.69; H, 6.51; N, 7.77. FTIR (KBr disc, cm–1, selected bands, strong (s), medium (m), weak (w)): 3029 (w), 2932
(m), 2855 (m), 1604 (m), 1488 (m), 1292 (s), 1136 (s), 958 (s), 797
(s), 654 (s). λmax, nm (DCM, ε, M–1 cm–1): 351 (7500). EPR (1:1 DCM:THF, 77 K): g = 9.0, 4.2.
15-crown-5⊃Ca–(μ-OH)–FeIIIMST]OTf
A solution of NMe4[n class="Chemical">FeIIMST] (50 mg, 0.061 mmol) and Ca⊃15-crown-5(OTf)2 (37 mg, 0.067 mmol) in 3 mL of DCM at room temperature was treated
with a solution of NMO (14 mg, 0.12 mmol) in 2 mL of DCM, resulting
in an immediate color change to orange. After 5 h, the reaction mixture
was filtered through Celite and the product was recrystallized twice
via pentane diffusion to give 60 mg (85%) of yellow-orange crystals.
Anal. Calcd for [15-crown-5⊃CaII–(μ-OH)–FeIIIMST]OTf·0.5CH2Cl2, C44.5H67CaClF3N4O15S4Fe: C, 44.00; H, 5.56; N, 4.61. Found: C, 44.13; H, 5.31; N, 4.50.
FTIR (KBr disc, cm–1, selected bands, strong (s),
medium (m), weak (w)): 3379 (m), 2937 (m), 2868 (m), 1604 (w), 1266
(s), 1144 (s), 1090 (s), 1031 (s), 955 (s), 811 (s), 659 (s), 638
(s). λmax, nm (DCM, ε, M–1 cm–1): 383 (6000). EPR (1:1 DCM:THF, 77 K): g = 9.4, 4.7, 4.2
NMe4[FeIITST(OH2)]
A solution of H3TST (0.200 g, 0.33 mmol) in 4 mL of anhydrous dimethylacetimide (DMA)
at room temperature was treated with 3 equiv of solid NaH (24 mg,
1.0 mmol), causing H2 evolution and precipitation of the
deprotonated ligand. After the evolution of H2 gas ceased,
Fe(OAc)2 (57 mg, 0.33 mmol) and NMe4OAc (44
mg, 0.33 mmol) were added to the heterogeneous mixture, which was
then stirred for 3 h. One equivalent (6 μL) of H2O was then added via syringe and the reaction mixture filtered through
a medium-porosity frit to remove 3 equiv of insoluble NaOAc (79 mg,
0.96 mmol). Vapor diffusion of Et2O into the pale yellow
filtrate gave the product as pale blue crystals in 90% yield. Anal.
Calcd for NMe4[FeIITST(OH2)], C31H47N5O7S3Fe:
C, 49.40; H, 6.28; N, 9.29. Found: C, 49.13; H, 6.23; N, 9.14. FTIR
(KBr disc, cm–1, selected bands, strong (s), medium
(m), weak (w)): 3257 (m), 3037 (w), 2896 (w), 2845 (m), 1599 (w),
1494 (m), 1246 (s), 1129 (s), 973 (s), 815 (s), 663 (s), 597 (m),
555 (s).
NMe4[FeIIITST(OH)]
A solution of NMe4[n class="Chemical">FeIITST(H2O)]
(0.100 g, 0.13 mmol) in 6 mL of DCM at room temperature was treated
with a solution of NMO (15 mg, 0.13 mmol) in 2 mL of DCM, causing
an immediate color change to red. The reaction was stirred for 4 h,
during which time the color faded to orange. After filtering through
Celite, the product was recrystallized twice by layering the DCM solution
under Et2O to give 60 mg (60%) of yellow-orange crystals.
Analy. Calcd for NMe4[FeIIITST(OH)], C31H46N5O7S3Fe: C, 49.46;
H, 6.16; N, 9.30. Found: C, 49.55; H, 6.00; N, 8.98. FTIR (KBr disc,
cm–1, selected bands, strong (s), medium (m), weak
(w)): 3450 (m), 3036 (w), 2962 (w), 2859 (m), 1599 (w), 1490 (m),
1270 (s), 1138 (s), 1091 (s), 962 (s), 816 (s), 666 (s), 553 (s).
