Solution and structural characterization of iron(II) complexes with ortho -halogenated phenolates: Insights into potential substrate binding modes in hydroquinone dioxygenases

10914 Inorg. Chem. 2010, 49, 10914–10929

DOI: 10.1021/ic101377u

Solution and Structural Characterization of Iron(II) Complexes

with Ortho-Halogenated Phenolates: Insights Into Potential Substrate Binding

Modes in Hydroquinone Dioxygenases

Sara S. Rocks,^†William W. Brennessel,^†Timothy E. Machonkin,*^,‡and Patrick L. Holland*^,†

^†Department of Chemistry, University of Rochester, Rochester, New York 14627, United States, and

^‡Department of Chemistry, Whitman College, Walla Walla, Washington 99362, United States

Received July 10, 2010

The new ligand cis,cis-1,3,5-tris-(E)-(tolylideneimino)cyclohexane (TACH-o-tolyl) forms a 1:1 complex with iron(II).

Addition of substituted phenolates forms 1:1:1 ligand:iron:phenolate complexes, which have been characterized both

in the solid state and in solution. There is complete binding of the phenolate to the complex only when there are ortho-

halogens on the phenolate. The tertiary complexes with ortho-halo-substituted phenolates exhibit short Fe-halogen

distances, and the complex containing a non-coordinating but similarly sized ortho-methyl phenolate has a significantly

different conformation and coordination geometry. Therefore, it is likely that the metal-halogen interaction stabilizes

the complexes. The iron(II)-halogen interaction in these complexes may explain the substrate specificity of PcpA

and LinE, enzymes that preferentially bind phenols and hydroquinones containing halogen substituents in ortho

positions.

Introduction

The degradation pathway for pentachlorophenol by

S. chlorophenolicum is shown in Scheme 1.^4,5The key oxidative

ring cleavage step is catalyzed by 2,6-dichlorohydroquinone

1,2-dioxygenase (PcpA), which is an iron(II) dependent

enzyme.^5,6γ-Hexachlorocylohexane, or lindane, is also catab-

olized through a chlorohydroquinone cleavage pathway

that includes the enzyme chlorohydroquinone 1,2-dioxygen-

ase, LinE.⁷Thus, the PCP and lindane pathways proceed

through the oxidative ring cleavage of a chlorinated hydro-

quinone catabolic intermediate. This is a key difference from

most bacterial arene degradation pathways, which proceed

through the oxidative ring cleavage of a catechol.⁸LinE and

PcpA are both non-heme iron(II)-containing dioxygenases,

like the well-studied extradiol catechol dioxygenase (EDO)

enzymes.⁸Site-directed mutagenesis and homology modeling

have determined that, like the EDOs, the iron(II) in PcpA is

ligated facially by two histidines and a glutamate (Figure 1).⁹

LinE has 51% sequence identity to PcpA, including the

The remediation of chlorinated pollutants is an area of

significant interest to the scientific community. Pentachloro-

phenol (PCP) is a carcinogen and a biocide that is still used in

the United States; however, its use has been restricted since

1987 to use as a wood preservative for power poles and

railroad ties.¹The stability and attendant longevity of such

chlorinated pollutants in the environment have created a

problem that spans generations. One strategy for dealing

with this problem is bioremediation, which requires the

isolation of organisms that have evolved the capability to

degrade undesirable molecules such as PCP. Some bacteria

can degrade chlorophenols, but their tolerance of different

chlorophenols is very specific to the degree of chlorination of

the ring.²Interestingly, some bacteria, such as Sphingobium

chlorophenolicum (formerly called Sphingomonas chloro-

phenolica) ATCC 39723, can use pentachlorophenol as their

sole carbon source.³

(4) (a) Cai, M.; Xun, L. Y. J. Bacteriol. 2002, 184, 4672–4680. (b) Dai,

M. H.; Copley, S. D. Appl. Environ. Microbiol. 2004, 70, 2391–2397.

(5) Xu, L.; Resing, K.; Lawson, S. L.; Babbitt, P. C.; Copley, S. D.

Biochemistry 1999, 38, 7659–7669.

(6) (a) Ohtsubo, Y.; Miyauchi, K.; Kanda, K.; Hatta, T.; Kiyohara, H.;

Senda, T.; Nagata, Y.; Mitsui, Y.; Takagi, M. FEBS Lett. 1999, 459, 395–

398. (b) Xun, L.; Bohuslavek, J.; Cai, M. Biochem. Biophys. Res. Commun.

1999, 266, 322–325.

(7) (a) Endo, R.; Kamakura, M.; Miyauchi, K.; Fukuda, M.; Ohtsubo,

Y.; Tsuda, M.; Nagata, Y. J. Bacteriol. 2005, 187, 847–853. (b) Miyauchi, K.;

Adachi, Y.; Nagata, Y.; Takagi, M. J. Bacteriol. 1999, 181, 6712–6719. (c) Nagata,

Y.; Endo, R.; Ito, M.; Ohtsubo, Y.; Tsuda, M. Appl. Microbiol. Biotechnol. 2007,

76, 741–752.

*To whom correspondence should be addressed. E-mail: machonte@

whitman.edu (T.E.M.), holland@chem.rochester.edu (P.L.H.).

(1) (a) Toxicological Profile for Pentachlorophenol; PB/2001/109106/AS;

U.S. Department of Health and Human Services, Agency for Toxic Substances

and Disease Registry: Atlanta, GA, 2001. (b) Reregistration Eligibility Decision

for Pentachlorophenol; EPA 739-R-08-008; U.S. Environmental Protection

Agency, Office of Pesticide Programs: Arlington, VA, 2008.

(2) Haggblom, M. M. FEMS Microbiol. Rev. 1992, 103, 29–72.

(3) (a) Saber, D. L.; Crawford, R. L. Appl. Environ. Microbiol. 1985, 50,

1512–1518. (b) Ederer, M. M.; Crawford, R. L.; Herwig, R. P.; Orser, C. S. Mol.

Ecol. 1997, 6, 39–49. (c) Nohynek, L. J.; Suhonen, E. L.; Nurmiaho-Lassila, E. L.;

Hantula, J.; Salkinoja-Salonen, M. Syst. Appl. Microbiol. 1996, 18, 527–538.

r

pubs.acs.org/IC

Published on Web 11/08/2010

2010 American Chemical Society

Article

Inorganic Chemistry, Vol. 49, No. 23, 2010 10915

of the protein, it is possible to understand the inherent

selectivity, binding modes, and reactivity of the metal ion.

Using this information, one may deconvolute the factors that

stem from the inherent chemistry of the metal center from

those that come from the protein active site pocket. For

PcpA, such data are essential in the effort to develop an

understanding of how this enzyme uses chlorinated sub-

strates without inactivation.

Catechol complexes of iron have been studied for many

years to provide insight into substrate binding, spectroscopy,

and mechanism of EDO enzymes.¹¹On the other hand, little

synthetic chemistry has been done to elucidate the binding of

hydroquinones (the substrates of PcpA and LinE) with

mononuclear iron complexes. A search of the Cambridge

Crystallographic Database (CCD) and Gmelin produced

only one example of a chlorinated hydroquinone on iron,

an unpublished structure of a diiron(III) porphyrin complex

bridged by tetrachlorohydroquinonate.¹²This example high-

lights a significant difficulty, which is that hydroquinones can

bridge multiple metal ions. In the only two known solid-state

structures containing an unsupported hydroquinone coordi-

nated to non-heme iron, the hydroquinone bridges between

metals.¹³

Figure 1. Proposed active site of PcpA based upon the resting state of

EDOs (left) compared with the iron(II) complex of a TACH-based

triimine ligand (right).

Scheme 1. Pathway for the Degradation of Pentachlorophenol in

S. chlorophenolicum^4,5

In the work presented below, the strategy for avoiding

hydroquinone bridging was to use phenols as proxies for

hydroquinones, which assumes that the para hydroxyl group

does not alter the influence of ortho-halogen substituents on

phenolate binding. The idea that phenols bind similarly to

hydroquinones in the enzyme is supported by the fact that

ortho-halogenated phenols are potent inhibitors of PcpA.¹⁴

Interestingly, even with the more common phenolate as a

ligand, iron(II) coordination chemistry is understudied.

Though many iron(III)-phenolate complexes are known,

iron(II) complexes with unsupported phenolate ligands are

comparatively rare (see Discussion below). The known iron-

phenolate complexes do not provide insight into the influence

of chlorine substituents on iron binding because there are no

crystallographically characterized mononuclear complexes

of iron(II) with chlorinated phenolates.¹⁵

The goal of this work was to investigate the fundamental

coordination chemistry of non-tethered hydroquinones and

phenols on a Fe(II) center supported by a tridentate ligand,

and to examine how ortho-substituents affect the binding

mode. These synthetic model compounds are expected to

elucidate the possible origins of substrate selectivity in LinE

and PcpA, and to expand our understanding of the basic

coordination chemistry of ortho-halogenated phenols on

iron(II).

conserved two histidines and glutamate, suggesting that ithas

a similar active site.⁷

Though EDO enzymes and the chlorohydroquinone diox-

ygenase enzymes have similar activities (oxidative ring

cleavage), metal ion cofactor (iron(II)), and active site ligands

(two histidines, one glutamate), there is a surprising differ-

ence between the different types of enzymes. Namely, EDO

enzymes are inactivated by chlorinated substrates,¹⁰while

PcpA and LinE are use chlorinated molecules as their native

substrates.^5-7It is interesting that iron(II) active sites with

similar geometry that catalyze similar reactions have such

different patterns of selectivity and inactivation. Understand-

ing the differences between catechol dioxygenases and PcpA

may help to elucidate the mechanisms through which speci-

ficity for chlorinated substrates can arise. This information,

in turn, may help scientists to design and utilize better systems

for bioremediation of chlorinated pollutants.

One way to explore the determinants of substrate selec-

tivity is through the synthesis of model complexes with the

same metal and geometry as the enzyme. By design, these

complexes lack the protein scaffold. By removing the influence

(8) (a) Bugg, T. D. H.; Winfield, C. J. Nat. Prod. Rep. 1998, 513–530.

(b) Solomon, E. I.; Brunold, T. C.; Davis, M. I.; Kemsley, J. N.; Lee, S.-K.;

Lehnert, N.; Neese, F.; Skulan, A. J.; Yang, Y.-S.; Zhou, J. Chem. Rev. 2000, 100,

235–349. (c) Bugg, T. D. H.; Lin, G. Chem. Commun. 2001, 941–952. (d) Vaillancourt,

F. H.; Bolin, J. T.; Eltis, L. Ring-Cleavage Dioxygenases. In Pseudomonas;

Ramos, J.-L., Ed.; Kluwer Academic/Plenum: New York, 2004; Vol. 3,

pp 359-396. (e) Vaillancourt, F. H.; Bolin, J. T.; Eltis, L. D. Crit. Rev. Biochem.

Mol. Biol. 2006, 41, 241–267. (f) Bugg, T. D. H.; Ramaswamy, S. Curr. Opin.

Chem. Biol. 2008, 12, 134–140.

Results

Ligand Design Strategy. cis,cis-1,3,5-Triaminocyclo-

hexane (TACH) is a well-precedented template for the

synthesis of multidentate ligands that are preorganized to

(9) Machonkin, T. E.; Holland, P. L.; Smith, K. N.; Liberman, J. S.;

Dinescu, A.; Cundari, T. R.; Rocks, S. S. J. Biol. Inorg. Chem. 2010, 10, 291–

301.

(10) (a) Klecka, G. M.; Gibson, D. T. Appl. Environ. Microbiol. 1981, 41,

1159–1165. (b) Bartels, I.; Knackmuss, H.-J.; Reineke, W. Appl. Environ.

Microbiol. 1984, 47, 500–505. (c) Cerdan, P.; Wasserfallen, A.; Rekik,

M.; Timmis, K. N.; Harayama, S. J. Bacteriol. 1994, 176, 6074–6081. (d)

(11) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L. Chem. Rev. 2004,

104, 939–986.

(12) Rheingold, A. L.; Miller, J. Private Communication to Cambridge

Structural Database, 2003.

(13) (a) Heistand, R. H.; Roe, A. L.; Que, L. Inorg. Chem. 1982, 21, 676–

681. (b) Maroney, M. J.; Day, R. O.; Psyris, T.; Fleury, L. M.; Whitehead, J. P.

Inorg. Chem. 1989, 28, 173–175.

(14) Doerner, A. E.; Machonkin, T. E. unpublished work.

(15) Cambridge Structural Database, v. 5.31 (November 2009 update):

Allen, F. H. Acta Crystallogr. 2002, B58, 380–388.

ꢀ

Vaillancourt, F. H.; Labbe, G.; Drouin, N. M.; Fortin, P. D.; Eltis, L. D. J. Biol.

Chem. 2002, 277, 2019–2027. (e) Dai, S.; Vaillancourt, F. H.; Maaroufi, H.;

Drouin, N. M.; Neau, D. B.; Snieckus, V.; Bolin, J. T.; Eltis, L. D. Nat. Struct.

Biol. 2002, 9, 934–939.

10916 Inorganic Chemistry, Vol. 49, No. 23, 2010

Rocks et al.

give fac coordination of three donor atoms. In the work

described here, we use imine donors because of the

electronic similarity to the biological histidine residues.

