Origins of stereoselectivity in optically pure phenylethaniminopyridine tris-chelates M(NN′)3n+ (M = Mn, Fe, Co, Ni and Zn)

View Article Online / Journal Homepage / Table of Contents for this issue

This article is published as part of the Dalton Transactions themed issue entitled:

Dalton Transactions 40th Anniversary

Guest Editor Professor Chris Orvig, Editorial Board Chair

University of British Columbia, Canada

Published in issue 40, 2011 of Dalton Transactions

Image reproduced with permission of Shinobu Itoh

Welcome to issue 40 of the 40th volume of Dalton Transactions-40/40! Articles in the issue include:

PERSPECTIVE:

Synthesis and coordination chemistry of macrocyclic ligands featuring NHC donor groups

Peter G. Edwards and F. Ekkehardt Hahn

Dalton Trans., 2011, 10.1039/C1DT10864F

FRONTIER:

The future of metal–organic frameworks

Neil R. Champness

Dalton Trans., 2011, DOI: 10.1039/C1DT11184A

ARTICLES:

Redox reactivity of photogenerated osmium(II) complexes

Jillian L. Dempsey, Jay R. Winkler and Harry B. Gray

Dalton Trans., 2011, DOI: 10.1039/C1DT11138H

Molecular squares, cubes and chains from self-assembly of bis-bidentate bridging ligands with

transition metal dications

Andrew Stephenson and Michael D. Ward

Dalton Trans., 2011, DOI: 10.1039/C1DT10263J

Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research

www.rsc.org/dalton

Dynamic Article Links

Dalton

Transactions

Cite this: Dalton Trans., 2011, 40, 10416

www.rsc.org/dalton

PAPER

Origins of stereoselectivity in optically pure phenylethaniminopyridine

tris-chelates M(NN¢)₃ⁿ⁺(M = Mn, Fe, Co, Ni and Zn)†

Suzanne E. Howson, Laura E. N. Allan, Nikola P. Chmel, Guy J. Clarkson, Robert J. Deeth, Alan D. Faulkner,

Daniel H. Simpson and Peter Scott*

Received 6th April 2011, Accepted 24th May 2011

DOI: 10.1039/c1dt10588d

One-pot reactions of 2-pyridinecarboxaldehyde, chiral phenylethanamines and Fe(II) give single

diastereomer fac diimine complexes at thermodynamic equilibrium so that no chiral separations are

required (d.r. > 200 : 1). The origins of this stereoselectivity are partly steric and partly a result of the

presence of three sets of inter-ligand parallel-offset p-stacking interactions. Mn(II), Co(II), Co(III), Ni(II)

and Zn(II) give similar fac structures, alongside the imidazole analogues for Fe(II). While most of the

complexes are paramagnetic, the series of molecular structures allows us to assess the inﬂuence of the

p-stacking present, and there is a strong correlation between this and the M–N bond length. Fe(II) is

close to optimal. For the larger Zn(II) ion, very weak p-stacking leads to poorer measured

stereoselectivity (NMR) but this is improved with increased solvent polarity. The mechanism of

stereoselection is further investigated via DFT calculations, chiroptical spectroscopy and the use of

synthetic probes.

Introduction

The development of methods to synthesise optically pure tris-

chelate octahedral coordination complexes has been of interest

since the beginning of coordination chemistry,¹but it is a

problem still far from solved. Single isomers are required to

develop the large number of potential applications in anion

recognition,^2,3molecular sensors,^4,5supramolecular chemistry,^6,7

DNA targeting,^8,9protein probes¹⁰and cancer therapy.¹¹

Fig. 1 Enantiomers of [Ru(bpy)₃]²⁺

.

Tris-chelate octahedral complexes with symmetrical (A–A)

type ligands, e.g. [Ru(bpy)₃]²⁺are generally formed as racemic

mixtures of the D and K enantiomers (Fig. 1).¹In some cases with

inert metals, e.g. Ru(II), these can be resolved by diastereomeric

crystallisation with chiral anions^12–15or by chromatographic

techniques, either using a chiral stationary phase¹⁶or by cation

exchange chromatography with chiral anions.^17–19Determination

of the optical purity can, however, be challenging and racemisation

is an ever-present issue.²⁰For the asymmetric (A–B) ligand class,

e.g. monosubstituted bipyridines,^21,22geometric fac and mer con-

ﬁgurations are introduced and, excluding any steric or electronic

effects, these are expected in a statistical 1 : 3 ratio. Separation

of the four isomers fac-D, mer-D, fac-K and mer-K, is again only

viable for inert complexes and is inherently low-yielding.²³

approaches to highly stereoselective syntheses, notably from the

Meggers group, involve the use of a temporary chiral auxiliary

ligand to favour one enantiomer.²⁴The conﬁguration at the metal

centre is retained on substitution of the auxiliary for the required

ligand in inert complexes only to form e.g. enantiomerically

pure tris-heteroleptic ruthenium(II) complexes with achiral 2,2¢-

bipyridines.^24,25Earlier approaches, such as by von Zelewsky, use

chiral multidentate ligands to preorganise the conﬁgurations at

the metal centres forming optically pure octahedral complexes

without the need for resolution.^26,27The syntheses are, however,

rather lengthy and the systems are synthetically inﬂexible.

An elegant preorganisational technique has been used by

the Fletcher^21,28,29and Weizman³⁰groups to allow the exclusive

synthesis of fac isomers in inert Ru(II) complexes. Tripodal

bipyridine ligands force the structures to adopt a fac conﬁguration.

This method can also control the absolute conﬁguration at the

metal through the use of an optically pure chiral tether, resulting

in a single diastereomer.²⁸

Stereoselective syntheses resulting in single isomers are thus

a key challenge in this area of coordination chemistry. Current

Department of Chemistry, University of Warwick, Gibbet Hill Road,

Coventry, UK, CV4 7AL. E-mail: peter.scott@warwick.ac.uk; Fax: +44

024 7657 2710; Tel: +44 024 7652 3238

† CCDC reference numbers 821020–821027. For crystallographic data in

CIF or other electronic format see DOI: 10.1039/c1dt10588d

For more labile metals, complete thermodynamic control of

the diastereoselection had not been achieved until our recent

10416 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

report that iron(II) complexes with simple optically pure diimine

ligands give optically and stereochemically pure fac isomers with

d.r. > 200 : 1.³¹We describe here an investigation into the origins

of this unprecedented stereo- and chemical selectivity in the

[ML₃]ⁿ⁺complexes (M = Fe, Co, Ni, Mn, Zn) via computational

and structural studies, and also the extension to imidazole

ligands. We have also recently described the related complexes

of achiral hydrazones, which are highly fac selective and for which

single enantiomers crystallise as conglomerates.³²We have also

commented on some recent related Cu(II) chemistry.³³

Synthesis of ligand precursors

Optically pure (R)-2-phenylglycinol (4) was synthesised via the re-

duction of (R)-2-phenylglycine using lithium aluminium hydride³⁴

or, more conveniently on a larger scale (> 15 g), using NaBH₄/I₂

via an adaptation of a literature procedure.³⁵Puriﬁcation via

recrystallisation from hot toluene was found to maintain the

optical purity of this compound to give an optical rotation value

of -25.99^◦(Lit. = -25.8^◦).³⁶In contrast, distillation was found

to result in a partial racemisation leading to a product with an

enantiomeric ratio of approximately 7 : 3 (R:S).

Scheme 1 One-pot syntheses of tris-pyridine/imine complexes of Fe(II).

Although protecting group strategies have been used in the

synthesis of ether derivatives of 2-phenylglycinol, we found that

the direct reaction of the alcohol with halides in the presence of

sodium hydride gave the corresponding methyl (5)³⁷and benzyl

(6)³⁸ethers in good yields.

Imidazole-2-carbaldehydes (9–11) were synthesised in high

yields from the corresponding N-substituted imidazole and

dimethylformamide in the presence of n-BuLi.³⁹

Iron(II) complexes

Fig.

fac,K_Fe,R_C-[FeL³][ClO₄]₂and (c) fac,D_Fe,R_C-[FeL⁴][ClO₄]₂.

2

¹H NMR spectra in CD₃CN of (a) [FeL²][ClO₄]₂, (b)

3

Following early observations by Busch,⁴⁰three-component ‘tem-

plate’ reactions between amine, aldehyde and metal source, rather

than simple substitution reactions with pre-formed ligands, are

commonly used in the synthesis of complexes of neutral imine

ligands. Indeed Burgess and co-workers isolated an ‘intermediate’

iron(II) species coordinated to two pyridine-imine ligands and one

2-acetylpyridine ligand.⁴¹We were surprised to ﬁnd, however,

that rather few tris-pyridine/imine complexes of Fe(II) have

appeared in the literature. We nevertheless used a one-pot strategy,

mixing the appropriate amine and 2-pyridinecarboxaldehyde into

a solution of Fe(ClO₄)₂·6H₂O in acetonitrile (Scheme 1). This led

to the immediate formation of intense purple solutions, from which

the diamagnetic complexes [FeL₃][ClO₄]₂were isolated.

3

slow on the ¹H NMR chemical shift timescale - as expected for a

low spin d⁶ion - with distinct resonances observed for the imine

units and diastereotopic benzylic CH₂groups (one set for C₃-

symmetric fac and three for C₁-symmetric mer). We reasoned that

this thermodynamic preference for the fac isomer could not merely

be a steric effect; the bulkier L¹gave a mer-rich system so reducing

the steric bulk to L²could not in itself encourage the system beyond

the statistical ratio of 1 : 3. We were thus encouraged to explore

the phenomenon further.

The commercially available (R)-a-methylbenzylamine (e.e. =

96%) (3, Scheme 2) also gave a purple crystalline complex in the

standard reaction. Given the long history of this approach de-

scribed above and also our own explorations of diastereoselection

We found that 4-aminophenol (1) gave a product that we could

crystallise readily and from the general reaction described above,

diamagnetic [FeL¹][ClO₄]₂was isolated in 79% yield. The ¹H

at labile metals^42–47we were astounded to ﬁnd that the H NMR

1

3

NMR spectrum of this complex clearly shows the presence of the

(racemic) fac and mer isomers but with a far higher selectivity

for mer (1 : 15 ratio) than is expected statistically (1 : 3). This

gave us our ﬁrst indication that the stereoselectivity in these

complexes could be modiﬁed using steric effects. Consequently, we

moved to the less bulky benzylamine 2, which gave diamagnetic

spectra showed only one set of coordinated ligand signals [Fig.

2 (b)]. The tris stoichiometry is conﬁrmed by the diamagnetic

nature of the Fe(II) system, microanalysis and mass spectrometric

data. Hence a single diastereoisomer with a C₃-symmetric fac

structure is formed. No further isomers were observed at accessible

temperatures or after prolonged storage at elevated temperature,

1

[FeL²][ClO₄]₂. The ¹H NMR spectrum of this complex indicates a

and on the basis that the only impurities present have H NMR

3

fac:mer ratio of 1 : 2 [Fig. 2 (a)] - unexpectedly rich in fac isomers -

and this did not change over time even after prolonged (7 d)

heating at 80 ^◦C in air. The exchange between these isomers is

integrations lower than those of the ¹³C satellites from the major

isomer, the thermodynamic diastereomer ratio must be >200 : 1.

By analogy with several molecular structure determinations of

This journal is

The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10417

©

Scheme 2 Single diastereomer complexes [FeLⁿ]²⁺(n = 3–8).

3

Fig. 3 Structure of one of the independent cations in the asymmetric

related compounds and following circular dichroism studies

1

unit of fac,D_Fe,R_C-[FeL⁴][ClO₄]₂· CH₃CN· H₂O (H atoms, counteri-

3

2

4

(vide infra), the structure fac,K_Fe,R_C-[FeL³][ClO₄]₂is assigned

3

ons and solvent molecules omitted for clarity). Thermal ellipsoids are

◦

(Scheme 2).

˚

shown at 50% probability. Selected bond lengths (A) and angles ( ):

The analogous reaction with racemic a-methylbenzylamine

(3) was investigated using one equivalent each of (S)-a-

methylbenzylamine and (R)-a-methylbenzylamine in the standard

procedure. After 24 h the ¹H NMR spectrum of the crude product

was found to be essentially identical to that of non-racemic

Fe(1)–N(5) 1.967(4), Fe(1)–N(3) 1.969(4), Fe(1)–N(6) 1.970(4), Fe(1)–N(1)

1.970(4), Fe(1)–N(4) 1.983(4), Fe(1)–N(2) 1.991(4), N(2)–C(6) 1.289(7);

N(5)–Fe(1)–N(6) 81.23(17), N(3)–Fe(1)–N(4) 81.65(18), N(1)–Fe(1)–N(2)

81.28(16).

The use of the methyl ether³⁷5 in the standard synthesis

fac,K_Fe,R_C-[FeL³][ClO₄]₂. Recrystallisation gave a pure sample,

3

led to the formation of fac,D_Fe,R_C-[FeL⁵][ClO₄]₂. This complex

which did not alter in diastereomeric composition over time, thus

conﬁrming that the thermodynamic product is homochiral.

The use of (R)-2-phenylglycinol (4) in the standard synthesis

led to the isolation of the corresponding highly crystalline trihy-

3

was also formed as a single diastereomer. The single-crystal

X-ray molecular structure (Fig. 4) is analogous to fac,D_Fe,R_C-

[FeL⁴][ClO₄]₂.

3

droxy complex fac,D_Fe,R_C-[FeL⁴][ClO₄]₂. The ¹H NMR spectrum

3

conﬁrmed d.r. >200 : 1 as for the L³complex [Fig. 2 (c)]. Single

crystals were obtained and a single-crystal X-ray structure was

determined.

The molecular structure of the cationic unit is shown in Fig. 3,

viewed along the approximate C₃axis, along with key bond lengths

and angles. The crystal selected for single-crystal X-ray diffraction

had the chiral space group P2₁and allowed the fac,D_Fegeometry to

be readily assigned. The asymmetric unit contains two complexes,

four perchlorate counterions, an acetonitrile molecule and a water

molecule (O23) modelled at half occupancy. The complex is

approximately octahedral with an average Fe–N bond length of

˚

1.97 A, which is close to the expected average (1.958 A) for this

class of low-spin Fe(II) complex.⁴⁸Each of the three pyridine units

forms a face-to-face p-stack with a phenyl unit on a neighbouring

ligand, with an average angle between the mean planes of the

pyridine and phenyl rings of 7.7^◦and an average centroid–centroid

˚

distance of 3.55 A. These p-stacks contribute to the unprecedented

stereoselectivity in this monometallic system (vide infra) and orient

the CH₂OH groups such that they are mutually directed to encircle

the C₃axis. Each OH group participates in H-bonding with an OH

group of another molecular unit, and this is accompanied by short

contacts with perchlorate anions and partially-occupied H₂O.

Alongside the issue of stereoselectivity in the tris(bidentate)

Fig.

4 Structure of the cation in the asymmetric unit of

fac,D_Fe,R_C-[FeL⁵][ClO₄]₂·2CH₃CN (H atoms, counterions and solvent

3

molecules omitted for clarity). Thermal ellipsoids are shown at 50%

probability. Selected bond lengths (A) and angles ( ): Fe(1)–N(6) 1.962(7),

Fe(1)–N(3) 1.962(7), Fe(1)–N(5) 1.969(6), Fe(1)–N(2) 1.974(7), Fe(1)–N(1)

1.978(7), Fe(1)–N(4) 1.981(7), N(2)–C(6) 1.264(11); N(5)–Fe(1)–N(6)

81.5(3), N(3)–Fe(1)–N(4) 81.2(3), N(1)–Fe(1)–N(2) 80.5(3).