λmax, nm (DCM, ε, M–1 cm–1): 355 (5500). EPR (1:1 DCM:THF, 77 K): g = 9.7, 4.3.
Ligand Isolation Studies
In a typical
experiment, a DCM solution of the reaction mixture was brought out
of the dry box and extracted with 1 M HCl. The organic layer was washed
with brine, dried over MgSO4, and filtered. The DCM solution
was then passed through a plug of silica, which was flushed with additional
DCM. The ligand was eluted from the silica with 5% MeOH in DCM and
the solvent removed under vacuum. The products were analyzed by electrospray
ionization mass spectrometry (ESI-MS) and NMR spectroscopy.
Electronic
Absorption Studies
In a typical experiment, a 0.2 mM stock
solution of the metal complex was prepared in the glove box, and 3
mL of the solution was transferred to a quartz cuvette, which was
sealed with a rubber septum. The cuvette was brought out of the glove
box and allowed to equilibrate in the sample holder at 25 °C
for 10 min before NMO was added as a 30 mM solution via syringe.
Substrate Oxidation Studies
In a typical experiment, a solution
of NMO was added in one portion to a solution containing the FeII complex and DHA. After 3 h, the solvent was evaporated to
dryness and the resulting yellow residue stirred in Et2O. Et2O was then filtered through Celite, passed through
a silica plug, and evaporated to give the DHA products as an off-white
residue. The residue was redissolved in CDCl3, and the
ratio of products was determined by integration of their signals in
the 1H NMR spectrum. The Fe-containing products were redissolved
in DCM and recrystallized by Et2O layering.
Physical Methods
Electronic absorption spectra were recorded in a 1.0 or 0.1 cm
quartz cuvette on a Cary 50 spectrophotometer or an 8453 Agilent UV–vis
spectrometer equipped with an Unisoku Unispeks cryostat. Negative
mode electrospray ionization mass spectra were collected using a Micromass
MS Technologies LCT Premier Mass Spectrometer. X-band (9.28 GHz) EPR
spectra were collected as frozen solutions using a Bruker EMX spectrometer
equipped with an ER041XG microwave bridge. IR spectra were recorded
on a Varian 800 Scimitar Series FTIR spectrometer as KBr disks or
as a solution using a Beckman liquid IR cell.
X-ray Crystallographic
Methods
A Bruker SMART APEX II diffractometer was used to
collect all data. The APEX2[8] program package
was used to determine the unit-cell parameters and for data collections.
The raw frame data was processed using SAINT[9] and SADABS[10] to yield the reflection
data file. Subsequent calculations were carried out using the SHELXTL[11] program. Structures were solved by direct methods
and refined on F2 by full-matrix least-squares
techniques. Analytical scattering factors[12] for neutral atoms were used throughout the analysis. Hydrogen atoms
were included using a riding model. Hydrogenatoms H(1) of NMe4[FeIIITST(OH)] and H(1) and H(2) NMe4[FeIIMST(OH2)] were located from a difference-Fourier
map and refined (x, y, z, and Uiso). Data sets of both NMe4[FeIIITST(OH)] and NMe4[FeIII–O–MST] contained several high residuals in the final
difference-Fourier map. It was not possible to determine the nature
of the residuals, although it is probable that a pentane or DCM solvent
molecule was present. The SQUEEZE routine in the PLATON[13] program package was used to account for the
electrons in the solvent-accessible voids. In the NMe4[FeIIMST(OH2)] structure, the (NMe4)+ counterion was disordered. Carbon atoms C(35)–C(40)
were included using multiple components with partial site-occupancy
factors.
