Investigating the oxidation of alkenes by non-heme iron enzyme mimics

Article abstract of DOI:10.1039/c2ob25834j

Iron is emerging as a key player in the search for efficient and environmentally benign methods for the functionalisation of C-H bonds. Non-heme iron enzymes catalyse a diverse array of oxidative chemistry in nature, and small-molecule complexes designed to mimic the non-heme iron active site have great potential as C-H activation catalysts. Herein we report the synthesis of a series of organic ligands that incorporate key features of the non-heme iron active site. Iron(ii) complexes of these ligands have been generated in situ and their ability to promote hydrocarbon oxidation has been investigated. Several of these systems promote the biomimetic dihydroxylation of cyclohexene at low levels, when hydrogen peroxide is used as the oxidant; allylic oxidation products are also observed. An investigation of ligand stability reveals formation of several breakdown products under the conditions of the oxidative turnover reactions. These products arise via oxidative decarboxylation, dehydration and deamination reactions. Taken together these results indicate that competing mechanisms are at play with these systems: biomimetic hydroxylation involving high-valent iron species, and allylic oxidation via Fenton chemistry and Haber-Weiss radical pathways.

Full text of DOI:10.1039/c2ob25834j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 7372
www.rsc.org/obc
PAPER
Investigating the oxidation of alkenes by non-heme iron enzyme mimics†
Sarah M. Barry,‡a Helge Mueller-Bunzb and Peter J. Rutledge*a
Received 1st May 2012, Accepted 29th June 2012
DOI: 10.1039/c2ob25834j
Iron is emerging as a key player in the search for efficient and environmentally benign methods for the
functionalisation of C–H bonds. Non-heme iron enzymes catalyse a diverse array of oxidative chemistry
in nature, and small-molecule complexes designed to mimic the non-heme iron active site have great
potential as C–H activation catalysts. Herein we report the synthesis of a series of organic ligands that
incorporate key features of the non-heme iron active site. Iron(II) complexes of these ligands have been
generated in situ and their ability to promote hydrocarbon oxidation has been investigated. Several of
these systems promote the biomimetic dihydroxylation of cyclohexene at low levels, when hydrogen
peroxide is used as the oxidant; allylic oxidation products are also observed. An investigation of ligand
stability reveals formation of several breakdown products under the conditions of the oxidative turnover
reactions. These products arise via oxidative decarboxylation, dehydration and deamination reactions.
Taken together these results indicate that competing mechanisms are at play with these systems:
biomimetic hydroxylation involving high-valent iron species, and allylic oxidation via Fenton chemistry
and Haber–Weiss radical pathways.
non-heme iron oxidases generate high-valent iron-oxo intermedi-
Introduction
ates (iron(IV)- or iron(V)-oxo), now generally accepted as
the active oxidising species in these systems.18–20 Such
Chemists have long looked to nature for inspiration in the
design of new oxidation catalysts.1–8 Non-heme iron
species have been spectroscopically characterized in several
enzymes catalyse a diverse range of oxidation reactions in
biology, including desaturation and cyclisation (oxidase activity),
hydroxylation and epoxidation (mono-oxygenase activity), dihy-
droxylation, and oxidative ring opening reactions (di-oxygenase
activity).5–14
Over the past decade, significant information has been gar-
nered concerning the structure and catalytic mechanism of non-
heme iron enzymes.5–8 X-Ray crystal structures reveal a con-
served active site which incorporates an iron(II) centre bound by
three protein-derived ligands, two histidines and one carboxylate,
occupying one face of the iron(II) coordination sphere
(Fig. 1).15,16 While this ‘2-His-1-carboxylate’ triad has emerged
as a common motif across the family, non-heme iron enzymes
utilise a range of different cofactors and mechanisms to drive
reductive cleavage of the O–O bond.5,6,17 In cleaving this bond,
non-heme iron enzymes, including taurine/α-ketoglutarate
dioxygenase (TauD),21 prolyl-4-hydroxylase22 and tyrosine
hydroxylase.23
Further mechanistic insight has been derived using small-
molecule active site models of non-heme iron catalysis.5–8 The
most effective of these biomimetic systems are based upon
neutral, multidentate ligands such as tris(2-pyridylmethyl)amine
(TPA),24 bispidines,25 chiral bis-pyrrolidines (PDP or
BPBP),26–28 and tetramethylguanidines (TMG3tren and
TMG2dien).29,30 When complexed with iron(II) and combined
with a suitable oxidant (e.g. hydrogen peroxide, tert-butylhydro-
peroxide or molecular oxygen), these mimics can carry out
stereoselective dihydroxylation, epoxidation and hydroxylation
reactions on a range of hydrocarbon substrates, with good to
excellent catalytic efficiency. Model complexes have also
been used to generate high-valent iron-oxo intermediates similar
to those proposed to mediate non-heme catalytic pathways
in vivo, allowing visualization of oxygen activation in
model systems by isolating and characterising iron(III)-hydroper-
oxo and iron(IV)-oxo intermediates,31–36 and recently an iron(V)
species.37
aSchool of Chemistry F11, The University of Sydney, NSW 2006,
Australia. E-mail: peter.rutledge@sydney.edu.au;
Fax: +61 2 9351 3329; Tel: +61 2 9351 5020
bSchool of Chemistry and Chemical Biology, University College Dublin,
Belfield, Dublin 4, Ireland
†Electronic supplementary information (ESI) available. CCDC 879768
and 879769. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2ob25834j
‡Current address: Department of Chemistry, University of Warwick,
Coventry, CV4 7AL, UK.
Towards the goal of constructing small-molecule iron-based
catalysts for hydrocarbon oxidation, we have recently reported a
range of ligands 2–4 designed to mimic the non-heme iron
active site (Fig. 1).38–41 These ligands combine with iron(II)
7372 | Org. Biomol. Chem., 2012, 10, 7372–7381
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 1 Generalized non-heme iron active site 1, and ligand architectures 2–4 previously constructed to mimic its structure and function (R1 = Et,
allyl; R2 = Ph; R3 = -1-(pyrrolidine-2-carboxylic acid), -OH or -((S)-1-phenylethyl)amino; X = solvent or substrate-derived ligand).38–41
acetate and hydrogen peroxide in methanol solvent to convert a
range of simple alkenes (C5–C8, cyclic and straight-chain) to
diols and allylic oxidation products.39,41
Results and discussion
Synthesis of ligands 2a–d
Ligands 2a–d were prepared from (S)-mandelic acid 7, pyridine-
2,6-dimethanol 8, and N-substituted anilines 9a–d (Scheme 1),
extending our previously reported synthesis of ligand 2b.38,39
Diol 8 was desymmetrised and converted to key bromide
intermediate 11 in five steps as reported previously.38 The stereo-
chemistry of the starting mandelic acid is preserved throughout
this sequence by employing the dioxolanone as both protecting
group and chiral auxiliary,43 which ultimately delivers ligands
2a–d in enantiomerically pure form. An X-ray crystal structure†
of 11 was solved (Fig. 3), confirming that the stereochemical
integrity is preserved through the alkylation–deprotection–bromi-
nation sequence, and that both stereocentres have the stereo-
chemistry expected.
Coupling of bromide 11 with the N-substituted anilines 9a–d
gives protected ligands 12a–d. N-Methylaniline 9a, N-ethylani-
line 9b and diphenylamine 9d are commercially available, N-
tert-butylaniline was made from N-tert-butylamine and bromo-
benzene following a literature procedure.45
Optimisation studies with the ethyl analogue 9b revealed that
using sodium hydride as base in DMF solvent returned no reac-
tion, while reactions with n-BuLi in THF were low-yielding
unless N,N′-dimethylpropyleneurea (DMPU) was included as co-
solvent.46 The amine is efficiently deprotonated at 0 °C in all
cases, but the optimum temperature for the coupling step varies
with the amine being coupled: −78 °C for 9a, −40 °C with 9b,
−5 °C for 9c and −15 °C for 9d (Scheme 1). Using these opti-
mised conditions good yields were achieved for the methyl
To better understand and thus expand the catalytic behaviour
of these ligands, we now report a detailed investigation of the
factors that influence activity in ligands of general structure 2
and the related ligands 5–7, which incorporate different elements
of the larger structure 2 (Fig. 2).