◦

˚

complex [FeL⁴][ClO₄]₂it is noteworthy that no bis(tridentate)

3

[FeL⁴][ClO₄]₂with k -N,N¢,O coordinating ligands is detected

2

at equilibrium. This issue is explored later.

10418 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

The use of the benzyl ether³⁸6 in the standard synthesis led to

a purple solution as usual but after 2 h at ambient temperature

1

the H NMR spectrum of the crude reaction mixture indicated

that the kinetic product thus isolated contained one fac and one

mer isomer in an approximate 1 : 1 ratio. This feature is best seen

with the four sets of doublet of doublets in the region between

3.60 and 3.20 ppm [Fig. 5 (a)], corresponding to one of the two

diastereotopic CHCH₂protons on the ligands. After heating this

◦

sample at 75 C for 72 h the exclusive (thermodynamic) product

fac,D_Fe,R_C-[FeL⁶][ClO₄]₂was formed as for the above complexes

3

[Fig. 5 (b)].

Fig. 6 Structure of the cationic unit in [FeL⁸][ClO₄]₂(H atoms

2

Fig. 5 ¹H NMR spectra in CD₃CN of crude samples of R_C-[FeL⁶][ClO₄]₂

and counterions omitted for clarity). Thermal ellipsoids are shown

3

at 50% probability. Selected bond lengths (A) and angles (^◦):

˚

after (a) 2 h at ambient temperature and (b) 72 h at 75 ^◦C.

Fe(1)–N(108) 2.1043(17), Fe(1)–O(211) 2.1118(17), Fe(1)–N(208)

2.1178(16), Fe(1)–N(201) 2.1565(19), Fe(1)–O(111) 2.1729(15),

Fe(1)–N(101) 2.1952(18), C(107)–N(108) 1.279(3); O(211)–Fe(1)–N(208)

As part of a study into the origins of the stereoselection,

we investigated the use of other chiral a-substituted amines.

Commercial (R)-1-cyclohexylethanamine (e.e. = 95%) (7) in the

standard complex synthesis gave a purple diamagnetic mixture

74.36(7),

O(211)–Fe(1)–N(201)

147.01(7),

N(208)–Fe(1)–N(201)

76.10(7), N(108)–Fe(1)–O(111) 74.25(6), N(108)–Fe(1)–N(101) 76.22(7),

O(111)–Fe(1)–N(101) 146.40(6).

[FeL⁷][ClO₄]₂containing one fac and one mer isomer in the

3

thermodynamic ratio 2.6 : 1. While this ratio still strongly favours

the fac isomer with respect to the statistically expected ratio of 1 : 3,

the thermodynamic stereoselectivity is substantially worse than

We have also investigated the use of 2-imidazole in the place of

2-pyridine in this system (Fig. 7). The one-pot reaction between 1-

methyl-1H-imidazole-2-carbaldehyde, (R)-a-methylbenzylamine

and Fe(ClO₄)₂·6H₂O in acetonitrile resulted in an intense purple

solution. The dropwise addition of ethyl acetate resulted in the

precipitation of a dark purple microcrystalline solid. In com-

parison to the 2-pyridinecarboxaldehyde systems, the resulting

that seen in the analogous fac,K_Fe,R_C-[FeL³][ClO₄]₂complex.

3

Ligand L⁸, derived from (S)-tert-leucinol (8), gave in contrast

to all the systems above, a paramagnetic crystalline red complex

mer,S_C-[FeL⁸][ClO₄]₂. The molecular structure as determined by

2

single-crystal X-ray crystallography is shown in Fig. 6. The crystal

selected for single-crystal X-ray diffraction had the chiral space

group P2₁. The asymmetric unit contains the Fe(II) complex with

two perchlorate counterions. As well as the nitrogen atoms, the

oxygen atoms are also chelated to the iron(II) centre, forming

complex [FeL⁹][ClO₄]₂was found to be paramagnetic by ¹H

3

NMR spectroscopy. Slow vapour diffusion of ethyl acetate into an

acetonitrile solution resulted in single crystals that were suitable

for single-crystal X-ray diffraction. The solid state structure of one

of the two crystallographically independent but chemically similar

iron(II) complexes is shown in Fig. 8.

3

a bis(k -N,N¢,O) structure. The alcohol hydrogen atoms were

located and were found to form short strong H bonds to the

˚

perchlorate oxygen atoms with an average distance of 1.97 A. The

˚

average Fe–N bond length is 2.14 A, close to the expected average

The crystal has the chiral space group P2₁and the asymmet-

ric unit contains two iron imidazolimine complexes with two

partially occupied acetonitrile molecules, in addition to four

interspatial perchlorate counterions. The structure shows three

ligands arranged around a central iron(II) atom in the fac,K_Fe,R_C-

48

˚

(2.13 A) for this class of high-spin Fe(II) complex. The angle

between the mean planes of the two pyridine/imine units is 78.96^◦.

The bond angles between the coordinating atoms making up the

distorted octahedron range from 74.25^◦to 112.61^◦as a result of

the constraints of the chelate. In addition to the single-crystal X-

ray studies, microanalysis and NMR studies indicated that the

[FeL⁹][ClO₄]₂conﬁguration, in which each of the three N-methyl

3

imidazole rings are oriented face-to-face with the phenyl ring of

the adjacent ligand. The average angle between the mean planes of

these rings (11.8^◦) is, however, notably larger than those observed

in the pyridine Fe(II) structures (ca 8^◦), and the average centroid–

3

bulk sample comprises this paramagnetic bis(k -N,N¢,O) complex

and no tris complex analogous to fac,D_Fe,R_C-[FeL⁴][ClO₄]₂was

3

detected in solution.

˚

Magnetic measurements were conducted in the range 400–

2 K for [FeL⁸][ClO₄]₂. The complex remains in the high spin

state over the measured temperature range and displays expected

paramagnetic behaviour. SQuID magnetometry data led to a m_eff

value of 5.30 m_Bat room temperature (298 K), which is as expected

for an octahedral high spin Fe(II) complex.⁴⁹

centroid bond length of 3.77 A is substantially longer (ca 3.55 A).

Also, unlike the pyridine analogues in which the Fe–N bond

lengths are all essentially equivalent, there is a distinct difference

2

˚

between the average Fe–N_imzbond length (1.96 A) and the average

˚

Fe–N_iminebond length (2.02 A). This has been observed in related

imidazole Schiff base complexes.^50,51

This journal is

The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10419

©

Fig. 9 Plot of magnetic moment vs. temperature for [FeL⁹][ClO₄]₂.

3

Fig. 7 Ligands L⁷–L¹¹.

suitable for single-crystal X-ray diffraction were obtained through

slow vapour diffusion of ethyl acetate into an acetonitrile solution.

The resulting molecular structure (Fig. 10) was of the fac,K_Fe,R_C-

[FeL¹¹₃][ClO₄]₂diastereomer. The steric hindrance between the

phenyl ring and the N-tert-butyl group results in the complete

disruption of the p-stacking observed in the less bulky N-methyl

analogue. As a result there is a substantial increase in the mean

angle between the planes of the phenyl ring and the imidazole

to 29.9^◦and an associated increase in the average centroid–

˚

centroid distance (4.22 A). Inspection of the Fe–N bond lengths

˚

again identiﬁes a larger distance for the Fe–N_iminebond (1.98 A)

˚

compared to that of the Fe–N_imz(1.94 A). Attempts at assessment

of the diastereoselectivity in the bulk sample through powder XRD

were hindered by broad and weak diffraction patterns.

Fig. 8 Structure of one of the independent cations in the asymmetric

1

2

unit of fac,K_Fe,R_C-[FeL⁹][ClO₄]₂· CH₃CN (H atoms, counterions and

3

solvent molecules omitted for clarity). Thermal ellipsoids are shown at

◦

˚

50% probability. Selected bond lengths (A) and angles ( ): Fe(1)–N(1A)

1.954(4), Fe(1)–N(1) 1.962(4), Fe(1)–N(1B) 1.963(4), Fe(1)–N(3) 2.011(4),

Fe(1)–N(3A) 2.014(4), Fe(1)–N(3B) 2.019(4); N(1)–Fe(1)–N(3) 80.49(16),

N(1A)–Fe(1)–N(3A) 81.09(17), N(1B)–Fe(1)–N(3B) 81.00(15).

SQuID measurements on

a polycrystalline sample of

[FeL⁹][ClO₄]₂show a gradual increase in magnetic moment from

3

0.9 to 1.7 m_Bas the temperature increases from 10 to 300 K (Fig.

9). While the magnetic moment for a high spin d⁶complex is

expected to fall within the range of 5.0–5.2 m_B,⁴⁹this reduced value

indicates that the complex undergoes partial conversion to high

spin over the measured range. The observation of paramagnetic

behaviour in the imidazole system is a result of the combined

effects of the reduced p donation and p acceptor ability of the

ligand.^52–55Complete spin crossover has previously been reported

in analogous tris(pyridylbenzimidazole) iron(II) complexes by

Boca et al.⁵⁶Our own attempt to incorporate a benzimidazole

group⁵⁷in this series led to L¹⁰, for which we were unable to detect

a tris complex of Fe(II), presumably for steric reasons. In contrast,

the system with an N-tert-butyl imidazole unit (L¹¹) gave a dark

purple paramagnetic solid on complexation with iron(II). Crystals

Fig. 10 Structure of the cation in the asymmetric unit of

fac,K_Fe,R_C-[FeL¹¹₃][ClO₄]₂·1.25CH₃CN (H atoms, counterions and sol-

vent molecules omitted for clarity). Thermal ellipsoids _◦are shown at

˚

50% probability. Selected bond lengths (A) and angles ( ): Fe(1)–N(7)

1.940(3), Fe(1)–N(4) 1.942(3), Fe(1)–N(1) 1.948(3), Fe(1)–N(3) 1.974(3),

Fe(1)–N(9) 1.985(3), Fe(1)–N(6) 1.989(3); N(1)–Fe(1)–N(3) 81.14(15),

N(4)–Fe(1)–N(6) 80.73(12), N(7)–Fe(1)–N(9) 80.47(12).

10420 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

Zinc(II) complexes

Colourless complexes of Zn(II) with L³and L⁴were readily isolated

via the use of zinc(II) tetraﬂuoroborate hydrate. In the case of

R_C-L³, the molecular structure (Fig. 11) of the compound is the

fac,K_Zn,R_C-[ZnL³][ClO₄]₂isomer as expected on the basis of the

3

above Fe(II) chemistry i.e. the structure recorded is the mirror

image of fac,D_Fe,R_C-[FeL⁴][ClO₄]₂.

3

Fig. 12 Change in the fac/mer ratio of K_Zn,R_C-[ZnL³][BF₄]₂with solvent

3

polarity.

Cobalt(II) complex

The addition of cobalt(II) chloride to a methanol solution con-

taining pre-formed L³led to the immediate formation of a bright

orange solution. Analytically pure crystals of the paramagnetic

complex [CoL³][PF₆]₂were readily isolated following addition of

3

sodium hexaﬂuorophosphate. Crystals suitable for single-crystal

X-ray crystallography were grown from the methanol ﬁltrate. The

molecular structure of this compound is shown in Fig. 13 and is

the fac,K_Co,R_C-[CoL³][PF₆]₂isomer, as expected from the Fe(II)

3

chemistry. The structure is analogous to the mirror image of

fac,D_Fe,R_C-[FeL⁴][ClO₄]₂.

3

Fig. 11 Structure of the cationic unit in fac,K_Zn,R_C-[ZnL³][ClO₄]₂·

3

CH₃CN·EtOAc (H atoms, counterions and solvent omitted for clar-

ity). Thermal ellipsoids are shown at 50% probability. Selected

◦

˚

bond lengths (A) and angles ( ): Zn(1)–N(1) 2.155(2), Zn(1)–N(5)

2.157(2), Zn(1)–N(2) 2.158(2), Zn(1)–N(3) 2.175(2), Zn(1)–N(6) 2.189(2),

Zn(1)–N(4) 2.199(2), N(2)–C(6) 1.268(4); N(1)–Zn(1)–N(2) 77.32(10),

N(3)–Zn(1)–N(4) 77.11(9), N(5)–Zn(1)–N(6) 76.82(10).

In solution, the ¹H NMR spectrum of R_C-[ZnL³][BF₄]₂at

3

298 K revealed the presence of two isomers, which on the

basis of the 3-fold symmetry of one isomer and computations

(vide infra) are assigned as fac,K_Zn(major) and mer,K_Zn(minor).

This poorer stereoselectivity gives us the opportunity to probe

the effect of solvent polarity on the diastereomeric ratio using

¹H NMR spectroscopy. The complex is not soluble in protic

media such as methanol but it is conveniently soluble in various

mixtures of acetonitrile (polarity index 5.8) and dichloromethane

(polarity index 3.1). Fig. 12 shows that the fac/mer ratio increases

with solvent polarity. This is discussed further in relation to

comparisons of crystallographic data.

Fig. 13 Structure of the cationic unit in fac,K_Co,R_C-[CoL³][PF₆]₂·

3

2MeOH (H atoms, counterions and solvent omitted for clarity). Thermal

˚

ellipsoids are shown at 50% probability. Selected bond lengths (A)

With the potentially tridentate L⁴, Zn(II) gave a complex mixture

of products, probably comprising various diastereomers of tris and

bis ligand species. However, in the crystallised solid microanalysis

was consistent with the presence of predominantly the bis ligand

complex [ZnL⁴][BF₄]₂, analogous to [FeL⁸][ClO₄]₂. Hence it

and angles (^◦): Co(1)–N(1) 2.125(4), Co(1)–N(3) 2.139(4), Co(1)–N(2)

2.152(4), Co(1)–N(6) 2.152(4), Co(1)–N(4) 2.163(4), Co(1)–N(5) 2.167(4),

N(2)–C(6) 1.259(6); N(1)–Co(1)–N(2) 77.61(15), N(3)–Co(1)–N(4)

77.66(15), N(5)–Co(1)–N(6) 77.20(15).

2

3

appears that with Zn(II), where k coordination of the ligand is

Magnetic measurements were conducted in the range 340–2 K

3

possible, for example with L³(k -N,N¢,O), a non-chemoselective

for [CoL³][PF₆]₂. The complex remains in the high spin state over

3

reaction is present.

this range and displays expected paramagnetic behaviour. The

This journal is

The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10421

©

SQuID magnetometry data leads to a m_effvalue of 4.86 m_Bat room

temperature; values between 4.7 and 5.2 m_Bare typically observed

for octahedral high spin Co(II) complexes.⁴⁹Evans’ NMR method

gave the same value in solution.

Cobalt(III) complex

Rather few tris-diimine complexes of Co(III) are known, pre-

sumably because in this ligand environment the Co(II) state is

expected to be stabilised. The synthesis of this compound from

cobalt(II) precursors via oxidation was unsuccessful. Use of an

oxidising agent, e.g. H₂O₂or Cp₂FePF₆, resulted in a mixture of

cobalt(III) and cobalt(II) compounds. Finally, the reaction between

the proligand and the cobalt(III) precursor Na₃[Co(NO₂)₆] gave

the desired complex, [CoL³][PF₆]₃. This complex was found to

3

be unstable and decomposed rapidly, thus preventing detailed

1

analysis on a pure sample. The major species in the H NMR

spectrum of [CoL³][PF₆]₃appears to be the fac isomer and

3

while other minor components were observed, the instability

of the system means that we are unable to exclude the strong

possibility that these are e.g. diamagnetic solvated bis complexes.