Results and Discussion
Reactivity of the [FeIIMST]− Complex with Dioxygen
We
previously reported the preparation of bimetallic complexes that
are supported by [MST]3–.[6,14] We
discovered that treating NMe4[FeIIMST] or NMe4[MnIIMST] with dioxygen in the presence of a second
metal ion resulted in formation of Fe(III)– or Mn(III)–hydroxide
complexes with the second metal ion coordinated through the hydroxide
ligand and two of the sulfonamido ligand arms of [MST]3– (Scheme 1).[6] In
the absence of a secondary metal ion, this reaction is sluggish, with
the Mn(II) complex reacting so slowly that completion of the reaction
could not be observed. The Fe(II) complex reacts faster than the Mn(II)
complex, and analysis of the reaction mixture after 5 h suggested
formation of the analogous monometallic NMe4[FeIIIMST(OH)] salt. The electrospray ionization mass spectrum (ESI-MS)
contained a negative ion peak whose mass-to-charge ratio (m/z) of 762 corresponds to the formulation
[FeMST(OH)]− (Figure S1, Supporting
Information), and based on charge balance, the Fe center must
be in the 3+ oxidation state. This oxidation state is supported by
EPR spectroscopy, which exhibits a rhombic signal with g values of 4.2 and 8.6 that are consistent with a high-spin Fe(III)
center (Figure S2, Supporting Information). Furthermore, a broad band was observed at a frequency of 3463
cm–1 in the FTIR spectrum of the reaction mixture,
which is consistent with a vibration from a hydroxo ligand (Figure 1, solid black trace, and Figure S3, Supporting Information).[15]
Scheme 1
Preparative Route to Heterobimetallic Complexes[6]
Figure 1
FTIR spectra showing
the change in the product from reaction of NMe4[FeIIMST] with O2. Spectra were collected on a 24 mM
DCM solution over the course of 7 days (dashed black) with the first
spectrum collected after 5 h (solid black).
FTIR spectra showing
the change in the product from reaction of NMe4[n class="Chemical">FeIIMST] with O2. Spectra were collected on a 24 mM
DCM solution over the course of 7 days (dashed black) with the first
spectrum collected after 5 h (solid black).
Attempts to crystallize the proposed Fe(III)–hydroxide
product instead resulted in isolation of the Fe(II)–aquo salt,
NMe4[FeIIMST(OH2)], whose molecular
structure was determined by XRD methods (Figure S4, Supporting Information). The isolated product presumably results
from instability of the initial Fe(III)–hydroxide species in
solution, which is supported by an FTIR study on the reaction mixture
after removal of the excess dioxygen. Over the course of 7 days, the
O–H vibration from the putative [FeIIIMST(OH)]− complex decreased in intensity with a concomitant
increase in a new O–H vibration at 3300 cm–1 that matches the vibration observed for the aquo ligand of independently
prepared [FeIIMST(OH2)]− (Figures 1, S5, and S6, Supporting Information). When 0.5 equiv of diphenylhydrazine (DPH) were added to the initial
reaction mixture, the O–H band corresponding to the hydroxo
ligand and the N–H vibrations of DPH were replaced by the O–H
band of the Fe(II)–aquo complex within 2 h.[16] This increase in the rate of [FeIIMST(OH2)]− formation suggests that the observed
instability of NMe4[FeIIIMST(OH)] results from
H-atom abstraction from the solvent or from an external substrate
to give NMe4[FeIIMST(OH2)].