Ligands 2 are designed so that the pyridine and tertiary amine
ligands model the histidine residues of the non-heme iron active
site, mandelic acid mimics the carboxylate and water ligands of
the natural system; the R group of the exocyclic amine is varied
by incorporating different aniline derivatives. This design has
been developed considering the propensity of iron(II) complexes
to undergo competing intermolecular oxidation reactions to give
bridged products such as Fe(III)–O–Fe(III) species. Thus two
pendant phenyl groups are included in the structure to increase
steric bulk and inhibit such intermolecular interactions
that would render the complex inactive. Ligands 2a–d are
tetradentate and designed to form iron(II) complexes in which
there are two vacant coordination sites for peroxide binding,
while also providing steric bulk around the metal centre. This
arrangement of vacant cis positions mimics the naphthalene
dioxygenase active site,42 and is in accordance with the accepted
thesis that two vacant sites cis to each other are absolutely
required for a small-molecule iron complex to effect dihydroxy-
lation using hydrogen peroxide as oxidant.24 Ligands 5, 6 and 7
represent the ‘component parts’ of the larger ligand 2, and have
been studied to investigate the influence of each part on
reactivity.
Fig. 2 Ligands used in this study.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7372–7381 | 7373

View Article Online
Scheme 1 Synthesis of ligands 2a–d; i. NaH, TBDMSCl, DCM, rt, 5 hours, 57%; ii. CBr4, PPh3, DCM, rt, 2 hours, 94%; iii. ((2S,4S)-2-(tert-
butyl)-5-oxo-4-phenyl-1,3-dioxolan-4-yl)lithium, THF, −78 °C, 6 hours, 91%; iv. TBAF, THF, rt, 90 minutes, 88%; v. CBr4, PPh3, DCM, rt, 2 hours,
77%; vi. a. n-BuLi–DMPU, MeNHPh 9a, THF, −78 °C, 5 hours, 84%; or b. n-BuLi–DMPU, EtNHPh 9b, THF, −40 °C, 5 hours, 85%; or c. n-BuLi–
DMPU, tBuNHPh 9c, THF, −5 °C, 5 hours, 40%; or d. n-BuLi, HNPh2 9d, THF, −15 °C, 4 hours, 82%; vii. a.–c. 1 M LiOH, THF, reflux, 17 hours
then 1 M HCl; or d. 1 M LiOH, MeOH, reflux, 17 hours then 1 M HCl (91–95%).
(12a, 84%), ethyl (12b, 85%) and phenyl (12d, 82%) analogues,
but the more hindered tert-butyl derivative 12c was less coopera-
tive and returned a lower yield (40%) despite efforts to further
optimise this reaction, presumably a result of steric effects.
The diphenyl compound 12d was isolated as a white crystal-
line solid after column chromatography and an X-ray crystal
structure† of 12d was solved (Fig. 4). This structure confirms the
cis relationship of the tert-butyl and phenyl substituents on the
dioxolanone ring, and that this five-membered ring sits over the
top of the pyridine nitrogen in the crystalline compound.
Deprotection of 12a–d proceeded smoothly in all cases and
gave excellent yields. The hydrolysis reaction was carried out in
aqueous base (1 M lithium hydroxide) and a minimum of
organic co-solvent. THF was used for 12a–c, however 12d
showed no reaction in water–THF even at reflux due to phase
separation, so water–methanol was used instead. The crude
lithium salt of the ligand is protonated during workup to give the
neutral product, which was separated from the lithium chloride
side product by triturating with DCM or chloroform.
tetrabromomethane as in our synthesis of mono-bromide 11.38
Then N-ethylaniline 9b was introduced using the same method
as for preparation of 12b to give 5 as a flakey crystalline solid in
reasonable yield (55%) over the two steps (Scheme 2a).
Accessing bidentate ligand 6 required a change in strategy, as
derivatives of pyridine-2-methanol 13 that bear pendant leaving
groups are prone to dimerise, forming pyrazinium salts.41,47,48
Thus attempts to brominate 13 using the same mild, neutral con-
ditions described above for alcohol 8 gave 6,12-dihydrodipyrido-
[1,2-a:1′,2′-d]pyrazine-5,11-diium dibromide as a crystalline
solid (data not shown), while the tosylate derivative of 13
decomposed to elimination products. These problems are
avoided by keeping the pyridine nitrogen protonated: treating
pyridine-2-methanol 13 with neat thionyl chloride as reported by
Winterfield and Flick affords 2-chloromethylpyridine as its
hydrochloride salt in excellent yield (93%).49 This salt is highly
hygroscopic and thus difficult to handle which compromises
the yield of the following, moisture-sensitive step. Nonetheless
reaction with the combination of n-BuLi and N-ethylaniline
completed the synthesis of ligand 6 in 37% yield over the two
steps. (Note that N-ethylaniline was combined with two equiva-
lents of n-BuLi to account for the hydrochloride salt, however
the base was not added directly to 2-chloromethylpyridine
hydrochloride before the nucleophile to avoid pyrazinium
formation.)
Synthesis of ligands 5 and 6
To make the triamine ligand 5, pyridine-2,6-dimethanol 8 was
first converted to the dibromide, using triphenylphosphine and
Fig. 3 X-Ray crystal structure of bromide 11, synthesised according to
the procedure we have reported previously38 and crystallised by slow
evaporation of a cyclohexane–ether solution; image generated using
Mercury CSD 2.0.44
Fig. 4 X-Ray crystal structure of diphenylamine derivative 12d; image
generated using Mercury CSD 2.0.44
7374 | Org. Biomol. Chem., 2012, 10, 7372–7381
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 2 Synthesis of ligands 5 and 6; i. CBr4, PPh3, DCM, rt, 16 hours, 75%; ii. n-BuLi–DMPU, EtNHPh (2 eq), THF, −40 °C, 5 hours, 73%;
iii. SOCl2, reflux, 16 hours, 93%; iv. n-BuLi, HNPhEt, THF–DMPU, −41 °C, 40%.
It is interesting to note that bromide 11 does not undergo
dimerization to a pyrazinium derivative: it is a white, crystalline
solid that does not show any sign of degradation even after
standing for several months at room temperature. This is presum-
ably a steric effect, and the X-ray crystal structure of 11 (Fig. 3)
shows clearly that the pyridine nitrogen is effectively shielded
by the bulk of the dioxolanone ring, which also appears to
hinder access to σ* of the C–Br bond.
ligand precursors (alcohol 10 and dioxolanone-protected ligand
12b) for comparison. The diol 15, alcohol 16, ketone 17 and
epoxide 19 products of cyclohexene oxidation are formed at low
levels. The products of ligand breakdown (alkyl anilines 9a–d
and benzaldehyde) are also observed; oxidative ligand break-
down is discussed below. While levels of dihydroxylation are
low, control experiments confirm that the ligand is required for
dihydroxylation to occur: control turnover reactions carried out
in the absence of ligand give no diol product.
Ligand 2b effects biomimetic dihydroxylation and allylic oxi-
dation of cyclohexene 14 and other alkenes under these con-
ditions as previously reported.39 None of the N-methyl, N-tert-
butyl and N-phenyl variants bring any increase in turnover
yields: the bulkier tert-butyl derivative 2c produces trace
amounts of diol 15, while the methyl (2a) and phenyl (2d)
derivatives produce no detectable level of the desired diol
product.