The stereoselectivity in this product is thus unknown.

Fig. 15 Structure of the cationic unit in fac,D_Ni,S_C-[NiL³][ClO₄]₂·

3

CH₃CN·EtOAc (H atoms, counterions and solvent omitted for clar-

ity). Thermal ellipsoids are shown at 50% probability. Selected

◦

˚

bond lengths (A) and angles ( ): Ni(1)–N(3) 2.082(3), Ni(1)–N(5)

A molecular structure of the cobalt(III) complex has nevertheless

been obtained (Fig. 14), and reveals that the isomer was fac,D_Co,S_C-

2.094(3), Ni(1)–N(1) 2.097(3), Ni(1)–N(2) 2.102(3), Ni(1)–N(4) 2.117(3),

Ni(1)–N(6) 2.138(3), C(6)–N(2) 1.275(4); N(1)–Ni(1)–N(2) 78.97(11),

N(3)–Ni(1)–N(4) 79.06(10), N(5)–Ni(1)–N(6) 79.12(10).

[CoL³][PF₆]₃, as expected from the Fe(II) system. The single-

3

crystal X-ray molecular structure is analogous to fac,D_Fe,R_C-

the basis of the above Fe(II) chemistry i.e. the structure recorded

[FeL⁴][ClO₄]₂.

3

is analogous to fac,D_Fe,R_C-[FeL⁴][ClO₄]₂.

3

Manganese(II) complex

The use of manganese(II) perchlorate hexahydrate with L³in the

standard procedure resulted in the formation of a yellow Mn(II)

complex. The molecular structure (Fig. 16) of this compound is

Fig. 14 Structure of the cationic unit in fac,D_Co,S_C-[CoL³][PF₆]₃·

3

3CH₃CN (H atoms, counterions and solvent omitted for clarity). Thermal

˚

ellipsoids are shown at 50% probability. Selected bond lengths (A) and

angles (^◦): Co(1)–N(201) 1.932(4), Co(1)–N(101) 1.934(4), Co(1)–N(308)

1.937(4), Co(1)–N(301) 1.939(4), Co(1)–N(108) 1.959(4), Co(1)–N(208)

1.962(4), N(108)–C(107) 1.285(7); N(101)–Co(1)–N(108) 83.73(19),

N(201)–Co(1)–N(208) 83.21(17), N(301)–Co(1)–N(308) 83.53(18).

Fig. 16 Structure of the cationic unit in fac,K_Mn,R_C-[MnL³][ClO₄]₂·

3

Nickel(II) complex

2CH₃CN (H atoms, counterions and solvent omitted for clarity).

Thermal ellipsoids are_◦shown at 50% probability. Selected bond

A light red-brown coloured paramagnetic complex of Ni(II) with

L³was readily isolated in a good yield via the use of nickel(II)

perchlorate hexahydrate. The molecular structure (Fig. 15) of this

compound is the fac,D_Ni,S_C-[NiL³][ClO₄]₂isomer as expected on

˚

lengths (A) and angles ( ): Mn(1)–N(6) 2.240(2), Mn(1)–N(3) 2.254(2),

Mn(1)–N(1) 2.258(3), Mn(1)–N(2) 2.273(2), Mn(1)–N(5) 2.284(3),

Mn(1)–N(4) 2.285(2), C(6)–N(2) 1.276(4); N(1)–Mn(1)–N(2) 73.91(9),

N(3)–Mn(1)–N(4) 73.98(9), N(5)–Mn(1)–N(6) 73.59(9).

3

10422 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

the fac,K_Mn,R_C-[MnL³][ClO₄]₂isomer as expected from the Fe(II)

The UV-vis spectrum of R_C-[FeL⁹][ClO₄]₂has overlap-

3

system, i.e. the Mn(II) structure is the mirror image of fac,D_Fe,R_C-

ping bands in the region 250–350 nm, associated with p–p*

transitions.⁵⁰The MLCT region has two distinct maxima at ca

547 nm (e = 9000 M^-1cm^-1) and 504 nm (e = 7300 M^-1cm^-1).

The spectrum of the cobalt(II) complex contains a very broad

signal corresponding to MLCT transitions at ca 340 nm (e =

2300 M^-1cm^-1), which appears as a shoulder on the p–p* bands

at ca 280 nm. This MLCT signal is signiﬁcantly less intense than

these transitions in the iron(II) analogue, which consistent with

the notably less intense colour of the crystals and solutions of R_C-

[CoL³][PF₆]₂compared with fac,K_Fe,R_C-[FeL³][ClO₄]₂. The lower

[FeL⁴][ClO₄]₂.

3

Spectroscopy

UV-vis absorbance spectra for the Fe(II), Zn(II), Co(II), Ni(II)

and Mn(II) complexes with ligand L³, and the iron(II) imida-

zolimine complex R_C-[FeL⁹][ClO₄]₂in acetonitrile are shown in

3

Fig. 17.

3

intensity and the shift to a higher frequency is consistent with the

poorer p-donor character of cobalt(II).⁵⁸At higher concentration

(1.0 mM), two further signals can be detected at ca 465 nm (e =

80 M^-1cm^-1) and 535 nm (e = 20 M^-1cm^-1) corresponding to

Laporte-forbidden but spin-allowed d-d transitions.

The spectrum of the manganese(II) complex also has a shoulder

on the p–p* bands corresponding to the MLCT transitions. As

expected, no d-d transitions are observed for this complex as they

are both Laporte and spin forbidden.

The spectrum of the nickel(II) complex has no MLCT bands as

expected. At higher concentration (1.0 mM), two signals can be

detected at ca 360 nm (e = 100 M^-1cm^-1) and 545 nm (e = 3 M^-1

cm^-1) corresponding to Laporte-forbidden but spin-allowed d-d

transitions.

The spectrum of the zinc(II) complex is comparably less

interesting than the spectra of the other metal complexes. The

absence of any bands above 320 nm is consistent with no

MLCT or d-d transitions as expected for a d¹⁰octahedral

complex.

The circular dichroism (CD) spectra for the Fe(II), Zn(II),

Co(II), Ni(II) and Mn(II) complexes with ligand L³, and the

iron(II) imidazolimine complex R_C-[FeL⁹][ClO₄]₂in acetonitrile

3

are shown in Fig. 18. The spectrum of the opposite enantiomer for

the iron(II) complex, fac,D_Fe,S_C-[FeL³][ClO₄]₂was also collected

3

for comparison. All peaks are at comparable wavelengths and

relative intensities to the corresponding bands in the UV-vis

absorbance spectra (vide supra).

The CD spectra of the iron(II) complexes with L³, fac,K_Fe,R_C-

and fac,D_Fe,S_C-[FeL³][ClO₄]₂, contain intense features spanning

3

the whole UV-vis region and are equal and opposite for the

enantiomeric pair. The intense bisignate curves centred around

574 nm result from strong exciton coupling in the MLCT

transitions (as well as small contributions from Laporte-forbidden

d-d transitions) and conﬁrm the formation of non-racemic chiral

metal centres.^59,60The third MLCT band is observed around 360

nm.^59,60These MLCT bands are also visible in the UV-vis spectrum

as three peaks (vide supra). All bands at lower wavelengths in the

CD spectrum (220–335 nm) result from p–p* transitions within

the ligands. Comparison of the CD spectra of the structurally

similar D and K [Fe(bpy)₃]²⁺enantiomers conﬁrms the electronic

transition assignments and the conﬁguration assignments, D_Fefor

S_C-L³and K_Fefor R_C-L³.^60,61

Fig. 17 UV-vis absorbance spectra of fac,K_Fe,R_C-[FeL³][ClO₄]₂,

3

R_C-[CoL³][PF₆]₂,

K_Zn,R_C-[ZnL³][BF₄]₂,

S_C-[NiL³][ClO₄]₂,

3

R_C-[MnL³][PF₆]₂and R_C-[FeL⁹][ClO₄]₂in acetonitrile. The path

3

length was 1.0 cm and the concentration was (a) 0.03 mM and (b) 1.0 mM

showing shoulders in Ni and Co spectra corresponding to d-d and/or

MLCT transitions.

All complexes have strong bands in the region 220–320 nm, cor-

responding to p–p* transitions in the ligands. The UV-vis spectrum

of fac,K_Fe,R_C-[FeL³][ClO₄]₂has strong MLCT transition bands

3

between 320 and 700 nm. The main MLCT band has two distinct

maxima at ca 567 nm (e = 8100 M^-1cm^-1) and 520 nm (e = 5800 M^-1

cm^-1). The slight structural deviation from a perfect octahedron

gives rise to an additional third MLCT band at ca 360 nm (e =

3200 M^-1cm^-1). The bands are at comparable wavelengths and

relative intensities to the corresponding peaks assigned to MLCT

transitions in the CD spectrum (vide infra).

The CD spectrum of the R_C-[FeL⁹][ClO₄]₂complex also

3

contains intense features spanning the whole UV-vis region, with

an intense bisignate curve centred around 553 nm consistent with

the predominance of the expected K_Feisomer in solution.

The CD spectra of R_C-[CoL³][PF₆]₂, K_Zn,R_C-[ZnL³][BF₄]₂,

3

R_C-[MnL³][BF₄]₂and S_C-[NiL³][BF₄]₂also have bands in the

3

This journal is

The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10423

©

Origins of stereo- and chemical selectivity

It is evident that for the above complex cations [FeLⁿ]²⁺(n =

3

2

3–6) the product contains only one of four possible k -N,N¢

diastereomers i.e. there is essentially perfect thermodynamic

stereoselection with respect to the absolute conﬁguration at the

metal.

For the L⁷complex the fac:mer ratio was found to be 2.6 : 1.

Although cyclohexyl is rather larger than phenyl and is not capable

of p-stacking, the isomeric ratio still substantially favours the fac

conﬁguration. Hence, steric factors also appear to promote the

formation of the observed major isomer.

For the bulkier L⁸, an analysis of a simple molecular mechanics

model indicates that the tris structure is sterically untenable, i.e.

tert-butyl is too large a group to be accommodated. Consequently

the less crowded bis(tridentate) structure is favoured. Clearly also

the presence of a third ligating atom promotes this structure as a

result of the chelate effect, and arguably the greater surprise is the

formation of tris(bidentate) fac,D_Fe,R_C-[FeL⁴][ClO₄]₂described

3

above.

The single crystal X-ray crystallographic studies above show

that the observed fac diastereomers display three concurrent p-

stacks. The mutual arrangement of the rings is close to what is

regarded as optimum within this parallel-displaced regime (as

opposed to CH-p) i.e. the rings are laterally displaced with the

C atom of one ring over the centroid of the other.⁶²The key

structural parameters determined for all the above complexes

are summarised in Table 1, focusing on the various parameters

associated with these p-stacks (Fig. 19).

Fig. 18 CD spectra of fac,K_Fe,R_C-[FeL³][ClO₄]₂, R_C-[CoL³][PF₆]₂,

3

K_Zn,R_C-[ZnL³][BF₄]₂,

S_C-[NiL³][ClO₄]₂,

R_C-[MnL³][PF₆]₂

and

3

R_C-[FeL⁹][ClO₄]₂in acetonitrile. The concentration was 0.03 mM

3

and the path length was 1.0 cm.

lower wavelength region (220–335 nm) as expected from p–p*

transitions within the chiral ligands. In addition, the spectrum

of R_C-[CoL³][PF₆]₂also contains weak bands at ca 395 nm and

3

520 nm corresponding to weak MLCT and d-d transitions.⁵⁸

This conﬁrms the presence of non-racemic chiral metal centres

in solution for this paramagnetic compound, although this

cannot be taken as a measure of stereoselectivity.³³The absence

of any observable MLCT bands in the zinc(II), nickel(II) and

manganese(II) complexes does not allow for any conclusions to

be made about the chirality at these metal centres.

Fig. 19 Deﬁning the p-stacking parameters in Table 1. Distances a =

average M–N, b = arene centroid–centroid, c = carbon (Py) to centroid

(Ph), d = carbon (Ph) to centroid (Py), e = between substituted carbon

atoms, f = between carbon atoms meta to substituted carbon atoms; angles

q = between ring planes, j = average chelate bite.

Table 1 Selected structural parameters for the iminopyridine complexes fac-[MLⁿ]ⁿ⁺[M = Fe(II), Co(III), Ni(II), Co(II), Zn(II) and Mn(II)]

3

Fe(II)

Co(III)

Ni(II)

Co(II)

Zn(II)

Mn(II)

Parameter

L⁴

L⁵

L³

˚

a/A

1.98

1.97

1.94

2.10

2.15

2.17

2.27

˚

b/A

3.48–3.64

3.28–3.50

3.24–3.45

3.15–3.25

3.36–3.71

5.41–11.14

81.36

3.51–3.63

3.29–3.47

3.28–3.43

3.19–3.23

3.40–3.65

5.03–11.21

81.10

3.49–3.73

3.27–3.53

3.26–3.35

3.14–3.21

3.49–3.62

9.73–12.30

83.49

3.64–4.00

3.45–3.72

3.26–3.57

3.38–3.46

3.30–3.91

5.83–13.98

79.05

3.75–4.45

3.54–4.22

3.35–3.70

3.48–3.75

3.43–4.27

3.78–11.21

77.49

3.67–4.08

3.50–3.83

3.30–3.61

3.47–3.55

3.33–3.97

3.58–14.12

77.08

4.09–4.26

3.82–4.05

3.54–3.98

3.66–3.84

3.71–4.27

6.37–15.13

73.83

˚

c/A

˚

d/A

˚

e/A

˚

f /A

q/^◦

j/^◦

10424 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

For the Fe(II) and Co(III) structures with relatively short

M–N distances (a), the relevant arene centroid-centroid distances

˚

(b) are in the range 3.48–3.73 A, close to the calculated optimum

62

˚

distance of 3.5 A. The relevant C–centroid distances (c and d)

are in the range 3.24–3.53 A, the inter-ring distances between the

˚

substituted C atoms (e) are structurally constrained to be at rather

˚

short distances of 3.14–3.25 A, while the contacts meta to these (f )

˚

are found in the range 3.36–3.71 A. Fig. 20 shows the favourable

arrangement of the p-stacked rings in one of the Fe(II) complexes

(the Co(III) structure is similar). The nature and energy of such

non-covalent bonding interactions is an area of intense study and

debate,⁶³not least because of their signiﬁcant contribution to the

stability of important supramolecular structures such as DNA and

proteins. Calculations analysed at the SCS-MP2/aug-cc-pVTZ

level of theory by Hohenstein and Sherrill allow us to estimate

that each contact in our system contributes ca 3 kcal mol^-1.⁶²

Fig. 21 The molecular structure of fac,K_Co,R_C-[CoL³]²⁺showing the

3

p-stacking. In all cases the view is with the relevant phenyl ring uppermost

and in the plane of the page. Ring centroids are shown as black spheres.

energies calculated from density functional calculations using

ADF 2008 (version 2008.01; see experimental section).⁶⁴The

optimised structures are shown in Fig. 22.