Reactivity
of the [FeIIMST]− Complex with Oxygen-Atom
Transfer Reagents
The bimetallic Fe(III)–hydroxide
complexes could also be generated using O-atom transfer reagents such
as 4-methylmorpholine-N-oxide (NMO) in place of
dioxygen. For instance, reaction of NMe4[FeIIMST] and CaII⊃15-crown-5(OTf)2 with
2 equiv[17] of NMO gave [(15-crown-5)CaII–(μ-OH)–FeIIIMST]OTf in 85%
crystalline yield, which is similar to the yield obtained from dioxygen
(60–70%, eq 1).[6b] In addition, the [FeIIMST]− complex
reacts with 2 equiv of NMO in the
absence of a second metal ion, as evidenced by a rapid color change
to orange. On the basis of the match in reactivity between dioxygen
and NMO in the bimetallic systems, the analogous [FeIIIMST(OH)]− complex was predicted to be the major
product formed from this reaction (eq 2). Evidence
for this product was again observed by FTIR spectroscopy,
in which an identical O–H vibration at 3460 cm–1 was replaced by a second O–H vibration at 3300 cm–1 after 9 days (Figures S7–S9, Supporting
Information). However, the ESI mass spectrum contained only
a minor peak corresponding to the [FeIIIMST(OH)]− ion, while a dominant ion peak was observed two mass units lower
at a m/z of 760 (Figure S10, Supporting Information). Structural determination
of the crystallized product via X-ray diffraction methods revealed
the source of the dominant molecular ion: an ortho methyl group on
an arm of the ligand had been hydroxylated to form an Fe(III) product
with a coordinated alkoxide ([FeIII–O–MST]−, Figure 2). This Fe–alkoxide
species was the major species formed and was isolated in 50% crystalline
yield.
Figure 2
Thermal ellipsoid diagram depicting [FeIII–O–MST]−. Ellipsoids are drawn at the 50% probability level,
and all hydrogen atoms are omitted for clarity. Selected bond lengths
(Angstroms) and angles (degrees): Fe1–N1, 2.358(2); Fe1–N2,
2.034(2); Fe1–N3, 2.030(2); Fe1–N4, 1.999(2); Fe1–O1,
1.805(1); O1–Fe1–N1, 171.09(6); N2–Fe1–N3,
122.11(7); N2–Fe1–N4, 115.18(7); N3–Fe1–N4,
108.08(7).
Thermal ellipsoid diagram depicting [FeIII–O–n class="Gene">MST]−. Ellipsoids are drawn at the 50% probability level,
and all hydrogen atoms are omitted for clarity. Selected bond lengths
(Angstroms) and angles (degrees): Fe1–N1, 2.358(2); Fe1–N2,
2.034(2); Fe1–N3, 2.030(2); Fe1–N4, 1.999(2); Fe1–O1,
1.805(1); O1–Fe1–N1, 171.09(6); N2–Fe1–N3,
122.11(7); N2–Fe1–N4, 115.18(7); N3–Fe1–N4,
108.08(7).
The Fe(III) center in [FeIII–O–MST]− has an N4O primary coordination environment in which all five donors are provided
by the [MST]3– ligand. The anionic nitrogen atoms
and neutral apical nitrogendonor provide the base of a distorted
trigonal bipyramid (τ = 0.82)[18] with
an average Fe–Neq distance of 2.021(2) Å and
an Fe–N1 distance of 2.358(2) Å. The deprotonated hydroxyl
group that resulted from functionalization of a mesityl group completes
the coordination sphere of the Fe center. In order to accommodate
the binding of the hydroxyl group, the functionalized mesityl group
twists above the Fe–Neq plane, whereas those on
the two unfunctionalized ligand arms point outward from the complex.
The oxygen atom of the deprotonated hydroxyl group tilts away from
the Fe–N1 vector toward N4 with an O1–Fe–N1 bond
angle of 171.09(6)° and an Fe–O1 distance of 1.805(1)
Å (see Table S2, Supporting Information, for additional metrical parameters).The reaction of NMe4[FeIIMST] with NMO was further probed by determining
the extent of oxidation of the [MST]3– ligand after
isolation of the ligand from the complex. The metal ion was removed
from the ligand in an aqueous acid workup to give a mixture of H3MST and oxidized ligand species, which were recovered in a
combined yield of 82% (eq 3). Analysis of the
ligand products by NMR spectroscopy provided an estimate of the yields
of the Fe(III)– hydroxide and Fe(III)–alkoxide products.