Oxidative turnover reactions
The combination of iron(II), a ligand and hydrogen peroxide can
give rise to a variety of different oxidation products from an
alkene substrate (Scheme 3). Biomimetic dihydroxylation (Path
A) promoted by an iron-ligand complex can convert cyclohexene
14 to the cis-diol 15 using
a
high-valent iron-oxo
intermediate.31–33 Competing reactions via radical mechanisms
(Path B) traditionally associated with Fenton and Haber–Weiss
pathways leads to the allylic oxidation products alcohol 16 and
ketone 17.50 The hydroperoxide 18 (a potential precursor of both
16 and 17) may also be formed via the radical path and isolated
under some circumstances, and the epoxide 19 can arise from
reaction of 18 with another molecule of the alkene 14.6,16
Reactions with the following ligand combinations were
carried out to probe which parts of ligand 2 are most important
to activity: amines 5 and 6, disodium mandelate 7·Na2, the pre-
cursor alcohol 10, and the dioxolanone-protected ligand 12b
(Table 1). With the tridentate ligand 5, the main products are
cyclohexenol 16, cyclohexenone 17 and N-ethylaniline 9b; trace
amounts of diol 15 are seen and no epoxide 19. The reaction
using amine 6 in tandem with disodium mandelate 7·Na2 and
iron(II) acetate gives benzaldehyde as the major product, from
the oxidative breakdown of the ligand; smaller amounts of
N-ethylaniline 9b were observed relative to the reactions of the
ligands 2a–d, and only trace amounts of diol 15 and allylic oxi-
dation products 16 and 17 were observed. Using amine 6 on its
own does not generate any diol product, while disodium mande-
late (7·Na2) as sole ligand produces almost half the amount of
diol produced by ligand 2b. The protected ligand 12b (in which
both oxygen atoms are masked) gives primarily N-ethylaniline
9b and allylic oxidation products, while the precursor alcohol 10
gives mostly allylic oxidation products, with trace amounts of
diol 15. These experiments suggest that the oxygen donors in the
ligand 2b are key to diol formation. Anilines are not particularly
Iron complexes of ligands 2a–d and 5–7 were formed in situ
by mixing the sodium salt of the ligand with an equimolar
amount of an iron(II) salt in degassed methanol at room tempera-
ture. The resulting yellow solution was diluted further with
methanol before the cyclohexene substrate was added. Finally
hydrogen peroxide (30% aqueous, diluted in methanol) was
introduced slowly via syringe pump (final peroxide concen-
tration: 70 mM). After reaction at room temperature overnight,
oxidation products were analysed by gas chromatography, by
comparison to authentic samples of the expected products.
Ligand variations
The results of turnover reactions with ligands 2, 5, 6 and 7 are
summarized in Table 1, along with data obtained using two
Scheme 3 Oxidation products from reaction of cyclohexene 14: biomimetic dihydroxylation to give cis-cyclohexane-1,2-diol 15 (Path A) vs.
Fenton-type reactivity to give allylic oxidation products alcohol 16, ketone 17 and sometimes hydroperoxide 18 and epoxide 19 (Path B). The epoxide
may be hydrolysed to the trans-diol, so the stereochemistry of diol product was confirmed unambiguously by spiking with authentic samples and sep-
aration of isomers on the polar BP-20 column (see Experimental for further details).
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7372–7381 | 7375

View Article Online
Table 1 Variation of product yield with ligands 2–12a,b
temperature, forming the iron-ligand complex and diluting as
above, then cooling the solution to 0, −41 or −78 °C prior to
addition of substrate and the oxidant (hydrogen peroxide, diluted
and added over 4 hours as detailed above). The reaction was
then stirred at the specified temperature for 12–14 hours before
being allowed to warm to room temperature over 6 hours. Reac-
tion at 0 °C gave product yields and ratios little changed from
the room temperature reaction. However at −41 °C and −78 °C,
only trace amounts of the allylic oxidation products were
observed and the cis-diol 15 was the major substrate-derived
product, formed in 0.71% yield at −41 °C (versus 0.63% at
room temperature). The other major product was the ligand
breakdown product N-ethylaniline 9b, formed at a reduced level
relative to the room temperature reaction (0.51% vs. 1.63%).
pH: The reaction is sensitive to pH, which is unsurprising
given the nature of the ligand. The ligand is deprotonated at its
carboxylic acid and adjacent alcohol prior to complexation with
iron(II) by treatment with sodium hydride, and the sodium salt of
the ligand stored in a dessicator overnight to remove any water.
The oxidation reaction was found to tolerate a slight excess of
sodium hydride (3–4 equivalents) however addition of 6 equiva-
lents resulted in no reaction. The addition of only one equivalent
gave reduced yields of diol product (<0.10 μmol) and no diol
product at all was observed in the presence of protonated ligand
or under acidic conditions.
Ligand
15c
16c
17c
19c
RNHPhc PhCHOc
2a
2b
2c
2d
5
—
0.63
<0.10
—
<0.10
—
0.10
0.21
0.42
0.10
0.32 0.10
1.51 <0.10
—
1.26
1.63
4.11
—
1.46
1.72
—
0.34
—
0.52
—
—
—
—
—
—
3.13
7.06
—
<0.10 <0.10
—
—
—
0.43
0.10
0.21
0.76
0.22
6d
d
7·Na2
0.13
0.22 0.13e
7·Na2 + 6 f <0.10 <0.10
0.10
—
10
0.13
<0.10
—
0.63
0.21
0.10
1.98 <0.10e
12b
—
0.10
0.44
—
—
—
—
g
—
a Initial ratio catalyst : H2O2 : substrate 1 : 10 : 1000 (2 (9.6 μmol) : iron(II)
acetate (9.6 μmol) : hydrogen peroxide (96 μmol) : cyclohexene
(9.6 mmol) : solvent methanol (30 mL)). See Experimental section for
more details. b Catalyst prepared in situ from ligand and iron(II) acetate
in methanol; reaction run in methanol. c Percentage yield relative to
H2O2, the limiting reagent. Turnover number (μmol product produced
per μmol catalyst) can be derived by dividing these percentage yields by
10. Percentage conversion of alkene = percentage yield/100. d Two
equivalents relative to Fe(OAc)2. e Hydrolysed to and identified as the
trans-diol product. f 7·Na2 (1 eq) and 6 (1 eq) were dissolved in
methanol and stirred while Fe(OAc)2 (1 eq) was added. g Control
contains no ligand. All values shown are the averages of at least three
repetitions.
Solvent: Acetonitrile is commonly used in iron-based oxi-
dations given the potential for competing solvent oxidation when
methanol is used. However the iron complexes derived from
ligands 2a–d are poorly soluble in acetonitrile and gave no diol
product when tested under the alternative conditions of iron(II)
acetate–acetonitrile.
In combination, these results demonstrate a very delicate
balance of steric and electronic factors which affect the oxidative
ability of this system.
basic amines due to the electron withdrawing effect of partial
delocalisation of the nitrogen lone pair into the phenyl ring, so
the truncated ligands 5 and 6 are likely to bind weakly to iron.
Considered together, these results suggest that the most likely
reason for the low levels of dihydroxylation seen with ligands 2
is weak binding of the aniline moiety to iron.
Variation in reaction conditions
In addition to the ligand, a range of other reaction variables were
investigated in the effort to improve diol yields.
Ligand degradation
Concentration: Varying the concentration of the complex and
substrate also influences reaction outcome. Diluting the solution
results in more diol formation and less allylic oxidation: diol for-
mation increases from <0.10% to 0.63% when the concentrations
of complex and substrate are lowered from 1.6 mM complex/
95 mM substrate to 0.32 mM/32 mM. The substrate is used in
1000-fold excess relative to the complex, increasing the likeli-
hood of an activated iron(II) complex reacting with a substrate
molecule in dilute solution, and at the same time reducing the
chances of an activated iron complex coming into contact with
an unactivated iron(II) complex or free iron(II), which could lead
to the breakdown of the peroxide species and generate radicals
in a Fenton-type reaction.51 The time period over which the
dilute oxidant (hydrogen peroxide, 30% aqueous) was added
was also varied to assess the effect on conversion. The addition
time was varied from 30 minutes to 24 hours using a syringe
pump. The yield of diol product increased with addition times
up to 4 hours but no further increase was achieved by extending
the addition time.
The same Fenton-type pathways that result in allylic oxidation of
the alkene substrate could also promote destruction of the pepti-
domimetic ligands, which would give rise to the observed by-
products N-alkylanilines 9a–d and benzaldehyde. Zs-Nagy and
Floyd have reported the quenching effects of amino acids on
hydroxyl radicals generated in the presence of iron(II) in Fenton-
type chemistry.53 This study, and subsequent work by Stadtman
and Berlett,54 demonstrated that amino acids can be oxidatively
degraded by the combination of iron(II) and hydrogen peroxide
via various pathways including oxidative decarboxylation, de-
amination and hydrogen abstraction. We believe similar pro-
cesses contribute to the low activity of our system, as ligands 2
are amino acids (although the amino and carboxy termini are
remote from each other).