Fig. 20 Molecular structure of fac,D_Fe,R_C-[FeL⁴]²⁺showing the offset

3

p-stacking. The view is with the relevant phenyl ring uppermost and in the

plane of the page. Ring centroids are shown as black spheres.

In comparison to the Fe(II)/Co(III) complexes, the average M–N

distances (a) in the Ni(II), Co(II), Zn(II) and Mn(II) complexes are

substantially longer and increase in that order (Table 1), resulting

in smaller chelate bite angles (j). The presence of these longer M–

N bonds moves the benzylic phenyl group away from the adjacent

pyridine ring so that the p-stacking interaction is reduced, as can

be seen in the centroid–centroid distances (b) and the two centroid–

C atom distances (c and d). The rather short C–C distances (e) (ca.

˚

3.2 A) in the Fe(II)/Co(III) complexes resulting from the structural

˚

constraints mentioned earlier are relaxed to 3.4–3.8 A for the other

metals studied, and this is accompanied by angles between the

˚

planes for the p-stacks (q) as low as 3.6 A for Zn(II). Of all these

parameters, however, we feel that the centroid–centroid distance

(b) gives the readiest assessment of the interaction since all the

other distances and angles have an impact on it.

Fig. 22 Optimised dicationic structures of (a) fac,D_Fe,S_C-[FeL³]²⁺, (b)

3

Fig. 21 shows that for the Co(II) complex there is essentially no

p-stacking, and the situation is similar for the larger metals Zn(II)

and Mn(II). This corresponds with our observation of poorer

stereoselectivity for Zn(II) than Fe(II). Most interesting, however,

is our observation via NMR spectroscopy previously (Fig. 12)

that increasing polarity improves the stereoselectivity in the Zn(II)

system. This is presumably because the hydrophobic p-stacking

is promoted, although we cannot exclude effects arising from a

difference in the overall polarity of the two isomers.

fac,K_Fe,S_C-[FeL³]²⁺, (c) mer,D_Fe,S_C-[FeL³]²⁺, (d) mer,K_Fe,S_C-[FeL³]²⁺

.

3

The observed isomer fac,D_Fe[Fig. 22 (a)] is essentially super-

imposable on the appropriate subunit of the molecular structure

of fac,D_Fe,R_C-[FeL⁴]²⁺. As expected, this isomer has the lowest

3

energy, followed by the mer,D_Feisomer [Fig. 22 (c)] ca 10 kcal

mol^-1higher. This is consistent with the observation by NMR

spectroscopy of a minor mer isomer in the R_C-[ZnL³][BF₄]₂

3

The relative energies of the relevant diastereomers of S_C-

system. The optimised mer,D_Festructure reveals extended p-

stacking involving each of the three ligands; a pyridine ring from

one ligand with phenyl rings either side, with each ring aligning

[FeL³]²⁺were investigated by computational means. The four

3

different possible (low spin) geometries were optimised and their

This journal is

The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10425

©

approximately parallel to the next [Fig. 22 (c)].‡ While we are not

aware of computed ﬁgures for the energy of such an extended p-

stacked system we estimate that it would provide substantially less

stabilisation to the structure than three offset parallel p-stacked

rings in the lower energy fac,D_Festructure.

The two optimised K_Festructures - fac,K_Fe[Fig. 22 (b)] and

mer,K_Fe[Fig. 22 (d)] - have similar energies, higher than fac,D_Feby

ca 25 kcal mol^-1. Interestingly the fac,D_Feisomer has three quite

optimal p-stacks similar to the observed fac,K_Feisomer.‡ However,

the three methyl groups point towards each other in the fac,K_Fe

isomer causing substantially unfavourable steric interactions. The

two mer structures are similarly related - the mer,K_Fecontains

additional unfavourable steric interactions between two methyl

groups pointing towards each other.

The high stability of the fac,D_Feisomer compared with the other

three possible isomers is therefore due to two factors. Firstly, the

D_Feisomers are favoured due to the lack of unfavourable steric

interactions between the methyl groups seen in the K_Festructures.

Secondly, the three sets of concurrent p-stacks seen exclusively in

the fac conﬁgurations signiﬁcantly contribute to the stability of

the fac isomer over the mer isomer.

Fig. 23 Steric effects and preferred orientations in complexes of Lⁿ.

of the stereogenic benzylic centre - only towards the reader in

Fig. 23 and therefore for the complex only in one helical sense.

The ligand L¹²is sterically similar to L³- both have only one

methyl group in the critical region - so the observation of excellent

thermodynamic fac-selectivity is not unexpected. An explanation

for the observation under kinetic control of 1 : 6.5 fac:mer for the

L

¹²complex is, however, a more complicated issue. The pathway to

mer isomers is clearly favoured kinetically over fac, but the detail

of this mechanism of formation is likely to remain unknown. In

addition, since the ketimine methyl group in L¹²may orient the

benzylic phenyl in either direction, this creates the possibility of

a number of conformational diastereomers in the mer complexes.

We do know, however, that while conversion to fac is inevitable, the

reaction is slow. We thus propose that the preorganising effect that

the ketimine methyl group has on the benzyl substituent towards

p-stacking is the cause of this kinetic stability and is the reason

why the mer isomers are observed in the kinetic product. We

made a related observation recently in tris-ketohydrazone Fe(II)

complexes, for which we were able to isolate enantiomerically

enriched complexes of achiral ligands with a measured optical

t¹₂> 1 h.³²Finally, we note that we were unable to detect Fe(II) tris

complexes of L¹³, which contains two methyl groups in the imine

region.

We noted above that during the synthesis of fac,D_Fe,R_C-

[FeL⁶][ClO₄]₂a metastable mer isomer was detected. We con-

3

sidered that this kinetic trap may arise as a result of multiple p-

stacks involving the benzyl groups, and although a number of such

possible structures could be built, these motifs were not conserved

in the converged structures.

The synthesis of [FeL²][ClO₄]₂resulted in a thermodynamic

3

product containing a mixture of the (racemic) fac and mer isomers

in a 1 : 2 ratio. The reasons behind the lower selectivity towards

the fac isomers compared with the chiral ligand complexes, e.g.

fac,K_Fe,R_C-[FeL³][ClO₄]₂, have been investigated as follows.

3

The substitution of 2-pyridinecarboxaldehyde for 2-

acetylpyridine in the standard reaction with benzylamine

(2) resulted in the formation of the expected tris ligand

[FeL¹²₃][ClO₄]₂complex. After 6 h at ambient temperature, the

1

racemic fac:mer ratio was determined by H NMR spectroscopy

to be 1 : 6.5. This ratio did not change signiﬁcantly after 2 d at

ambient temperature but on heating the solution to 80 ^◦C for 3 d

essentially complete conversion to the fac isomers was observed,

alongside small amounts of decomposition products.

Conclusions

A range of optically-pure single diastereomer diimine complexes

of Fe(II) can be made very readily using 2-pyridinecarboxaldehyde

and simple phenylethanamines as the source of chirality. Since

these chelate low spin d⁶compounds are relatively inert, stereo-

chemical interconversions are slow - some even on the synthetic

timescale - and the diastereomeric ratios can be determined by ¹H

NMR spectroscopy to be >200 : 1 in all cases.

The origins of this stereoselectivity were investigated via NMR

spectroscopy, single-crystal X-ray crystallography and DFT calcu-

lations. For a given absolute conﬁguration of the phenethanamine

unit, one fac structure is stabilised by three sets of inter-ligand

parallel-offset p-stacking interactions while at the same time

avoiding unfavourable steric interactions. The stability of the single

fac diastereomer is nevertheless ﬁnely balanced: similar complexes

with other 3d metals can be synthesised but do not necessarily

show the same degree of diastereoselectivity. While quantiﬁcation

of the stereoselection in the paramagnetic systems is not feasible,

the molecular structures of all compounds in the series Mn–Ni,

Zn allow a qualitative analysis. The key issue is the M–N bond

length, which for Fe(II) places the ligand substituents in essentially

Fig. 23 depicts our reasoning to account for the unusual

substituent steric effects and accompanying p-stacking observed in

these compounds. In complexes of L², the benzyl group is relatively

free to rotate about the N–C bond although, as noted by Busch

in related hydrazone complexes, the particular rotamer shown for

L²in Fig. 23 is untenable.⁶⁵For L³the rotamer observed in the

molecular structures of the tris-complexes is as shown in Fig.

23, with the H atom pointing towards the sterically encumbered

region about the Fe atom, and the Ph and Me substituent bonds

bifurcated by the imine C–H. This orients the phenyl group rather

ideally for p-stacking (vide supra), but - as a result of the presence

‡ (letters refer to Fig. 19)(a) fac,D_Fe,S_C-[FeL³]²⁺a = 1.94 A, b = 3.83–

˚

3

˚

3.97 A, c = 3.67–3.84 A, d =_◦3.48–3.61 A, e = 3.28–3.37 A, f = 3.88–4.10 A,

◦

q = 15.64–19.71 , j = 81.81 .(b) fac,K_Fe,S_C-[FeL³]²⁺a = 1.97 A, b = 3.58–

˚

3

˚

3.69 A, c = 3.44–3.77 A, d =_◦3.42–3.70 A, e = 3.19–3.40 A, f = 3.68–4.08 A,

◦

q = 11.84–14.83 , j = 82.07 .(c) mer,D_Fe,S_C-[FeL³]²⁺a = 1.95 A, b = 3.45,

˚

3

◦

3

2+

˚

3.56 A, q = 5.70, 9.46 , j = 81.84 .(d) mer,KFe,SC-[FeL ] a = 1.96 A,

3

◦

˚

b = 3.38, 3.57 A, q = 6.68, 7.24 , j = 82.56 .

10426 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

ideal positions for the p-stacking. For diamagnetic Zn(II) the

larger radius leads to very weak p-stacking and poorer measured

stereoselectivity, although it is proposed that the improvement in

diastereoselection with increased solvent polarity may be due to

the promotion of this hydrophobic interaction.

In the Fe(II) system, the complex formed with L⁷demonstrates

that steric effects play an important role alongside the p-stacking,

and the bis(tridentate) structure formed with L⁸reminds us that

on the basis of the chelate effect the observation of only the

tris(bidentate) structure for the Fe(II) complex of L⁴is unexpected.

Evidently the energetic beneﬁt of the three p-stacks (ca 10 kcal

mol^-1) goes at least some way to offsetting the entropic penalty of

retaining the third ligand.

Imidazole-based compounds give similar structures to the

pyridines in the solid state and indeed substantial diastereomeric

excess in solution, but the complexes are paramagnetic and the

actual stereoselectivity is unknown.

The traditional approaches to the isomer problem of

tris(bidentate) octahedral structures more commonly use inert

metals such as Ru(II) so that racemisation or epimerization rates

are negligible. In the current appproach we rely on the accessibility

of stereochemical interconversions so that the system is not

trapped in some kinetic product mixture (e.g. the L⁶system)

while at the same time ensuring that one isomer is far more

thermodynamically stable than the others. As a result only one

isomer is present in measurable quantities.

performed by Warwick Analytical Services, Coventry, UK and

MEDAC Ltd, Surrey, UK.

UV-vis absorbance spectra were recorded using a Jasco V-

660 spectrometer. Measurements were collected in a 1 cm path-

length quartz cuvette and the standard parameters used were:

bandwidth 1 nm, response time 1 s, wavelength scan range 200–

800 nm, data pitch 0.2 nm, scanning speed 100 nm min^-1and

accumulation 2. CD spectra were measured on a Jasco J-815

spectrometer, which was calibrated conventionally using 0.060%

ACS for intensity and a holmium ﬁlter for wavelength, and

also against our recently introduced Na[Co(EDDS)] system.⁶⁶

Measurements were collected using a 1 cm path-length quartz

cuvette and the standard parameters used were: bandwidth 1 nm,

response time 1 s, wavelength scan range 200–800 nm, data pitch

0.2 nm, scanning speed 100 nm min^-1and accumulation 4.

The crystal data were collected using a Siemens SMART CCD

˚

single crystal diffractometer using a Mo-Ka (l = 0.71073 A) or

˚

Cu-Ka (l = 1.54184 A) radiation source. Structures were solved

by direct methods using SHELX (TREF)^67,68with additional

light atoms found by Fourier methods. Crystal reﬁnement was

performed using SHELX97.⁶⁸

Optical rotation measurements were performed on a Perkin

Elmer Polarimeter 341 by Warwick Analytical Services, Coventry,

UK. In all cases the following parameters were used: solvent

methanol, temperature 20 ^◦C, pathlength 100 mm, wavelength

589 nm.

The simplicity and ready accessibility of the stereochemically

pure system described here is allowing us to synthesise further

derivatives and introduce functionality, which we will report in

due course.

Magnetic measurements were performed on a Quantum Design

MPMS 5S SQuID magnetometer, using an external ﬁeld of 1000 G

in the temperature range 2–400 K. Samples were placed in gelatine

capsules. Diamagnetic corrections were calculated using Pascal’s

constants.⁶⁹

Density functional optimisations were carried out using the

Amsterdam Density Functional program (version 2008.01).⁶⁴

Starting points for geometry optimisations were taken from crys-

tallographic data where available, and where unavailable, starting

structures were created from existing crystallographic fragments.

Solution structures were optimised relative to acetonitrile (vide

infra) using a triple-z plus polarisation basis set (TZP) on all atoms

with the OPBE functional and Grimme’s empirical correction

for dispersion.⁷⁰Small frozen cores⁷¹were used throughout.

Calculations used integration level 5 (as deﬁned by ADF) with

Experimental

General considerations

All solvents and chemicals purchased from commercial sources

(Sigma-Aldrich, Acros, Fisher Scientiﬁc or Alfa Aesar) were

used without further puriﬁcation. Deuterated solvents were pur-

chased from Sigma-Aldrich and Cambridge Isotope Laboratories.

Sodium hydride dispersion in mineral oil was placed in a Schlenk

vessel under an inert atmosphere and washed three times with

diethyl ether to remove the oil. The sodium hydride powder was

then dried and stored in the dry box.

˚

convergence criteria of e = 0.0001 a.u., rad = 0.005 A and grad =

0.001 a.u. A^-1for the total binding energy, Cartesian displacement

˚

Where appropriate, reactions were carried out under argon

using a dual manifold argon/vacuum line and standard Schlenk

techniques or MBraun dry box. THF was pre-dried over sodium

wire and then heated to reﬂux for 3 d under dinitrogen over

potassium and degassed before use. Dried THF was stored in

a glass ampoule under argon. All glassware and cannulae were

stored in an oven at > 375 K.

and energy gradient respectively. Acetonitrile solvent effects were

included based on the conductor-like screening model (COSMO)

implemented in ADF.⁷²Non-bonded radii used were H = 1.350 A,

˚

C = 1.700 A, N = 1.608 A, Fe = 1.858 A. A dielectric constant

˚

of 37.5 (acetonitrile) and an outer cavity radius of 2.76 A were

further used to parameterise the COSMO solvation cavity.

NMR spectra were recorded on Bruker Spectrospin

300/400/500 MHz spectrometers. Routine NMR assignments

Synthesis

1

were conﬁrmed by H–¹H (COSY) and ¹³C–¹H (HMQC) corre-

(R)-2-Phenylglycinol (4).

