Alternatives to pyridinediimine ligands: Syntheses and structures of metal complexes supported by donor-modified α-diimine ligands

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

PAPER

www.rsc.org/dalton | Dalton Transactions

Alternatives to pyridinediimine ligands: syntheses and structures of metal

complexes supported by donor-modiﬁed a-diimine ligands

Benjamin M. Schmiege,^aMichael J. Carney,*^aBrooke L. Small,^bDeidra L. Gerlach^aand Jason A. Halfen^a

Received 12th February 2007, Accepted 20th March 2007

First published as an Advance Article on the web 16th April 2007

DOI: 10.1039/b702197f

This report describes the synthesis and characterization of metal halide complexes (M = Mn, Fe, Co)

supported by a new family of pendant donor-modiﬁed a-diimine ligands. The donor (N, O, P, S)

substituent is linked to the a-diimine by a short hydrocarbon spacer forming a tridentate,

mer-coordinating ligand structure. The tridentate ligands are assembled from monoimine precursors,

the latter being synthesized by selective reaction with one carbonyl group of the a-dione. While

attempts to separately isolate tridentate ligands in pure form were unsuccessful, metal complexes

supported by the tridentate ligand are readily synthesized in-situ, by forming the ligand in the presence

of the metal halide, resulting in a metal complex which subsequently crystallizes out of the reaction

mixture. Metal complexes with NNN, NNO, NNP and NNS donor sets have been prepared and

examples supported by NNN, NNP and NNS ligands have been structurally characterized. In the solid

state, NNN and NNP ligands coordinate in a mer fashion and the metal complexes possess distorted

square pyramidal structures and high spin (S = 2) electronic conﬁgurations. Compounds with NNS

coordination environments display a variety of solid state structures, ranging from those with unbound

sulfur atoms, including chloride bridged and solvent ligated species, to those with sulfur weakly bound

to the metal center. The extent of sulfur ligation depends on the donor ability of the crystallization

solvent and the substitution pattern of the arylthioether substituent.

ligands are formed via sequential and selective reaction of bulky

Introduction

anilines with the a-dione carbonyl groups; examples of such lig-

ands are shown as A–C^8–10in Scheme 1. Here the ability to modify

Over the past decade, ligands with imine functional groups

have found wide use in oleﬁn polymerization catalysis. For

the oligomer product distribution, as measured by the Schulz–

example, Brookhart demonstrated that nickel(II) and palladium(II)

Flory constant, K, has been well documented. In a rather dramatic

complexes supported by a-diimine ligands are effective a-oleﬁn

case, the iron(II) catalyst with ligand B switches from producing

polymerization catalysts in the presence of methylalumoxane

oligomers to polyethylene by replacing the hydrogen 2,6-aryl

(MAO). Of particular note, nickel and palladium diimine catalysts

substituents on one ring with methyl groups. Finally, PBI-related

polymerize ethylene to an unusual highly branched polyethylene

(via a chain walking mechanism)¹and palladium diimine cata-

lysts are able to copolymerize a-oleﬁns and functionalized vinyl

ligands, formed by placing heteroatoms in the ligand backbone,

have also been explored. Selected examples are shown as ligands

D–G in Scheme 1.^11–13Such ligand modiﬁcations often have a large

monomers.²More recently, Gibson³and Brookhart⁴separately

impact on subsequent catalyst activity and oligomer properties,

showed that iron(II) halides supported by pyridinebisimine (PBI)

ligands display exceptional reactivity towards ethylene, producing

a-oleﬁn oligomers or high density polyethylene depending on

the size and location of substituents on the imino-aryl rings.

Since the initial reports by Gibson and Brookhart, which detailed

the impact that imino-aryl substituents have on catalyst activity

and product properties, more recent studies have examined the

effect of heteroatom (e.g. Cl, Br, I⁵; F⁶; OMe, CF₃) substituted

particularly on the Schulz–Flory constant. The ability to vary

K over a fairly wide range, with, in most cases, relatively small

changes in ligand structure, has fueled our interest in this area.

Control over K is particularly important in an industrial setting

where the ability to modify the product distribution via catalyst

modiﬁcation can provide a-oleﬁn producers with another handle

to optimize the value of the oligomeric product exiting the reactor.

We are especially interested in the use of alternative tridentate

7

imino-aryl groups on iron(II)–PBI catalyst activity and oligomer

ligands that resemble the overall coordination geometry of PBI

properties. Of particular relevance was the ability to alter the

systems, but which allow for more substantial ligand (electronic

product distribution from high molar mass polymers to low molar

and steric) variation. In particular, we wished to introduce

mass oligomers by varying the substitution pattern of the aryl ring.

alternative donors (O, P, S) into a tridentate ligand environment

In addition, iron(II) complexes supported by asymmetric ver-

and we targeted strategies that allowed us to separately modify the

sions of PBI ligands have also been studied. These asymmetric

donor heteroatom and its substituents. Moreover, since most PBI

and a-diimine ligands reported to date are comprised of arylimine

functional groups, we also sought routes to more electronically

diverse alkyl-substituted imines. This report outlines the syntheses

and structures of divalent metal complexes supported by mer-

^aDepartment of Chemistry, University of Wisconsin-Eau Claire, 105 Garﬁeld

Avenue, Eau Claire, WI, USA 54702. E-mail: carneymj@uwec.edu; Fax: +1-

715-836-4979; Tel: +1-715-836-3500

^bChevron Phillips Chemical Company, LP, 1862 Kingwood Drive, Kingwood,

TX, 77339, USA

ligating acenaphthene-diimine ligands, which have been modiﬁed

This journal is

The Royal Society of Chemistry 2007

Dalton Trans., 2007, 2547–2562 | 2547

©

View Article Online

Scheme 1

with an additional pendant donor (N, O, P, S) atom. The metal

complexes described herein are supported by unique asymmet-

ric, donor-modiﬁed acenaphthene-diimine ligands incorporating

stable n-alkylimine functional groups.

Results and discussion

Preparation of monoimine ligand precursors

We set out to prepare mer-ligating tridentate ligands comprised

of a donor-modiﬁed a-diimine (DM-DI in Scheme 2). Such

ligands would provide the desired mer-coordination geometry,

but would allow for extensive ligand variation by modifying the

appended donor atom and its substituents. In contrast to the more

symmetrical PBI donor atom arrangement, wherein the imine

nitrogens ﬂank a central pyridine ring, the DM-DI ligand positions

the imines side-by-side with the pendant donor atom off to one

side. Thus, ligands with various heteroatom donors and donor

substituents, and with donors attached to the a-diimine by linking

groups of various lengths, are accessible with the appropriate

synthetic schemes. The ability to manipulate the ligand structure in

a controlled fashion allows us to explore the impact of such ligand

variations on metal complex structure and catalytic performance.