Subjecting ligand 2b to ‘turnover’ conditions without an
alkene substrate (i.e. combining with iron(II) acetate and adding
hydrogen peroxide at the same concentrations as for alkene oxi-
dation reactions) gave a mixture of products. This was shown by
LCMS (ESI†) to include compounds 20–23, the products of oxi-
dative decarboxylation (20), dehydration (21), deamination (22)
and decarboxylation plus deamination (23) (Scheme 4). None of
Temperature: Fenton and Fenton-type reactions are tempera-
ture-dependent and occur with greater efficiency at higher temp-
eratures.52 Thus turnover reactions were performed at low
7376 | Org. Biomol. Chem., 2012, 10, 7372–7381
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 4 Oxidative breakdown products of the ligand 2b, the products of oxidative decarboxylation (20), dehydration (21), deamination (22) and
decaboxylation–deamination (23), analogous to reaction pathways observed previously for the breakdown of simple amino acids under similar
conditions.53,54
the starting ligand 2b survived after 17 hours under these con-
ditions. Control experiments subjecting ligand 2b to dilute
hydrogen peroxide without iron returned most of the ligand
(>80%) unreacted plus low levels of amino-ketone 20 (ESI†).
A substantial portion of the oxidising potential introduced in
these reactions (which employ a 10-fold excess of hydrogen per-
oxide relative to the iron complex) must also be lost elsewhere.
Oxidation of the solvent methanol is the most likely explanation,
although the likely oxidation products were not directly observed
during this study. Future work will focus on the use of alternative
iron salts (e.g. iron(II) triflate, iron(II) tetrafluoroborate) to
improve complex solubility and allow turnover reactions to be
performed in acetonitrile solutions.
Mechanistic conclusions
We propose that the oxidation reactions observed with ligands 2,
5, 6 and 7 follow a mechanism akin to those proposed by Que
and others for the reaction of polyamine–iron complexes, hydro-
gen peroxide and alkene substrates.31–35 Thus reaction begins
with formation of an iron-peroxo intermediate (LnFeIII–OOH,
24). This intermediate can promote the allylic oxidation path-
ways itself or break down via several competing pathways
(Scheme 5), giving rise to a number of possible oxidants: (1)
LnFe–OOH 24 itself; (2) the peroxy radical (˙OOH) formed via
homolysis of the Fe–O bond in 24 (reaction a); (3) the hydroxy
radical ˙OH, generated by bimolecular reaction of two peroxy
radicals (reaction b); (4) an iron(IV)-oxo species LnFeIVvO 25,
generated by homolysis of the O–O bond in 24 (reaction c); (5)
an iron(V)-oxo species LnFeVvO 26, arising from heterolysis of
the O–O bond in 24 (reaction d).16
Experimental
General experimental
Solvents were purified and distilled under nitrogen prior to use:
dichloromethane (DCM) and methanol from calcium hydride;
tetrahydrofuran (THF) from sodium metal and benzophenone
ketyl radical. Alkene substrates were distilled from calcium
hydride and stored over 4 Å molecular sieves. When required,
solvents were degassed using the freeze–thaw method in three
cycles. Thin layer chromatography (TLC) was carried out on alu-
minium-backed plates, pre-coated with Merck silica gel 60 F254
;
plates were visualised using an ultraviolet lamp and/or by stain-
ing with ninhydrin solution (0.3% w/v in 97 : 3 n-BuOH–AcOH)
or phosphomolybdic acid (0.01% phosphomolybdic acid,
0.001% ceric sulphate, H2O–H2SO4–EtOH 20 : 1 : 40) and
heating. Retention factors are quoted to the nearest 0.05. Flash
chromatography was carried out using Ajax Finechem silica gel
230–400 mesh. Melting points were recorded on a GallenKamp
melting point apparatus. Optical rotations were measured on a
Perkin Elmer model 341 Polarimeter with a sodium lamp at
589 nm in a 1.0 or 0.5 dm cell at the concentration (g per
100 mL) and temperature indicated. Infrared spectra were
measured on a Shimadzu Bio-Rad 8400S on a sodium chloride
The nature of Ln dictates which of these pathways predomi-
nate and controls the outcome of the reaction. With the ligands
used in this study, pathways a and b appear to be dominant,
leading to various oxidation products via radical mechanisms.
Formation of the cis-diol product 15 indicates that an alternative
‘biomimetic’ reaction via 25 (or 26) is also operating under these
conditions, but only at very low levels.
Furthermore we postulate that the breakdown of ligand 2b to
the products 20–23 follows—at least in part—an intramolecular
mechanism via 25 or 26. The hydroxy and peroxy radical inter-
mediates generated via reactions a and b offer one route to pro-
ducts 20–23, via mechanisms analogous to those characterised
by Stadtman and Berlett.54 However under all the conditions
investigated in the current study, release of N-ethylaniline 9b via
oxidative deamination of ligand 2b occurs to a greater extent
than allylic oxidation of cyclohexene, even though the alkene is
present in 1000-fold excess. The amine product 9b is observed
even under conditions in which the allylic oxidation reaction is
all but abolished (reduced temperature). One explanation for
these observations is that at least one step of the oxidative degra-
dation of ligand 2b occurs while the ligand is bound to the iron
centre: intramolecular iron-mediated oxidation would explain
why even in high substrate concentration, oxidation of the ligand
occurs preferentially. It would also explain why when the Fenton
pathway has all but been eliminated, ligand oxidation occurs and
cis-diol yield is not substantially increased.
Scheme 5 Potential breakdown pathways for the LnFeIII–OOH inter-
mediate 24, generating five potential oxidants in non-heme iron
systems.16
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7372–7381 | 7377

View Article Online
plate (oils), KBr disc, or in solution in the solvent indicated;
absorbances are quoted in wavenumbers (cm−1) and recorded as
strong (s), medium (m), weak (w) and broad (br). NMR spectra
were recorded on Bruker DPX 200, DPX 300 and DPX 400
spectrometers; chemical shifts are reported in parts per million
(ppm) from tetramethylsilane (TMS) and referenced to TMS or
1221 (s); δH (300 MHz, CDCl3): 0.88 (9H, s, C(CH3)3), 3.11
(3H, s, NCH3), 3.37 (1H, d, J = 14.5 Hz, 1 of CH2CC6H5), 3.69
(1H, d, J = 14.5 Hz, 1 of CH2CC6H5), 4.61 (2H, s, CH2NC6H5),
4.79 (1H, s, CHC(CH3)3), 6.60–6.98 (3H, m, 3 of NC6H5), 6.98
(1H, d, J = 7.5 Hz, CHPyr), 7.06 (1H, d, J = 7.5 Hz, CHPyr), 7.21
(2H, t, J = 7.5 Hz, 2 of NC6H5), 7.35 (3H, m, 3 of CC6H5), 7.52
(1H, t, J = 7.5 Hz, CHPyr), 7.77 (2H, m, 2 of CC6H5); δC
(75 MHz, CDCl3): 23.6, 34.9, 39.2, 48.4, 58.6, 82.2, 109.6,
112.1, 116.7, 119.2, 122.5, 125.0, 128.0, 128.3, 129.2, 137.1,
139.1, 149.1, 155.1, 159.2, 173.4; m/z (ES+): 883 (25%, [2M +
Na]+), 453 (100%, [M + Na]+), 431 (65%, [MH]+); HRMS
1
the appropriate residual solvent peak. H NMR data are quoted
as the signal (ppm), relative integral, multiplicity (singlet s,
doublet d, triplet t, quartet q, double doublet dd, multiplet m)
and coupling constant J (quoted in hertz to the nearest 0.5 Hz).