The ability to control the level of steric protection around

the metal is an additional asset in this class of ligands,

since TACH can be condensed with a variety of different

aldehydes.¹⁶These factors made a TACH-based imine an

attractive choice of ligand with which to synthesize a 1:1

ligand:iron(II) complex that would have additional co-

ordination sites on iron for phenolate binding.

Scheme 2. Synthesis of TACH-o-tolyl and its Iron(II) Complex

The reaction of cis,cis-1,3,5-tris-(E)-benzylideneami-

nocyclohexane (TACH-benz) with copper(II), zinc(II),

and nickel(II) has been reported, and the use of hydrated

metal salts led to partial hydrolysis of the ligand.¹⁷In the

work presented here, rigorous exclusion of water during

the reaction of Fe(II) with TACH-benz prevented ligand

hydrolysis. Preliminary reactions of TACH-benz with

iron(II) triflate in CD₃CN (using techniques similar to

those described below) indicated that addition of only one

molar equivalent of ligand led to formation of a 2:1

ligand:iron(II) complex, and therefore the phenyl groups

did not provide sufficient steric bulk.¹⁸A TACH-based

ligand with mesityl groups instead of phenyl groups has

been reported to form a 1:1 complex with Cu(II),¹⁹but

when we treated an isosteric ligand containing 2,6-

dimethylphenyl substituents (cis,cis-1,3,5-tris-(E)-xylyli-

deneaminocyclohexane, TACH-xyl) with iron(II), ¹H NMR

spectra indicated incomplete binding. We surmised that

this ligand was too bulky to form the desired 1:1 complex

of tridentate ligand and iron(II).¹⁸

electronic configuration. The number of acetonitrile sol-

vent molecules bound to iron is unknown.

A second equivalent of TACH-o-tolyl was added to a

1:1 mixture of ligand and iron(II) triflate in CD₃CN, and

the changes in the ¹H NMR signal intensities were deter-

mined with a capillary integration standard of cobalto-

cene. The paramagnetically shifted ¹H NMR signals

did not increase in intensity, the free ligand signals did

increase in intensity, and no new signals were observed.

From the absence of new signals and lack of increased

intensity of observed paramagnetic signals, we conclude

that a 2:1 complex does not form between 1 and iron(II)

triflate in CH₃CN. Therefore, using an ortho-methyl

group on the benzylidene arm provides enough steric

bulk to prevent 2:1 ligand:metal complexes, yet not so

much steric hindrance that metal binding is prevented in

CH₃CN. Anaerobic electrospray mass spectra of a solu-

tion generated from 1 and iron(II) triflate contained a

signal at m/z 640 that corresponds to [Fe(TACH-o-tolyl)-

(OTf)]^þ(Scheme 2) and is consistent with the formation

of a 1:1 ligand:Fe^2þcomplex. The ¹⁹F NMR spectrum has

a single broadened peak near the position of free triflate,

suggesting that the triflate anions rapidly exchange on

and off of the metal. On the basis of these measurements,

we formulate this complex as (TACH-o-tolyl)Fe(OTf)₂,

with the triflate ions exchanging between outer-sphere

and inner-sphere in solution. Since the complex partially

dissociates the TACH-o-tolyl ligand in solution, it exists

as a mixture and therefore was not fully characterized.

This mixture was typically formed in situ for subsequent

reactions with phenolates; as seen below, addition of

certain phenolates or hydroquinonates led to complete

coordination of the TACH-o-tolyl ligand.

Therefore, we prepared a new tridentate ligand with

intermediate steric demands that would allow adequate

binding of the metal ion while still precluding the forma-

tion of the bis-ligand complex. This compound, cis,cis-

1,3,5-tris-(E)-(tolylideneimino)cyclohexane (TACH-o-

tolyl, 1), was prepared by condensation of TACH and

ortho-tolualdehyde and isolated in 78% yield (Scheme 2).

Compound 1 was combined with iron(II) triflate in

1

CD₃CN, and the resulting solution was analyzed by H

NMR spectroscopy. Nine signals are anticipated in com-

plexes of 1, assuming a C₃axis of symmetry through the

ligand. In the H NMR spectrum of the solution, there

1

are five highly shifted signals at δ 301, 268, 18.3, -3.5,

and -12.6 ppm, and four signals are observed in the

diamagnetic region (0-10 ppm) amid smaller signals for

unbound ligand. The relative integrations of bound and

unbound ligand in the H NMR spectrum indicate that

¹H NMR Studies of Reactions of (TACH-o-tolyl)Fe-

(OTf)₂with Hydroquinones and Phenols. A variety of

substituted phenols, as well as unsubstituted phenol and

methylhydroquinone, were each deprotonated to form

solutions of phenolate and hydroquinonate anions. (In

the remainder of the paper, “phenolate” will refer to either

a 1:1 mixture of phenol and triethylamine or a sodium

phenolate. The choice of deprotonation protocol led to

1

roughly 70% of the ligand is bound to iron. The wide

chemical shift dispersion in the H NMR spectrum of

the complex indicates that the iron(II) has a high-spin

1

(16) (a) Boxwell, C. J.; Bhalla, R.; Cronin, L.; Turner, S. S.; Walton, P. H.

J. Chem. Soc., Dalton Trans. 1998, 2449–2450. (b) Cronin, L.; Foxon, S. P.;

Lusby, P. J.; Walton, P. H. J. Biol. Inorg. Chem. 2001, 6, 367–377. (c) Nairn,

A. K.; Bhalla, R.; Foxon, S. P.; Liu, X. M.; Yellowlees, L. J.; Gilbert, B. C.;

Walton, P. H. J. Chem. Soc., Dalton Trans. 2002, 1253–1255. (d) Archibald,

S. J.; Foxon, S. P.; Freeman, J. D.; Hobson, J. E.; Perutz, R. N.; Walton, P. H.

J. Chem. Soc., Dalton Trans. 2002, 2797–2799.

(17) (a) Greener, B.; Cronin, L.; Wilson, G. D.; Walton, P. H. J. Chem.

Soc., Dalton Trans. 1996, 401–403. (b) Greener, B.; Moore, M. H.; Walton, P. H.

Chem. Commun. 1996, 27–28. (c) Cronin, L.; Greener, B.; Foxon, S. P.; Heath,

S. L.; Walton, P. H. Inorg. Chem. 1997, 36, 2594–2600.

1

no apparent differences in the H NMR signals of the

resulting complexes.) When each phenolate and hydro-

quinonate was reacted with a mixture of iron(II) triflate

and 1 in CD₃CN, the resultant H NMR spectrum was

1

consistent with the formation of the desired 1:1:1 ligand:

iron:phenolate or ligand:iron:hydroquinonate complex.

The ¹H NMR spectra are shown in Figure 2. For complexes

of some of the phenolates, the electrospray mass spectrum

was collected under rigorously anaerobic conditions, and

(18) Further details: Rocks, S. S. Ph.D. Thesis, University of Rochester,

Rochester, New York, 2009.

(19) Cushion, M.; Ebrahimpour, P.; Haddow, M. F.; Hallett, A. J.;

Mansell, S. M.; Orpen, A. G.; Wass, D. F. Dalton Trans. 2009, 1632–1635.

Article

Inorganic Chemistry, Vol. 49, No. 23, 2010 10917

The remaining signals overlap to such an extent that

further assignments are difficult. Even though they could

not be fully assigned, the ¹H NMR spectra indicate that a

new species forms in the presence of 2-methylhydroqui-

nonate, and from the integrations of non-overlapping

signals, the observed species is consistent with the 1:1:1

iron(II):TACH-o-tolyl:2-methylhydroquinonate complex.

Unfortunately, this complex is unstable and decomposes

to unidentified species at room temperature within hours,

preventing further characterization.

Because of the instability of the hydroquinonate com-

plex, subsequent reactions used phenolates, isosteric

hydroquinone analogues that are likely to bind to iron(II)

in PcpA because they are known inhibitors of enzymatic

activity.¹⁴Mixing 1, Fe(OTf)₂2CH₃CN, and phenolate

3

in CD₃CN yielded a bright yellow solution. Analysis by

¹H NMR spectroscopy showed that binding of phenolate

was incomplete (Figure 2). The highest intensity signals

corresponded to (TACH-o-tolyl)Fe(OTf)₂and to a sec-

ond species identified as “Fe(phenolate)_n”.²⁰There was

also a new species formed with an NMR signature that is

consistent with the desired 1:1:1 complex. For the latter

species, 11 signals were clearly observed while 12 are

expected. Because of the congestion of the spectrum in the

area from δ 11 to -3 ppm, determining the exact number

of signals that correspond to each product is difficult.

Two signals at δ 279 and 265 ppm are assigned as the

imine proton and the R-CH of the cyclohexane ring, as

above. Likewise, a large signal that integrates to 9 protons

is observed at δ -12.3 ppm and corresponds to the methyl

groups of the tolylidene arms. Signals at δ 59, 39, and -37

ppm that integrated to 2H, 2H, and 1H, respectively, were

tentatively assigned as the meta, ortho, and para protons

of the bound phenolate in the 1:1:1 complex. This assign-

ment was confirmed by mixing Fe(TACH-o-tolyl)(OTf)₂

with the deuterated phenolate C₆D₅O^-in CD₃CN, which

gave a spectrum that was identical except for the absence

of signals at these chemical shifts. Overall, the relative

integrations and the approximate number of observed

signals in the new species are consistent with formation of

the desired 1:1:1 iron:ligand:phenolate product, although

this is not the only species in solution.

Figure 2. ¹H NMR spectra of products from adding deprotonated

2-methylhydroquinone (MeHQ) or phenols to Fe(OTf)₂2CH₃CN and

1 in CD₃CN at 25 °C. Proton signals corresponding to the bound

phenolates are denoted with an asterisk (*). In the products from MeHQ

and phenol, the peaks from (TACH-o-tolyl)Fe(OTf)₂(•) and “Fe-

(phenolate)_n” (P) are indicated, and the bound phenolate/hydroquino-

nate peaks (*) were assigned by deuterating the substrates.

3

these spectra showed a parent ion corresponding to the

expected [(TACH-o-tolyl)Fe(phenolate)]^þcation (see Ex-

perimental Section). Crystal structures of some of the

complexes (given in the next section) verify that the com-

plexes are monomers in the solid state with a single

TACH-o-tolyl ligand and a single phenolate coordinated

to iron(II).

Mixing Fe(OTf)₂2CH₃CN, 1, and methylhydroqui-

3

nonate in CH₃CN resulted in a bright orange solution and

a small amount of white precipitate. The ¹H NMR spec-

trum showed the formation of a paramagnetic species

with two highly shifted signals at δ 297 and 264 ppm and

the remaining signals lying between δ 40 and -40 ppm, as

well as small amounts of unidentified impurities (Figure 2).

The two signals at δ 297 and 264 ppm are tentatively as-

signed as the imine proton and the R-CH of the cyclo-

hexane ring, since these protons are closest to the para-

magnetic iron(II) center and are therefore expected to

experience the largest Fermi contact shift. A large signal

at δ -12.5 ppm, while overlapping with several smaller

signals, integrates to approximately 9 protons and is

assigned as the methyl protons of the tolyl groups. The

signals corresponding to the hydroquinone ring protons

were identified by performing the same reaction with

methylhydroquinone-d₃that was deuterated at the three

ring positions. In the ¹H NMR spectrum of the deuterated

complex, signals at δ 29, 31, and 39 ppm were absent,

showing that the corresponding resonances in the pro-

tiated analogue derive from the hydroquinonate ligand.

To investigate the influence of ortho-substituents on

binding, a variety of substituted phenolates were added to

Fe(TACH-o-tolyl)(OTf)₂. Three different chlorinated

phenols (2-chloro-, 2,6-dichloro-, and 2,3,4-trichlorophe-

nol), 2,6-dibromophenol, and 2-methylphenol were each

deprotonated with Et₃N and reacted with Fe(OTf)₂and 1

in CD₃CN. The 2-methylphenolate complex behaved

similarly to the parent phenolate, in that formation of

the 1:1:1 complex was not complete. However, the ¹H NMR

spectra showed a single iron-containing product with each of

the halogenated phenolates, consistent with complete binding

of the phenolate and TACH-o-tolyl ligands. The significance

of this important difference in binding will be discussed in

more detail below.

1

The H NMR spectra of the complexes formed by

reaction of Fe(TACH-o-tolyl)(OTf)₂with each of the

(20) The latter species was not characterized in detail, but was recognized

from its presence in control reactions between Fe(OTf)₂and phenolate

without a supporting ligand. Additional spectra are shown in the Supporting

Information.

10918 Inorganic Chemistry, Vol. 49, No. 23, 2010

Rocks et al.

substituted phenolates show very similar patterns of

signals (Figure 2). The ¹H NMR spectrum of each com-

plex with a substituted phenolate shows two signals

shifted downfield to between δ 289 and 313 ppm, which

are assigned to the imine proton and R CH of the cyclo-

hexane ring. The signals from the methyl groups of the

tolylidene arms integrate to 9H and are located between

δ -9 and -14 ppm in each case. The phenolate ¹H NMR

signals in each of the substituted phenolate complexes can

be assigned by integration as well as by comparison

between complexes with different substitution patterns.