For our targeted asymmetric ligands, the appropriate synthetic

schemes require the introduction of imine functional groups by

selective reaction of one carbonyl group of the a-dione at a

time. A recent report outlines a preparative scheme to synthesize

asymmetric a-diimines,¹⁴however no examples of ligands with

pendant donor atoms were described. Similar synthetic schemes

have been used to prepare the asymmetric PBI ligands mentioned

previously (ligands A–C in Scheme 1).¹⁵

Scheme 2

Path A, successful formation of the donor-modiﬁed monoimine

would seemingly favor subsequent derivatization (using a bulky

aniline) at the more accessible carbonyl rather than at the

just-introduced imine functionality. Unfortunately, attempts to

prepare the initial donor-modiﬁed monoimines by reaction of

acenaphthenequinone with one equivalent of a pendant donor-

modiﬁed primary alkyl amine, such as 2-(aminomethyl)pyridine,

2-(aminoethyl)pyridine or N,N -dimethylethylenediamine did not

yield isolable material. Instead a complex product mixture

resulted.¹⁶Analysis of the mixture by H NMR suggested that,

1

among other products, both diimine and monoimine products

were formed, even when reaction conditions were employed to

favor mono-substitution (e.g. slow addition of one equivalent of

the amine to acenaphthenequinone). In one instance, reaction of

acenaphthenequinone with one equivalent of the more sterically

encumbered (and less basic) 2-phenylthioaniline did lead to

selective monoimine formation (Scheme 3, compound 1). The

1

In the present case, two possible synthetic approaches were

envisioned starting with acenaphthenequinone, these approaches

are outlined in Scheme 2. The ﬁrst (Path A) involves selective

reaction of one carbonyl group with a donor-modiﬁed amine,

followed by reaction of the remaining carbonyl with a bulky aniline

derivative. The second route (Path B) reverses the order of addition

of the bulky aniline derivative and the donor-modiﬁed amine. For

isolated yellow powder displayed a complex solution H NMR

spectrum at room temperature, perhaps due to the presence of

conformational isomers that interconvert slowly relative to the

NMR timescale. Elemental analysis, GC-MS data and an X-

ray crystal structure determination conﬁrmed the monoimine

formulation. The molecular geometry of 1 is shown in Fig. 1.

Selected interatomic distances are listed in Table 2. The structure

2548 | Dalton Trans., 2007, 2547–2562

This journal is

The Royal Society of Chemistry 2007

©

View Article Online

Thus, the formation of 1 demonstrates that monoimines with

pendant donor atoms can form selectively. However, for successful

ligation to a metal complex, rigid phenyl linking groups, such

as that in 1, are not compatible with an acenaphthenediimine

backbone. Closer inspection of 1 in Fig. 1 indicates that unfavor-

able steric interactions between the acenaphthene ring (C2) and

the imino-aryl ring (C18) would prevent the imino-aryl ring and

the acenaphthene ring from adopting the coplanar arrangement

necessary to complex a metal with mer-ligation geometry. Indeed,

we have not been successful in forming a metal complex with the

thioether sulfur in 1 bound to the metal (vide infra).

Scheme 3

An alternative route (Path B in Scheme 2) to the desired a-

diimine ligand family involves initial monoimine formation using

a bulky aniline derivative, followed by subsequent addition of a

pendant donor-modiﬁed amine to the remaining carbonyl group.

In this case, we found that the monoimines are readily prepared

by reacting acenaphthenequinone with stoichiometric amounts

of sterically bulky aniline derivatives (Scheme 4). Slow addition

of a dilute ethanol solution of the appropriate aniline derivative

(1 equivalent) to a heated (60 ^◦C) slurry of acenaphthenequinone

in ethanol provides monoimines 2–7 in moderate to good yields.

The ligand precursors were isolated by ﬁltration of unreacted ace-

naphthenequinone and cooling of the ﬁltrate to afford crystalline

orange solids, which were judged to be of acceptable purity by

Fig. 1 Thermal ellipsoid representation (35% probability boundaries) of

the X-ray crystal structure of monoimine 1 with hydrogen atoms omitted

for clarity.

˚

=

of 1 shows normal C N (1.272(3) A) and C O (1.209(3) A) bond

lengths and is clearly consistent with the monoimine formulation.

The structure also shows the thioether aryl substituent nestled

underneath the acenaphthene ring in a coplanar arrangement

˚

(centroid to centroid distance = 3.66 A), suggesting some level

of p-stacking interactions between the two aromatic systems.

Scheme 4

Table 1 Summary of X-ray crystallographic data for monoimine 1 and NNN- and NNP-ligated metal complexes^a

Complex

1

(3a)FeCl₂

(2c)FeCl₂·0.5MeCN (2d)FeCl₂

(2a)MnCl₂

(3g)FeCl₂

Empirical formula

FW

Crystal system

C₂₄H₁₅NOS

365.43

Orthorhombic

Pbca

14.488(2)

15.586(2)

16.105(1)

90

3636.6(7)

8

1.335

C₂₈H₃₃Cl₂FeN₃

538.32

C₂₈H_24.50Cl₂FeN_3.50C₂₆H₂₇Cl₂FeN₃O C₂₄H₂₅Cl₂MnN₃

C₄₂H₄₃Cl₂FeN₄P

761.52

Orthorhombic

536.76

Triclinic

¯

P1

524.26

Triclinic

¯

P1

481.31

Monoclinic

C2/c

Orthorhombic

P2₁2₁2₁

9.336(2)

12.5150(8)

23.120(3)

90

2701.3(7)

4

1.324

Space group

Pna2₁

˚

a/A

11.348(2)

15.047(3)

16.062(3)

85.13(5)

77.34(4)

76.89(6)

2604.3(8)

4

12.309(1)

14.355(3)

14.553(2)

89.42(2)

77.51(1)

77.48(1)

2449.1(6)

4

19.107(2)

16.5813(7)

14.816(1)

90

102.74(1)

90

4578.4(6)

8

1.397

10.690(2)

30.399(6)

11.998(2)

90

3898.9(12)

4

1.297

˚

b/A

˚

c/A

◦

a/

b/

◦

c /

3

˚

V/A

Z

d

_calcd/Mg m⁻³

1.369

1.422

Crystal size/mm

0.26 × 0.24 × 0.24 0.40 × 0.38 × 0.18 0.42 × 0.26 × 0.26

0.48 × 0.08 × 0.06 0.38 × 0.36 × 0.20 0.48 × 0.16 × 0.14

Abs. Coeff./mm⁻¹

0.191

49.94

1.0–1.0

6225

0.777

55.00

0.8728–0.7463

6489

5874

0.807

0.858

0.826

0.600

49.98

0.995–0.839

7128

3609

◦

2h max/

49.98

1.0–1.0

9637

49.90

1.0–0.692

8980

53.94

1.0–0.9514

5125

Transmission range

No. of reﬂns collected

No. of indep reﬂns

3195

9134

8545

4975

No. of obsd reﬂections 1887

4548

313

0.0440 (0.0853)