Method 1. (Adapted from prep for (S)-tert-Leucinol)³⁵

A dry round-bottom Schlenk charged with sodium borohydride

(31.21 g, 0.83 mol) and (R)-2-phenylglycine (50.00 g, 0.33 mol)

was ﬂushed with argon and dry THF (400 mL) was added. The

ﬂask was cooled to 0 ^◦C using an ice-water bath before iodine

(83.75 g, 0.33 mol) in dry THF (150 mL) was added dropwise

lation experiments where necessary. The spectra were internally

referenced using the residual protio solvent (CDCl₃, CD₃CN

etc.) resonance relative to tetramethylsilane (d = 0 ppm). ESI

mass spectra were recorded on Bruker Esquire 2000 and Bruker

MicroTOF spectrometers. Infra-Red spectra were measured using

a Perkin-Elmer FTIR spectrometer. Elemental analyses were

This journal is

The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10427

©

to the solution. A reﬂux condenser and nitrogen bubbler were

ﬁtted and the reaction was stirred at ambient temperature for 2 h,

before heating to reﬂux (80 ^◦C) overnight. Methanol (400 mL) was

added slowly until the solution became clear. All solvents were then

removed under reduced pressure to leave a white paste, which was

dissolved in aqueous 20% potassium hydroxide solution (600 mL)

and stirred at ambient temperature overnight. The product was

extracted into dichloromethane (5 ¥ 250 mL), dried over sodium

sulfate and the solvent removed under reduced pressure to leave

the crude product. The product was recrystallised from hot toluene

to yield a white crystalline solid. Yield = 23.33 g, 0.17 mol, 52%.

MS (ESI) m/z 135.0 [M–NH₂]⁺.

IR v cm^-13028 w, 2888 m, 1603 w, 1493 m, 1453 m, 1355 w, 1194

m, 1111 s, 968 m, 844 m, 758/700 s.

Elemental analysis found (calculated for C₉H₁₃NO)% C 71.01

(71.49), H 8.65 (8.67), N 8.92 (9.26).

Optical rotation -34.04^◦(5.963 g/100 mL).

2-(Benzyloxy)-1-phenylethanamine (6)³⁸. (R)-2-Phenylglycinol

(4) (1.00 g, 7.3 mmol, 1.00 eq.) was dissolved in dry THF (15 mL)

and was added dropwise to a stirred suspension of sodium hydride

(0.36 g, 14.9 mmol, 2.05 eq.) in dry THF (10 mL). The solution

was stirred for 1 h at ambient temperature under partial vacuum.

Benzyl bromide (0.91 mL, 7.7 mmol, 1.05 eq.) was added dropwise

over 10 min and the solution was stirred for 1 h at ambient

temperature under argon. At this point, the solution was heated to

reﬂux (65 ^◦C) under partial vacuum for a further 3 h before cooling

to ambient temperature, followed by the addition of brine (40 mL).

The product was extracted into diethyl ether (4 ¥ 60 mL), dried

over sodium sulfate and the solvent was removed under reduced

pressure to leave a yellow oil (crude yield = 1.42 g). This crude

product was puriﬁed by two Ku¨gelrohr distillations, the ﬁrst at

125 ^◦C (under high vacuum) to remove unreacted benzyl bromide

and (R)-2-phenylglycinol, and the second to give the product, a

slightly yellow liquid (bp 165 ^◦C under high vacuum). Puriﬁed

yield = 0.93 g, 4.1 mmol, 56%.

Method 2³⁴. Lithium aluminium hydride (4.90 g, 129 mmol,

1.95 eq.) was suspended in dry THF (200 mL) under argon at

0

^◦C. Solid (R)-2-phenylglycine (10.00 g, 66.15 mmol, 1.00 eq.)

was added in small portions. The mix_◦ture was stirred at 0 ^◦C for

1 h, then slowly heated to reﬂux (80 C) overnight. A saturated

potassium carbonate solution (75 mL) was added very slowly to

the mixture, which was cooled in an ice/water bath. The mixture

was ﬁltered and the solvents were removed from the ﬁltrate under

reduced pressure. The crude yellow solid was recrystallised from

hot toluene to yield a white crystalline solid. Yield 5.74 g, 42 mmol,

63%.

¹H NMR (400 MHz, 298 K, CDCl₃) d_H7.30–7.18 (5H, m, Ph),

3

2

3.97 (1H, dd, J_HH= 4.0 Hz, 8.5 Hz, CH), 3.66 (¹H, dd, J_HH

=

11.0 Hz, ³J_HH= 4.0 Hz, CH₂), 3.48 (1H, dd, ²J_HH= 11.0 Hz, ³J_HH

¹H NMR (300 MHz, 298 K, CDCl₃) d_H7.33–7.16 (10H, m, Ph),

8.5 Hz, CH₂), 2.05 (2H, s, NH₂).

1

¹³C{ H} NMR (100 MHz, 298 K, CDCl₃) d_C142.7 (Ph), 128.7

(Ph), 127.5 (Ph), 126.5 (Ph), 68.0 (CH₂), 57.4 (CH).

MS (ESI) m/z 138.0 [M+H]⁺, 121.0 [M–NH₂]⁺.

IR v cm^-12835 m, 1604 m, 1497 m, 1453 m, 1361 w, 1197 w,

1077 m, 1047 m, 978 m, 882 m, 755 s, 700 s.

3

4.49 (2H, s, CH₂Ph), 4.18 (1H, dd, J_HH= 4.0 Hz, 9.0 Hz, CH),

2

3

3.54 (1H, dd, J_HH= 9.0 Hz, J_HH= 4.0 Hz, CH₂), 3.38 (1H, t,

²J_HH/³J_HH= 9.0 Hz, CH₂), 1.69 (2H, s, NH₂).

1

¹³C{ H} NMR (75 MHz, 298 K, CDCl₃) d_C142.5 (Ph), 138.2

(Ph), 128.5 (Ph), 127.8 (Ph), 127.7 (Ph), 127.6 (Ph), 127.4 (Ph),

126.9 (Ph), 76.7 (CH₂), 73.3 (CH₂Ph), 55.6 (CH).

MS (ESI) m/z 228.0 [M+H]⁺, 250.0 [M+Na]⁺.

IR v cm^-13029 w, 2857 w, 1603 w, 1494 m, 1453 m, 1355 w, 1206

w, 1090 s, 1027 m, 843 m, 735 s, 697 s.

Elemental analysis found (calculated for C₈H₁₁NO)% C 69.79

(70.04), H 8.08 (8.08), N 10.18 (10.21).

Melting point 76–78 ^◦C (Lit. mp 76–79 ^◦C).⁷³

Optical rotation -25.99^◦(6.619 g/100 mL) [Lit. = -25.8^◦(6.6

g/100 mL)].³⁶

Elemental analysis found (calculated for C₁₅H₁₇NO)% C 78.88

(79.26), H 7.58 (7.54), N 6.09 (6.16).

2-Methoxy-1-phenylethanamine (5)³⁷. (R)-2-Phenylglycinol

(4) (2.00 g, 14.6 mmol, 1.00 eq.) was dissolved in dry THF

(20 mL) and was added dropwise to a stirred suspension of

sodium hydride (0.72 g, 30.0 mmol, 2.05 eq.) in dry THF (10 mL).

The solution was stirred for 1 h at ambient temperature under

partial vacuum. Iodomethane (0.96 mL, 15.4 mmol, 1.05 eq.) was

added dropwise and the solution was stirred for 1 h at ambient

temperature under argon. At this point, the solution was heated

to reﬂux (65 ^◦C) under partial vacuum for a further 2 h before

cooling to ambient temperature, followed by the addition of

brine (40 mL). The product was extracted into diethyl ether (4 ¥

60 mL), dried over sodium sulfate and the solvent was removed

under reduced pressure to leave a yellow oil (crude yield = 1.80 g).

This crude product was puriﬁed by Ku¨gelrohr distillation to give

a clear oil (bp 70 ^◦C under high vacuum). Puriﬁed yield = 1.38 g,

9.1 mmol, 63%.

Optical rotation -18.87^◦(3.015 g/100 mL).

N-Methyl-1H-imidazole-2-carbaldehyde (9)³⁹. Methylimida-

zole (1.0 g, 12.2 mmol) was dissolved in dry THF (20 mL). n-

Butyllithium (4.5 mL, 2.5 M in hexanes) was added dropwise at

-78 ^◦C. The resulting solution was stirred for 1 h, after which

dry DMF (2.8 mL, 36.6 mmol) was added and the solution was

stirred for a further 3 h. The solution was then warmed to ambient

temperature and stirred overnight. The reaction was quenched

by the addition of distilled water (20 mL) and extracted into

chloroform (3 ¥ 100 mL). The organic layers were combined, dried

over sodium sulfate, ﬁltered and the solvent was removed under

reduced pressure to give a yellow oil. On cooling to -30 ^◦C a white

solid formed and was isolated by ﬁltration and washed with diethyl

ether (30 mL) followed by pentane (30 mL) to give the product as

a waxy white solid. Yield 1.12 g, 10.17 mmol, 83%.

¹H NMR (300 MHz, 298 K, CDCl₃) d_H7.33–7.16 (5H, m, Ph),

3

2

4.12 (1H, dd, J_HH= 4.0 Hz, 9.0 Hz, CH), 3.43 (1H, dd, J_HH

=

¹H NMR (400 MHz, 298 K, CDCl₃) d_H9.79 (1H, s, HC O),

7.24 (1H, s, Imz), 7.08 (1H, s, Imz), 3.99 (3H, s, Me).

8.5 Hz, ³J_HH= 4.0 Hz, CH₂), 3.32–3.26 (4H, m, CH₂, CH₃), 1.67

1

(2H, s, NH₂).

¹³C{ H} NMR (75 MHz, 298 K, CDCl₃) d_C182.01 (C O),

1

¹³C{ H} NMR (75 MHz, 298 K, CDCl₃) d_C142.6 (Ph), 128.4

143.62 (Imz), 131.36 (Imz), 127.33 (Imz), 34.86 (Me).

(Ph), 127.4 (Ph), 126.8 (Ph), 79.0 (CH₂), 58.9 (CH₃), 55.4 (CH).

MS (ESI) m/z 111.3 [M+H]⁺, 133.2 [M+Na]⁺.

10428 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

N-tert-Butylimidazole⁷⁴. Glyoxal (9.6 g of glyoxal dimer de-

hydrate, 0.14 mol), tert-butylamine (10.0 g, 0.14 mol), methanol

(150 mL) and distilled water (25 mL) were heated to 70 ^◦C, at which

point formaldehyde (12 mL, 37% aqueous solution, 0.14 mol) was

added followed by the dropwise addition of ammonia (10 mL, 28

wt%, 0.14 mol). The resulting solution was stirred for a further 6

h at 70 ^◦C. The solution was cooled to ambient temperature and

the solvents were removed under reduced pressure to give a dark

yellow oil, which was extracted into dichloromethane (150 mL)

and was washed with distilled water (3 ¥ 150 mL). The organic

layer was dried over sodium sulfate and the solvent was removed

under reduced pressure to give a yellow oil. The product was

puriﬁed by Ku¨gelrohr distillation to give a colourless oil (bp 70 ^◦C

under high vacuum). Yield 8.2 g, 66.0 mmol, 47%.

mer isomer), 8.43 (0.2H, d, ³J_HH= 9 Hz, Py fac isomer), 8.38 (1H,

3 3

d, J_HH= 7.5 Hz, Py mer isomer), 8.33 (0.2H, t, J_HH= 9 Hz, Py

fac isomer), 8.25 (1H, t, ³J_HH= 9 Hz, Py mer isomer), 8.06 (1H, t,

³J_HH= 9 Hz, Py mer isomer), 7.94 (3H, m, Py mer isomer, 2 ¥ OH

mer isomer), 7.73 (0.2H, t, ³J_HH= 8.5 Hz, Py fac isomer), 7.65 (2H,

m, 2 ¥ Py mer isomer), 7.56 (2H, m, 2 ¥ Py mer isomer), 7.48 (2H,

m, 2 ¥ Py mer isomer), 7.45 (0.2H, s, OH fac isomer), 7.43 (0.2H,

3

d, J_HH= 7.5 Hz, Py fac isomer), 7.35 (1H, s, OH mer isomer),

6.77 (2H, d, ³J_HH= 9.5 Hz, Ph mer isomer), 6.68 (0.4H, d, ³J_HH

=

9.5 Hz, Ph fac isomer), 6.58 (6H, m, Ph mer isomer), 6.42 (2H, d,

³J_HH= 9.5 Hz, Ph mer isomer), 6.15 (2H, d, ³J_HH= 9.5 Hz, Ph mer

isomer), 5.54 (0.4H, d, ³J_HH= 9.5 Hz, Ph fac isomer).

1

¹³C { H} NMR (100 MHz, 298 K, CD₃CN) d_C170.8 (C N),

170.7 (C N), 170.6 (C N), 158.8 (Py/Ph), 158.3 (Py/Ph), 158.0

(Py/Ph), 157.5 (Py/Ph), 157.5 (Py/Ph), 157.4 (Py/Ph), 156.7

(Py/Ph), 156.6 (Py), 155.9 (Py), 155.9 (Py), 155.4 (Py/Ph), 155.4

(Py/Ph), 138.6 (Py), 138.3 (Py), 138.2 (Py), 131.0 (Py), 129.7 (Py),

129.3 (Py), 128.6 (Py), 128.2 (Py), 128.1 (Py), 123.7 (Ph), 123.2

(Ph), 123.2 (Ph), 116.0 (Ph), 115.6 (Ph), 115.2 (Ph). All labelled

peaks are the mer isomer. The fac isomer is too weak to distinguish.

MS (ESI) m/z 199.2 [L+H]⁺.

¹H NMR (400 MHz, 298 K, CDCl₃) d_H7.42 (1H, s, Imz), 6.90

(2H, m, Imz), 1.42 (9H, s, ^tBu).

1

¹³C{ H} NMR (75 MHz, 298 K, CDCl₃) d_C134.0 (Imz), 128.7

(Imz), 116.1 (Imz), 54.4 (^tBu), 30.3 (^tBu).

MS (ESI) m/z 125.2 [M+H]⁺, 147.3 [M+Na]⁺.

N-tert-Butyl-1H-imidazole-2-carbaldehyde (11). This com-

pound has previously been synthesised by an alternative method.⁷⁵

To a stirred solution of 1-tert-butylimidazole (4.0 g, 32.0 mmol)

in THF (20 mL) n-butyllithium (20 mL, 2.5 M in hexanes) was

added dropwise at -78 ^◦C. The resulting solution was allowed to

warm to ambient temperature and stirred for 3 h. Dry DMF (3.0

mL, 38.7 mmol) was added and the resulting yellow solution was

stirred overnight. The reaction was quenched by the addition of

water (20 mL) and extracted into chloroform (3 ¥ 100 mL). The

organic layers were collected and dried over sodium sulfate, ﬁltered

and the solvents removed under reduced pressure to give a yellow

oil. The product was puriﬁed by Ku¨gelrohr distillation to give a

colourless oil (70 ^◦C under high vacuum). Yield 3.2 g, 20.0 mmol,

63%.

IR n cm^-1: 3327 w, 1592 m, 1507 m, 1473 m, 1448 m, 1354 w,

1275 m, 1226 m, 1042 s, 834 m, 767 s.

Elemental analysis found (calculated for C₃₆H₃₀O₁₁Cl₂FeN₆)%

C 50.13 (50.90), H 3.61 (3.56), N 9.51 (9.89).