13C NMR data (50, 75 or 100 MHz) are quoted as the signal
(ppm) to the nearest 0.1 ppm. Low resolution mass spectra were
recorded on a Finnigan LCQ MS Detector (ESI); major peaks
are recorded as mass to charge (m/z) followed by the height of
the peak as a percentage of the base peak and assignment. High
resolution mass spectra were recorded on a Bruker Apex II
FTICR mass spectrometer. LCMS was carried out using a
Thermo Separation Products SCM 1000 vacuum membrane
degasser, Thermo Separation Products Spectra System, P4000
pump and AS3000 autosampler, a Phenomenex Luna C18(2)
column (2.0 × 150 mm ID, 5 μm), and a ThermoQuest Finnigan
LCQ DecaMass spectrometer. GC analyses were performed on a
Hewlett-Packard 5890 Series II gas chromatograph using an
HP-1ms column (30 m × 0.25 mm ID, 0.25 μm; ramping 70 to
160 °C at 12 °C min−1) and a Hewlett-Packard 5890A gas chro-
matograph with a BP-20 column (25 m × 0.22 mm ID, 0.25 μm;
ramping 40 to 220 °C at 12 °C min−1), each equipped with split/
splitless capillary inlet and flame ionisation detector (FID) and
controlled using ChemStation software. GCMS was carried out
using a Finnigan PolarisQ fitted with a ZB–5 column (15 m ×
0.25 mm ID, 0.10 μm; ramping 40 to 120 °C at 10 °C min−1 or
40 to 260 °C at 20 °C min−1) connected to a Thermo-Finnigan
Polaris Q ion trap mass spectrometer.
+
(ES+): C27H31N2O3 ([MH+]) requires 431.2335, found
431.2345.
(2S,5S)-2-tert-Butyl-5-((6-((tert-butyl(phenyl)amino)methyl)-
pyridin-2-yl)methyl)-5-phenyl-1,3-dioxolan-4-one 12c. tert-
Butylaniline 9c (43 mg, 0.28 mmol) in THF (4.0 mL) was
cooled to −41 °C and a solution of n-BuLi in hexane (1.42 M,
0.24 mL, 0.35 mmol) was added via syringe. The solution was
stirred at −5 °C for 45 minutes then added via cannula to a solu-
tion of the bromide 11 (100 mg, 0.24 mmol) in THF (4.0 mL)
also at −5 °C. The resulting yellow solution was stirred at −5 °C
for 4 hours, monitored by TLC (cyclohexane–ethyl acetate 5 : 1).
The solution was allowed to warm to room temperature then
poured onto half-saturated aqueous ammonium chloride solution.
The aqueous phase was then extracted with DCM (3 × 4 mL).
The organic phases were combined and washed with brine
(3 mL), and dried (MgSO4). The solvent was removed in vacuo
to give an orange oil which was purified by column chromato-
graphy (cyclohexane–ethyl acetate 12 : 1) to afford the title com-
pound as a pale yellow oil (45 mg, 40%). Rf 0.5 (cyclohexane–
ethyl acetate, 5 : 1); [α]2D0 −43.1 (CHCl3, c 0.65); νmax (NaCl):
3065 (w), 2963 (s), 2961 (s), 1785 (s), 1590 (m), 1574 (m),
1214 (s); δH (300 MHz, CDCl3): 0.81 (9H, s, CHC(CH3)3), 1.18
(9H, s, NC(CH3)3), 3.25 (1H, d, J = 14.5 Hz, 1 of CH2CC6H5),
3.42 (1H, d, J = 14.5 Hz, 1 of CH2CC6H5), 4.50 (3H, s,
CH2NC6H5, CHC(CH3)3), 6.87 (1H, d, J = 7.5 Hz, CHPyr),
7.10–7.39 (8H, m, 3 of CC6H5 and 5 of NC6H5), 7.47 (1H, t,
J = 7.5 Hz, CHPyr), 7.51 (1H, d, J = 7.5 Hz, CHPyr), 7.73 (2H,
m, 2 of CC6H5); δC (75 MHz, CDCl3): 23.6, 28.4, 34.9, 48.7,
56.1, 57.1, 82.3, 109.5, 120.8, 121.8, 124.5, 124.9, 127.9,
128.2, 128.2, 129.3, 136.6, 139.4, 149.8, 153.8, 162.2, 173.3;
m/z (ES+): 967 (5%, [2M + Na]+), 473 (100%, [MH]+); HRMS
Synthesis of ligands
(2S,5S)-2-tert-Butyl-5-((6-((methyl(phenyl)amino)methyl)-
pyridin-2-yl)methyl)-5-phenyl-1,3-dioxolan-4-one 12a. A solu-
tion of N-methylaniline 9a (0.16 mL, 1.49 mmol) in THF
(10 mL) was cooled to 0 °C and a solution of n-BuLi in hexane
(1.5 M, 1.19 mL, 1.78 mmol) was added via syringe. The sol-
ution was stirred at 0 °C for 20 minutes before DMPU (0.22 mL,
1.78 mmol) was added and the solution was stirred for a further
20 minutes. The now yellow solution was cooled to −78 °C and
added via cannula to a solution of the bromide 11 (0.50 g,
1.24 mmol) in THF (10 mL) also at −78 °C. The resulting
yellow solution was stirred at −78 °C for 5 hours, monitored by
TLC (cyclohexane–ethyl acetate 5 : 1). The solution was allowed
to warm to room temperature slowly overnight then poured onto
half-saturated aqueous ammonium chloride solution. The
aqueous phase was extracted with diethyl ether (3 × 5 mL). The
organic phases were combined, washed with brine (4 mL), and
dried (MgSO4). The solvent was removed in vacuo to give a
crude yellow oil, which was purified by column chromatography
(cyclohexane–ethyl acetate 20 : 1) to afford the title compound
as a pale yellow semi-solid (0.38 g, 84%). Rf 0.46 (cyclo-
hexane–ethyl acetate 5 : 1); [α]2D0 −53.8 (CHCl3, c 0.93); νmax
(NaCl, cm−1): 2978 (s), 2952 (s), 1787 (s), 1448 (m), 1383 (m),
+
(ES+): C30H37N2O3 ([MH]+) requires 473.2804, found:
473.2819.
(2S,5S)-2-tert-Butyl-5-((6-((diphenylamino)methyl)pyridin-2-
yl)methyl)-5-phenyl-1,3-dioxolan-4-one 12d. A solution of
diphenylamine 9d (0.35 g, 2.08 mmol) in THF (12.0 mL) was
cooled to 0 °C and a solution of n-BuLi in hexane (1.5 M,
1.7 mL, 2.5 mmol) was added via syringe. The solution was
stirred at 0 °C for 20 minutes, before DMPU (0.3 mL,
2.5 mmol) was added and the solution was stirred for a further
20 minutes. The resulting yellow solution was cooled to −15 °C
then added via cannula to a solution of the bromide 11 (0.70 g,
1.73 mmol) in THF (12.0 mL) also at −15 °C. The resulting
yellow solution was stirred at −15 °C for 5 hours, monitored by
TLC (cyclohexane–ethyl acetate 5 : 1). The solution was allowed
to warm to room temperature overnight, then poured onto half-
7378 | Org. Biomol. Chem., 2012, 10, 7372–7381
This journal is © The Royal Society of Chemistry 2012

View Article Online
saturated aqueous ammonium chloride solution. The aqueous
phase was extracted with DCM (3 × 4 mL) and the organic
phases were combined, washed with brine (5 mL), and dried
(MgSO4), then concentrated in vacuo to give an orange oil. The
crude product was purified by column chromatography (cyclo-
hexane–ethyl acetate 10 : 1) to afford the title compound as a
white solid (0.70 g, 82%). Rf 0.5 (cyclohexane–ethyl acetate
5 : 1); m.p. 105–108 °C; [α]2D0 −65.7 (CHCl3, c 0.82); νmax
(NaCl): 3060 (w), 2962 (s), 2961 (s), 1793 (s), 1590 (m), 1577
(m), 1178 (s); δH (300 MHz, CDCl3): 0.86 (9H, s, C(CH3)3),
3.36 (1H, d, J = 14.5 Hz, 1 of CH2CC6H5), 3.67 (1H, d, J =
14.5 Hz, 1 of CH2CC6H5), 4.77 (1H, s, CHC(CH3)3), 5.06 (2H,
s, CH2NC6H5), 6.60–6.94 (3H, m, 2 of NC6H5, 1 of CC6H5),
7.04 (4H, m, 4 of NC6H5), 7.20–7.41 (8H, m, 2 of CC6H5, 4 of
NC6H5, 2 × CHPyr), 7.49 (1H, t, J = 7.5 Hz, CHPyr), 7.76 (2H,
d, J = 1.5 Hz, 2 of CC6H5); δC (75 MHz, CDCl3): 23.6, 35.0,
48.3, 58.3, 82.2, 109.8, 119.5, 120.6, 121.7, 122.6, 125.0, 128.0,
128.3, 129.4, 137.2, 139.1, 147.8, 155.1, 159.5, 173.5; m/z
(ES+): 515 (35%, [M + Na]+), 493 (100%, [MH]+); HRMS
6.87–6.95 (2H, m, 2 of NC6H5), 7.06–7.33 (7H, m, 3 of CC6H5,
3 of NC6H5, CHPyr), 7.53–7.64 (2H, m, 2 × CHPyr), 7.55 (2H,
m, 2 of CC6H5); δC (75 MHz, CDCl3): 26.2, 42.1, 56.5, 56.6,
78.5, 115.4, 118.0, 120.3, 122.2, 124.0, 125.1, 125.9, 127.1,
134.5, 141.0, 147.0, 155.6, 158.1, 177.2; m/z (ES+): 427 (50%,
[M + Na]+), 411 (45%, [M + Li]+), 405 (100%, [MH]+); HRMS
(ES+): C25H28N2O3Li+ requires 411.2260, found 411.2266.