1.017

0.251, −0.275

6314

623

0.0552 (0.1322)

1.019

0.645, −0.597

5260

599

0.0680 (0.1460)

1.004

0.693, −0.618

4244

275

0.0312 (0.0761)

1.039

0.246, −0.287

2575

453

0.0503 (0.0930)

1.018

0.592, −0.452

No. of variables

244

R1 (wR2)^b[I > 2r(I)]

Goodness-of-ﬁt (F²)

0.0473 (0.0863)

0.990

−

−3

˚

Diff. Peaks/e A

0.153, −0.174

^aSee Experimental section for additional data collection, reduction, and structure solution and reﬁnement details. ^bR1 = R ꢀF_o|−|F_cꢀ/R |F_o|; wR2 =

2

[R[w(F_o−F_c²)²]]^1/2where w = 1/r²(F_o²) + (aP)²+ bP.

This journal is

The Royal Society of Chemistry 2007

Dalton Trans., 2007, 2547–2562 | 2549

©

View Article Online

¹H and ¹³C NMR spectroscopy and elemental analysis.

Monoimines 2–7 encompass a range of aryl ring substituents and

were prepared with the expectation that, as in the case of Ni-

diimine^1,2and Fe–PBI^3,4catalysts, the aryl substitution pattern

would have a signiﬁcant impact on catalytic performance.

With monoimines 2–7 in hand, we turned our attention to

the preparation of asymmetric diimine ligands by reaction of

the remaining carbonyl group with a donor-modiﬁed primary

amine. The acenaphthene backbone structure necessarily limited

us to pendant donors linked via short alkyl spacers as these

spacers would minimize unfavorable steric interactions, such

as those observed in 1, between the linking group and the

acenaphthenediimine backbone system. Unfortunately, numerous

attempts to prepare pure donor-modiﬁed a-diimine ligands were

unsuccessful. Our efforts were thwarted by the tendency of the

(inherently more basic) donor-modiﬁed alkylamine to react at

the imine and carbonyl groups in 2–7. While NMR data were

consistent with formation of some of the desired donor-modiﬁed a-

diimine, examination of the reaction mixtures invariably revealed

substantial quantities of the free substituted aniline, indicating

that reaction was also occurring at the imino group. Furthermore,

attempts to isolate and purify the desired donor-modiﬁed a-

diimine by column chromatography (silica) were unsuccessful

1

due to decomposition (conﬁrmed by H NMR spectroscopy) of

the imines back to acenaphthenequinone, a reaction apparently

facilitated by the relatively acidic silica chromatographic media.

We have not extensively explored the use of other chromatographic

supports to purify the a-diimine ligands. Instead we turned our

attention to an in-situ technique to prepare the a-diimine ligand

in the presence of the metal halide, leading to the desired metal

complex in one step.

In-situ synthesis of metal complexes supported by donor modiﬁed

a-diimine ligands

Due to the difﬁculties in obtaining pure samples of pendant donor-

modiﬁed a-diimine ligands, we explored a synthetic route that

avoids ligand isolation and puriﬁcation, and instead produces

the donor-modiﬁed ligand and the metal complex in one step.

Utilizing this strategy, shown in Scheme 5, we have prepared

Scheme 5

2550 | Dalton Trans., 2007, 2547–2562

This journal is

The Royal Society of Chemistry 2007

©

View Article Online

Scheme 6

a fairly extensive family of metal complexes utilizing a-diimine

ligands modiﬁed with N, O, P and S pendant donor atoms. It is

unclear whether the metal halide facilitates ligand formation or

whether the metal simply sequesters the ligand once it forms. In

any event, crystallization of the metal complex out of the reaction

mixture undoubtedly serves to drive formation of the desired metal

complex. In a representative synthetic example, treatment of one

equivalent of 2,6-methyl substituted monoimine 2 in dry butanol

with one equivalent each of N,N-dimethylethylenediamine (a) and

anhydrous iron(II) chloride yields, after overnight heating at 60 ^◦C,

a dark green microcrystalline solid of complex (2a)FeCl₂in 80%

yield. Subsequent ﬁltration and drying provided analytically pure

material. Similar synthetic procedures have been used to produce

the array of metal complexes shown in Scheme 6.

Similar to metal–PBI complexes, the metal complexes in

Scheme 6 display magnetic moments consistent with high spin

metal centers, and, when sufﬁciently soluble, possess param-

agnetically shifted ¹H NMR spectra. Unfortunately, low com-

pound solubility precluded obtaining reliable NMR spectra for

all compounds. In addition, the inherent asymmetry of these

molecules, coupled with the broad, and sometimes overlapping

resonances, complicated the signal assignments. For these reasons,

detailed chemical shift assignments are not provided. However,

the observed resonances are useful for compound identiﬁcation

and, when possible, the chemical shifts are included in the

Experimental section. Metal complexes have been characterized

by FT-IR and UV-visible spectroscopies, elemental analysis, and

magnetic susceptibility; data for each compound can be found in

the Experimental section.

Solid state structures

To better understand the nature of metal–ligand binding, we

determined the X-ray crystal structures of several derivatives. We

were especially interested in the detailed coordination geometry

of the metal complex and the extent of metal–pendant donor

interactions. X-Ray crystallographic data for (3a)FeCl₂, (2c)FeCl₂,

(2d)FeCl₂, (2a)MnCl₂, and (3g)FeCl₂are presented in Table 1;

important distances and angles for these compounds are collected

in Table 2. In all cases, the metal complexes adopt 5-coordinate ge-

ometries, which can be described as distorted square pyramidal or

trigonal bipyramidal structures depending on the pendant donor

and the imino-aryl substituents. For all structurally characterized

=

complexes, the formal C N double bond character of the imino

=

groups is preserved, as evidenced by C N bond lengths in the

˚

range from 1.266(7) A to 1.304(9) A.