The signals corresponding to the TACH-o-tolyl ligand

are at very similar chemical shifts in all five complexes,

and the proton signals in the substituted phenolates are

all at similar chemical shifts as those observed with the

unsubstituted phenolate. In the 2-chlorophenolate complex

(labeled 3 below), the meta protons are at δ 68 and 60 ppm,

the ortho proton at 12 ppm, and the para proton at -30 ppm.

In the 2,6-dichlorophenolate complex (labeled 2 below), the

meta and para protons are at δ 62 and -29 ppm, and in

the 2,6-dibromophenolate complex (labeled 5 below), the

meta and para protons are at δ 60 and -30 ppm. In the

2-methylphenolate complex (labeled 6 below), the meta

protons are at δ 63 and 57 ppm, the ortho proton at

26 ppm, the para proton at -39 ppm, and the methyl

protons lie at δ 77 ppm. In the analogous 2,3,4-trichlor-

ophenolate complex (which was characterized by ¹H

NMR spectroscopy only; see Supporting Information),

the meta and ortho protons are at δ 60 and -2 ppm,

respectively.

In all of these complexes, the signals with the largest

downfield shifts (264-313 ppm) are assigned to the imine

proton and the R CH of the cyclohexane ring. Hyperfine

shifts of this magnitude have been observed for the

RHCdN-CR₂H moiety in mononuclear high-spin iron-

(II) complexes with imine ligands, and were ascribed to

significant delocalization of unpaired spin density onto

the imine. The -CR₂H proton has been observed at 195 ppm

at 23 °C in a four-coordinate complex,²¹and the imine

proton was observed at 460 ppm at 55 °C in a six-

coordinate iron(II) complex.²²In a ligand with a six-

membered π system, such as a phenolate, the usual

pattern is to observed alternating upfield and downfield

shifts of the ring protons, because of the alternating sign

of the unpaired spin density.²³In these phenolate com-

plexes, this pattern is observed for the meta and para

protons, which are observed at 59 to 68 and -29 to -39

ppm, respectively, but not for the ortho protons, which

are observed at 39 to -2 ppm. This is probably due to a

pseudocontact shift. A significant pseudocontact shift is

expected for high-spin iron(II) systems,²³especially for

the phenolate ortho protons, which are close to the iron-

(II) center. Modest differences in the distance and the

orientation with respect to the magnetic susceptibility

tensor can lead to large variations in the pseudocontact

shift, which could explain the large variation in the

observed chemical shifts of the ortho protons in these

complexes.

Several notable conclusions can be drawn from these

¹H NMR studies. First, the solution-generated Fe(TACH-

o-tolyl)(OTf)₂species binds a variety of phenols as well as

methylhydroquinone. The adducts give similar ¹H NMR

spectra in each case. Second, each of the three ortho-

halogenated phenols completely forms the 1:1:1 complex

in solution, while the unsubstituted phenol, 2-methylphe-

nol, and methylhydroquinone give incomplete binding.

This observation suggests stronger binding affinity of the

halogenated phenolate ligands to the TACH-o-tolyl-iron-

(II) complex. Third, the ¹H NMR spectra indicate that the

TACH-o-tolyl ligand in each of these complexes has C₃

symmetry in solution on the NMR time scale, and the

equivalence of the meta protons in the unsubstituted

phenolate, the 2,6-dichlorophenolate, and 2,6-dibromo-

phenolate complexes indicates fast rotation about the

Fe-O and O-C bonds on the ¹H NMR time scale (this

issue will be revisited below).

X-ray Crystallographic Studies of (TACH-o-tolyl)Fe-

(phenolate) Complexes. The complexes were difficult

to crystallize, which often prevented the acquisition of

accurate microanalytical data for solids (the carbon analysis

was often about 1% low, possibly from co-crystallization

with salts). Despite the potential presence of impurities in

the bulk sample, single crystals of a number of complexes

were obtained and analyzed by X-ray diffraction.

Crystals of [Fe(TACH-o-tolyl)(2,6-dichlorophenolate)]-

OTf (2) were obtained by two different methods. The first

method used sodium 2,6-dichlorophenolate, concentra-

tion of the sample in hot tetrahydrofuran (THF) and slow

diffusion of pentane vapor at -35 °C for two months. The

second method used 2,6-dichlorophenol deprotonated

with triethylamine, concentration of a THF solution

at room temperature, and slow diffusion of Et₂O vapor

at -35 °C for two months. While both methods produced

single crystals of 2, the crystals had different space groups

(P2₁/c from the method with no heat and P2₁2₁2₁from the

heated method). The P2₁/c structure had an asymmetric

unit containing a single cation/anion pair while the asym-

metric unit of the other structure (P2₁2₁2₁) contained two

discrete cation/anion pairs. Thus, there are three unique

structures that provide information on bond lengths and

angles for [Fe(TACH-o-tolyl)(2,6-dichlorophenolate)]OTf,

Table 1, labeled 2A, 2B, and 2C where 2A is from the P2₁/c

structure and 2B and 2C are from the P2₁2₁2₁structure. An

ORTEP diagram of molecule 2B is shown in Figure 3.

Two different methods were also used to crystallize the

2-chlorophenolate complex. In the first method, a mix-

ture of 1, Fe(OTf)₂2CH₃CN and 2-chlorophenol depro-

3

tonated with triethylamine was stirred in THF for 15 min

and then Et₂O vapor was diffused into the solution at -35 °C

over the course of two weeks. Small yellow needles of

[Fe(TACH-o-tolyl)(2-chlorophenolate)]OTf (3) formed.

The crystals diffracted well, and the refined structure

showed no solvent molecules in the unit cell. The molecule

is disordered over two positions with the ligand tolylidene

arms arranged in two orientations (1:1 ratio). In both

conformers, the chloride of the phenolate is pointed

toward the iron(II). An ORTEP diagram of the major

conformer is shown in Figure 4. Bond lengths and angles

of interest are listed in Table 2.

(21) Torzilli, M. A.; Colquhoun, S.; Kim, J.; Beer, R. H. Polyhedron 2002,

21, 705–713.

(22) Weber, B.; Walker, F. A. Inorg. Chem. 2007, 46, 6794–6803.

(23) Bertini, I.; Luchinat, C.; Parigi, G. Solution NMR of Paramagnetic

Molecules; Elsevier: Amsterdam, 2001.

In the second crystallization method, a yellow solu-

tion in THF/Et₂O generated from iron(II) triflate, 1, and

Article

Inorganic Chemistry, Vol. 49, No. 23, 2010 10919

Table 1. Relevant Bond Lengths and Angles from the X-ray Structures of [Fe(TACH-o-tolyl)(2,6-dichlorophenolate)]OTf (2)

A

B

C

˚

Bond Lengths (A)

Fe1-O1

Fe1-N3

Fe1-N1

Fe1-N2

Fe1-Cl1

1.865(3)

2.061(5)

2.132(4)

2.070(5)

3.1224(17)

Fe1-O1

Fe1-N3

Fe1-N1

Fe1-N2

1.885(3)

2.086(4)

2.143(4)

2.074(4)

2.8898(15)

Fe2-O2

Fe2-N6

Fe2-N4

Fe2-N5

Fe2-Cl3

1.879(3)

2.084(4)

2.135(4)

2.078(5)

2.9877(16)

Fe1-Cl1

Bond Angles (deg)

Fe1-O1-C31_phen

N3-Fe1-N1

N2-Fe1-N1

N3-Fe1-N2

O1-Fe1-N3

O1-Fe1-N1

O1-Fe1-N2

N1-Fe1-Cl1

140.1(3)

Fe1-O1-C31

N3-Fe1-N1

N2-Fe1-N1

N3-Fe1-N2

O1-Fe1-N3

O1-Fe1-N1

O1-Fe1-N2

N1-Fe1-Cl1

131.8(3)

Fe2-O2-C67

N6-Fe2-N4

N5-Fe2-N4

N6-Fe2-N5

O2-Fe2-N6

O2-Fe2-N4

O2-Fe2-N5

N4-Fe2-Cl3

137.3(4)

88.44(16)

92.66(16)

97.81(17)

130.76(16)

127.84(16)

110.40(17)

164.28(12)

97.13(16)

83.49(16)

97.45(17)

110.67(15)

119.67(15)

139.40(16)

163.88(11)

86.21(16)

95.38(17)

97.55(18)

138.84(16)

119.96(16)

109.65(17)

167.05(12)

Figure 4. Solid state structure of conformer A of [Fe(TACH-o-tolyl)-

(2-chlorophenolate)]OTf (3) with ellipsoids shown at 50% probability.

Hydrogen atoms and triflate anions are omitted for clarity.

Figure 3. One of three independent molecules in the solid state structure

of [Fe(TACH-o-tolyl)(2,6-dichlorophenolate)]OTf, labeled molecule 2B,

with ellipsoids shown at 50% probability. Hydrogen atoms and the

triflate anion are omitted for clarity.

Table 2. Relevant Bond Lengths and Angles from the X-ray Structure of

[Fe(TACH-o-tolyl)(2-chlorophenolate)]OTf (3)

conformer A

˚

conformer B

2-chlorophenol deprotonated with triethylamine was

concentrated by heating, and then stored at -35 °C for

10 days to yield yellow blocks. Surprisingly, the crystals

were not of the expected product, 3, but were instead of

[Fe(cis-TACH-o-tolyl)(2-chlorophenolate)(THF)]OTf (4)

(Figure 5). In the cis-TACH-o-tolyl ligand, one imine

CdN bond has isomerized to the cis stereoisomer, and

thus it differs from the all-trans geometry observed in 2

and 3. Also, unlike the structures of 2 and 3, here the iron

has an additional ligand: a THF molecule bound between

the two trans tolylidene arms. The 2-chlorophenolate is

disordered over two positions, where 91% of the time the

chlorine is pointed outside the pocket and away from the

metal and 9% of the time the chlorine is turned inside the

binding pocket and toward the metal. Bond lengths and

angles of interest are listed in Table 3.

Bond Lengths (A)

Fe-O1/Fe⁰-O1⁰

Fe-N2/Fe⁰-N2⁰

Fe-N3/Fe⁰-N3⁰

Fe-N1/Fe⁰-N1⁰

Fe-Cl1/Fe⁰-Cl1⁰

1.915(7)

2.097(5)

2.100(5)

2.101(5)

2.929(7)

1.914(7)

2.098(5)

2.096(5)

2.103(5)

3.010(8)

Bond Angles (deg)

Fe-O1-C31_phen/ Fe⁰-O1⁰-C31⁰_phen

N3-Fe-N2/N3⁰-Fe⁰-N2⁰

N3-Fe-N1/N3⁰-Fe⁰-N1⁰

N2-Fe-N1/N2⁰-Fe⁰-N1⁰

O1-Fe-N2/O1⁰-Fe⁰-N2⁰

O1-Fe-N3/O1⁰-Fe⁰-N3⁰

O1-Fe-N1/O1⁰-Fe⁰-N1⁰

N1-Fe-Cl1/N1⁰-Fe⁰-Cl1⁰

120.7(13)

97.5(5)

91.0(4)

118.6(12)

97.3(5)

83.1(5)

83.8(5)

91.0(4)

124.7(8)

133.6(8)

110.0(7)

168.5(5)

128.5(8)

127.8(8)

115.1(8)

163.4(5)

It was also possible to grow crystals of the 2,6-dibromo-

phenolate complex (5) by vapor diffusion of diethyl ether

into a dimethoxyethane solution. The refined structure is

shown in Figure 6, and distances and angles are in Table 4.

As in the above structure, one arm of the TACH-o-tolyl

ligand has isomerized. However, the space that is created

around the iron atom is filled not with a solvent molecule,

but with one of the two bromo substituents of the pheno-

late with an iron-bromine bond distance of 2.8414(3) A.

Combining 1 with iron(II) triflate and 2-methylphe-

nolate resulted in an orange solution. Crystals of [Fe-

(TACH-o-tolyl)(2-methylphenolate)]OTf (6) were obtained

˚

10920 Inorganic Chemistry, Vol. 49, No. 23, 2010

Rocks et al.

Figure 5. Solidstate structure ofthe majorconformer of[Fe(cis-TACH-

o-tolyl)(2-chlorophenolate)(THF)]OTf (4) with ellipsoids shown at 50%

probability. Hydrogen atoms and triflate anion are omitted for clarity.

Table 3. Relevant Bond Lengths and Bond Distances for [Fe(cis-TACH-o-

tolyl)(2-chlorophenolate)(THF)]OTf (4) and for [Fe(cis-TACH-o-tolyl)-

(phenolate)(THF)]OTf (7)

Figure 6. Solid state structure of [Fe(cis-TACH-o-tolyl)(2,6-dibromo-

phenolate)]OTf (5) with ellipsoids shown at 50% probability. Hydrogen

atoms and triflate anion are omitted for clarity.