The hydroxylated ligand 1, which is isolated from [FeIII–O–MST]−, makes up 56% of the ligand products, while unfunctionalized H3MST, which we propose is isolated from [FeIIIMST(OH)]− or [FeIIMST(OH2)]−, makes up 25% (Chart 2). The remaining 19%
of the products from [MST]3– consists of three species
that have been further oxidized beyond hydroxylation of the ortho
metal group. Two of the species are assigned to cyclization of one
arm through the sulfonamidenitrogen atom and the hydroxylated ortho
carbon atom of the activated ligand arm. Of these two products, one
retained the hydroxyl functionality (12%, 2) and the
other was further oxidized to the carbonyl (5%, 3). In
the final ligand species, two of the ligand arms contained the cyclized
carbonyl product (2%, 4). Note that no further ligand
oxidation is observed from reaction of pure [FeIII–O–MST]− with NMO; only the singly hydroxylated ligand product 2 was observed after isolation of the ligand products from
the reaction (eq 4). Moreover, free H3MST and 2 show no reactivity with NMO, which suggests
that the ligand must be coordinated to the metal center in order to
become activated.
Chart 2
Ligand
Products Isolated from the Reaction Shown in Eq 3
Ligand products from reaction of NMe4[FeIIMST] with NMO in the presence of Ca2+ ions (eq 1) were also analyzed after extraction of the ligand
from the metal complex. None of the oxidized ligand products shown
in Chart 2 were observed. Other than a small
amount of an unidentified ligand product (<5% overall), only unfunctionalized
H3MST was isolated. One possibility for this lack of ligand
oxidation could be preassociation of a Ca2+ ion within
the secondary coordination sphere of the [FeIIMST]− complex, which positions the mesityl groups far enough
away from the metal center to prevent hydroxylation upon addition
of NMO.
Reactivity of the [FeIITST]− Complex
Oxidation of the ligand observed from reaction of NMe4[n class="Chemical">FeIIMST] with NMO highlights a limitation of complexes
of [MST]3– in oxidation reactions, and modification
of the ligand is required to prevent this undesirable reactivity.
We therefore replaced the susceptible mesityl groups of the ligand
with tolyl groups ([TST]3–) in order to eliminate
this pathway and redirect the reactivity toward external substrates
with C–H bonds.[19]
Unlike [FeIIMST]−, the four-coordinate [FeIITST]− complex could not be cleanly isolated due
to partial coordination of adventitious water to form [FeIITST(OH2)]−, which could not be removed
during purification. Therefore, [FeIITST(OH2)]− was independently prepared and used as the
starting complexes for all oxidation reactions. In order to verify
that the presence of an aquo ligand would not influence the reactivity
relative to [FeIIMST]−, the analogous
[FeIIMST(OH2)]− complex was
also prepared and its oxidation with O2 and NMO were investigated
(eqs 5a and 5b). The [FeIIMST(OH2)]− complex reacted
with O2 and NMO in a manner similar to that of [FeIIMST]−, and a similar ratio of ligand oxidation
products was observed from reaction with excess NMO.[20]Oxidation of the [FeIITST(OH2)]− complex with NMO resulted in initial formation
of a red species that faded to orange over the course of several minutes
(eq 6). When this reaction was monitored optically, an intermediate with peaks at 380 and 895
nm appeared and then converted to a new species containing a single
optical feature at 352 nm (Figure 3). Spectroscopic
and analytical data are consistent with this final species being [FeIIITST(OH)]−, which was isolated in 60% crystalline
yield.[21] In addition, no evidence for
formation of the hydroxylated ligand product was observed in the analysis
of the isolated ligand after the oxidation reaction of [FeIITST]−.
Figure 3
Electronic absorption spectra for oxidation
of a 0.2 mM DCM solution of NMe4[FeIITST(H2O)] by NMO at 25 °C showing (A) conversion of the Fe(II)
complex (dashed black) to the intermediate species (solid black) and
(B) further reaction of the intermediate to the final NMe4[FeIIITST(OH)] product (dotted black). (Inset of B) Decay
of the low-energy band in a 5 mM DCM solution.