Complexes supported by tridentate NNN ligands

Crystals suitable for X-ray crystallography were grown by slow

cooling or slow evaporation of acetonitrile solutions of the

respective compounds. The molecular structures of iron com-

plexes (3a)FeCl₂, (2c)FeCl₂, (2d)FeCl₂and manganese complex

This journal is

The Royal Society of Chemistry 2007

Dalton Trans., 2007, 2547–2562 | 2551

©

View Article Online

(2a)MnCl₂are shown in Fig. 2–5, respectively. Each compound

is square pyramidal, with varying degrees of distortion towards

trigonal bipyramidal geometry. In the square pyramidal descrip-

tion, the three nitrogen atoms and one chlorine atom reside in

the basal plane with the other chlorine occupying the apical

position. The square pyramidal descriptions are supported by

values of the shape-deﬁning parameter, s, which are less than

0.43 in all cases, s = 1.0 describes a perfectly trigonal bipyramidal

structure.^17,18The four basal atoms are coplanar to within 0.126 A

˚

for (3a)FeCl₂, 0.026 A and 0.085 A for (2c)FeCl₂(two independent

˚

molecules), and 0.110 A for (2a)MnCl₂. For each complex, the

metal atom is raised out of the plane by 0.53 A, 0.58 A and

˚

0.56 A (two independent molecules), and 0.73 A, respectively. For

(2d)FeCl₂, deviations from the mean plane are more signiﬁcant

˚

(max. deviation = 0.833 and 0.417 A for the two independent

˚

molecules), with Fe raised out of this plane by 0.523 and 0.648 A,

Fig. 3 Thermal ellipsoid representation (35% probability boundaries) of

the X-ray crystal structure of (2c)FeCl₂with hydrogen atoms omitted for

clarity.

respectively. Among these four NNN complexes, (2d)FeCl₂can

also be described as a distorted trigonal bipyramid, with N(1),

Cl(1) and Cl(2) occupying the trigonal plane and N(2) and N(3)

positioned in the axial positions. In such a depiction, the Fe atom

resides nearly in the plane formed by N(1), N(2), and N(3) and

this plane almost perfectly bisects the Cl–Fe–Cl angle.

Fig. 4 Thermal ellipsoid representation (35% probability boundaries) of

the X-ray crystal structure of (2d)FeCl₂with hydrogen atoms omitted for

clarity.

Fig. 2 Thermal ellipsoid representation (35% probability boundaries) of

the X-ray crystal structure of (3a)FeCl₂with hydrogen atoms omitted for

clarity.

Complexes with NNN ligation share features in common with

structurally characterized iron(II)–PBI complexes. For example,

Fe–PBI complexes typically display shorter distances from Fe

˚

to the central pyridyl nitrogen (0.070 A to 0.162 A shorter for

seven structurally characterized complexes)^3,4a,5,6than to either

of the imino nitrogens, presumably due to the ligand’s enforced

mer coordination geometry. A similar trend is observed for the

new NNN ligated complexes reported here, the Fe–N distance

to the central imino nitrogen is shorter (ranging from 0.027 to

˚

0.317 A shorter) than to either of the other nitrogen donors. Other

structural similarities include the tendency of the imino-aryl

rings to orient nearly orthogonal to the plane containing the

metal and nitrogen atoms. Such an arrangement minimizes

steric interactions between aryl substituents and the remainder

of the molecule. Furthermore, many PBI-supported complexes

Fig. 5 Thermal ellipsoid representation (35% probability boundaries) of

the X-ray crystal structure of (2a)MnCl₂with hydrogen atoms omitted for

clarity.

2552 | Dalton Trans., 2007, 2547–2562

This journal is

The Royal Society of Chemistry 2007

©

View Article Online

19,20

display elongation along the axial Fe–Cl bond, with this

˚

iron(II)-phosphine complexes (typically 2.23–2.30 A).

Also

˚

bond typically 0.03 to 0.05 A longer than its equatorial

consistent with its ﬁve-coordinate geometry, the Fe(II)–P distance

counterpart. This elongation is also observed in (3a)FeCl₂and

in one of the two molecules in the asymmetric units of both

(2c)FeCl₂and (2d)FeCl₂. However, for (2c)FeCl₂and (2d)FeCl₂,

the second molecule in each asymmetric unit displays slight axial

in (3g)FeCl₂falls between that observed for high spin octahedral

21,22

˚

(typically 2.58 to 2.71 A)

and tetrahedral (typically 2.41 to

23,24

˚

2.47 A)

compounds. As with structurally characterized NNN

complexes, (3g)FeCl₂displays axial Fe–Cl elongation, with the

˚

compression, with the axial Fe–Cl bond being 0.016 and 0.014 A

˚

axial bond being 0.026 A longer than its equatorial counterpart.

shorter than the respective equatorial bond. The origin of this

reversal is unclear as there are no obvious intra- or intermolecular

interactions that would cause the observed axial compression.

Complexes (3a)FeCl₂(Fig. 2) and (2a)MnCl₂(Fig. 5) provide a

direct comparison between Fe(II) and Mn(II) chloride complexes

supported by nearly equivalent NNN ligands. As expected, the

Mn–N and Mn–Cl distances (Table 2) are slightly longer than the

corresponding Fe–N and Fe–Cl distances due to the slightly larger

radius of the manganese(II) ion. Other metrical parameters are

similar, except that, in (2a)MnCl₂, there is no discernable difference

However, in contrast to NNN ligated complexes, the distance from

˚

Fe to the central imino nitrogen (2.181(6) A) is not statistically

˚

different from that to the adjacent imino nitrogen (2.166(6) A).

The relatively long Fe–P bond apparently allows the central imino

nitrogen to pull back slightly, resulting in nearly equivalent Fe–

N(imino) distances.

Complexes supported by tridentate NNS ligands

Complexes supported by a-diimine ligands with pendant thioether

donors (NNS ligands) display far more variety in their coordi-

nation geometries. In-situ reactions between monoimine ligand

1, FeCl₂and several bulky aniline derivatives failed to produce

isolable metal complexes. However, when 1 was combined with one

equivalent of FeCl₂and N,N-dimethylethylenediamine, complex

(1a)FeCl₂was isolated. A crystal suitable for X-ray crystal-

lography was grown by diffusing pentane into a concentrated

dichloromethane solution. X-Ray data collection parameters are

given in Table 3 and selected distances and angles are provided in

Table 4. The molecular structure of (1a)FeCl₂is shown in Fig. 7,

which clearly shows the presence of a dangling thioether group and

a ﬁrmly bound N,N-dimethylamine donor. The bonding parame-

ters of (1a)FeCl₂are very similar to those of (3a)FeCl₂(Fig. 2 and

Table 2), including a short Fe–N bond to the central imino nitrogen

˚

between axial and equatorial Mn–Cl distances (2.3524(5) A and

˚

2.3495(5) A), in contrast to the axial elongation noted above for

(3a)FeCl₂.

Finally, the morpholine derivative (2d)FeCl₂possesses a com-

˚

paratively long Fe–N (pendant donor) bond (2.358(5) A and

˚

2.356(5) A for the two independent molecules) relative to the other

structurally characterized NNN complexes. The reason for this

elongation is unclear as there do not appear to be any obvious

interactions between the morpholine ring and the remainder of

the molecule.