1

2

[FeL²][ClO₄]₂· CH₃CN. Yield = 26%. Ratio is 1 : 2 fac:mer.

3

¹H NMR (400 MHz, 298 K, CD₃CN) d_H9.08 (1H, s, CH

N

mer), 8.49 (3H, s, CH N fac), 8.38 (1H, s, CH N mer), 8.30 (2H,

m, Py mer), 8.22 (1H, s, CH N, mer), 8.17 (1H, t, ³J_HH= 7.0 Hz,

3

Py mer), 8.02 (1H, t, J_HH= 7.0 Hz, Py mer), 7.98-6.80 (50H, m,

2

Py/Ph mer and fac), 5.48 (3H, d, J_HH= 12.5 Hz, CH₂fac), 5.38

(1H, d, ²J_HH= 12.5 Hz, CH₂mer), 5.06 (3H, d, ²J_HH= 12.5 Hz, CH₂

2

fac), 4.94 (1H, d, J_HH= 12.5 Hz, CH₂mer), 4.81 (1H, d, J_HH

=

2

¹H NMR (300 MHz, 298 K, CDCl₃) d_H9.77 (1H, s, HC O),

7.34 (1H, d, ³J_HH= 1.0 Hz, Imz), 7.23 (1H, d, ³J_HH= 1.0 Hz, Imz),

1.65 (9H, s, ^tBu).

12.5 Hz, CH₂mer), 4.49 (1H, d, J_HH= 12.5 Hz, CH₂mer), 4.38

2

(1H, d, J_HH= 12.5 Hz, CH₂mer), 4.07 (1H, d, J_HH= 12.5 Hz,

CH₂mer).

¹³C{ H} NMR (100 MHz, 298 K, CD₃CN) d_C174.0 (C N),

1

¹³C{ H} NMR (75 MHz, 298 K, CDCl₃) d_C181.2 (C O), 144.7

(Imz), 130.5 (Imz), 124.2 (Imz), 58.7 (^tBu), 29.6 (^tBu).

153.7 (Py/Ph), 138.2 (Py/Ph), 128.9 (Py/Ph), 128.9 (Py/Ph),

128.7 (Py/Ph), 128.6 (Py/Ph), 128.1 (Py/Ph), 127.9 (Py/Ph), 126.9

(Py/Ph), 65.2 (CH₂), 64.4 (CH₂), 63.6 (CH₂), 62.6 (CH₂). Many

peaks due to the same carbon in the fac/mer isomers overlap.

MS (ESI) m/z 153.2 [M+H]⁺, 175.2 [M+Na]⁺.

General procedure for [FeLⁿ][ClO₄]₂complexes (n = 1–9,11). 2-

3

MS (ESI) m/z 322.1 [FeL²]²⁺.

Pyridinecarboxaldehyde (3.0 eq.) and iron(II) perchlorate hexahy-

drate (1.0 eq.) were dissolved in acetonitrile to form a red solution

containing the iron(II) tris(2-pyridinecarboxaldehyde) complex.

The appropriate amine (1–7) (3.0 eq.) was added and immediately

the solution turned purple. This was stirred overnight before

ethyl acetate/diethyl ether was added dropwise until signs of

crystallisation. The purple crystals were ﬁltered and dried in vacuo.

3

IR v cm^-11614 w, 1590 w, 1556 w, 1496 m, 1471 m, 1444 m, 1297

w, 1239 w, 1072 s, 741 s, 698 s, 622 s.

Elemental analysis found (calculated for C₄₀H_37.5Cl₂FeN_6.5O₈)%

C 55.22 (55.60), H 4.39 (4.37), N 10.59 (10.54).

fac,K_Fe,R_C-[FeL³][ClO₄]₂. Yield = 64%.

3

¹H NMR (400 MHz, 298 K, CD₃CN) d_H8.75 (3H, s, HC N),

If necessary, the product [FeLⁿ][ClO₄]₂, was recrystallised from

3

7.83 (3H, t, J_HH= 7.0 Hz, Py), 7.47 (3H, d, J_HH= 7.0 Hz, Py),

acetonitrile and ethyl acetate. Yields are unoptimised -analysis of

the crude reaction mixtures indicated that reactions were essen-

tially complete. Procedures involving imidazole-2-carbaldehydes

were conducted similarly.

3

7.33 (3H, t, J_HH= 7.0 Hz, Py), 7.12 (3H, t, J_HH= 8.5 Hz, Ph),

3

6.99 (6H, t, J_HH= 8.5 Hz, Ph), 6.74 (3H, d, J_HH= 7.0 Hz, Py),

6.59 (6H, d, ³J_HH= 8.5 Hz, Ph), 5.36 (3H, quartet, ³J_HH= 4.0 Hz,

CH), 1.98 (9H, d, ³J_HH= 4.0 Hz, CH₃).

[FeL¹][ClO₄]₂. Yield = 0.67 g, 79%. Ratio is 1 : 15 fac:mer.

¹³C{ H} NMR (100 MHz, 298 K, CD₃CN) d_C170.8 (C N),

1

3

¹H NMR (400 MHz, 298 K, CD₃CN) d_H9.30 (1H, s, HC

N

158.2 (Py), 153.6 (Py), 139.5 (Ph), 138.3 (Py), 129.1 (Py), 129.0

(Ph), 127.8 (Py), 127.3 (Ph), 124.4 (Ph), 69.1 (CH), 25.4 (CH₃).

mer isomer), 9.02 (1H, s, CH N mer isomer), 8.88 (0.2H, s,

HC N fac isomer), 8.80 (1H, s, HC N mer isomer), 8.54 (1H,

MS (ESI) m/z 343.14 [FeL³]²⁺.

3

d, J_HH= 9 Hz, Py mer isomer), 8.48 (1H, d, J_HH= 7.5 Hz, Py

IR v cm^-11706 m, 1614 w, 1444 m, 1386 w, 1070 s, 757 s, 701 s.

This journal is The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10429

©

2

Elemental analysis found (calculated for C₄₂H₄₂Cl₂FeN₆O₈)% C

56.81 (56.96), H 5.02 (4.78), N 9.08 (9.49).

The fac,D_Fe,S_Cisomer was prepared similarly.

CH), 4.79 (3H, d, J_HH= 11.0 Hz, CH₂Ph), 4.76 (3H, d, J_HH=

2

11.0 Hz, CH₂Ph), 4.14 (3H, t, J_HH/³J_HH= 10.0 Hz, CH₂), 3.28

(3H, dd, ³J_HH= 9.5 Hz, ³J_HH= 3.0 Hz, CH₂).

¹³C{ H} NMR (100 MHz, 298 K, CD₃CN) d_C171.0 (C N),

1

2

1

4

fac,D_Fe,R_C-[FeL⁴][ClO₄]₂· CH₃CN· H₂O. Yield = 47%.

3

158.5 (Ph/Py), 153.4 (Py), 138.3 (Ph/Py), 137.5 (Ph/Py), 134.6

(Ph/Py), 128.9 (Ph/Py), 128.8 (Ph/Py), 128.6 (Ph/Py), 128.4

(Ph/Py), 128.2 (Ph/Py), 127.8 (Ph/Py), 127.4 (Ph/Py), 125.8 (Ph),

73.5 (CH₂Ph), 72.1 (CHCH₂), 71.1 (CH).

¹H NMR (400 MHz, 298 K, CD₃CN) d_H8.97 (3H, s, HC N),

3

7.73 (3H, t, J_HH= 7.0 Hz, Py), 7.42 (3H, d, J_HH= 7.0 Hz, Py),

3

7.21 (3H, t, J_HH= 7.0 Hz, Py), 7.11 (3H, t, J_HH= 8.5 Hz, Ph),

7.00 (6H, t, ³J_HH= 8.5 Hz, Ph), 6.82 (9H, m, Py/Ph), 5.77 (3H, dd,

³J_HH= 11.0 Hz, ³J_HH= 1.5 Hz, CH), 4.26 (3H, m, CH₂), 4.10 (3H,

m, OH), 3.92 (3H, m, CH₂).

MS (ESI) m/z 502.20 [FeL⁶]²⁺.

3

IR v cm^-11613 w, 1454 m, 1362 w, 1241 w, 1084 s, 758 s, 701 s.

Elemental analysis found (calculated for C₆₅H₆₃Cl₂FeN₇O₁₁)%

C 62.85 (62.71), H 4.99 (5.10), N 7.67 (7.88).

¹³C{ H} NMR (100 MHz, 298 K, CD₃CN) d_C171.4 (C N),

1

158.6 (Py), 153.4 (Py), 138.2 (Py), 135.1 (Ph), 128.7 (Py), 128.5

(Ph), 127.6 (Ph), 127.4 (Py), 125.7 (Ph), 73.3 (CH), 64.6 (CH₂).

K_Fe,R_C-[FeL⁷][ClO₄]₂·H₂O. Yield = 24%. Ratio is 2.6 : 1

3

MS (ESI) m/z 607.10 [(FeL⁴)(ClO₄)]⁺.

fac:mer.

2

IR v cm^-13463 w, 1614 w, 1473 m, 1452 m, 1241 w, 1080 s, 757

¹H NMR (400 MHz, 298 K, CD₃CN) d_H9.32 (1H, s, HC N,

mer), 9.25 (1H, s, HC N, mer), 9.00 (1H, s, HC N, mer), 8.91

(3H, s, HC N, fac), 8.43–7.45 (21H, m, Py mer/fac), 6.85 (3H,

d, J_HH= 6.0 Hz, Py, fac), 3.89 (3H, m, CH, fac), 3.59 (1H, m,

CH mer), 3.45 (1H, m, CH mer), 3.27 (1H, m, CH mer), 1.66–0.54

s, 701 s.

Elemental

analysis

found

(calculated

for

3

C₄₃H₄₄Cl₂FeN_6.5O_11.25)% C 53.30 (53.88), H 4.57 (4.63), N

9.34 (9.50).

Crystallography (CCDC refcode COWFAW³¹):

(84H, m, CH₃/Cy, mer/fac).

1

2

1

4

fac,D_Fe,R_C-[FeL⁴][ClO₄]₂· CH₃CN· H₂O

¹³C{ H} NMR (100 MHz, 298 K, CD₃CN) d_C170.3 (C N),

1

3

C₄₃H₄₄Cl₂FeN_6.5O_11.25

,

M_r

=

958.60, monoclinic, P2₁, pink

159.3 (Py), 154.6 (Py), 138.9 (Py), 129.8 (Py), 128.3 (Py), 69.2

(CH), 40.5 (CH₃), 30.8 (Cy), 26.3 (Cy), 25.6 (Cy), 23.4 (Cy).

plate 0.30 ¥ 0.20 ¥ 0.02 mm, a = 18.5532(6), _◦b = 12.1579(2), c =

◦

3

1

˚

19.1141(6) A, a = 90 , b = 90.7860(10) , g = 90 , U = 4311.1(2) A ,

All peaks on ¹³C { H} NMR are due to the major isomer (fac,K).

˚

Z = 4, T = 120(2) K, radiation Mo-Ka (l = 0.71073 A), 44 867

The peaks for the mer,K isomer are too small to be observed.

total reﬂections, 15 037 unique (R_int= 0.1029), R₁= 0.0557 (obs.

data), wR₂= 0.1209 (all data), GooF 1.012, Flack 0.004(16).

MS (ESI) m/z 352.21 [FeL⁷]²⁺.

3

IR v cm^-11731 w, 1612 w, 1445 m, 1395 w, 1241 m, 1077 s, 766 m.

Elemental analysis found (calculated for C₄₂H₆₂Cl₂FeN₆O₉)% C

55.03 (54.73), H 6.67 (6.78), N 8.89 (9.12).

fac,D_Fe,R_C-[FeL⁵][ClO₄]₂·2CH₃CN. Yield = 75%.

3

¹H NMR (400 MHz, 298 K, CD₃CN) d_H8.89 (3H, s, HC N),

7.69 (3H, td, ³J_HH= 7.5 Hz, ⁴J_HH= 1.5 Hz, Py), 7.38 (3H, d, ³J_HH

=

mer,S_C-[FeL⁸][ClO₄]₂. Despite the stoichiometry used, analy-

2

7.5 Hz, Py), 7.17 (3H, m, Py), 7.09 (3H, t, ³J_HH= 7.5 Hz, Ph), 6.98

sis indicated that only two ligands were coordinated to each iron(II)

centre. Yield = 0.28 g, 0.42 mmol, 63%.

3

(6H, t, J_HH= 7.5 Hz, Ph), 6.77 (3H, d, J_HH= 5.5 Hz, Py), 6.71

(6H, d, ³J_HH= 7.5 Hz, Ph), 5.81 (3H, dd, ³J_HH= 3.5 Hz, 10.5 Hz,

MS (ESI) m/z 207.15 [L⁸+H]⁺.

CH), 4.23 (3H, t, J_HH/³J_HH= 10.5 Hz, CH₂), 3.67 (9H, s, CH₃),

IR v cm^-13225 w, 1600 m, 1481 m, 1370 m, 1225 m, 1046 s,

786 s.

2

3.55 (3H, dd,²J_HH= 10.5 Hz, ³J_HH= 3.5 Hz, CH₂).

¹³C{ H} NMR (100 MHz, 298 K, CD₃CN) d_C171.1 (C N),

Elemental analysis found (calculated for C₂₄H₃₆Cl₂FeN₄O₁₀)%

C 43.48 (43.20), H 5.46 (5.44), N 8.59 (8.40).

Crystallography (CCDC refcode COWFEA³¹):

mer,S_C-[FeL⁸][ClO₄]₂C₂₄H₃₆Cl₂FeN₄O₁₀, M_r= 667.32, mono-

1

158.5 (Py), 153.5 (Py), 138.3 (Py), 134.7 (Ph), 129.0 (Py), 128.9

(Ph), 127.8 (Ph), 127.5 (Py), 125.6 (Ph), 74.0 (CH₂), 70.9 (CH),

58.2 (CH₃).

2

MS (ESI) m/z 388.16 [FeL⁵]²⁺, 635.14 [(FeL⁵)(ClO₄)]⁺.

clinic, P2₁, red block 0.16 ¥ 0.15 ¥_◦0.13 mm, a = 9.5751(2), b _◦=

3

2

◦

IR v cm^-11613 w, 1473 m, 1451 m, 1236 w, 1078 s, 758 s, 699 s.

Elemental analysis found (calculated for C₄₉H₅₄Cl₂FeN₈O₁₁)%

C 55.29 (55.64), H 4.89 (5.15), N 10.12 (10.59).

Crystallography:

15.1762(3), c = 10.7704(2) A, a = 90 , b = 110.5290(10) , g = 90 ,

˚

3

˚

U = 1465.69(5) A , Z = 2, T = 120(2) K, radiation Mo-Ka (l =

0.71073 A), 16 635 total reﬂections, 6382 unique (R_int= 0.0313),

R₁= 0.0269 (obs. data), wR₂= 0.0654 (all data), GooF 1.046, Flack

˚

fac,D_Fe,R_C-[FeL⁵][ClO₄]₂·2CH₃CN C₄₉H₅₄Cl₂FeN₈O₁₁, M_r

=

0.006(12).

3

1057.75, orthorhombic, P2₁2₁2₁, purple block 0.14 ¥ 0.06 ¥ 0.05

1

2

◦

[FeL⁹][ClO₄]₂· CH₃CN. Yield 0.42 g, 0.46 mmol, 51%.