Electronic absorption spectra for oxidation
of a 0.2 mM DCM solution of NMe4[FeIITST(H2O)] by NMO at 25 °C showing (A) conversion of the Fe(II)
complex (dashed black) to the intermediate species (solid black) and
(B) further reaction of the intermediate to the final NMe4[FeIIITST(OH)] product (dotted black). (Inset of B) Decay
of the low-energy band in a 5 mM DCM solution.The molecular structure of NMe4[FeIIITST(OH)], determined using X-ray diffraction methods, revealed similar
coordination properties of the Fe center as the Fe(III)–alkoxide
species of [MST]3– (Figure 4). The [FeIIITST(OH)]− ion consists
of a five-coordinate Fe(III) center in trigonal bipyramidal geometry
(τ = 0.82) that is established by the four nitrogendonors of
[TST]3– and a terminal hydroxo ligand. The average
Fe–Neq distance of 2.035(2) Å and the Fe–N1
distance of 2.329(2) Å are similar to the distances in [FeIII–O–MST]− (2.021 and 2.358
Å). The Fe–O1 distance is slightly longer at 1.831(1)
Å compared to 1.803 Å, and the oxygen atom of the hydroxo
ligand is also tilted out of the Fe–N1 vector with an O1–Fe–N1
angle of 173.56(6)°. In contrast to the Fe(III)–alkoxide
complex of MST3–, one sulfonamido oxygen atom on
each of the three ligand arms of [FeIIITST(OH)]− points nearly parallel to the Fe1–O1 vector, forming a negatively
polarized fence around the hydroxo ligand. A short distance (2.743
Å) between the oxygen atom of the hydroxo ligand and one of these
sulfonamideoxygen atoms (O2) is suggestive of an intramolecular hydrogen-bonding
interaction between these two groups (see Table S3, Supporting Information, for additional metrical parameters).
Figure 4
Thermal
ellipsoid diagram depicting [FeIIITST(OH)]− (bond lengths in Angstroms and angles in degrees). Ellipsoids are
drawn at the 50% probability level, and only the hydroxo hydrogen
atom is shown. The NMe4+ counterion is omitted
for clarity. Fe1–N1, 2.329(2); Fe1–N2, 2.053(2); Fe1–N3,
2.031(2); Fe1–N4, 2.022(2); Fe1–O1, 1.831(1); O1···O2,
2.743; O1–Fe1–N1, 173.56(6); N2–Fe1–N3,
124.46(7); N2–Fe1–N4, 110.42(6); N3–Fe1–N4,
111.26(6).
Thermal
ellipsoid diagram depicting [FeIIITST(OH)]− (bond lengths in Angstroms and angles in degrees). Ellipsoids are
drawn at the 50% probability level, and only the hydroxohydrogen
atom is shown. The NMe4+ counterion is omitted
for clarity. Fe1–N1, 2.329(2); Fe1–N2, 2.053(2); Fe1–N3,
2.031(2); Fe1–N4, 2.022(2); Fe1–O1, 1.831(1); O1···O2,
2.743; O1–Fe1–N1, 173.56(6); N2–Fe1–N3,
124.46(7); N2–Fe1–N4, 110.42(6); N3–Fe1–N4,
111.26(6).The mechanism for oxidation of
[FeIITST(OH2)]− with NMO is
still under investigation, but observation of an absorbance band at
λmax = 895 nm indicates that the reaction may have
involved an Fe(IV)–oxo intermediate. There is a growing body
of data to suggest that synthetic nonheme Fe(IV)–oxo complexes
have optical features between 800 and 900 nm that arise from d–d
transitions, and these features appear to be independent of spin state
and molecular structure.[2c,2d,2f] For example, we previously characterized the related trigonal bipyramidal
Fe(IV)–oxo complex [FeIVH3buea(O)]− ([H3buea, tris[(N′-tert-butylureaylato)-N-ethylene]aminato),
which exhibits a band at 808 nm.[22] Similarly,
the Fe(IV)–oxo species supported by the macrocyclic ligand
1,4,8,11-tetramethyl-1,4,8,11-tetraaza-cyclotetradecane (TMC) exhibits
a band at 820 nm in the optical spectrum despite having a different
coordination geometry (tetragonal) and spin state (S = 1).[23] Reactivity of this putative Fe(IV)–oxo
intermediate with a C–H bond in the solvent could possibly
be the source of the characterized [FeIIITST(OH)]− product.Observation of a putative Fe(IV)–oxo species
that did not react with the ancillary tripodal ligand suggested that
we might be able to intercept this reactive intermediate to activate
a C–H bond on an external substrate. Indeed, reaction of [FeIITST(OH2)]− with NMO in the presence
of 1 equiv of dihydroanthracene (DHA) in dichloromethane resulted
in 20% conversion to the oxidized products anthracene (A), 9,9′,10,10′-tetrahydro-9,9′-bianthracene
(B), and anthraquinone (C, Figures 5 and S18, Supporting Information).[24] For comparison, no conversion of
DHA was observed for identical reactions with NMe4[FeIIMST] or NMe4[FeIIMST(OH2)]. When reaction of the TST complex was conducted in acetonitrile
instead of dichloromethane, 50% of the DHA was converted to oxidized
products (Figure 5). The NMe4[FeIIITST(OH)] product was crystallized from this reaction in
90% yield.