Complexes supported by tridentate NNP ligands

Crystals of NNP-ligated complex (3g)FeCl₂were grown by slow

evaporation of an acetonitrile solution, which resulted in two

acetonitrile molecules in the unit cell. The structure of complex

(3g)FeCl₂is shown in Fig. 6; selected distances and angles can

be found in Table 2. The 5-coordinate geometry of (3g)FeCl₂

is adequately described as a distorted square pyramid, with

˚

(2.090(6) A) and slight elongation of the axial Fe–Cl bond. Thus,

the external thioether substituent causes only minor perturbations

to the now familiar square pyramidal NNN structure. The inability

of sulfur to bind to iron in (1a)FeCl₂is likely due to the

unfavorable steric interactions noted previously between aryl and

acenaphthene groups, but may also be due to inherently poor

coordinating ability of a thioether sulfur. In order to probe this

latter possibility, thioethers linked by alkyl spacers to the ace-

naphthenediimine backbone are required. The necessary thioether

˚

the Fe atom raised 0.606 A out of the mean basal plane, the

maximum deviation of the four basal atoms from their mean

˚

plane is 0.18 A. Consistent with the observed high spin character

of (3g)FeCl₂(S = 2, l_eff= 4.7 l_B), the Fe(II)–P distance of

˚

2.497(2) A is signiﬁcantly longer than that exhibited by low spin 4-

coordinate (square planar) or low-spin 6-coordinate (octahedral)

Fig. 6 Thermal ellipsoid representation (35% probability boundaries) of

the X-ray crystal structure of (3g)FeCl₂with hydrogen atoms omitted for

clarity.

Fig. 7 Thermal ellipsoid representation (35% probability boundaries) of

the X-ray crystal structure of (1a)FeCl₂with hydrogen atoms omitted for

clarity.

This journal is

The Royal Society of Chemistry 2007

Dalton Trans., 2007, 2547–2562 | 2553

©

View Article Online

Table 3 Summary of X-ray crystallographic data for NNS-ligated metal complexes^a

Complex

(1a)FeCl₂

(2i)FeCl₂(MeCN)·1.5 MeCN

[(2i)FeCl₂]₂·CH₂Cl₂

(3j)FeCl₂·0.5 THF

Empirical formula

FW

crystal system

C₂₈H₂₅Cl₂FeN₃S

562.32

C₃₃H_30.50Cl₃FeN_4.50

684.38

Monoclinic

S

C₂₉H₂₅Cl₅FeN₂S

666.67

Triclinic

¯

P1

10.095(1)

13.842(1)

20.837(3)

92.33(2)

92.33(1)

92.78(2)

2903.2(5)

4

C₃₆H₄₀Cl₂FeN₂O_0.50

667.51

Monoclinic

S

Trigonal (rhombohedral)

¯

space group

R3 (hexagonal axes)

P2₁/n

P2₁/c

˚

a/A

38.587(6)

10.062(2)

90

120

12975(4)

18

1.295

10.464(2)

25.493(3)

12.312(2)

90

95.58(2)

90

3268.8(9)

4

1.391

11.8801(13)

17.027(2)

16.671(2)

90

92.91(2)

90

3367.9(7)

4

1.316

˚

b/A

˚

c/A

˚

a/A

˚

b/A

˚

c /A

3

˚

V/A

Z

d

_calcd/Mg m⁻³

1.525

Crystal size/mm

0.40 × 0.20 × 0.10

0.42 × 0.40 × 0.36

0.42 × 0.18 × 0.03

0.30 × 0.22 × 0.20

Abs. Coeff./mm⁻¹

0.801

1.075

0.697

2h max/^◦

51.96

53.94

50.94

49.42

Transmission range

No. of reﬂns collected

No. of indep reﬂns

No. of obsd reﬂections

No. of variables

0.9242–0.7400

5914

5649

2587

316

0.0820 (0.2008)

1.007

0.696, −0.479

0.7615–0.7297

7488

7106

4077

399

0.0741 (0.1331)

1.005

0.458, −0.855

1.0–0.7815

12 053

11 373

6848

689

0.0665 (0.1442)

1.008

1.057, −0.778

1.000–0.8197

6026

5731

4239

412

0.0438 (0.0957)

1.016

0.373, −0.324

R1 (wR2)^b[I > 2r(I)]

Goodness-of-ﬁt (F²)

−

−3

˚

Diff. Peaks/e A

^aSee Experimental section for additional data collection, reduction, and structure solution and reﬁnement details. ^bR1 = R ꢀF_o|−|F_cꢀ/R |F_o|; wR2 =

[R[w(F_o−F_c²)²]]^1/2where w = 1/r²(F_o²) + (aP)²+ bP.

2

amine precursors, 2-(4-chlorophenylthio)ethylamine and 2-(3,5-

dimethylphenylthio)ethylamine, are readily prepared by reaction

of 2-bromoethylamine hydrobromide with the appropriate thiol in

the presence of excess K₂CO₃.²⁵Distillation of the crude oils (under

vacuum) provided thioether amines that were judged to be pure

by GC-MS as well as ¹H and ¹³C NMR spectroscopic data. In-situ

reactions, as outlined in Scheme 5, using these thioether amines

provided the sulfur containing metal complexes diagrammed in

Scheme 6. Three derivatives were found to be acceptable for

X-ray crystallographic analysis: (2i)FeCl₂(MeCN), [(2i)FeCl₂]₂,

and (3j)FeCl₂. Data collection parameters are contained in

Table 3 and important distances and angles are presented in

Fig. 8 Thermal ellipsoid representation (35% probability boundaries)

of the X-ray crystal structure of (2i)FeCl₂(MeCN) with hydrogen atoms

omitted for clarity.

Table 4. Crystals of complex (2i)FeCl₂(MeCN) were grown by

slow evaporation of an acetonitrile solution, giving rise to the

solid state structure shown in Fig. 8, with the a-diimine ligand

bound in a bidentate fashion and acetonitrile occupying the ﬁfth

coordination site at iron. Thus, the strong r-donating acetonitrile

effectively blocks sulfur from coordinating to iron (the Fe–S

Fe–S interactions when (2i)FeCl₂is dissolved in non-coordinating

solvents. We considered that the 4-chlorophenyl substituent might

withdraw electron density from sulfur and contribute to weak

sulfur ligation, and that aryl substituents with electron donating

groups might enhance the nucleophilicity of sulfur. To test this

hypothesis, we prepared the 3,5-dimethylphenyl substituted

thioether and used it to synthesize metal complex (3j)FeCl₂.