3

˚

mm, a = 13.2870(4), b = 18.8293(7), c = 20.0225(8) A, a = 90 , b =

90 , g = 90 , U = 5009.3(3) A , Z = 4, T = 120(2) K, radiation Mo-

◦

Elemental

analysis

found

(calculated

for

3

˚

C₄₀H_46.50Cl₂FeN_9.50O₈)% C 52.38 (52.50), H 5.33 (5.12), N

˚

Ka (l = 0.71073 A), 25 270 total reﬂections, 8665 unique (R_int

=

14.60 (14.54).

0.1236), R₁= 0.0869 (obs. data), wR₂= 0.2018 (all data), GooF

1.118, Flack 0.09(4).

IR v cm^-13152 w, 1577 m, 1454 m, 1425 m, 1290 m, 1181 w,

1025 s.

fac,D_Fe,R_C-[FeL⁶][ClO₄]₂·CH₃CN. Yield = 36%

Crystallography:

3

1

2

¹H NMR (400 MHz, 298 K, CD₃CN) d_H8.88 (3H, s, HC N),

7.68–7.61 (9H, m, Py/Ph), 7.42–7.31 (12H, m, Py/Ph), 7.12 (3H,

t, ³J_HH= 6.5 Hz, Py), 7.02 (3H, t, ³J_HH= 7.0 Hz, Ph), 6.86 (6H, t,

fac,K_Fe,R_C-[FeL⁹][ClO₄]₂· CH₃CN C₄₀H_46.50Cl₂FeN_9.50O₈, M_r=

3

915.12, Monoclinic, P2₁, purple block 0.20 ¥ 0.15 ¥ 0.15 mm,

◦

˚

a = 12.2800(3), b = 17.6405(4), c = 20.8359(5) A, a = 90 , b =

◦

3

³J_HH= 7.0 Hz, Ph), 6.68 (3H, d, J_HH= 6.5 Hz, Py), 6.53 (6H, d,

92.794(2) , g = 90 , U = 4508.20(19) A , Z = 4, T = 298(2) K,

˚

3

³J_HH= 7.0 Hz, Ph), 5.74 (6H, dd, ³J_HH= 10.0 Hz, J_HH= 3.0 Hz,

radiation Cu-Ka (l = 1.54184 A), 15 340 total reﬂections, 10 206

˚

10430 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

unique (R_int= 0.0270), R₁= 0.0470 (obs. data), wR₂= 0.1337 (all

data), GooF 1.071, Flack 0.005(4).

¹H NMR (400 MHz, 298 K, CD₃CN) d_H8.21 (3H, s, HC N),

3 4

8.07 (3H, td, J_HH= 8.0 Hz, J_HH= 2 Hz, Py), 7.53 (3H, m, Py),

3 3

7.45 (3H, d, J_HH= 8.0 Hz, Py), 7.40 (3H, d, J_HH= 5.5 Hz, Py),

[FeL¹¹₃][ClO₄]₂1.25(CH₃CN). Yield = 0.42 g, 0.39 mmol, 65%.

3

7.09 (3H, t, J_HH= 7.0 Hz, Ph), 6.92 (6H, t, J_HH= 7.0 Hz, Ph),

6.63 (6H, d, ³J_HH= 7.0 Hz, Ph), 5.45 (3H, quartet, ³J_HH= 7.5 Hz,

CH), 1.62 (9H, d, ³J_HH= 7.5 Hz, CH₃).

Elemental

analysis

found

(calculated

for

C_50.50H_66.75Cl₂FeN_10.25O₈)% C 56.03 (56.57), H 6.07 (6.28), N

12.01 (13.39).

UV-vis in CH₃CN (l, nm; e, M^-1cm^-1) 274 (30333), 536 (2117).

IR v cm^-12984 w, 1731 m, 1558 m, 1441 m, 1373 m, 1264 m,

1234 m, 1159 m, 1078 s, 915 w, 846 w.

These are the major peaks observed, which correspond to the

fac,K isomer. There are also small broad peaks corresponding to

the peaks observed for the mer,K isomer, which sharpen at lower

temperatures.

Crystallography:

1

¹³C{ H} NMR (100 MHz, 298 K, CD₃CN) d_C161.6 (C N),

fac,K_Fe,R_C-[FeL¹¹₃][ClO₄]₂·1.25(CH₃CN)

147.6 (Py), 146.1 (Ph/Py), 141.4 (Py), 140.4 (Ph/Py), 129.9

(Ph/Py), 129.2 (Ph/Py), 128.7 (Ph/Py), 127.6 (Ph), 125.5 (Ph),

64.1 (CH), 22.6 (CH₃).

C_50.50H_66.75Cl₂FeN_10.25O₈, M_r= 1072.14, Orthorhombic, P2₁2₁2₁,

purple block 0.40 ¥ 0.20 _◦¥ 0.15 mm, a = 12.0388(4), b = 17.7032(6),

◦

3

˚

c = 25.8524(6) A, a = 90 , b = 90 , g = 90 , U = 5509.8(3) A , Z =

1

All peaks on ¹³C{ H} NMR spectrum are from the major

˚

4, T = 100(2) K, radiation Mo-Ka (l = 0.71073 A), 33 650 total

isomer (fac,K). The peaks for the mer,K isomer are too weak to be

observed.

reﬂections, 15 181 unique (R_int= 0.0563), R₁= 0.0580 (obs. data),

wR₂= 0.1424 (all data), GooF 0.814, Flack 0.001(17).

MS (ESI) m/z 211.0 [L³+H]⁺.

fac-[FeL¹²₃][ClO₄]₂. 2-Acetylpyridine (0.34 mL, 3.0 mmol, 3.0

eq.) and iron(II) perchlorate hexahydrate (0.36 g, 1.0 mmol, 1.0 eq.)

were dissolved in acetonitrile to form a red solution containing the

iron(II) tris(2-acetylpyridine) complex. Benzylamine (2) (0.33 mL,

3.0 mmol, 3.0 eq.) was added and immediately the solution turned

purple. This was stirred for 6 h at ambient temperature and a ¹H

NMR spectrum was measured for a sample of the crude reaction

mixture (1 : 6.5 fac:mer). The solution was stirred for a further 2 d

IR v cm^-11647 w, 1598 m, 1446 m, 1392 w, 1050 s, 761 s, 702 s.

Elemental analysis found (calculated for C₄₂H₄₂B₂F₈N₆Zn)% C

57.51 (57.99), H 4.91 (4.87), N 9.47 (9.66).

Crystallography:

fac,K_Zn,R_C-[ZnL³][ClO₄]₂·CH₃CN·EtOAc C₄₈H₅₃Cl₂N₇O₁₀Zn,

3

M_r= 1024.24, monoclinic, P2₁, colourless block 0.35 ¥ 0.35 ¥

˚

0.20 mm, a = 12.12947(12), _◦b = 13.37998(12), c = 15.44345(15) A,

◦

3

˚

a = 90 , b = 103.3800(10) , g = 90 , U = 2438.32(4) A , Z =

2, T = 100(2) K, radiation Cu-Ka (l = 1.54184 A), 49 726 total

1

˚

at ambient temperature and a H NMR spectrum was measured

for a second sample of the crude reaction mixture (1 : 6.5 fac:mer).

The solution was then stirred at 80 ^◦C for 3 d. A ¹H NMR spectrum

measured for a sample of the crude reaction mixture after this

time showed the fac-isomer exclusively (with a small amount of

reﬂections, 9264 unique (R_int= 0.0421), R₁= 0.0487 (obs. data),

wR₂= 0.1323 (all data), GooF 1.045, Flack 0.007(14).

R_C-[ZnL⁴][BF₄]₂. MS (ESI) m/z 227.1 [L⁴+H]⁺.

2

decomposition).

IR v cm^-11661 m, 1604 m, 1447 w, 1313 m, 1219 w, 1050 s, 773

s, 705 s.

3

¹H NMR (300 MHz, 298 K, CD₃CN) d_H7.88 (3H, td, J_HH

=

4

3

7.5 Hz, J_HH= 1.0 Hz, Py), 7.56 (3H, d, J_HH= 7.5 Hz, Py), 7.38

Elemental analysis found (calculated for C₂₈H₂₈B₂F₈N₄O₂Zn)%

C 48.48 (48.63), H 4.08 (4.08), N 8.05 (8.10).

3

(3H, t, J_HH= 5.5 Hz, Py), 7.10 (3H, t, J_HH= 7.5 Hz, Ph), 6.95

3

(6H, t, J_HH= 7.5 Hz, Ph), 6.76 (3H, d, J_HH= 5.5 Hz, Py), 6.42

(6H, d, ³J_HH= 7.5 Hz, Ph), 5.68 (3H, d, ²J_HH= 15.0 Hz, CH₂), 4.88

(3H, d, ²J_HH= 15.0 Hz, CH₂), 2.35 (9H, s, CH₃).

R_C-[CoL³][PF₆]₂·2MeOH. Ligand L³(0.50 g, 2.38 mmol, 3.0

3

eq.) was dissolved in methanol (25 mL) and stirred at ambient

temperature as cobalt(II) chloride (0.10 g, 0.79 mmol, 1.0 eq.)

in methanol (5 mL) was added causing an immediate colour

change from yellow to orange. The solution was stirred at ambient

temperature for 2 h. A solution of NaPF₆in water (2 mL) was

added, causing the immediate formation of an orange precipitate.

The precipitate was ﬁltered, washed with methanol and dried in

vacuo. Yield = 0.55 g, 0.53 mmol, 67%.

1

¹³C{ H} NMR (75 MHz, 298 K, CD₃CN) d_C154.6 (Py/Ph),

138.9 (Py/Ph), 134.2 (Py/Ph/ketimine), 130.1 (Py/Ph/ketimine),

129.8 (Py/Ph), 129.1 (Py/Ph), 128.6 (Py/Ph), 128.2 (Py/Ph), 127.2

(Py/Ph/ketimine), 126.0 (Py/Ph), 60.6 (CH₂), 18.1 (CH₃).

MS (ESI) m/z 343.1 [FeL¹²₃]²⁺.

IR v cm^-11602 w, 1496 w, 1472 w, 1453 w, 1379 w, 1330 w, 1256

w, 1080 s, 929 w, 770 m, 741 m, 701 m.

Elemental analysis found (calculated for C₄₂H₄₂Cl₂N₆O₈)% C

56.41 (56.96), H 4.63 (4.78), N 9.26 (9.49).

MS (ESI) m/z 344.64 [CoL³]²⁺.

3

IR v cm^-11639 w, 1598 w, 1494 w, 1446 w, 1391 w, 1321 w, 1234

w, 1162 w, 1113 w, 1072 w, 1019 w, 982 w, 925 w, 829 s, 780 m, 760

m, 702 m.

General procedure for [ZnL³][BF₄]₂and [ZnL⁴][BF₄]₂Com-

3

2

plexes. 2-Pyridinecarboxaldehyde (3.0 eq.) and the appropriate

amine (3,4) (3.0 eq.) were dissolved in acetonitrile to form a

yellow solution containing the ligand. Zinc(II) tetraﬂuoroborate

hydrate (H₂O = 6–7) (1.0 eq.) was added and no colour change

was observed. This solution was stirred overnight at ambient

temperature before ethyl acetate/diethyl ether was added dropwise

until signs of crystallisation. The white crystals were ﬁltered and

dried in vacuo. Yields are unoptimised - analysis of the crude

reaction mixtures indicates that reactions are essentially complete.

Elemental analysis found (calculated for C₄₄H₅₀CoF₁₂N₆O₂P₂)%

C 50.04 (50.63), H 4.31 (4.83), N 8.29 (8.05).

Crystallography:

fac,K_Co,R_C-[CoL³][PF₆]₂·2MeOH C₄₄H₅₀CoF₁₂N₆O₂P₂, M_r

=

3

1043.77, monoclinic, P2₁, yellow block 0.26 ¥ 0.15 ¥ 0.09 mm,

◦

˚

a = 11.6650(5), b = 13.5956(5), c = 15.3936(7) A, a = 90 , b =

◦

3

˚

95.949(2) , g = 90 , U = 2428.16(18) A , Z = 2, T = 120(2) K,

radiation Mo-Ka (l = 0.71073 A), 33 640 total reﬂections, 11 096

˚

unique (R_int= 0.0782), R₁= 0.0669 (obs. data), wR₂= 0.1828 (all

data), GooF 0.986, Flack 0.044(18).

K_Zn,R_C-[ZnL³][BF₄]₂. Yield = 26%. Ratio is 6.5 : 1 fac:mer.

3

This journal is

The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10431

©

S_C-[CoL³][PF₆]₃·3CH₃CN. The solvents were degassed before

IR n cm^-12993 w, 2287 w, 2249 w, 1981 w, 1641 m, 1597 m, 1494

m, 1483 w, 1443 m, 1392 m, 1324 m, 1272c w, 1232 w, 1072 s, 1101

s, 983 m, 921 m, 782 m,757 s.

3

use. The ligand and Na₃[Co(NO₂)₆] were dried in vacuo for 2 h

before use. Ligand L³(0.50 g, 2.38 mmol, 3.0 eq.) was dissolved in

a mixture of acetonitrile (40 mL) and water (3 mL). Na₃[Co(NO₂)₆]

(0.32 g, 0.79 mmol, 1.0 eq.) was added as a solid and immediately

the colour changed to orange. The solution was stirred overnight

at ambient temperature and then sodium perchlorate (0.29 g,

2.38 mmol, 3.0 eq.) was added, and the solution was stirred for

a further 4 h at ambient temperature. After this time the solvent

was removed under reduced pressure and the resulting crude solid

was dissolved in DCM (60 mL) and washed with water (15 mL).

The DCM layer was concentrated until signs of precipitation were

observed (ca 10 mL). The resulting precipitate was ﬁltered and

dried in vacuo. Yield = 0.17 g, 0.14 mmol, 17%.

Crystallography:

fac,K_Mn,R_C-[MnL³][ClO₄]₂·2CH₃CN C₄₆H₄₈Cl₂MnN₈O₈, M_r=

3

966.76, orthorhombic, P2₁2₁2₁, yellow block 0.30 ¥ 0.20 ¥ 0.20

◦

˚

mm, a = 13.9695(2), b = 17.3214(3), c = 19.2264(3) A, a = 90 , b =

◦

3

˚

90 , g = 90 , U = 4652.23(12) A , Z = 4, T = 100(2) K, radiation

Mo-Ka (l = 0.71073 A), 54 447 total reﬂections, 12 920 unique

˚

(R_int= 0.0522), R₁= 0.0591 (obs. data), wR₂= 0.1292 (all data),

GooF 1.078, Flack 0.038(16).

Acknowledgements

This complex was found to be unstable and decomposed rapidly

The Oxford Diffraction Gemini XRD system was obtained

through the Science City Advanced Materials project: Creating

and Characterising Next Generation Advanced Materials project,

with support from Advantage West Midlands (AWM) and part

funded by the European Regional Development Fund (ERDF).

We thank the EPSRC and the University of Warwick for ﬁnancial

support. We thank the EPSRC National Crystallography Service

(University of Southampton, UK) for data collection.

thus preventing detailed analysis on a pure sample.