4

7

Table 4. Relevant Bond Lengths and Bond Distances for [Fe(cis-TACH-o-

tolyl)(2,6-dibromophenolate)(THF)]OTf (5)

˚

Bond Lengths (A)

Fe1-O1

Fe1-N1

Fe1-N2

Fe1-N3

Fe1-O2

1.9202(9)

2.1414(10)

2.1758(10)

2.1296(10)

2.2007(8)

1.891(2)

2.148(2)

2.179(2)

2.119(2)

2.231(2)

5

˚

Bond Lengths (A)

Fe1-O1

Fe1-N3

Fe1-N1

Fe1-N2

Fe1-Br1

1.894(1)

2.100(1)

2.117(1)

2.079(1)

2.8414(3)

Bond Angles (deg)

N1-Fe1-N2

N2-Fe1-N3

N3-Fe1-N1

O1-Fe1-N1

O1-Fe1-N2

O1-Fe1-N3

O2-Fe1-O1

Fe1-O1-C31

O2-Fe1-N2

83.16(4)

91.08(4)

92.28(4)

135.49(7)

94.03(6)

132.22(7)

88.11(6)

139.99(18)

174.73(4)

86.56(9)

89.07(9)

93.60(9)

128.22(10)

96.91(9)

137.92(10)

87.51(8)

156.7(2)

174.74(8)

Bond Angles (deg)

Fe1-O1-C31_phen

N3-Fe1-N1

N2-Fe1-N1

N3-Fe1-N2

O1-Fe1-N3

O1-Fe1-N1

O1-Fe1-N2

N1-Fe1-Br1

130.80(9)

86.43(5)

91.80(5)

97.29(5)

137.97(5)

100.29(5)

123.60(5)

174.99(3)

from vapor diffusion of Et₂O into a concentrated solution

of 6 in dimethoxyethane over three weeks at -35 °C. The

refined structure is shown in Figure 7, and relevant bond

angles and distances are listed in Table 5. The iron atom in

this structure is pseudotetrahedral with only four bonds.

In contrast to the halogenated phenolates, the methyl

group points away from the metal, in a less sterically

crowded environment.

A concentrated yellow solution of 1, iron(II) triflate,

and phenolate in THF/Et₂O was heated, then stored

at -35 °C for three days to yield yellow blocks. Similar

to the case of the 2-chlorophenolate complex that was

prepared from a hot THF/Et₂O solution, the crystals

were not of the expected [Fe(TACH-o-tolyl)(phenolate)]-

OTf complex, but instead were [Fe(cis-TACH-o-tolyl)-

(phenolate)(THF)]OTf (7) (Figure 8; metrical parameters

in Table 3). Once again, one imine CdN bond in the

TACH-o-tolyl has isomerized to the cis stereochemistry,

and a THF molecule is ligated to the iron atom.

solution is possible, the conditions for crystallization

(high concentration, low temperature) favor dimer for-

mation, while the crystal structures exclusively show

mononuclear complexes. Also, electrospray mass spectra

of selected complexes showed the presence of monomers.

Therefore, for the remainder of the paper we describe the

complexes in the context of a monomeric formulation.

The crystal structures of 2 show approximate C_ssym-

metry, with two of the tolylidene arms in similar orienta-

tions having the 2,6-dichlorophenolate sandwiched

between them, and the third tolylidene arm in a very dif-

ferent orientation. Furthermore, in the crystal structures

of 2 the meta protons of the phenolate are inequivalent.

1

However, the room-temperature H NMR spectrum of

2 shows that the TACH-o-tolyl ligand has C₃symmetry

on the NMR time scale, and the equivalence of the two

meta protons on the phenolate indicates fast rotation of

the phenolate on the NMR time scale. When the solution

of 2 was cooled to -40 °C, two of the TACH-o-tolyl signals

between δ30 and 40 ppm split into two peaks each (Figure 9).

1

Solution Ligand Dynamics and Isomerization. The H

NMR studies by themselves cannot definitively assign the

nuclearity of these complexes in solution; however, the

crystal structures indicate that they are monomers in the

solid state. Although a monomer-dimer equilibrium in

Article

Inorganic Chemistry, Vol. 49, No. 23, 2010 10921

Figure 7. Solid state structure of [Fe(TACH-o-tolyl)(2-methylphenola-

te)]OTf (6) with ellipsoids shown at 50% probability. Hydrogen atoms

and triflate anions are omitted for clarity.

1

Figure 9. Low-temperature H NMR spectra of the solution resulting

from the mixture of iron(II) triflate, 1, and sodium 2,6-dichlorophenolate

in CD₃CN, showing the signals for the cyclohexane β protons of the

TACH-o-tolyl ligand. The lowest temperature (decoalesced) spectrum is

at the top. The change in chemical shifts arises from the usual temperature

dependence ofthe hyperfineshifts expected for anS = 2 complex. The full

spectra are shown in the Supporting Information.

Table 5. Relevant Bond Angles and Bond Lengths for [Fe(TACH-o-tolyl)-

(2-methylphenolate)] OTf (6)

6

˚

Bond Lengths (A)

Fe-O1

Fe-N2

Fe-N3

Fe-N1

1.858(3)

2.094(3)

2.078(3)

2.075(3)

Bond Angles (deg)

Fe-O1-C31_phen

N3-Fe-N2

N3-Fe-N1

N2-Fe-N1

O1-Fe-N2

O1-Fe-N3

O1-Fe-N1

125.6(2)

95.73(12)

94.15(12)

87.08(12)

112.30(12)

117.38(12)

139.55(12)

Figure 10. Two views of a proposed solution structure of 2, shown

without tolyl groups for clarity. A mirror plane containing the phenolate

and bisecting the cyclohexyl ring is responsible for two different cyclo-

hexane β proton environments, and it explains the decoalescence of the

two ¹H NMR signals between δ 30 and 40 ppm and the 2:1 integration

pattern of the decoalesced signals.

between each set of tolylidene arms. At low temperature,

this motion around the Fe-O bond is slowed, and the

phenolate is trapped between a single set of ligand arms

(as found in the crystal structure) on the NMR time scale.

Without rotation around the iron-oxygen bond, there is

only one plane of symmetry, which runs through the

metal center and bisects the cyclohexane ring of the

TACH ligand (Figure 10). This causes there to be two

inequivalent types of cyclohexane ring β protons, leading

to low-temperature splitting into the observed 2:1 inte-

gration pattern. Note, however, that there is no evidence

for decoalescence of the phenolate meta proton signals.

Therefore, the rotation of the phenolate O-C bond in

solution remains rapid on the NMR time scale down

to -40 °C, in contrast to the crystal structures, where a

single phenolate orientation is observed.

Figure 8. Solid state structure of [Fe(cis-TACH-o-tolyl)(phenolate)-

(THF)]OTf (7) with ellipsoids shown at 50% probability. Hydrogen

atoms and triflate anion are omitted for clarity.

Using the deconvolution function of TopSpin,²⁴it was

possible to calculate the integrations of the overlapped signals

to be roughly 2:1 for each set of split peaks.

These data can be interpreted through the following

model. At room temperature, all three tolylidene arms are

equivalent, because the phenolate can sample the space

1

Other H NMR studies examined the nature of the

isomerization of the imine NdC bonds in the tolylidene

arms. The [Fe(cis-TACH-o-tolyl)(2-chlorophenolate)]OTf

complex (4) was analyzed by ¹H NMR spectroscopy and

compared to the spectrum of 3 in CD₃CN, discussed

above. Since 4 has one cis-imine arm and two trans-imine

arms, it no longer has C₃symmetry, and is predicted to

(24) Topspin, 1.3; Bruker: Rheinstetten, Germany, 2005.

10922 Inorganic Chemistry, Vol. 49, No. 23, 2010

Scheme 3. Isomerization of the TACH-o-tolyl Ligand

Rocks et al.

Hexadentate, trianionic TACH-based ligands have been

coordinated to iron(III) to form neutral, coordinatively

saturated complexes.^16c,25Though 1 is a new TACH-

based ligand, it is closely related to mesityl and benzyl

substituted ligands that have been complexed to Cu and

Ni.^17,19To date, the only examples of an iron(II)

complex bound to a TACH-based ligand use the parent

triaminocyclohexane.²⁶

Onepersistentproblem thatplaguedthe studiesreported

here was that TACH-o-tolyl could be displaced by strong

donors. For example, addition of excess phenolate caused

loss of the tridentate ligand. To explain this observation,

note that the favored conformation of the free triimine has

all threesubstituentsinequatorialpositions. Tocoordinate

to a metal, these substituents must reach axial positions.

Therefore, despite the advantages of the chelate effect,

there is an unfavorable enthalpic contribution to metal

binding by this tridentate ligand.

have at most C_ssymmetry. Consistent with this lower-

symmetry point group, the ¹H NMR spectrum of 4 con-

tains more signals than the spectrum of 3 (see Supporting

Information). The imine protons and the cyclohexane R-CH

are no longer all equivalent, and instead of two highly

downfield signals integrating to 3H each, there are four

signals, two that integrate to 1H and two that integrate to

2H. This observation is consistent withthe expected mirror

plane of symmetry in the molecule. The remaining highly

overlapped signals between δ 18 and -6 ppm are not

amenable to further characterization. The unique methyl

group of the isomerized tolylidene is identified at δ -4.2 ppm

from its integration. In solution, the THF molecule bound to

the metal in the solid state structure of 4 is presumably

displaced by the more strongly donating CD₃CN, although

the congestion of signals around δ 2 and 4 ppm in the ¹H

NMR spectrum precludes any attempt at assigning peaks

to bound solvent. Without considering bound solvent,

One advantage to the TACH-based triimine ligands is

that the ligand synthesis gives imines that each have a

trans stereochemistry. Upon metal coordination, trans

imine substituents are constrained to form “walls”

around the remaining binding sites, which restrict the

number of additional ligands. The crystal structures of 2,

3, and 6 show that the imine substituents surround the

fourth donor to the iron(II) center, and prevent pheno-

lates from bridging to a second metal. Variable-tempera-

ture ¹H NMR studies suggest that in 2, the movement of

the fourth ligand around the binding site is restricted. We

assume that Fe-O bond rotation is somewhat hindered in

complexes 3 and 6 as well, though we could not reach

temperatures low enough to cause decoalescence of peaks

in their ¹H NMR spectra. Overall, TACH-imine ligands

are excellent at sterically protecting the phenolate binding

site, and give mononuclear, high-spin iron(II) complexes.

However, certain difficulties are present because the

trans-imines are thermodynamically disfavored in the

metal complexes. Walton and co-workers have reported

the hydrolysis of the imine CdN bonds in benzylidene

substituted TACH type ligands;¹⁷however in the work

reported here, hydrolysis was prevented by the stringent

exclusion of water. Although we were able to circumvent

hydrolysis as a problem with TACH-imine based com-

plexes, we uncovered another problem: imine isomeriza-

tion. As evidenced by the solid-state structures of 4, 5, and 7,

one tolylidene arm of the TACH-o-tolyl ligand can isomer-

ize from trans to cis. Monitoring the transformation to these

species by ¹H NMR spectroscopy showed that the isomer-

ization of imines from trans to cis is thermodynamically

favored. Control studies showed that 1 isomerizes only in

the presence of metal. We are not aware of previous

examples in the literature of arm isomerization in transition-

metal complexes of TACH-imine ligands. Qualitative

measurements showed that ortho-substituents on the

phenolate slowed the arm isomerization.

1

21 H NMR signals are expected and 20 distinct signals

are observed. The approximate number of signals and the

integration values of the non-overlapping peaks are con-

sistent with the solid-state structure.

To further investigate the isomerization of TACH-o-tolyl,

a CD₃CN solution containing 1, iron(II) triflate and 2-chlor-

ophenolate was heated at 60 °C, and ¹H NMR spectra were

taken periodically. Initially, the spectrum corresponded to

that of 3. Over several hours, ¹H NMR signals correspond-

ing to 3 diminished in intensity while signals corresponding

to 4 began to emerge, with a half-life of about an hour.

Cooling the solution back to room temperature did not

1

change the ratio of products by H NMR spectroscopy.

Samples of 4 in CD₃CN did not convert to 3 even after

several days in solution at room temperature. The thermal

stability of 4 was confirmed by heating a sample in CD₃CN

to 65 °C for 10 min, but no changes were observed in the ¹H

NMR spectrum. Thus, the isomerization of the imine in one

tolylidene arm is favorable and irreversible under these

conditions (Scheme 3).

While the 2-chlorophenolate complex could be crystallized

with the TACH-o-tolyl ligand in the desired all trans orienta-

tion as well as the cis-isomerized form, the unsubstituted

phenolate complex was crystallized only with the tridentate

1

ligand in the cis-isomerized form. Accordingly, H NMR

studies (see Supporting Information) show that “Fe(TACH-

o-tolyl)(phenolate)”, which displays the C₃symmetry indica-

tive of the all trans ligand geometry, converts in 3 h at 60 °Cto

the isomerized ligand complex, 7, which has C_ssymmetry.

The signals corresponding to (TACH-o-tolyl)Fe(OTf)₂and

Fe(phenolate)_xremain unchanged.

Synthesis of Iron(II) Phenolate Complexes with a Bulky

Supporting Ligand. Though the coordination of phenolate-

like species to non-heme iron(II) complexes is important

Discussion

(25) (a) Bollinger, J. E.; Mague, J. T.; Oconnor, C. J.; Banks, W. A.;

Roundhill, D. M. J. Chem. Soc., Dalton Trans. 1995, 1677–1688. (b) Bollinger,

J. E.; Mague, J. T.; Roundhill, D. M. Inorg. Chem. 1994, 33, 1241–1242.

(26) Yang, J. Y.; Shores, M. P.; Sokol, J. J.; Long, J. R. Inorg. Chem. 2003,

42, 1403–1419.