When crystallized from tetrahydrofuran (pentane diffusion),

crystals suitable for X-ray diffraction were obtained. The

molecular structure of (3j)FeCl₂, shown in Fig. 10, indeed shows

a monomeric square pyramidal complex, with a long Fe–S bond

˚

distance in (2i)FeCl₂(MeCN) is >5.6 A). Also of note, with the

compound freed from the constraints imposed by a tridentate

ligand arrangement, (2i)FeCl₂(MeCN) adopts a pseudo-trigonal

bipyramidal structure (s = 0.91) with nearly identical Fe–Cl bond

˚

lengths (2.3058(15) and 2.3161(14) A).

We reasoned that crystallization of (2i)FeCl₂from a poorly

coordinating solvent might allow sulfur ligation; crystallization of

(2i)FeCl₂from dichloromethane (via pentane diffusion) provided

X-ray quality crystals. However, rather than the hoped for sulfur-

ligated compound, the solid state structure, shown in Fig. 9, is

that of a chloride-bridged dimer, [(2i)FeCl₂]₂. Thus, at least in

the solid state, chloride is a more effective donor to iron than the

thioether sulfur atom. Although sulfur does not coordinate to

iron in the solid state, we cannot rule out the presence of weak

˚

of 2.7126(10) A as compared to other structurally characterized

iron–thioether complexes. A search of the Cambridge Structural

Database (November 2005 release)²⁶reveals 271 compounds

containing iron–thioether bonds—the majority of these being

iron(II) complexes. This collection has a mean Fe–S bond length of

2554 | Dalton Trans., 2007, 2547–2562

This journal is

The Royal Society of Chemistry 2007

©

View Article Online

Fig. 9 Thermal ellipsoid representation (35% probability boundaries) of

the X-ray crystal structure of [(2i)FeCl₂]₂with hydrogen atoms omitted for

clarity.

Fig. 10 Thermal ellipsoid representation (35% probability boundaries)

of the X-ray crystal structure of (3j)FeCl₂with hydrogen atoms omitted

for clarity.

˚

2.300 A (min 2.141 A, max 2.806 A) and only two complexes in the

database have longer Fe–S bonds than that in (3j)FeCl₂. One, an

iron(II) compound supported by a PBI-based dithia macrocycle,

{[(2,15-dimethyl-7,10-dithia-3,14,20-triazabicyclo(14.3.1)eicosa-

1(20),2,14,16,18-pentaene-N,N^ꢁ,N^ꢁꢁ,S)]FeCl(MeOH)}ClO₄, pos-

27

˚

sesses the longest Fe–S bond observed to date (2.806 A). The

second, {[N,N^ꢁ-bis(2-(2-mercaptophenylthio)ethyl)piperazine]Fe},

ligated by a piperazine-based thiolate–thioether ligand system,

28

˚

displays 2.755 A Fe–S(thioether) linkages. Thus, the Fe–S bond

in (3j)FeCl₂is unusually long and is indicative of a relatively weak

Fe–S interaction. We presume that acetonitrile would ligate to

iron, analogous to (2i)FeCl₂(MeCN), if (3j)FeCl₂were recrystal-

lized from this solvent.

The successful isolation of (3j)FeCl₂suggests that other

monomeric, sulfur-bound complexes would form with appro-

priately chosen alkyl- or aryl-substituted thioethers. We are

continuing to examine aryl- and alkylthioether derivatives to

further tune the nucleophilicity of sulfur and to better understand

the role of Fe–S ligation in ethylene oligomerization reactions

catalyzed by iron complexes. This work will be the subject of future

This journal is

The Royal Society of Chemistry 2007

Dalton Trans., 2007, 2547–2562 | 2555

©

View Article Online

reports. In the meantime, we have discovered that soft donors (like

P and S) have a large impact on the catalytic performance of these

metal complexes, both in terms of activity and oligomer product

properties. Our initial ethylene polymerization results in this area

are described in detail elsewhere.²⁹

134.32, 132.28, 131.07, 129.08, 40.84, 38.29. EI mass spectrum,

m/z 187 [M⁺].

2-(3,5-Dimethylphenylthio)ethylamine. To a stirring mixture

of 1.93 g (9.4 mmol) of 2-bromoethylamine hydrobromide and

3.00 g (21.7 mmol) of K₂CO₃in 30 mL of CH₂Cl₂was added

1.00 g (7.2 mmol) of 3,5-dimethylbenzenethiol. The mixture was

stirred at room temperature under argon for 2 d. The mixture was

washed twice with distilled water, dried with MgSO₄and ﬁltered.

Volatiles were removed in vacuo leaving a slightly cloudy yellow

oil. Distillation of the oil under reduced pressure (0.10 Torr) at

80–95 ^◦C produced 0.547 g (41.7%) of a clear liquid which was

identiﬁed as the desired product by its ¹H and ¹³C NMR spectra.

¹H NMR (400 MHz, CDCl₃): d = 6.98 (2 H, s), 6.82 (1 H, s), 2.98

Summary

We have prepared ﬁve-coordinate metal(II) halide complexes

supported by a new family of donor-modiﬁed a-diimine lig-

ands. Compounds with NNN, NNO, NNP and NNS ligands

have been prepared and selected examples of NNN, NNP and

NNS derivatives have been structurally characterized by X-ray

crystallography. All isolated metal complexes possess high spin

electronic conﬁgurations. Whereas the structurally characterized

NNN and NNP complexes invariably display square pyramidal

coordination geometries, NNS-ligated complexes exhibit a range

of coordination behaviors (from dimeric, to solvent-ligated, to

weakly sulfur-ligated) depending on the crystallization solvent and

the apparent nucleophilicity of the thioether sulfur atom.

1

(2 H, t), 2.89 (2 H, t), 2.27 (6 H, s), 1.55 (2 H, br s). ¹³C{ H} NMR

(100 MHz, CDCl₃): d = 138.58, 135.10, 128.21, 127.50, 41.00,

38.07, 21.29. EI mass spectrum, m/z 181 [M⁺].

Synthesis of mono-imine ligands

(E)-2-(2-(Phenylthio)phenylimino)acenaphthylen-1(2H)-one (1).

An ethanol (200 mL) solution of acenaphthenequinone (10.0 g,

54.9 mmol) was treated with 1 mL of formic acid, followed by

slow, dropwise addition (over approx. 8 h) of a solution of 2-

(phenylthio)aniline (11.0 g, 54.9 mmol) in 120 mL of ethanol.

The resulting mixture was stirred at 60 ^◦C overnight, cooled

to room temperature and ﬁltered to remove unreacted acenaph-

Experimental

General considerations

Unless otherwise stated, all operations were carried out under

argon in a glovebox or using standard Schlenk techniques.