Crystallography:

fac,D_Co,S_C-[CoL³][PF₆]₃·3CH₃CN C₄₈H₅₁CoF₁₈N₉P₃, M_r

=

3

1247.82, monoclinic, P2₁, pink plate 0.16 ¥ 0.06 ¥ 0.02 mm, a =

◦

˚

11.978, b = 11.769, c = 18.739 A, a = 90 , b = 92.09 , g = 90 , U =

3

˚

2639.9 A , Z = 2, T = 120(2) K, radiation Mo-Ka (l = 0.71073 A),

47 585 total reﬂections, 11 835 unique (R_int= 0.0919), R₁= 0.0669

(obs. data), wR₂= 0.1382 (all data), GooF 1.170, Flack 0.08(1).

S_C-[NiL³][ClO₄]₂·CH₃CN·EtOAc.

2-Pyridinecarboxalde-

Notes and references

3

hyde (0.88 g, 8.19 mmol, 3 eq.) and (S)-a-methylbenzylamine

(0.99 g, 8.19 mmol, 3 eq.) were dissolved in acetonitrile (10 mL)

and stirred at ambient temperature for 30 min. A solution of

Ni(ClO₄)₂·6H₂O (1.00 g, 2.73 mmol, 1 eq.) in acetonitrile (5 mL)

was added resulting in a red solution. The reaction was stirred

for a further 2 h before ethyl acetate was added dropwise. The

product was left to crystallise overnight. The red/brown crystals

were collected by vacuum ﬁltration and dried in air. Yield 1.59 g,

1.56 mmol, 57%.

1 A. Werner and A. Vilmos, Z. Anorg. Allg. Chem., 1899, 21, 145–158.

2 C. R. Rice, Coord. Chem. Rev., 2006, 250, 3190–3199.

3 P. A. Gale and R. Quesada, Coord. Chem. Rev., 2006, 250, 3219–3244.

4 M. H. Keefe, K. D. Benkstein and J. T. Hupp, Coord. Chem. Rev., 2000,

205, 201–228.

5 V. Balzani, A. Juris, M. Venturi, S. Campagna and S. Serroni, Chem.

Rev., 1996, 96, 759–834.

6 F. R. Keene, Chem. Soc. Rev., 1998, 27, 185–194.

7 F. R. Keene, Coord. Chem. Rev., 1997, 166, 121–159.

8 K. E. Erkkila, D. T. Odom and J. K. Barton, Chem. Rev., 1999, 99,

2777–2796.

9 C. Metcalfe and J. A. Thomas, Chem. Soc. Rev., 2003, 32, 215–224.

10 K. K.-W. Lo, K. H.-K. Tsang, K.-S. Sze, C.-K. Chung, T. K.-M. Lee,

K. Y. Zhang, W.-K. Hui, C.-K. Li, J. S.-Y. Lau, D. C.-M. Ng and N.

Zhu, Coord. Chem. Rev., 2007, 251, 2292–2310.

11 M. J. Clarke, Coord. Chem. Rev., 2002, 232, 69–93.

12 F. P. Dwyer and E. C. Gyarfas, J. Proc. R. Soc. N. S. W., 1949, 83,

174–176.

13 X. Hua and A. von Zelewsky, Inorg. Chem., 1995, 34, 5791–5797.

14 C. Hiort, P. Lincoln and B. Norden, J. Am. Chem. Soc., 1993, 115,

3448–3454.

15 S. Bodige, A. S. Torres, D. J. Maloney, D. Tate, G. R. Kinsel, A. K.

Walker and F. M. MacDonnell, J. Am. Chem. Soc., 1997, 119, 10364–

10369.

16 W. R. Browne, D. Hesek, J. F. Gallagher, C. M. O’Connor, J. S. Killeen,

F. Aoki, H. Ishida, Y. Inoue, C. Villani and J. G. Vos, Dalton Trans.,

2003, 2597–2602.

MS (ESI) m/z 344.1 [NiL³]²⁺, 239.1 [NiL³]²⁺.

3

2

IR v cm^-11736 w, 1641 w, 1599 m, 1493 w, 1446 w, 1390 w,

1373 w, 1320 w, 1235 w, 1082 s, 1019 w, 982 w, 925 w, 779 w, 760 m,

702 m.

Elemental analysis found (calculated for C₄₈H₅₃Cl₂N₇NiO₁₀)%

C 56.66 (56.66), H 5.35 (5.25), N 10.34 (9.64).

Crystallography:

fac,D_Ni,S_C-[NiL³][ClO₄]₂·CH₃CN·EtOAc C₄₈H₅₃Cl₂N₇NiO₁₀,

3

M_r= 1017.58, monoclinic, P2₁, red block 0.25 ¥ 0.25 ¥ 0.20 mm,

◦

˚

a = 12.1724(3), b = 1_◦3.3579(2), c = 15.4379(3) A, a = 90 , b =

◦

3

˚

103.8310(10) , g = 90 , U = 2437.38(8) A , Z = 2, T = 120(2) K,

˚

radiation Mo-Ka (l = 0.71073 A), 28 903 total reﬂections, 11 043

unique (R_int= 0.0445), R₁= 0.0515 (obs. data), wR₂= 0.1258 (all

data), GooF 1.055, Flack 0.044(11).

17 J. Lacour, S. Torche-Haldimann, J. J. Jodry, C. Ginglinger and F.

Favarger, Chem. Commun., 1998, 1733–1734.

18 N. C. Fletcher, P. C. Junk, D. A. Reitsma and F. Richard Keene, J.

Chem. Soc., Dalton Trans., 1998, 133–138.

19 J. G. Collins, J. R. Aldrich-Wright, I. D. Greguric and P. A. Pellegrini,

Inorg. Chem., 1999, 38, 5502–5509.

20 S. Torelli, S. Delahaye, A. Hauser, G. Bernardinelli and C. Piguet,

Chem.–Eur. J., 2004, 10, 3503–3516.

R_C-[MnL³][ClO₄]₂·2CH₃CN. Mn(ClO₄)₂·6H₂O (0.15 g,

3

0.4 mmol) was add to a stirred solution of 2-pyridinecar-

boxaldehyde (0.14 g, 1.3 mmol) and (R)-a-methylbenzylamine

(0.16 mL, 1.3 mmol) in acetonitrile (20 mL) at ambient

temperature and stirred for 4 h. The resulting yellow solution

yielded the desired product as a yellow crystalline solid on the

addition of ethyl acetate. Yield 66%.

21 N. C. Fletcher, M. Nieuwenhuyzen and S. Rainey, J. Chem. Soc., Dalton

Trans., 2001, 2641–2648.

22 A. Grabulosa, M. Beley and P. C. Gros, Eur. J. Inorg. Chem., 2008,

1747–1751.

23 N. C. A. Baker, N. McGaughey, N. C. Fletcher, A. V. Chernikov, P. N.

Horton and M. B. Hursthouse, Dalton Trans., 2009, 965–972.

24 E. Meggers, Chem.–Eur. J., 2010, 16, 752–758.

Elemental analysis found (calculated for C₄₆H₄₈Cl₂MnN₈O₈)%

C 57.50 (57.15), H 5.10 (5.00), N 11.20 (11.59).

10432 | Dalton Trans., 2011, 40, 10416–10433

This journal is

The Royal Society of Chemistry 2011

©

25 L. Gong, S. P. Mulcahy, K. Harms and E. Meggers, J. Am. Chem. Soc.,

2009, 131, 9602–9603.

53 P. Gu¨tlich, Y. Garcia and H. A. Goodwin, Chem. Soc. Rev., 2000, 29,

419–427.

54 R. J. Dosser, W. J. Eilbeck, A. E. Underhill, P. R. Edwards and C. E.

Johnson, J. Chem. Soc. A, 1969, 810–816.

55 B. J. Childs, D. C. Craig, M. L. Scudder and H. A. Goodwin, Inorg.

Chim. Acta, 1998, 274, 32–41.

26 P. Hayoz, A. Von Zelewsky and H. Stoeckli-Evans, J. Am. Chem. Soc.,

1993, 115, 5111–5114.

27 A. von Zelewsky and O. Mamula, J. Chem. Soc., Dalton Trans., 2000,

219–231.

28 N. C. Fletcher, C. Martin and H. J. Abraham, New J. Chem., 2007, 31,

1407–1411.

29 N. C. Fletcher, M. Nieuwenhuyzen, R. Prabarahan and A. Wilson,

Chem. Commun., 2002, 1188–1189.

30 H. Weizman, J. Libman and A. Shanzer, J. Am. Chem. Soc., 1998, 120,

2188–2189.

31 S. E. Howson, L. E. N. Allan, N. P. Chmel, G. J. Clarkson, R. van

Gorkum and P. Scott, Chem. Commun., 2009, 1727–1729.

32 C. P. Sebli, S. E. Howson, G. J. Clarkson and P. Scott, Dalton Trans.,

2010, 39, 4447–4454.

33 S. E. Howson and P. Scott, Dalton Trans., 2011, 40, 4332–4333.

34 Y. Hsiao and L. S. Hegedus, J. Org. Chem., 1997, 62, 3586–3591.

35 D. A. Evans, G. S. Peterson, J. S. Johnson, D. M. Barnes, K. R. Campos

and K. A. Woerpel, J. Org. Chem., 1998, 63, 4541–4544.

36 K. Sutowardoyo, M. Emziane and D. Sinou, Tetrahedron Lett., 1989,

30, 4673–4676.

37 R. N. Bream, S. V. Ley, B. McDermott and P. A. Procopiou, J. Chem.

Soc., Perkin Trans. 1, 2002, 2237–2242.

38 Z. Y. Chang and R. M. Coates, J. Org. Chem., 1990, 55, 3475–3483.

39 F. J. LaRonde and M. A. Brook, Inorg. Chim. Acta, 1999, 296, 208–221.

40 M. C. Thompson and D. H. Busch, J. Am. Chem. Soc., 1962, 84, 1762–

1763.

56 R. Boca, P. Baran, L. Dlha´n, J. Sima, G. Wiesinger, F. Renz, U. El-

Ayaan and W. Linert, Polyhedron, 1997, 16, 47–55.

57 M. J. Plater, P. Barnes, L. K. McDonald, S. Wallace, N. Archer, T.

Gelbrich, P. N. Horton and M. B. Hursthouse, Org. Biomol. Chem.,

2009, 7, 1633–1641.

58 A. Escuer, T. Comas, R. Vicente and J. Ribas, Transition Met. Chem.,

1993, 18, 42–44.

59 A. Rodger and B. Norde´n, Circular Dichroism and Linear Dichroism,

Oxford University Press, Oxford, 1997.

60 M. Ziegler and A. von Zelewsky, Coord. Chem. Rev., 1998, 177, 257–

300.

61 J. Hidaka and B. E. Douglas, Inorg. Chem., 1964, 3, 1180–1184.

62 E. G. Hohenstein and C. D. Sherrill, J. Phys. Chem. A, 2009, 113,

878–886.

63 S. Grimme, Angew. Chem., Int. Ed., 2008, 47, 3430–3434.

64 E. J. B. A. Baerends, C. Bo, P. M. Boerrigter, L. Cavallo, L. Deng, R.

M. Dickson, D. E. Ellis, L. Fan, T. H. Fischer, C. Fonseca Guerra, S.

J. A. van Gisbergen, J. A. Groeneveld, O. V. Gritsenko, F. E. Harris,

P. van den Hoek, H. Jacobsen, G. van Kessel, F. Kootstra, E. van

Lenthe, V. P. Osinga, P. H. T. Philipsen, D. Post, C. C. Pye, W. Ravenek,

P. Ros, P. R. T. Schipper, G. Schreckenbach, J. G. Snijders, M. Sola,

D. Swerhone, G. te Velde, P. Vernooijs, L. Versluis, O. Visser, E. van

Wezenbeek, G. Wiesenekker, S. K. Wolff, T. K. Woo and T. Ziegler,

adf2008.01, Scientiﬁc Computing and Modelling NV: Free University,

Amsterdam, 2008.

41 S. Alshehri, J. Burgess, J. Fawcett, D. R. Russell and A. M. Shaker,

Transition Met. Chem., 2000, 25, 691–694.

42 I. J. Munslow, A. J. Clarke, R. J. Deeth, I. Westmoreland and P. Scott,

Chem. Commun., 2002, 1868–1869.

65 M. A. Robinson and D. H. Busch, Inorg. Chem., 1963, 2, 1171–1177.

66 A. Damianoglou, E. J. Crust, M. R. Hicks, S. E. Howson, A. E.

Knight, J. Ravi, P. Scott and A. Rodger, Chirality, 2008, 20, 1029–

1038.

43 P. D. Knight and P. Scott, Coord. Chem. Rev., 2003, 242, 125–143.

44 E. J. Crust, A. J. Clarke, R. J. Deeth, C. Morton and P. Scott, Dalton

Trans., 2004, 4050–4058.

45 I. J. Munslow, A. R. Wade, R. J. Deeth and P. Scott, Chem. Commun.,

67 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 1990,

2004, 2596–2597.

46, 467–473.

46 I. Westmoreland, I. J. Munslow, A. J. Clarke, G. Clarkson and P. Scott,

Organometallics, 2004, 23, 5066–5074.

68 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,

64, 112–122.

69 G. A. Bain and J. F. Berry, J. Chem. Educ., 2008, 85, 532–536.

70 S. Grimme, J. Comput. Chem., 2004, 25, 1463–1473.

71 E. J. Baerends, D. E. Ellis and P. Ros, Theor. Chim. Acta, 1972, 27,

339–354.

72 C. C. Pye and T. Ziegler, Theor. Chem. Acc., 1999, 101, 396–408.

73 J. M. Janey, T. Iwama, S. A. Kozmin and V. H. Rawal, J. Org. Chem.,

2000, 65, 9059–9068.

74 R. E. Cowley, R. P. Bontchev, E. N. Duesler and J. M. Smith, Inorg.

Chem., 2006, 45, 9771–9779.

47 I. Westmoreland, I. J. Munslow, A. J. Clarke, G. Clarkson, R. J. Deeth

and P. Scott, J. Organomet. Chem., 2006, 691, 2228–2236.

48 A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G. Watson and

R. Taylor, J. Chem. Soc., Dalton Trans., 1989, S1–S83.

49 F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, John

Wiley & Sons, Inc., New York, 1988.

50 F. Tuna, M. R. Lees, G. J. Clarkson and M. J. Hannon, Chem.–Eur. J.,

2004, 10, 5737–5750.

51 D. Pelleteret, R. Clerac, C. Mathoniere, E. Harte, W. Schmitt and P. E.

Kruger, Chem. Commun., 2009, 221–223.

75 C. D. Bedford, R. N. Harris, R. A. Howd, D. A. Goff, G. A. Koolpe,

M. Petesch, I. Koplovitz, W. E. Sultan and H. A. Musallam, J. Med.

Chem., 1989, 32, 504–516.

52 J. A. Cowan, Inorganic Biochemistry: An Introduction, Wiley-VCH,

New York, 1997.

This journal is

The Royal Society of Chemistry 2011

Dalton Trans., 2011, 40, 10416–10433 | 10433

©

Article Doi

Origins of stereoselectivity in optically pure phenylethaniminopyridine tris-chelates M(NN′)3n+ (M = Mn, Fe, Co, Ni and Zn)

DOI: 10.1039/c1dt10588d

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c1dt10588d

Full text of DOI:10.1039/c1dt10588d

Products guided by the article

R&D Labs maybe for 127180-88-5

Relevant to this article

Hot Product