Tridentate Imine Ligands Based on TACH. We report

here the first examples of iron(II) complexed to a triimine

that is based on the triaminocyclohexane (TACH) core.

Article

Inorganic Chemistry, Vol. 49, No. 23, 2010 10923

for understanding enzymes that process these aromatics,

the coordination chemistry of high-spin, non-heme iron-

(II) complexes with phenolates has primarily used pheno-

lates that are tethered as part of a multidentate ligand.²⁷

Iron(II) complexes with unsupported phenolates are

much less common, and require ligands that constrain the

iron coordination sphere. The most common supporting

ligands are based on porphyrin.²⁸Phenoxide has been

crystallization. Nonetheless, since there are no mono-

nuclear hydroquinone complexes with iron(II) in the

literature, the solution characterization reported herein

is an important first step in accumulating structural

information on how substituted hydroquinone substrates

may bind to the iron(II) center in the enzymes LinE and

PcpA. In particular, the phenolate complexes provide

critical insights into the role of the ortho substituent, as

will be described below.

3þ

attached to a [4Fe-4S] cluster in the Fe₂^2þFe₂state.²⁹

Dinuclear complexes have been reported with 2,4,6-tri-

t-butylphenolate as a bridging ligand and as both a brid-

ging and terminal ligand.³⁰A series of four-coordinate

tris(pyrazolyl)borate-supported iron(II) phenolate com-

Comparison of the Structures of the (TACH-o-tolyl)Fe-

(phenolate) Complexes. The structures of the iron pheno-

late complexes fall into two distinct categories: (1) com-

pounds 2, 3, 5, and 6 have only the TACH-o-tolyl ligand

and phenolate coordinated to the iron(II) center, and (2)

compounds 4 and 7 that have an additional solvent

molecule bound to the iron(II) center.

1

plexes were characterized by H NMR, IR, MCD, and

elemental analysis.^31,32A crystal structure was reported

for one of these, with pentafluorophenolate.²⁷The known

chemistry of halogenated phenolates on iron(II) is even

more scarce. To our knowledge there are no iron(II)

2-chlorophenolate or 2,6-dibromophenolate complexes

reported in the literature. Iron(II) complexes coordinated

with 2,6-dichlorophenolate have been reported, but not

characterized fully.^31,32No crystallized iron 2-methylphe-

nolate complexes have been reported in the literature, and

only iron(III) 2-methylphenolate complexes have been

reported thus far.³³Thus, there is a lack of information in

the literature regarding the behavior of halogenated

phenolates on iron(II) that could provide insight into

the origin of substrate specificity in PcpA and LinE.

The work reported herein shows that the [Fe(TACH-o-

tolyl)]^2þsystem provides a platform for the synthesis of

complexes with a variety of substituted phenolates. Though

these complexes in some cases could not be purified to

In complexes 2, 3, 5, and 6, the structure of the TACH-

iron(II) unit is highly conserved. In the structures of the

metal-TACH-imine ligands in the literature, the average

˚

M-N_iminebond distance is 2.06(7) A, regardless of metal

identity or TACH ligand substitution.^16-19This is similar

to the values observed in the TACH-o-tolyl iron(II)

complexes reported here, where the average Fe-N bond

˚

distance of the trans TACH-o-tolyl complexes is 2.09(2) A

˚

and the values range from 2.061 A (Fe-N3 in 2A) to 2.143

˚

A (Fe-N1 in 2B), with Fe-N1 being slightly longer than

the other two distances in all of the structures but 6. The

three N-Fe-N angles are distinctly different in each

structure. One of the N-Fe-N angles is always signifi-

cantly larger than the other two, and in each case, these

two nitrogen atoms hold the tolylidene arms that sand-

wich the phenolate (N2 and N3). This N-Fe-N angle

ranges from 95.7° in 6 to 97.8° in 2A. The other two

N-Fe-N angles show somewhat more variability, and

range from 83.1° (for N3⁰-Fe⁰-N1⁰in 3B) to 97.1° (for

N3-Fe-N1 in 2B). Overall, the Fe-N bond lengths and

N-Fe-N angles in these complexes appear to be con-

strained by the rigidity of the cyclohexane ring, and many

aspects of the geometry of the bound Fe(II) in these

complexes are dictated by the core of the tridentate ligand.

In each of these four compounds (2, 3, 5, and 6), there is

a close interaction between the phenolate aromatic ring

and two o-tolyl groups of the supporting TACH-o-tolyl

1

analytical purity, we have obtained H NMR solution

characterization for complexes with five different substi-

tuted phenolates as well as unsubstituted phenolate and

2-methylhydroquinonate. Crystal structures of four of

the substituted phenolate complexes were obtained, with

chloro-, bromo-, and methyl- substituents at the ortho

positions. Each iron-containing cation is paired with an

outer-sphere triflate, which was evident from the crystal

structure as well as the characteristic bands in the infrared

spectrum near 1263, 1154, and 1030 cm^{-1 34}

A crystal

.

structure was also obtained for the unsubstituted pheno-

late complex, albeit with the tolylidene arm isomerized to

the cis-orientation. The 2-methylhydroquinonate com-

plex was unstable in solution and resisted all attempts at

˚

ligand (minimum ring-ring distances of 3.2-3.5 A).

These are well within the distance usually accepted as

π-stacking.³⁵This observation raises the issue of whether

the strength of π-stacking influences the binding of

different phenolates. The nature of electronic effects on

π-stacking has been the source of recent controversy:

though the traditional view holds that electron-with-

drawing substituents increase the strength of π-stacking

interactions, gas-phase and computational studies have

questioned the generality and trends in this interaction.³⁶

In the compounds described here, there is no compelling

evidence for significant differences in π-stacking, because

the shortest ring-ring distances are similar in the dichloro-

phenolate complex 2 as in the methylphenolate complex 6

(27) There are more than 100 crystallographically characterized iron(II)

complexes with tethered phenolates as part of supporting multidentate

ligands, primarily derived from salen. Selected examples: (a) Jameson,

G. B.; March, F. C.; Robinson, W. T.; Koon, S. S. J. Chem. Soc., Dalton

Trans. 1978, 185–191. (b) Cini, R. Inorg. Chim. Acta 1983, 73, 146–152.

(c) Kayal, A.; Lee, S. C. Inorg. Chem. 2002, 41, 321–330; see also ref 21.

(28) (a) Ainscough, E. W.; Addison, A. W.; Dolphin, D.; James, B. R.

J. Am. Chem. Soc. 1978, 100, 7585–7591. (b) Phillippi, M. A.; Shimomura, E. T.;

Goff, H. M. Inorg. Chem. 1981, 20, 1322–1325.

(29) Weigel, J. A.; Holm, R. H. J. Am. Chem. Soc. 1991, 113, 4184–4191.

(30) Bartlett, R. A.; Ellison, J. J.; Power, P. P.; Shoner, S. C. Inorg. Chem.

1991, 30, 2888–2894.

(31) Ito, M.; Amagai, H.; Fukui, H.; Kitajima, N.; Moro-oka, Y. Bull.

Chem. Soc. Jpn. 1996, 69, 1937–1945.

˚

(both 3.2 A).

(32) Pavel, E. G.; Kitajima, N.; Solomon, E. I. J. Am. Chem. Soc. 1998,

120, 3949–3962.

(33) (a) Richard, M. J.; Shaffer, C. D.; Evilia, R. F. Electrochim. Acta

1982, 27, 979–983. (b) Arasasingham, R. D.; Balch, A. L.; Hart, R. L.;

Latosgrazynski, L. J. Am. Chem. Soc. 1990, 112, 7566–7571.

(34) Johnston, D. H.; Shriver, D. F. Inorg. Chem. 1993, 32, 1045–1047.

(35) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525–

5534.

(36) A review of different viewpoints may be found in: Wheeler, S. E.;

€

McNeil, A. J.; Muller, P.; Swager, T. M.; Houk, K. N. J. Am. Chem. Soc.

2010, 132, 3304–3311.

10924 Inorganic Chemistry, Vol. 49, No. 23, 2010

Rocks et al.

In 4 and 7, where one of the tolylidene arms has

isomerized to the cis orientation, the iron(II) is trigonal

bipyramidal with a THF bound between the two trans

tolylidene ligand arms. The axial ligands in the trigonal

bipyramid are the THF oxygen atom (O2) and the nitro-

gen atom from the isomerized imine (N2). The trigonal

bipyramidal geometry is demonstrated by the nearly linear

O2-Fe-N2 angles of 174.73(4)° in 4 and 174.74(8)° in 7,

and by the fact that N3, N1, and O1 compose a plane with

the sum of their angles equaling 360.0(1)° in 4 and

359.7(2)° in 7. Instead of stacking between the trans toly-

lidenes as seen in all of the structures that lack a bound

solvent, the phenolate is bent into the open space created

by the isomerized cis-tolylidene. Reducing steric conges-

tion likely offsets the small energetic stabilization lost

from the disruption of π-stacking found in the structures

of the trans-TACH-o-tolyl complexes.

Table 6. Comparison between Fe-Halogen Distances in TACH-o-tolyl-iron(II)

Complexes Reported Here^a

[Fe(TACH-o-tolyl)(___)]^þ

Fe-halogen

distance (A)

Δ average

Fe-halogen (A)

OTf^-

˚

2-chlorophenolate

2.929(7)

3.010(8)

3.122(2)

2.890(2)

2.988(2)

2.8414(3)

0.72(6)

0.80(6)

0.91(6)

0.68(6)

0.78(6)

0.49(4)

2,6-dichlorophenolate A

2,6-dichlorophenolate B

2,6-dichlorophenolate C

2,6-dibromophenolate

^aThese may be compared to the average Fe-halogen bond length in

˚

the Cambridge Structural Database (2.21(6) A for Fe-Cl and 2.35(4) A

for Fe-Br).⁴⁰The second column gives the difference between the

Fe-halogen distance in the TACH-o-tolyl supported complexes and

the average Fe-halogen bond length.

bonding interaction is likely.³⁹Table 6 compares the

Fe-Cl(chloroarene) distances of the TACH-o-tolyl-Fe(II)

complexes to the average Fe-Cl(chloride) bond length

Evidence for Secondary Bonding Interactions of Halo-

genated Phenolates with Iron(II). LinE and PcpA are

unique in their ability to cleave chlorinated hydroqui-

nones. What drives this substrate selectivity is not well

understood.⁹The basic coordination chemistry of halo-

genated phenols with iron(II) has been left unexplored in

the literature. The complexes herein represent the first ortho-

chlorophenolate-iron(II) complexes to be crystallographi-

cally characterized.³⁷Compound 5 is the first example of an

˚

for four-coordinate iron(II) chloride complexes of 2.21 A

40

(standard deviation 0.06 A). The TACH-o-tolyl sup-

˚

ported iron-phenolate complexes exhibit Fe-Cl distances

ranging from 2.890(2) to 3.122(2) A, and therefore they

˚

meet Wulfsberg’s criterion for a secondary interaction.

The evidence for secondary interactions between iron and

the bromine substituent in the 2,6-dibromophenolate

complex is even more convincing, with a Fe-Br distance

˚

less than 0.5 A longer than the average from the Cam-

1

iron 2,6-dibromophenolate complex. Interestingly, our H

˚

bridge Structural Database (2.35 A; standard deviation

NMR studies indicate that ortho-halophenolates bind

completely to TACH-ligand iron complexes while unsub-

stituted phenolate and 2-methylphenolate do not. These

results suggest that the presence of a halogen adjacent to the

hydroxyl group of a phenolate may stabilize its iron(II)

complexes.

˚

0.04 A). Previous workers have seen that the strength of

secondary interactions increases with heavier halogen

atoms.^38dThere is only one reported example of an ortho-

bromophenolate complex that exhibits a metal-halogen

secondary bond.⁴¹

There are other structural differences between haloge-

nated (2, 3, 5) and non-halogenated (6) phenolate com-

plexes. For example, there are significant differences in

the Fe-O(aryl) distance among the different complexes.

The increased stability of the iron(II) phenolate com-

plexes with ortho-halogen substituents, combined with

the structural evidence that shows the halogen substitu-

ents oriented toward the iron(II) center in all of the

structures with all-trans-TACH ligands, suggests that

an iron-halogen interaction may be important in these

complexes. Electronic interactions between metals and

the chlorine atoms of chloroarenes have been reported in

the literature.³⁸Wulfsberg and co-workers have postu-

lated that when the distance between the metal (M) and

˚

The Fe-O1 distances vary from 1.865(3) to 1.915(7) A for

complexes 2, 3, and 5, while non-halogenated 6 has the

˚

shortest distance at 1.858(3) A. This is suggestive of a lower

coordination number in the absence of the iron-halogen

interaction. Additionally, the N1-Fe-O1 angles for com-

plexes 2, 3, and 5 vary from 100.29(5)° to 127.84°, while the

angle in 6 is anomalous at 139.55(12)°. This difference can

also be attributed to an iron-halogen interaction that

constrains the Fe-O-C angle.