Tetrahydrofuran, diethylether, dichloromethane, acetonitrile and

pentane were puriﬁed by standard drying procedures and stored

over activated molecular sieves prior to use. Other reagents were

obtained commercially from Aldrich Chemical Company or Acros

Organics and used as received. Atlantic Microlab, Inc. (Norcross,

GA) performed elemental analyses. Proton and NMR spectra

were obtained using either JEOL Eclipse or Bruker AVANCE II

400 MHz spectrometers operating at room temperature. GC-MS

spectra were obtained using electron impact (EI) on an HP 5890

gas chromatograph coupled to an HP 5970 mass selective detector.

Magnetic susceptibilities were determined at room temperature us-

ing a Johnson Matthey magnetic susceptibility balance. Electronic

absorption spectra were recorded on a Hewlett-Packard 8453

diode array spectrophotometer (190–1100 nm range). Samples

for IR were dispersed in KBr and spectra were recorded on a

Nicolet 5DXC spectrometer with a diffuse reﬂectance (DRIFTS)

attachment.

◦

thenequinone. The ﬁltrate was cooled to −10 C overnight. The

red–orange solid that deposited was ﬁltered, washed with ether and

dried to yield 6.68 g (33.3%) of product. The room temperature ¹H

NMR spectrum appears to be a superposition of spectra (approx.

10 : 1 ratio), perhaps resulting from conformers that interconvert

slowly on the NMR timescale due to hindered rotation about the

imino aryl substituent. The following NMR data correspond to

1

the predominant isomer. H NMR (400 MHz, CDCl₃): d = 8.14

(2 H, t), 7.98 (1 H, d), 7.79 (1 H, t), 7.43-7.18 (5 H, m), 7.03-

1

6.8 (4 H, m), 6.91(2 H, d). ¹³C{ H} NMR (100 MHz, CDCl₃):

d = 189.29, 160.10, 150.32, 143.20, 134.74, 132.41, 132.10, 131.28,

130.93, 130.62, 129.55, 128.95, 128.87, 128.27, 128.09, 127.95,

127.17, 125.90, 125.78, 123.40, 122.12, 118.27. Anal. Calc. (Found)

for C₂₄H₁₅NOS: C, 78.88 (78.47); H, 4.14 (4.10); N, 3.83 (3.83). EI

mass spectrum, m/z 365 [M⁺]. Single crystal X-ray crystallography

conﬁrmed the monoimine structure.

(2E)-2-[(2,6-Dimethylphenyl)imino]acenaphthylen-1(2H )-one

(2). An ethanol (200 mL) solution of acenaphthenequinone

(10.0 g, 55 mmol) was treated with 0.5 mL of formic acid, followed

by slow, dropwise addition (over approx. 12 h) of a solution of

2,6 dimethylaniline (6.8 mL, 55 mmol) in 60 mL of ethanol. The

resulting mixture was stirred at 60 ^◦C overnight, cooled to room

temperature and ﬁltered. The ﬁltrate was cooled to 0 ^◦C. After

3 d, the orange solid that deposited was ﬁltered, washed with

cold methanol and dried, yielding 4.91 g of pure product. Slow

evaporation of the remaining ﬁltrate yielded an additional 5.87 g of

orange solid for a total yield of 10.78 g (69%). ¹H NMR (400 MHz,

CDCl₃): d = 8.18 (2 H, d), 8.00 (1 H, d), 7.82 (1 H, t), 7.43 (1 H,

Synthesis of 2-(thioether)ethylamines

2-(4-Chlorophenylthio)ethylamine. To a stirring mixture of

4.07 g (20.0 mmol) of 2-bromoethylamine hydrobromide and

6.00 g (43.4 mmol) of K₂CO₃in 30 mL of CH₂Cl₂was added

2.89 g (20.0 mmol) of 4-chlorobenzenethiol. The mixture was

stirred at room temperature under argon for 2 d. The mixture was

washed twice with distilled water, dried with MgSO₄and ﬁltered.

Volatiles were removed in vacuo leaving a slightly cloudy yellow

oil. Distillation of the oil under reduced pressure (0.10 Torr) at

80–100 ^◦C produced 1.99 g (54.7%) of a clear liquid which was

identiﬁed as the desired product by its ¹H and ¹³C NMR spectra.

¹H NMR (400 MHz, CDCl₃): d = 7.25 (4 H, m), 2.97 (2 H, t), 2.89

1

t), 7.15-7.05 (3 H, m), 6.70 (1 H, d), 2.04 (6 H, s). ¹³C{ H} NMR

(100 MHz, CDCl₃): d = 189.64, 160.07, 148.51, 142.95, 132.11,

1

(2 H, t), 1.28 (2 H, br s). ¹³C{ H} NMR (100 MHz, CDCl₃): d =

130.97, 130.91, 129.41, 128.49, 128.32, 128.28, 127.74, 124.60,

2556 | Dalton Trans., 2007, 2547–2562

This journal is

The Royal Society of Chemistry 2007

©

View Article Online

124.33, 122.53, 122.19, 17.89. Anal. Calc. (Found) for C₂₀H₁₅NO:

C, 84.19 (83.61); H, 5.30 (5.52); N, 4.91 (5.20). EI mass spectrum,

m/z 285 [M⁺].

(400 MHz, CDCl₃): d = 8.17 (2 H, d), 7.98 (1 H, d), 7.81 (1 H,

t), 7.53 (1 H, d), 7.42 (1 H, t), 7.22 (2 H, m), 6.85 (1 H, d), 6.77

1

(1 H, d), 1.32 (9 H, s). ¹³C{ H} NMR (100 MHz, CDCl₃): d =

189.66, 158.42, 149.69, 143.46, 139.70, 132.06, 131.06, 130.81,

129.06, 128.19, 127.97, 126.92, 126.63, 126.34, 125.37, 123.80,

122.11, 118.25, 35.46, 29.82. Anal. Calc. (Found) for C₂₂H₁₉NO:

C, 84.31 (83.82); H, 6.11 (6.32); N, 4.47 (4.46). EI mass spectrum,

m/z 313 [M⁺].