(S)-3-(6-((Diphenylamino)methyl)pyridin-2-yl)-2-hydroxy-2-
phenyl propanoic acid 2d. Following the procedure outlined
above for preparation of 2a from 12a, but using methanol
(0.2 mL) as the co-solvent in place of THF, 12d (0.22 g,
0.5 mmol) was converted to 2d as a white solid (0.18 g, 90%).
Rf 0.2 (DCM–MeOH 10 : 1); m.p. 112–117 °C; [α]2D0 −14.0
(CHCl3, c 0.60); δH (300 MHz, CD3OD): 3.64 (1H, d, J =
14.5 Hz, 1 of CH2CC6H5), 3.53 (1H, d, J = 14.5 Hz, 1 of
CH2CC6H5), 4.57 (2H, s, CH2NC6H5), 6.65 (2H, m, 2 of
NC6H5), 7.05 (4H, d, 1 Hz, 4 of NC6H5), 7.12–7.29 (9H, m, 3
of CC6H5, 4 of NC6H5, 2 × CHPyr), 7.54 (1H, t, J = 7.0 Hz,
CHPyr), 7.76 (2H, d, J = 1.0 Hz, CC6H5); δC (75 MHz, CD3OD–
CDCl3): 45.6, 59.4, 82.1, 121.2, 122.1, 123.2, 125.2, 127.7,
128.7, 129.6, 130.9, 140.1, 144.0, 149.2, 159.5, 159.6, 179.3;
m/z (ES+): 447 (60%, [M + Na]+), 425 (100%, [MH]+); HRMS
(ES+): C32H33N2O3 ([MH]+) requires 493.2491, found
+
493.2491.
(S)-2-Hydroxy-3-(6-((methyl(phenyl)amino)methyl)pyridin-
2-yl)-2-phenylpropanoic acid 2a. Aqueous lithium hydroxide
solution (1 M, 0.15 mL, 0.15 mmol) was added to a solution of
12a (50 mg, 0.12 mmol) in THF (0.1 mL) producing two
phases. The reaction mixture was heated at reflux (105 °C) for
17 hours during which time it became homogeneous. The result-
ing solution was allowed to cool to room temperature and
adjusted to pH 7 using aqueous hydrochloric acid (1 M). The
solvent was removed in vacuo to give a pale yellow solid which
was triturated with DCM (3 × 3 mL). The DCM phase was
reduced in vacuo to afford the crude product as a pale yellow
foam, which was triturated with ether (3 × 0.5 mL). The solvent
was decanted to give the title compound as an off-white solid
(43 mg, 81%); Rf 0.5 (DCM–MeOH 10 : 1); m.p. 76–82 °C;
[α]2D0 −33.2 (CHCl3, c 0.50); νmax (CHCl3): 3355 (w), 1604 (s),
1573 (m), 1504 (m), 1456 (w), 1447 (w), 1371 (m); δH
(300 MHz, CD3OD): 3.06 (3H, s, NCH3), 3.52 (1H, d, J = 14.0
Hz, 1 of CH2CC6H5), 3.72 (1H, d, J = 14.5 Hz, 1 of
CH2CC6H5), 4.56 (2H, s, CH2N), 6.60–6.67 (3H, m, 3 of
NC6H5), 6.94 (1H, d, J = 7.5 Hz, CHPyr), 7.10–7.27 (6H, m, 3
of CC6H5, CHPyr, 2 of NC6H5), 7.49 (1H, t, J = 7.5 Hz, CHPyr),
7.69–7.72 (2H, m, 2 of CC6H5); δC (75 MHz, CDCl3): 39.5,
44.8, 58.65, 81.0, 112.6, 117.0, 118.8, 123.7, 126.5, 127.8,
128.4, 129.6, 137.8, 143.6, 149.3, 158.0, 159.3, 179.9; m/z
(ES+): 385 (100%, [M + Na]+), 363 (100%, [MH]+); HRMS
(ES+): C27H25N2O3 ([MH]+) requires 425.1865, found
+
425.1874.
N,N′-(Pyridine-2,6-diylbis(methylene))bis(N-ethylaniline)
5.
N-Ethylaniline 9b (0.12 mL, 0.90 mmol) was dissolved in THF
(2.0 mL) and the solution was cooled to 0 °C. n-BuLi (1.5 M,
0.66 mL, 1.0 mmol) was added and the solution was stirred for
20 minutes at 0 °C. DMPU (0.12 mL, 1.0 mmol) was added and
the solution was stirred for a further 20 minutes, then added via
cannula to a solution of 2,6-dibromomethylpyridine (100 mg,
0.4 mmol)41 in THF (4.0 mL) cooled to −41 °C. The solution
was stirred at −41 °C for 16 hours after which time it was
poured onto half-saturated aqueous ammonium chloride solution
(3 mL). The aqueous phase was washed with ethyl acetate (3 ×
10 mL) and the organic layers combined and washed with brine,
dried (MgSO4) and reduced in vacuo to yield an orange oil. The
crude product solidified on standing and was recrystallised from
ethyl acetate to afford the title compound as a white solid
(100 mg, 73%); Rf 0.3 (hexane–ethyl acetate, 5 : 1); m.p.:
151–156 °C; νmax (CHCl3): 3002 (w), 2970 (m), 1600 (s), 1573
(s), 1500 (s), 1348 (s), 1245 (m); δH (200 MHz, CDCl3): 1.17
(6H, t, J = 7.0 Hz, 2 × (NCH2CH3)), 3.46 (4H, q, J = 7.0 Hz, 2
× (NCH2CH3)), 4.60 (4H, s, 2 × (CH2NC6H5)), 6.60–6.65 (6H,
m, 6 of 2 × (NC6H5)), 7.01–7.17 (6H, m, 2 of 2 × (NC6H5), 2 ×
CHPyr), 7.44 (1H, t, J = 7.5 Hz, CHPyr); δC (50 MHz, CDCl3):
12.8, 46.1, 56.8, 112.6, 116.8, 119.3, 129.8, 137.9, 148.6, 160.0;
m/z (ES+): 346 (100%, [MH]+), 225 (27%, [M − PhNEt]+);
+
(ES+): C22H23N2O3 ([MH]+) requires 363.1709, found
363.1710.
HRMS (ES+): C23H28N3 ([MH]+)requires: 346.2283, found
+
(S)-3-(6-((tert-Butyl(phenyl)amino)methyl)pyridin-2-yl)-2-
hydroxy-2-phenylpropanoic acid 2c. Following the procedure
outlined above for preparation of 2a from 12a, 12c (0.26 g,
0.6 mmol) gave 2c as an off-white solid (0.20 g, 90%). Rf 0.2
(DCM–MeOH 10 : 1); m.p. 98–100 °C; [α]2D0 −56.5 (CHCl3, c
0.69); νmax (CHCl3): 3200 (w), 1629 (s), 1596 (s), 1575 (m),
1491 (m), 1460 (m), 1365 (s); δH (300 MHz, CDCl3): 1.08 (9H,
s, NC(CH3)3), 3.35 (1H, d, J = 14.0 Hz, 1 of CH2CC6H5), 3.43
(1H, d, J = 14.5 Hz, 1 of CH2CC6H5), 4.33 (2H, s, CH2NC6H5),
346.2279.