˚

the chlorine of the chloroarene is within 1.0 A of the

average M-chloride distance for that metal, a secondary

It is also possible to show that the tetrahedral geometry

of the methyl-substituted compound 6 is distorted toward a

trigonal bipyramid in 2, 3, and 5. There are two ways to

probe the extent to which the geometry of the Fe-N2-N3-

O1 unit is trigonalized. First, for a trigonal (bi)pyramidal

geometry, atoms N2, N3, and O1 compose a plane, and

the sum of the N2, N3, and O1 angles about the iron(II)

center should be 360°. This sum is 358.9(2)° in the 2,6-

dibromophenolate complex 5, making it the most trigonal.

(37) There are two crystallographically characterized iron(III) phenolate

compounds with 2,6-chloro substituents. (a) Koch, S. A.; Millar, M. J. Am.

Chem. Soc. 1982, 104, 5255–5257. (b) Helms, A. M.; Jones, W. D.; McLendon,

G. L. J. Coord. Chem. 1991, 23, 351–359.

(38) Examples: (a) Meyer, R.; Gagliardi, J.; Wulfsberg, G. J. Mol. Struct.

1983, 111, 311–316. (b) Wulfsberg, G.; Yanisch, J.; Meyer, R.; Bowers, J.; Essig,

M. Inorg. Chem. 1984, 23, 715–719. (c) Colsman, M. R.; Newbound, T. D.;

Marshall, L. J.; Noirot, M. D.; Miller, M. M.; Wulfsberg, G. P.; Frye, J. S.;

Anderson, O. P.; Strauss, S. H. J. Am. Chem. Soc. 1990, 112, 2349–2362.

(d) Kulawiec, R. J.; Crabtree, R. H. Coord. Chem. Rev. 1990, 99, 89–115. (e) Garcia,

M. P.; Jimenez, M. V.; Cuesta, A.; Siurana, C.; Oro, L. A.; Lahoz, F. J.; Lopez,

J. A.; Catalan, M. P.; Tiripicchio, A.; Lanfranchi, M. Organometallics 1997, 16,

1026–1036. (f) Poignant, G.; Nlate, S.; Guerchais, V.; Edwards, A. J.; Raithby,

P. R. Organometallics 1997, 16, 124–132. (g) Wulfsberg, G.; Parks, K. D.;

Rutherford, R.; Jackson, D. J.; Jones, F. E.; Derrick, D.; Ilsley, W.; Strauss, S. H.;

Miller, S. M.; Anderson, O. P.; Babushkina, T. A.; Gushchin, S. I.; Kravchenko,

E. A.; Morgunov, V. G. Inorg. Chem. 2002, 41, 2032–2040.

(40) Our values come from a search of terminal Fe-halide complexes with

four-coordinate iron, from structures without disorder, with R < 10%, in

the Cambridge Crystallographic Database (November 2009). Similar num-

bers are quoted in the literature: Orpen, A. G.; Brammer, L.; Allen, F. H.;

Kennard, O.; Watson, D. G.; Taylor, R. J. Chem. Soc., Dalton Trans. 1989,

(39) Richardson, M. F.; Wulfsberg, G.; Marlow, R.; Zaghonni, S.;

McCorkle, D.; Shadid, K.; Gagliardi, J.; Farris, B. Inorg. Chem. 1993, 32,

1913–1919.

S1–S83.

::

€

(41) Camurlu, P.; Yilmaz, A.; Tatar, L.; Kisakurek, D.; Ulku, D. Cryst.

Res. Technol. 2005, 40, 271–276.

Article

Inorganic Chemistry, Vol. 49, No. 23, 2010 10925

This sum is somewhat smaller for the 2-chloro- and 2,6-

dichlorophenolate complexes, ranging from 339.0(3)° to

356(1)°, while in methylphenolate complex 6 this sum is

much smaller at 325.4(2)°. The second measure of devia-

tion from a perfect trigonal (bi)pyramidal geometry is the

N3-Fe-N2-O1 torsion angle. A value of 180° indicates

that O1 lies directly in the same plane as the N3-Fe-N2

unit. The N3-Fe-N2-O1 torsion angle of 5 is the closest

to the ideal value at 169.8(1)°. This torsion angle is

somewhat smaller in the 2-chloro- and 2,6-dichlorophe-

nolate complexes, ranging from 134.4(3)° to 160(1)°. In

contrast, the N3-Fe-N2-O1 torsion angle for 6 is only

122.7(2)°. Thus, the halogen-substituted complexes are

more trigonal than the 2-methylphenolate complex, sug-

gesting that the halogen influences the geometry at iron-

(II), distorting the tetrahedral geometry toward a five-

coordinate trigonal bipyramidal geometry in which one

of the five bonds to the iron(II) center is a secondary

bonding interaction with the halogen. This is shown

clearly in Figure 11, where the cyclohexyl rings of com-

plexes 2, 3, 5, and 6 have been overlaid. The ortho-halogen

atoms of the phenolates in 2, 3, and 5 are similarly placed,

while the 2-methylphenolate lies drastically out of the N2,

N3, O1 plane.

The 2,6-dibromophenolate complex 5 is the most tri-

gonal bipyramidal of the halophenolate complexes. For

example, the N1-Fe-Br angle in 5 is closest to linear at

174.99(3)°. The N1-Fe-Cl angles in the 2,6-dichloro-

and 2-chlorophenolate complexes are somewhat smaller,

ranging from 163.4(5)° to 168.5(5)°. Another measure is

the τ₅value, where a τ₅value of 1.0 represents a perfect

trigonal bipyramidal geometry and 0.0 represents a

square pyramidal geometry.⁴²The τ₅values of these

complexes range from 0.408(3) for 2C to 0.617(1) for 5.

(These values are somewhat less than 1.0 because of the

constraints of the chelating ligand, which prevent the

complex from achieving perfect 90°/120° angles.)

Thus, comparison of the structures of 2, 3, 5 versus 6

illustrates that ortho-halogen substitution on the pheno-

late ring alters the placement of the phenolate. The proxi-

mity of the halogen to the metal center and its location as

one of the positions in a trigonal bipyramidal geometry

are consistent with a halogen-iron interaction. The different

orientation of the ortho-substituent in the 2-methylpheno-

late complex (6);despite the similar steric demands of a

methyl versus a chloro-substituent;suggest that an iron-

halogen interaction is responsible for the halogen orien-

tation and the coordination geometry in 2, 3, and 5 rather

than sterics or crystal packing. The shorter distance of the

Fe-halogen interaction and the closer agreement to

trigonal bipyramidal geometry in the 2,6-dibromopheno-

late complex (5) compared to any of the structures of

2-chloro- and 2,6-dichlorophenolate complexes are parti-

cularly noteworthy. This difference may indicate that sec-

ondary bond formation to an iron(II) is especially favor-

able for softer Lewis bases such as bromine substituents.

Although secondary iron-halogen interactions most

coherently explain the trends seen in this work, it is important

to consider other possible reasons for the observed trend of

halogenated phenolates binding more strongly to the iron

Figure 11. Comparison of the solid-state structures of 2,6-dichlorophe-

nolate 2B (green), 2-chlorophenolate 3A (orange), 2,6-dibromophenolate

5 (purple), and 2-methylphenolate 6 (black). The TACH rings are over-

laid, and tolylidene arms are removed for clarity. This shows that the

halogen substituent undergoing the secondary bonding interaction is

always in nearly the same position, and also that the 2-methylphenolate

complex 6 (in black) has a significantly different conformation than the

ortho-halogenated phenolates.

atom than the non-halogenated phenolates. First, one

could envision π-bonding between the phenolate oxygen

lone pairs and half-filled d orbitals on the iron ion.

However, iron-to-oxygen π-donation would be expected

to be greatest with electron-donating phenolate substitu-

ents, and in our study electron-withdrawing phenolates

were bound most strongly. On the other hand, electron-

withdrawing substituents are known to increase the

strength of polar M-X σ-bonds by stabilizing the partial

negative charge on X.⁴³Thus, this effect might play a role

in the trend observed here. Second, π-stacking between

TACH-o-tolyl aromatic groups and the phenolates is

present and could contribute. However, as mentioned

above, the ring-ring distances are similar for all com-

plexes of the trans-TACH-o-tolyl ligand, suggesting that

differences in π-stacking strength are minimal. Naturally,

we cannot rule out energetic influences on π-stacking that

do not significantly influence the structures. Third, differ-

ences in the enthalpies of solvation of the different

phenolates provides another possible contribution to

the binding enthalpies; however, the solvation enthalpies

are not known for the halogenated phenolates studied

here. Each of these differences (secondary bonding, the

strength of the Fe-O bond, π-stacking, and solvation)

would have to be considered to provide a complete

account of the thermodynamic differences between the

binding of phenolates with and without ortho-halogen

substituents.

Biological Relevance. The active site structure of PcpA

has not been crystallographically determined; however,

(43) (a) Erikson, T. K. G.; Bryan, J. C.; Mayer, J. M. Organometallics

(42) Addison, A. W.; Rao, T. N.; Reedijk, J.; Vanrijn, J.; Verschoor, G. C.

1988, 7, 1930–1938. (b) Holland, P. L. Comments Inorg. Chem. 1999, 21, 115–

J. Chem. Soc., Dalton Trans. 1984, 1349–1356.

129.

10926 Inorganic Chemistry, Vol. 49, No. 23, 2010

Rocks et al.

site-directed mutagenesis and homology modeling indi-

cate that the iron(II) ion is ligated to two histidines and a

glutamate as in EDOs.⁹In the model complexes, these

biological metal donors are mimicked by a symmetric tris-

imine ligand. Though the electronic properties are cer-

tainly not identical between the chelating ligand used here

and the biological donor set, the TACH-based ligand

benefits from o-tolyl substituents that protect the binding

pocket, preventing bridging phenolates and giving a

mononuclear iron(II) complex with a geometry that

should be considered as possible in the enzyme-substrate

adduct. The most significant result in the studies reported

here is that a close iron-halogen contact is present, and

contributes to the strength of phenolate binding. This

suggests that the chlorinated hydroquinones that are

oxidatively cleaved by the hydroquinone dioxygenases

could bind to the non-heme iron(II) site of the enzyme

through a weak interaction with the ortho chlorine of the

substrate with the metal. Because the substrate of hydro-

quinone dioxygenases has only one coordinating oxygen

atom, there is a site on the iron(II) for this secondary

interaction, whereas the two coordinating oxygen atoms

in the extradiol catechol dioxygenase enzymes do not

leave space for an interaction between the metal ion and

the halogen substituents. We suggest that in PcpA and

LinE, the weak iron-chlorine interaction may orient the

hydroquinone in the binding pocket for productive clea-

vage and provide tighter binding of the substrate to the

enzyme.

to promote chemoselective and regioselective C-C bond

cleavage.

Experimental Section

General Considerations. All manipulations were carried out

under an N₂atmosphere using standard Schlenk techniques or

in an M. Braun glovebox maintained at or below 1.0 ppm of O₂.

Acetonitrile, pentane, dichloromethane, diethyl ether, and THF

were dried using activated alumina and “deoxygenizer” columns

from Glass Contour Co. (Laguna Beach, CA) prior to use. Glass-

ware was dried at 150 °C overnight. Celite was dried at 200 °C

under vacuum prior to use.

Deuterated acetonitrile was degassed using the freeze-

˚

pump-thaw method and then dried over 4 A molecular sieves

three successive times. Deuterated THF and dimethoxyethane

were dried over CaH₂, then over sodium metal and then vacuum

distilled into a storage container before use. NMR spectra in

CD₃CN, CD₂Cl₂, and C₄D₈O (Cambridge Isotopes) were

recorded using a Bruker Avance 400 (400 MHz) and Bruker

Avance 500 (500 MHz) instruments. Acquisition parameters

were designed to suppress diamagnetic signals with long relaxa-

tion times, whereas signals from high-spin iron complexes have

relaxation times <10 ms. Typical parameters: acquisition time=

132 ms; preacquisition delay time=4.5 μs; delay time=400 ms;

pulse width=3.15 μs; sweep width=3.0ꢀ10⁵Hz; number of

scans = 64. All peaks are singlets unless otherwise specified.

Frequencies were referenced to the signal of CD₂HCN at δ

1.94 ppm, CDHCl₂at δ 5.32 ppm, and C₄D₇HO at δ 3.58 and

1.73 ppm. ¹⁹F NMR spectra were referenced to an external

standard of hexafluorobenzene at δ -164.9 ppm.

Electronic spectra (shown in Supporting Information) were

recorded from 200 to 700 nm on a Cary 50 UV-visible spectro-

photometer using quartz cuvettes of 1 cm optical path length.

Air sensitive electrospray mass spectrometry used a gastight

Hamilton syringe and direct injection of dilute solution samples

into the electrospray chamber of an Agilent LC/MS. IR spectra

of anhydrous KBr pellets and solid samples were recorded on a

Shimadzu 8400S FT-IR spectrometer. Elemental analyses were

determined at the microanalysis facilities at the University of

Illinois, Urbana-Champaign or at the University of Rochester.

Reagents, unless otherwise noted, were purchased from

Aldrich or Alfa Aesar and used without purification. All phenols

were sublimed under vacuum or, if liquid, distilled, degassed,

Conclusions

The new ligand TACH-o-tolyl can be synthesized in high

yields, and it forms a 1:1 ligand:metal complex with iron(II).