Figure 5
Oxidation products of dihydroathracene: (A) anthracene, (B) 9,9′,10,10′-tetrahydro-9,9′-bianthracene,
and (C) anthraquinone and percent conversion obtained from reaction
in DCM and acetonitrile.
Oxidation products of dihydroathracene: (A) n class="Chemical">anthracene, (B) 9,9′,10,10′-tetrahydro-9,9′-bianthracene,
and (C) anthraquinone and percent conversion obtained from reaction
in DCM and acetonitrile.
Conclusions
In this article, we showed that reaction
of NMe4[FeIIMST] with dioxygen produced an Fe(III)–hydroxide
complex that is analogous to the bimetallic complexes formed in the
presence of a secondary metal ion. However, unlike the bimetallic
Fe(III)–hydroxide products, the NMe4[FeIIIMST(OH)] species is not stable in solution and converts to NMe4[FeIIMST(OH2)]. The reactivity of [FeIIMST]− alone further deviates from the reactivity
in the presence of Ca2+ ions when the oxygen-atom transfer
reagent NMO is used as the oxidant. While the same bimetallic Fe(III)–hydroxide
complex was isolated from NMO and Ca2+ as from dioxygen,
a new product was isolated when NMO was reacted with NMe4[FeIIMST] alone. This product was determined to be an
Fe(III)–alkoxide species that resulted from activation of a
C–H bond in a mesityl group of the ligand (BDEC–H ≈ 88 kcal mol–1). This result highlights
the oxidizing power of this system as well as a limitation of the
[MST]3– ligand system in generating high-valent
iron species. Oxidation of the ligand was prevented via substitution
of the mesityl groups with the tolyl derivative, and the C–H
bonds of an external substrate could then be activated. An intermediate
species was observed in the reaction between NMe4[FeIITST(OH2)] and NMO, which was postulated to be
an Fe(IV)–oxo species. The Fe(III)–hydroxide complex
was identified as the metal-containing product of this reaction.
Authors: Aram Joel Panay; Michael Lee; Carsten Krebs; J Martin Bollinger; Paul F Fitzpatrick Journal: Biochemistry Date: 2011-02-16 Impact factor: 3.162
Authors: Yohei Sano; Nathanael Lau; Andrew C Weitz; Joseph W Ziller; Michael P Hendrich; A S Borovik Journal: Inorg Chem Date: 2017-11-07 Impact factor: 5.165
Authors: Victoria F Oswald; Justin L Lee; Saborni Biswas; Andrew C Weitz; Kaustuv Mittra; Ruixi Fan; Jikun Li; Jiyong Zhao; Michael Y Hu; Esen E Alp; Emile L Bominaar; Yisong Guo; Michael T Green; Michael P Hendrich; A S Borovik Journal: J Am Chem Soc Date: 2020-06-24 Impact factor: 15.419
Authors: Justin L Lee; Victoria F Oswald; Saborni Biswas; Ethan A Hill; Joseph W Ziller; Michael P Hendrich; A S Borovik Journal: Dalton Trans Date: 2021-06-15 Impact factor: 4.390