N-Ethyl-N-(pyridin-2-ylmethyl)aniline 6. N-Ethylaniline 9b
(0.81 mL, 1.46 mmol) was dissolved in THF (4.0 mL) and
stirred at 0 °C while n-BuLi (1.5 M, 1.95 mL, 2.92 mmol) was
added via syringe. The resulting yellow solution was stirred at
0 °C for 20 minutes, then DMPU (0.35 mL, 1.46 mmol) was
added and stirring continued for a further 10 minutes. The sol-
ution was cooled to −78 °C and added dropwise via cannula to a
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7372–7381 | 7379

View Article Online
suspension of 2-(chloromethyl)pyridine (200 mg, 1.22 mmol)49
in THF (4.0 mL). The reaction mixture was allowed to slowly
warm to room temperature while stirring overnight. The clear,
orange solution that resulted was poured onto half-saturated
aqueous ammonium chloride solution (5 mL). The aqueous
phase was extracted with ethyl acetate (3 × 5 mL) and the com-
bined organic extracts were washed with brine (5 mL), dried
(MgSO4) and reduced in vacuo to give the crude product as an
orange oil. The product was purified by column chromatography
(hexane–ethyl acetate 5 : 1) to give the pure product as a bright
yellow oil (100 mg, 40%); Rf 0.15 (hexane–ethyl acetate 5 : 1);
νmax (CHCl3): 3089 (w), 2950 (s), 1600 (s), 1585 (s) 1492 (s),
1350 (s), 1272 (m); δH (200 MHz, CDCl3): 1.14 (3H, t, J =
7.0 Hz, CH2CH3), 3.44 (2H, q, J = 7.0 Hz, CH2CH3), 4.53 (2H,
s, NCCH2), 6.55–6.61 (3H, m, 3 of NC6H5), 7.01–7.13 (4H, m,
2 of NC6H5, 2 × CHPyr), 7.47 (1H, dt, J = 7.5 Hz, 8.0 Hz,
CHPyr), 8.49 (1H, q, J = 4.0 Hz, CHPyr); δC (50 MHz, CDCl3):
12.6, 46.1, 56.7, 112.6, 116.8, 120.4, 121.2, 129.6, 137.2, 148.5,
149.9, 160.2; m/z (ES+): 213 (100%, [MH]+); HRMS (ES+):
[C14H17N2]+ requires 213.1392, found 213.1382.
acetate (1.7 mg, 9.5 μmol) in degassed, dry methanol (0.1 mL)
was added to a solution of the disodium salt of the ligand
(4.0 mg, 9.5 μmol) in methanol (0.2 mL) under argon to give a
yellow solution. The reaction was stirred for 20 minutes after
which time it was diluted with methanol (30 mL) and the alkene
substrate (9.5 mmol) was added. Hydrogen peroxide (30%
aqueous, 97 μmol) in methanol (1.0 mL) was added by syringe
pump over 4 hours. The reaction was stirred at room temperature
for 24 hours. The solution was concentrated in vacuo and diluted
with ethyl acetate (∼2 mL) and passed through a short plug of
silica to remove insolubles. An internal standard (n-decane) was
added and products were analysed by GC or GCMS. Products
were identified unambiguously in GC and GCMS analyses by
comparison and spiking with authentic samples. Analysis on the
HP-1s column achieved separation of products 15–19, anilines
9a–d and benzaldehyde; further analysis using the BP-20
column was required to confirm the stereochemistry of the cyclo-
hexane-1,2-diol product 15 (Rt = 10.84 minutes for the cis-
isomer 15 and 11.33 minutes for trans-cyclohexane-1,2-diol).
We have reported the synthesis and characterisation of com-
pounds 2b and 12b previously.39 Synthetic procedures and data
for these compounds are included in the ESI† for reference.
Ligand degradation experiments
a. Iron, ligand and oxidant: The sodium salt 2b-Na (2.0 mg,
5.0 μmol) was dissolved in methanol (0.1 mL). Iron(II) acetate
(0.85 mg, 5.0 μmol) in methanol (0.2 mL) was added, giving a
yellow solution which was stirred at room temperature for
45 minutes. The solution was diluted with methanol (7.2 mL)
and stirred at room temperature while hydrogen peroxide (30%
aqueous, 50 μmol) in methanol (0.5 mL) was added over
4 hours. The reaction was stirred at room temperature for a
Oxidative turnover reactions
Ligand (0.13 mmol) was dissolved in anhydrous DCM
(0.75 mL) and treated with NaH (12 mg, 0.53 mmol). After stir-
ring for 45 minutes at room temperature the solvent was
removed in vacuo to give a white powder. A solution of iron(II)
Table 2 Crystal data and structure refinement
Compound no.
11
12d
Empirical formula
Formula weight
Temperature
C20H22NO3Br
404.30
100(2) K
C32H32N2O3
492.60
113(2) K
0.71073 Å
Wavelength
0.71073 Å
Crystal system
Space group
Unit cell dimensions
Monoclinic
P21 (#4)
a = 9.8965(17) Å α = 90°
b = 8.9206(15) Å β = 113.274(2)°
c = 11.355(2) Å γ = 90°
920.9(3) Å3
Orthorhombic
P212121 (#19)
a = 10.0288(13) Å α = 90°
b = 15.960(2) Å β = 90°
c = 16.609(2) Å γ = 90°
2658.5(6) Å3
4
Volume
Z
Density (calculated)
Absorption coefficient
F(000)
Crystal size
θ range for data collection
Index ranges
Reflections collected
Independent reflections
Completeness to θmax
Absorption correction
Max. and min. transmission
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2σ(I)]
R indices (all data)
Flack×parameter
2
1.458 Mg m−3
1.231 Mg m−3
0.079 mm−1
2.250 mm−1
416
1048
1.00 × 0.80 × 0.50 mm3
1.95–28.31°
0.50 × 0.50 × 0.50 mm3
1.77–26.00°
−12 < = h < = 12, −11 < = k < = 11, −14 < = l < = 14
7231
3936 [R(int) = 0.0359]
92.9%
−12 < = h < = 12, −19 < = k < = 19, −20 < = l < = 20
41 199
2952 [R(int) = 0.0235]
100.0%
Semi-empirical from equivalents
0.3991 and 0.1088
Semi-empirical from equivalents
0.9616 and 0.8218
Full-matrix least-squares on F2
3936/1/229
Full-matrix least-squares on F2
2952/0/337
1.004
1.039
R1 = 0.0438, wR2 = 0.1007
R1 = 0.0490, wR2 = 0.1020
0.047(10)
R1 = 0.0291, wR2 = 0.0733
R1 = 0.0297, wR2 = 0.0740
— (Friedel pairs merged)
0.236 and −0.138 e Å−3
Largest diff. peak and hole
1.102 and −0.802 e Å−3
7380 | Org. Biomol. Chem., 2012, 10, 7372–7381
This journal is © The Royal Society of Chemistry 2012

View Article Online
15 E. L. Hegg and L. Que Jr., Eur. J. Biochem., 1997, 250, 625.
16 M. Costas, K. Chen and L. Que, Coord. Chem. Rev., 2000, 200–202,
517.
17 C. He and Y. Mishina, Curr. Opin. Chem. Biol., 2004, 8, 201.
18 J. M. Bollinger Jr and C. Krebs, J. Inorg. Biochem., 2006, 100, 586.
19 J. England, Y. Guo, E. R. Farquhar, V. G. Young Jr, E. Münck and
L. Que Jr, J. Am. Chem. Soc., 2010, 132, 8635.
20 S. Ye and F. Neese, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 1228.
21 J. C. Price, E. W. Barr, B. Tirupati, J. M. Bollinger and C. Krebs,
Biochemistry, 2003, 42, 7497.
further 16 hours then concentrated in vacuo. Products were ana-
lysed by LCMS (water–acetonitrile with 0.2% TFA; gradient
elution from 90–100% acetonitrile).
b. Iron-free control (ligand and oxidant): The sodium salt
2b-Na (2.0 mg, 5.0 μmol) was dissolved in methanol (7.5 mL)
and treated to hydrogen peroxide (30% aqueous, 50 μmol) in
methanol (0.5 mL) as above. The reaction mixture was concen-
trated in vacuo and analysed by LCMS.