When combined with iron(II) triflate and substituted pheno-

lates, TACH-o-tolyl forms 1:1:1 ligand:iron:phenolate com-

plexes, which have been characterized both in the solid state

and in solution. Despite being tridentate, the neutral ligand is

labile when complexed with iron(II) and phenolate. When

heated to moderate temperatures, a tolylidene arm in the

1:1:1 iron:TACH-o-tolyl:phenolate complexes can isomerize

from trans to cis geometry, increasing the size of the binding

pocket at the metal. In the isomerized complexes, the metal

can bind a solvent molecule, while in the all trans ligand

complexes, there is no coordinated solvent.

The complexes containing TACH-o-tolyl with ortho-halo-

substituted phenolates exhibit Fe-halogen distances that are

short enough to qualify as secondary bonding interactions.

The geometries of these complexes also reflect the interaction

between the iron and halogen and are best described as

trigonal bipyramidal, while the complex containing a non-

coordinating ortho-methyl substituent is four-coordinate.

Further, the ortho-methyl group is pointed away from the

metal, despite the similar steric demands of the methyl and

chloro substituents. In solution, only phenolates with ortho-

halo substituents yielded complete binding to the complex,

suggesting that a metal-halogen interaction stabilizes the

complexes as well. The iron(II)-halogen secondary bond in

these complexes may point toward an explanation for the

substrate specificity of PcpA and LinE, enzymes whose

function is to oxidatively cleave chlorinated substrates. In

these enzymes, a metal-chlorine interaction may improve

substrate binding and orient the substrate in the active site

˚

and stored over fresh 4 A sieves prior to use. 2-Methylhydro-

quinone was crystallized from dry Et₂O and dried under

vacuum. It was essential that all phenols and hydroquinones be

purified and dried. Iron(II) triflate was used as its bis-acetonitrile

adduct, and was prepared according to a literature method.⁴⁴

For synthesis of phenolate complexes for X-ray crystallographic

studies, the sublimed, dried phenols were generally treated with

one molar equivalent of triethylamine to generate [HNEt₃]-

[phenolate] solutions in either THF (more often) or acetonitrile

(in a few cases). In these reactions, the use of highly purified

THF (freshly passed through activated alumina, even after

drying and degassing through standard techniques) was essen-

tial for the effective synthesis of iron(II) complexes.

Synthesis of cis,cis-1,3,5-Tris-(E)-(ortho-tolylideneimino)cyclo-

hexane (TACH-o-tolyl (1)). Method 1. Potassium hydroxide

(157 mg, 2.90 mmol) was dissolved in H₂O (1 mL) and added to

TACH 3HBr (315 mg, 0.850 mmol). The solution was stirred

3

until the TACH salt dissolved producing a colorless solution.

Ortho-tolualdehyde (0.300 mL, 2.45 mmol) was added to diethyl

ether (2.5 mL), and the mixture was added to the rapidly stirring

aqueous solution. The biphasic mixture was stirred rapidly

overnight, resulting in a white solid. H₂O (10 mL) was added

(44) Hagadorn, J. R.; Que, L.; Tolman, W. B. Inorg. Chem. 2000, 39,

6086–6090.

Article

Inorganic Chemistry, Vol. 49, No. 23, 2010 10927

to the reaction mixture, and the product was extracted with

chloroform (3 ꢀ 20 mL). The combined organic layers were

washed with H₂O (2ꢀ10 mL), dried with Na₂SO₄, filtered, and

solvent was removed in vacuo for at least 6 h to yield a white

powder (300 mg, 0.69 mmol, 81%). The product may be further

purified by crystallization under N₂from hot, anhydrous acet-

onitrile (12 mL) yielding a colorless microcrystalline solid (260 mg,

0.597 mmol, 70%). Attaining analytical purity required subsequent

washing with an aqueous EDTA solution, perhaps removing small

amounts of adventitious Zn(II). ¹H NMR (400 MHz, CDCl₃): δ

8.69 (s, 3H, HC=N), 7.87 (d, 3H, o-ArH, J=7.6 Hz), 7.62 (m,

6H, two ArH), 7.16 (d, 3H, ArH, J = 7.2 Hz), 3.60 (m, 3H,

cyclohexyl), 2.50 (s, 9H, CH₃), 2.13 (dd, 3H, cyclohexyl, J =

23 Hz, J=12 Hz), 1.93 (m, 3H, cyclohexyl). ¹³C{¹H} NMR (500

MHz, CDCl₃): δ 158.1 (CdN), 137.6 (Ar), 134.6 (Ar), 130.8

(Ar), 130.2 (Ar), 127.8 (Ar), 126.3 (Ar), 67.0 (cyclohexyl), 41.4

(cyclohexyl), 19.5 (CH₃). IR (solid sample): 1632 (s), 1599 (w),

1575 (w), 1483 (w), 1458 (w), 1440 (w), 1387 (w), 1376 (w), 1343

(w), 1285 (w), 1220 (w), 1156 (w), 1120 (w), 1086 (w), 1016 (w),

973 (w), 951 (w), 940 (w), 873 (w), 857 (w), 754 (s), 746 (s), 716

(m), 696 (w), 667 (w), 636 (w), 575 (w) cm^-1. Elem. Anal. Calcd:

C, 82.72; H, 7.64; N, 9.65. Found: C, 82.78; H, 7.67; N, 9.70.

Method 2. This procedure is adapted from the literature.¹⁹

yellow solid was washed with Et₂O (5 mL) and dried under

vacuum (43.1 mg, 0.059 mmol, 60%). Single crystals suitable for

analysis by X-ray diffraction were obtained from vapor diffusion

of Et₂O into a concentrated THF solution at -35 °C. ¹H NMR

(500 MHz, CD₃CN): δ 308 (3H, imine H or cyclohexane

RC-H), 290 (3H, imine H or cyclohexane RC-H), 67.7 (1H,

m-phenolate H), 59.8 (1H, m-phenolate H), 37.6 (3H, cyclohex-

ane βC-H), 34.6 (3H, cyclohexane βC-H), 11.6 (1H, o-pheno-

late), 10.7 (3H, tolyl H), 7.2 (3H, tolyl H), 3.1 (3H, tolyl H), -13.0

(9H, tolyl Me), -29.5 (1H, p-phenolate), -42.2 (3H, tolyl H)

ppm. UV-vis: see Supporting Information. As this complex

readily isomerizes, see [Fe(cis-TACH-o-tolyl)(2-chloropheno-

late)(THF)]^þOTf^-(4) for additional characterization.

[Fe(cis-TACH-o-tolyl)(2-chlorophenolate)(THF)]OTf (4). Fe-

(OTf)₂2CH₃CN (45 mg, 0.10 mmol) was dissolved in CH₃CN

3

(4 mL) and added to 1 (45 mg, 0.10 mmol). Sodium 2-chlor-

ophenolate (16 mg, 0.10 mmol) was dissolved in CH₃CN (4 mL)

and added to the TACH-ligand iron solution. The bright yellow

solution was stirred overnight. The solution was filtered through

Celite and concentrated to less than 1 mL. Diethyl ether (3 mL)

was added resulting in a white precipitate, and the yellow solu-

tion was filtered and solvent removed in vacuo. The yellow-

orange oil was extracted with Et₂O (2ꢀ10 mL) leaving behind a

white powder. The Et₂O washes were combined, and Et₂O was

removed under vacuum resulting in a yellow-orange powder

(64 mg, 0.070 mmol, 70%). Et₂O vapor diffusion into a con-

centrated THF solution gave X-ray quality crystals (44 mg,

0.052 mmol, 52%). As discussed in the text, the ¹H NMR

spectrum contains regions of extreme signal congestion, there-

fore signal assignments and integrations are not all specified. ¹H

NMR (500 MHz, CD₃CN): δ 260 (2H, imine H or cyclohexane

R-CH), 253 (1H, imine H or cyclohexane R-CH), 243 (2H, imine

H or cyclohexane R-CH), 239 (1H, imine H or cyclohexane

R-CH), 54.8, 51.4, 50.62, 15.7 (2H), 10.2, 9.00, 8.25, 6.06 (1H),

5.09 (1H), 4.20, 3.66, 1.13, 0.98, -1.37 (6H, Me), -4.17 (3H,

Me), -28.3 (2H) ppm. IR (KBr): 3055 (w), 2974 (m, br), 2935

(m, br), 2890 (m, br), 1620 (m), 1600 (m), 1580 (m), 1477 (s), 1463

(m), 1439 (m), 1426 (w), 1385 (w), 1316 (s), 1265 (s), 1224 (s),

1160 (s), 1121 (m), 1030 (s), 983 (w), 944 (w), 866 (m). ES-MS:

m/z = 618.10 [M]^þ, (calcd 690.25, loss of THF=618.20). Elem.

Anal. Calcd: C, 58.61; H, 5.40; N, 5.00. Found: C, 57.11; H, 5.35;

N, 4.69. The disagreement indicates that small amounts of

impurities are present.

Toluene (25 mL) was added to TACH 3HBr (535 mg, 1.44

3

mmol) in a round-bottom flask followed by ortho-tolualdehyde

(0.500 mL, 4.32 mmol) and triethylamine (0.600 mL, 4.32

mmol). The reaction was refluxed for 18 h using a Dean-Stark

trap to remove water by azeotropic distillation. The solution

was allowed to cool to room temperature, water (25 mL) was

added to the solution, and the product was extracted with

chloroform (3ꢀ25 mL). The combined organics were dried with

Na₂SO₄, filtered, and solvent removed under reduced pressure

for at least 6 h to yield a white powder (460 mg, 1.06 mmol,

78%). The material could be further purified by washing the

solid with Et₂O, dissolving in hot, dry acetonitrile, and cooling

to room temperature to precipitate. The product was spectro-

scopically identical to that from Method 1.

[Fe(TACH-o-tolyl)(2,6-dichlorophenolate)]OTf (2). Fe(OTf)₂

3

2CH₃CN (45 mg, 0.10 mmol) was dissolved in CH₃CN (4 mL)

and added to 1 (44 mg, 0.10 mmol). Sodium 2,6-dichlorophenolate

(19 mg, 0.10 mmol) was dissolved in CH₃CN (4 mL) and added to

the solution containing iron and ligand. The solution immediately

turned brilliant yellow and was stirred for 2 h. The sample was

concentrated to 4 mL, filtered through Celite, and all solvent was

removed in vacuo. The solid was washed with diethyl ether (7 mL)

and dried under vacuum yielding a bright yellow powder (77 mg,

0.096 mmol, 96%). Judging from ¹H NMR spectroscopy, the

yellow powder contains some of the cis-TACH-o-tolyl complex.

To separate isomers, crystallization is necessary. Crystals suitable

for analysis by X-ray diffraction were obtained from vapor diffu-

sion of Et₂O into a concentrated THF solution at -35 °C. ¹HNMR

(500 MHz, CD₃CN): δ 308 (3H, imine H or cyclohexane RC-H),

292 (3H, imine H or cyclohexane RC-H), 61.6 (2H, m-phenolate

H), 35.8 (3H, cyclohexane βC-H), 33.0 (3H, cyclohexane βC-H),

11.2 (3H, tolyl H), 9.1 (3H, tolyl H), 7.3 (3H, tolyl H), -12.3 (9H,

tolyl Me), -28.9 (1H, p-phenolate H), -47.7 (3H, tolyl H) ppm. IR

(KBr): 3063 (w), 2924 (w, br), 2860 (w), 1621 (m), 1598 (m), 1462

(s), 1442 (w), 1309 (m), 1266 (s), 1223 (s), 1154 (m), 1119 (w), 1031

(m), 873(w). UV-vis: see Supporting Information. ES-MS: m/z =

652.10 [M]^þ, (calcd 652.16). Elem. Anal. Calcd: C, 55.38; H, 4.52;

N, 5.24. Found: C, 55.03; H, 4.48; N, 5.28.

[Fe(cis-TACH-o-tolyl)(2,6-dibromophenolate)]OTf (5). Fe-

(OTf)₂2CH₃CN (98.2 mg, 0.225 mmol) was dissolved in

3

CH₃CN (6 mL) and added to 1 (98.2 mg, 0.225 mmol). 2,6-

Dibromophenol (56.7 mg, 0.225 mmol) was dissolved in

CH₃CN (17 mL) and Et₃N (31 μL, 0.22 mmol) was added to

the solution. Addition of the 2,6-dibromophenolate solution

to the iron solution resulted in an immediate bright orange

colored solution, and the solution was stirred for 90 min.

The solution was filtered and solvent removed in vacuo. The

orange solid was dissolved in DME (2 mL), and yellow crystals

were obtained through diffusion of Et₂O at room temperature

1

overnight (138 mg, 0.154 mmol, 69%). H NMR (500 MHz,

CD₃CN) for initial complex before isomerization to the cis

form: δ 292 (3H, imine H or cyclohexane RC-H), 286 (3H,

imine H or cyclohexane RC-H), 60.2 (2H, m-phenolate H),

31.4 (3H, cyclohexane βC-H), 29.2 (3H, cyclohexane

βC-H), 10.5 (3H, tolyl H), 9.3 (3H, tolyl H), 7.5 (3H, tolyl

H), -11.0 (9H, tolyl Me), -30.3 (1H, p-phenolate H), -41

(3H, tolyl H) ppm. IR (KBr): 3057 (w), 3018 (w), 2976 (w),

2959 (w), 2934 (m), 2916 (m), 1612 (s), 1568 (m), 1454 (s), 1423

(s), 1260 (vs), 1152 (s), 1119 (m), 1031 (s), 867 (s), 851 (m), 775

(m), 748 (s), 716 (s), 636(s), 573 (w). UV-vis: see Supporting

Information. Elem. Anal. Calcd: C, 49.85; H, 4.07; N, 4.71.