22 L. M. Hoffart, E. W. Barr, R. B. Guyer, J. M. Bollinger and C. Krebs,
Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 14738.
23 B. E. Eser, E. W. Barr, P. A. Frantom, L. Saleh, J. M. Bollinger, C. Krebs
and P. F. Fitzpatrick, J. Am. Chem. Soc., 2007, 129, 11334.
24 K. Chen and L. Que Jr., Angew. Chem., Int. Ed., 1999, 38, 2227.
25 J. Bautz, P. Comba, C. L. de Laorden, M. Menzel and G. Rajaraman,
Angew. Chem., Int. Ed., 2007, 46, 8067.
26 M. S. Chen and M. C. White, Science, 2007, 318, 783.
27 K. Suzuki, P. D. Oldenburg and L. Que, Angew. Chem., Int. Ed., 2008,
47, 1887.
28 M. S. Chen and M. C. White, Science, 2010, 327, 566.
29 J. England, M. Martinho, E. R. Farquhar, J. R. Frisch, E. L. Bominaar,
E. Münck and L. Que, Angew. Chem., Int. Ed., 2009, 48, 3622.
30 J. England, Y. Guo, K. M. Van Heuvelen, M. A. Cranswick, G. T. Rohde,
E. L. Bominaar, E. Münck and L. Que, J. Am. Chem. Soc., 2011, 133,
11880.
31 J. U. Rohde, J. H. In, M. H. Lim, W. W. Brennessel, M. R. Bukowski,
A. Stubna, E. Munck, W. Nam and L. Que, Science, 2003, 299, 1037.
32 B. B. Bassan, M. R. A. Blomberg, P. E. M. Siegbahn and L. Que Jr.,
Angew. Chem., Int. Ed., 2005, 44, 2939.
Crystallography
Crystal data were collected using a Bruker SMART APEX CCD
area detector diffractometer. A full sphere of reciprocal space
was scanned by phi-omega scans. Pseudo-empirical absorption
correction based on redundant reflections was performed by the
program SADABS.55 The structures were solved by direct
methods using SHELXS-97 (X-2) and refined by full matrix
least-squares on F2 for all data using SHELXL-97.56 Friedel
pairs were merged in 12d. Hydrogen atoms were added at calcu-
lated positions and refined using a riding model. Their isotropic
thermal displacement parameters were fixed to 1.2 times (1.5
times for methyl groups) the equivalent one of the parent atom.
Anisotropic thermal displacement parameters were used for all
non-hydrogen atoms. Further refinement details are listed in
Table 2.
33 Y.-M. Lee, S. Hong, Y. Morimoto, W. Shin, S. Fukuzumi and W. Nam,
J. Am. Chem. Soc., 2010, 132, 10668.
34 S. Hong, Y.-M. Lee, W. Shin, S. Fukuzumi and W. Nam, J. Am. Chem.
Soc., 2009, 131, 13910.
35 A. Thibon, J. England, M. Martinho, V. G. Young, J. R. Frisch,
R. Guillot, J.-J. Girerd, E. Münck, L. Que and F. Banse, Angew. Chem.,
Int. Ed., 2008, 47, 7064.
36 L. Que, Acc. Chem. Res., 2007, 40, 493.
37 I. Prat, J. S. Mathieson, M. Güell, X. Ribas, J. M. Luis, L. Cronin and
M. Costas, Nat. Chem., 2011, 3, 788.
38 J. A. Krall, P. J. Rutledge and J. E. Baldwin, Tetrahedron, 2005, 61,
137.
39 S. M. Barry and P. J. Rutledge, Synlett, 2008, 2172.
40 V. J. Dungan, Y. Ortin, H. Mueller-Bunz and P. J. Rutledge, Org. Biomol.
Chem., 2010, 8, 1666.
Acknowledgements
This work was supported by the Irish Research Council for
Science, Engineering and Technology (IRCSET) via an Embark
Award postgraduate scholarship to SMB, the Centre for Syn-
thesis and Chemical Biology at University College Dublin under
the Programme for Research in Third Level Institutions (PRTLI)
administered by the HEA, and the University of Sydney.
41 V. J. Dungan, S. M. Wong, S. M. Barry and P. J. Rutledge, Tetrahedron,
2012, 68, 3231.
42 A. Karlsson, J. V. Parales, R. E. Parales, D. T. Gibson, H. Eklund and
S. Ramaswamy, Science, 2003, 299, 1039.
43 D. Seebach and R. Naef, Helv. Chim. Acta, 1981, 64, 2704.
44 C. F. Macrae, I. J. Bruno, J. A. Chisholm, P. R. Edgington, P. McCabe,
E. Pidcock, L. Rodriguez-Monge, R. Taylor, J. van de Streek and
P. A. Wood, J. Appl. Crystallogr., 2008, 41, 466.
45 A. R. Johnson, C. C. Cummins and S. Gambarotta, Inorg. Synth., 1998,
32, 123.
46 E. Juaristi, P. Murer and D. Seebach, Synthesis, 1993, 1243.
47 F. Sorm and L. Sedivy, Collect. Czech. Chem. Commun., 1948, 13,
289.
48 M. R. Bryce, J. G. Eaves, D. Parker, J. A. K. Howard and O. Johnson,
J. Chem. Soc., Perkin Trans. 2, 1985, 433.
49 K. Winterfeld and K. Flick, Arch. Pharm., 1956, 289, 448.
50 D. T. Sawyer, A. Sobkowiak and T. Matsushita, Acc. Chem. Res., 1996,
29, 409.
51 S. Goldstein and D. Meyerstein, Acc. Chem. Res., 1999, 32, 547.
52 E. Neyens and J. Baeyens, J. Hazard. Mater., 2003, 98, 33.
53 L. Zs-Nagy and R. A. Floyd, Biochim. Biophys. Acta, 1984, 790, 238.
54 E. R. Stadtman and B. S. Berlett, J. Biol. Chem., 1991, 266, 17201.
55 Bruker, SADABS, Bruker AXS Inc, Madison, Wisconsin, USA, 2001.
56 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 2008,
A64, 112.
References
1 D. H. R. Barton, Chem. Soc. Rev., 1996, 25, 237.
2 P. Gamez, P. G. Aubel, W. L. Driessen and J. Reedijk, Chem. Soc. Rev.,
2001, 30, 376.
3 J. Piera and J. E. Backvall, Angew. Chem., Int. Ed., 2008, 47, 3506.
4 S. I. Murahashi and D. Zhang, Chem. Soc. Rev., 2008, 37, 1490.
5 T. D. H. Bugg, Tetrahedron, 2003, 59, 7075.
6 M. Costas, M. P. Mehn, M. P. Jensen and L. Que, Chem. Rev., 2004, 104,
939.
7 W. Nam, Acc. Chem. Res., 2007, 40, 522.
8 P. C. A. Bruijnincx, G. van Koten and R. J. M. Klein Gebbink, Chem.
Soc. Rev., 2008, 37, 2716.
9 A. Daruzzaman, I. J. Clifton, R. M. Adlington, J. E. Baldwin and
P. J. Rutledge, ChemBioChem, 2006, 7, 351.
10 M. L. Neidig and E. I. Solomon, Chem. Commun., 2005, 5843.
11 M. M. Abu-Omar, A. Loaiza and N. Hontzeas, Chem. Rev., 2005, 105,
2227.
12 W. Ge, I. J. Clifton, J. E. Stok, R. M. Adlington, J. E. Baldwin and
P. J. Rutledge, J. Am. Chem. Soc., 2008, 130, 10096.
13 W. Ge, I. J. Clifton, A. R. Howard-Jones, J. E. Stok, R. M. Adlington,
J. E. Baldwin and P. J. Rutledge, ChemBioChem, 2009, 10, 2025.
14 A. R. Grummitt, P. J. Rutledge, I. J. Clifton and J. E. Baldwin, Biochem.
J., 2004, 382, 659.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 7372–7381 | 7381