Found: C, 49.63; H, 3.88; N, 4.52.

[Fe(TACH-o-tolyl)(2-chlorophenolate)]OTf (3). Fe(OTf)₂

3

2CH₃CN (43 mg, 0.099 mmol) was dissolved in CH₃CN (3 mL)

and added to 1 (43 mg, 0.099 mmol). Sodium 2-chlorophenolate

(15 mg, 0.099 mmol) was dissolved in acetonitrile (2 mL) and

added to the ligand-iron solution. The bright yellow solution

was stirred for 30 min, and concentrated to less than 0.5 mL.

Addition of Et₂O (2 mL) caused precipitation of a yellow,

microcrystalline solid. After decanting the supernatant, the

[Fe(TACH-o-tolyl)(2-methylphenolate)]OTf (6). Fe(OTf)₂

2CH₃CN (108 mg, 0.25 mmol) was dissolved in CH₃CN (6 mL)

3

10928 Inorganic Chemistry, Vol. 49, No. 23, 2010

Rocks et al.

Table 7. Crystal and Data Parameters for the X-ray Crystal Structures of 2-7

2A

2B, 2C

3

4

5

6

7

empirical formula

C₃₇H₃₆Cl₂F₃C₃₇H₃₆Cl₂

Fe N₃O₄S F₃Fe N₃O₄S

802.50

C₃₇H₃₇Cl F₃C₄₉H₆₁Cl F₃

Fe N₃O₄S Fe N₃O₇S

768.06 984.37

C₃₇H₃₆Br₂F₃C₃₈H₄₀F₃

Fe N₃O₄S Fe N₃O₄S

891.42 747.64

C₄₅H₅₆F₃

Fe N₃O₆S

879.84

100.0(1)

0.71073

monoclinic

P2₁/n

15.789(5)

11.047(3)

25.260(8)

90

94.905(5)

90

formula weight

temperature (K)

˚

wavelength (A)

802.50

100.0(1)

0.71073

monoclinic

P2₁/c

14.686(4)

16.131(4)

15.192(4)

90

100.0(1)

0.71073

orthorhombic

P2₁2₁2₁

14.6868(15)

16.0815(16)

32.540(3)

90

100.0(1)

0.71073

monoclinic

P2₁/n

16.9571(17)

11.3491(11)

19.949(2)

90

112.264(2)

90

3552.9(6)

4

yellow,

100.0(1)

0.71073

triclinic

P1

10.2579(6) A

11.6864(7)

20.0868(13)

81.558(1)

84.955(1)

84.044(1)

2362.5(3)

2

100.0(1)

0.71073

triclinic

100.0(1)

0.71073

monoclinic

P2₁/c

crystal system

space group

P1

˚

a (A)

10.6512(11)

12.1541(12)

15.2230(15)

92.183(2)

97.622(2)

112.503(2)°

1796.1(3)

2

12.5196(15)

18.836(2)

15.5583(19)

90

108.099(2)

90

˚

b (A)

˚

c (A)

R (deg)

β (deg)

γ (deg)

volume (A )

Z

101.686(5)

90

3

˚

3524.5(16)

4

7685.5(13)

8

3487.4(7)

4

4390(2)

4

yellow-orange,

needle

crystal color,

morphology

crystal size (mm)

yellow,

plate

yellow-orange,

plate

yellow-orange, yellow, block

block

orange,

needle

plate

0.26 ꢀ 0.20

0.26 ꢀ 0.22

0.26 ꢀ 0.20

0.36 ꢀ 0.30

0.22 ꢀ 0.18

0.16 ꢀ 0.06

0.36 ꢀ 0.16

ꢀ 0.04

ꢀ 0.08

ꢀ 0.26

ꢀ 0.12

ꢀ 0.04

ꢀ 0.05

data/restraints/

parameters

6232/0/463

15710/19/996

6290/127/625

20481/16/512

17055/0/463

7131/0/455

11784/6/550

GOF on F²

0.963

0.995

R1 = 0.0586

1.043

R1 = 0.0702

1.005

R1 = 0.0461

1.022

R1 = 0.0359,

0.984

R1 = 0.0582,

1.009

R1 = 0.0578

final R indices [I > 2σ(I )] R1 = 0.0514

wR2 = 0.0985 wR2 = 0.0787

R1 = 0.1394 R1 = 0.1228

wR2 = 0.1394 wR2 = 0.0979

wR2 = 0.1842 wR2 = 0.1176 wR2 = 0.0812 wR2 = 0.0897 wR2 = 0.1353

R1 = 0.1336 R1 = 0.0629 R1 = 0.0565, R1 = 0.1376, R1 = 0.1116

wR2 = 0.2328 wR2 = 0.1247 wR2 = 0.0879 wR2 = 0.1120 wR2 = 0.1583

R indices (all data)

and added to 1 (108 mg, 0.25 mmol). 2-Methylphenol (27 mg,

0.25 mmol) was dissolved in CH₃CN (6 mL) and Et₃N (35 μL,

0.25 mmol) was added to the solution. Addition of the 2-methyl-

phenolate solution to the iron solution resulted in an immediate

bright orange colored solution, and the solution was stirred for

30 min. The solution was filtered and solvent removed in vacuo.

The orange solid was dissolved in DME (2 mL), and orange

crystals were obtained through diffusion of Et₂O at room

temperature overnight (two crops, 90 mg, 0.18 mmol, 72%).

¹H NMR (500 MHz, CD₃CN): δ 299 (3H, imine H or cyclo-

hexane RC-H), 283 (3H, imine H or cyclohexane RC-H), 76.7

(3H, 2-methylphenolate CH₃), 62.7 (1H, m-phenolate H), 57.2

(1H, m-phenolate H), 28.9 (3H, cyclohexane βC-H), 26.8 (3H,

cyclohexane βC-H), 26.0 (1H, o-phenolate H), 11.8 (3H, tolyl

H), 8.8 (3H, tolyl H), 7.1 (3H, tolyl H), -9.7 (9H, tolyl

Me), -23.0 (3H, tolyl H), -39.3 (1H, p-phenolate H) ppm. IR

(KBr): 3062 (w), 3039 (w), 3008 (w), 2948 (w), 2922 (w), 2895 (w),

1618 (m), 1597 (m), 1484 (m), 1455 (w), 1444 (w), 1426 (w), 1401

(w), 1265 (s), 1250 (s), 1169 (m), 1148 (m), 1124 (w), 1085 (w),

1064 (w), 1043 (m), 1031 (s), 872 (w), 858 (w), 750 (m), 715 (w).

Elem. Anal. Calcd: C, 60.96; H, 5.52; N, 5.61. Found: C, 60.07;

H, 5.25; N, 5.74.

¹H NMR of [Fe(cis-TACH-o-tolyl)(phenolate)(CH₃CN)]^þ

OTf^-(500 MHz, CD₃CN): δ 252 (2 signals, 1H þ 2H, imine H

or cyclohexane R-CH), 242 (1H, imine H or cyclohexane R-CH),

237 (2H, imine H or cyclohexane R-CH), 54.7, 52.0 (2 signals,

1H þ 2H, phenolate), 39.0 (1H, phenolate), 32.5 (1H), 13.1, 10.4,

8.70, 8.29, 7.78, 7.19, 5.96, 4.46, 3.59, 2.45, -0.30, -3.40, -5.10, -15.1

(2H), -27.5 (1H), -30.43 (2H, phenolate) ppm. IR (KBr): 3063

(w), 3022 (w), 2925 (w), 2885 (w), 1621 (m), 1593 (m), 1587 (m),

1487 (m), 1458 (w), 1443 (w), 1423 (w), 1385 (w), 1266 (s), 1154

(m), 1121 (w), 1031 (m), 922 (w), 919 (w), 860 (m), 755 (s), 695

(w). UV-vis: see Supporting Information. ES-MS: m/z =

584.15 [M]^þ, (calcd 656.29, loss of THF=584.24) Elem. Anal.

Calcd: C, 61.12; H, 5.75; N, 5.22. Found: C, 59.97; H, 5.99, N,

5.04. The disagreement indicates that small amounts of impu-

rities are present.

If the reagents are mixed and the ¹H NMR spectrum is

collected quickly before isomerization, [Fe(TACH-o-tolyl)-

(phenolate)]^þOTf^-can be observed. ¹H NMR of [Fe(TACH-

o-tolyl)(phenolate)]^þOTf^-(500 MHz, CD₃CN): δ 279 (3H,

imine H or cyclohexane R-CH), 265 (3H, imine H or cyclohex-

ane R-CH), 58.7 (2H, m-phenolate), 38.9 (2H, o-phenolate), 20.6

(2 overlapping signals), -12.3 (9H, tolyl Me), -36.9 (1H, p-

phenolate) ppm, diamagnetic region overlaps with free TACH-

o-tolyl and Fe(TACH-o-tolyl)(OTf)₂.

[Fe(cis-TACH-o-tolyl)(phenolate)(THF)]^þOTf^-(7). Fe(OTf)₂

3

2CH₃CN (36 mg, 0.084 mmol) was dissolved in CH₃CN (7 mL)

and added to 1 (37 mg, 0.085 mmol). Sodium phenolate (9.8 mg,

0.084 mmol) was dissolved in CH₃CN (4 mL) and the solution

added to the TACH-ligand iron solution. The bright yellow solu-

tion was stirred for 1.5 h and then concentrated to 2 mL. Diethyl

ether (5 mL) was added resulting in white precipitate that was

removed by filtering over Celite. Solvent was removed in vacuo

leaving a yellow powder. The yellow colored compound was

extracted from the solid by washing with Et₂O (2ꢀ7 mL). Solvent

was removed from the combined Et₂O washes resulting in a

yellow-orange powder (42 mg, 0.057 mmol, 68%). There are

small amounts of trans-TACH-o-tolyl iron-phenolate complex

in the product. Heating drives the mixture completely to the cis

isomer. Crystals suitable for X-ray diffraction were obtained

from a concentrated solution of THF/Et₂O, made by heating,

with Et₂O vapor diffusion at -35 °C. As discussed in the text, the

¹H NMR spectrum contains many overlapping signals, there-

fore not all signal assignments and integrations can be specified.

X-ray Crystallography. Each crystal was placed onto the tip

of a glass fiber and mounted on a Bruker SMART Platform

diffractometer equipped with an APEX II CCD area detector.

All data were collected at 100.0(1) K using MoKR radiation

(graphite monochromator). For each sample a preliminary set

of cell constants and an orientation matrix were determined

from reflections harvested from three orthogonal wedges of

reciprocal space. Full data collections were carried out with

frame exposure times of 25-120 s at detector distances of 4 or

5 cm. Randomly oriented regions of reciprocal space were

surveyed for each sample: three or four major sections of frames

were collected with 0.50-1.00° steps in ω at different j settings

and detector positions of -33 or -38° in 2θ. The intensity data

were corrected for absorption,⁴⁵and final cell constants were

€

(45) Sheldrick, G. M. SADABS, version 2008/1; University of Gottingen:

€

Gottingen, Germany, 2008.

Article

Inorganic Chemistry, Vol. 49, No. 23, 2010 10929

calculated from the xyz centroids of approximately 4000 strong

reflections from the actual data collection after integration.⁴⁶

Structures were solved using SIR97⁴⁷and refined using SHELXL-

97.⁴⁸Direct-methods solutions were calculated which provided

most non-hydrogen atoms from the difference Fourier map.

Least squares (on F²)/difference Fourier cycles located the

remaining non-hydrogen atoms. Non-hydrogen atoms were

refined with anisotropic displacement parameters, and hydro-

gen atoms were placed in ideal positions and refined as riding

atoms with relative isotropic displacement parameters. Details

of each structure are given in Table 7.

Acknowledgment. This research was supported by grants

from the American Chemical Society-Petroleum Research Fund

(ACS-PRF 44410-G3 to T.E.M.) and the National Science Foun-

dation (CHE-0951999 to T.E.M. and CHE-0911314 to P.L.H.).

Analytical data from the University of Rochester used the CENTC

Elemental Analysis Facility, funded by NSF CHE-0650456.

(46) SAINT, version 7.60A; Bruker AXS: Madison, WI, 2008.

(47) Altomare, A.; Burla, M.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi,

A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. SIR97: A new program for

solving and refining crystal structures; Istituto di Cristallografia, CNR: Bari,

Italy, 1999.

Supporting Information Available: Crystallographic data is

given in CIF format; further experimental details are given in

Figures S1-S7. This material is available free of charge via the

Internet at http://pubs.acs.org.

(48) Allen, F. H. Acta Crystallogr. 2002, B58, 380–388.

Article Doi

Solution and structural characterization of iron(II) complexes with ortho -halogenated phenolates: Insights into potential substrate binding modes in hydroquinone dioxygenases

DOI: 10.1021/ic101377u

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ic101377u

Full text of DOI:10.1021/ic101377u

Products guided by the article

R&D Labs maybe for 1257225-98-1

Relevant to this article

Hot Product