(2E)-2-[(2,6-Diisopropylphenyl)imino]acenaphthylen-1(2H)-one

(3). An ethanol (65 mL) solution of acenaphthenequinone (2.0 g,

11 mmol) was treated with 1 mL of formic acid, followed by

slow, dropwise addition (over approx. 8 h) of a solution of 2,6

diisopropylaniline (1.6 mL, 8.3 mmol) in 65 mL of ethanol. The

resulting mixture was stirred at 60 ^◦C overnight, cooled and

ﬁltered. After removal of solvent under vacuum, the resulting

orange solid was dissolved in ether, ﬁltered and cooled to −10 ^◦C

overnight. The orange solid that deposited was ﬁltered, washed

with cold ether and dried. The ﬁltrate was again cooled to −10 ^◦C

overnight and more orange solid was isolated, giving a total yield

(E)-2-(2-Ethylphenylimino)acenaphthylen-1(2H)-one (6). An

ethanol (70 mL) solution of acenaphthenequinone (2.00 g,

11.0 mmol) was treated with 1 mL of formic acid, followed

by dropwise addition (over approx. 5 h) of a solution of 2-

ethylaniline (1.0 mL, 8.3 mmol) in 70 mL of ethanol. The

resulting orange mixture was stirred at 60 ^◦C overnight. Unreacted

acenaphthenequinone was removed by ﬁltration and the ﬁltrate

was cooled to −10 ^◦C overnight, yielding 0.100 g of orange

solid. Further concentration of the ﬁltrate and cooling to −10 ^◦C

deposited an additional 0.652 g of orange product, giving a total

1

of 1.91 g (68.5%). H NMR (400 MHz, CDCl₃): d = 8.18 (2 H,

m), 7.99 (1 H, d), 7.81 (1 H, t), 7.39 (1 H, t), 7.25 (3 H, m), 6.62

1

(1 H, d), 2.82 (2 H, m), 1.16 (6 H, d), 0.88 (6 H, d). H NMR

(400 MHz, MeCN-d₃): d = 8.27 (1 H, d), 8.12 (2 H, m), 7.87 (1 H,

1

yield of 0.752 g (31.9%). H NMR (400 MHz, CDCl₃): d = 8.17

t), 7.45 (1 H, t), 7.30 (3 H, m), 6.60 (1 H, d), 2.79 (2 H, septet),

1

1.10 (6 H, d), 0.89 (6 H, d). ¹³C{ H} NMR (100 MHz, CDCl₃):

(2 H, d), 7.99 (1 H, d), 7.82 (1 H, t), 7.42 (1 H, t), 7.36 (1 H,

m), 7.25 (2 H, m), 6.87 (1 H, m), 6.84 (1 H, d), 2.54 (2 H, q),

d = 189.67, 160.54, 146.53, 143.11, 135.30, 132.23, 131.07, 130.94,

129.41, 128.34, 128.17, 127.70, 125.06, 123.56, 123.39, 122.28,

28.43, 23.47. Anal. Calc. (Found) for C₂₄H₂₃NO: C, 84.42 (84.34);

H, 6.79 (6.81); N, 4.10 (4.41). EI mass spectrum, m/z 341 [M⁺].

1

1.05 (3 H, t). ¹³C{ H} NMR (100 MHz, CDCl₃): d = 189.69,

159.44, 149.14, 143.36, 132.80, 132.12, 131.04, 130.79, 129.23,

129.07, 128.24, 128.06, 127.22, 126.58, 125.36, 123.49, 122.20,

117.00, 24.57, 14.38. Anal. Calc. (Found) for C₂₀H₁₅NO: C, 84.19

(83.73); H, 5.30 (5.35); N, 4.91 (4.91). EI mass spectrum, m/z 285

[M⁺].

(2E)-2-[(2,5-Di-t-butylphenyl)imino]acenaphthylen-1(2H)-one

(4). An ethanol (65 mL) solution of acenaphthenequinone

(2.00 g, 11.0 mmol) was treated with 1 mL of formic acid, followed

by slow, dropwise addition (over approx. 8 h) of a solution of

2,5-di-t-butylaniline (1.69 g, 8.25 mmol) in 65 mL of ethanol.

The resulting mixture was stirred at 60 ^◦C overnight, cooled and

ﬁltered to remove unreacted acenaphthenequinone. After removal

of solvent under vacuum, the resulting orange solid was dissolved

in ether, ﬁltered and cooled to −10 ^◦C overnight. The orange solid

that deposited was ﬁltered, washed with cold ether and dried.

The ﬁltrate was again cooled to −10 ^◦C overnight and additional

product was isolated, giving a total yield of 2.13 g (70%). ¹H NMR

(400 MHz, CDCl₃): d = 1.24 (9 H, s), 1.31 (9 H, s), 6.80 (1 H, d),

6.88 (1 H, d), 7.23 (1 H, d), 7.40 (1 H, t), 7.44 (1 H, d), 7.81 (1 H,

(E)-2-(2-Isopropyl-6-methylphenylimino)acenaphthylen-1(2H)-

one (7). An ethanol (50 mL) solution of acenaphthenequinone

(2.00 g, 11.0 mmol) was treated with 1 mL of formic acid,

followed by dropwise addition (over approx. 5 h) of a solution

of 2-isopropyl-6-methylaniline (1.3 mL, 8.3 mmol) in 70 mL

of ethanol. The resulting orange mixture was stirred at 60 ^◦C

overnight. Unreacted acenaphthenequinone was removed by

ﬁltration and solvent was removed in vacuo. The orange residue

was dissolved in diethylether, ﬁltered and reduced in volume until

incipient crystallization. Cooling this solution to −10 ^◦C overnight

deposited 0.594 g of orange solid. Further concentration of the

ﬁltrate and cooling to −10 ^◦C deposited an additional 0.094 g of

orange product, giving a total yield of 0.684 g (26.5%). ¹H NMR

(400 MHz, CDCl₃): d = 8.18 (2 H, d), 7.99 (1 H, d), 7.82 (1 H, t),

7.42 (1 H, t), 7.26 (1 H, d), 7.18-7.12 (2 H, m), 6.67 (1 H, d), 2.92

1

t), 7.97 (1 H, d), 8.17 (2 H, d). ¹³C{ H} NMR (100 MHz, CDCl₃):

d = 189.78, 158.33, 149.41, 149.24, 143.43, 136.93, 132.03, 131.09,

130.86, 129.03, 128.79, 128.16, 127.82, 126.59, 123.62, 122.07,

121.85, 115.98, 35.03, 34.43, 31.19, 29.87. Anal. Calc. (Found)

for C₂₆H₂₇NO: C, 84.51 (84.48); H, 7.37 (7.46); N, 3.79 (3.82). EI

mass spectrum, m/z 369 [M⁺].

1

(1 H, septet), 2.00 (3 H, s), 1.20 (3 H, d), 0.89 (3H, s). ¹³C{ H}

NMR (100 MHz, CDCl₃): d = 189.61, 160.32, 147.42, 143.00,

136.00, 132.13, 131.00, 130.90, 129.38, 128.37, 128.28, 128.15,

127.78, 124.70, 124.04, 123.68, 122.80, 122.20, 28.48, 23.79, 22.89,

18.07. Anal. Calc. (Found) for C₂₂H₁₉NO: C, 84.31 (83.81); H,

6.11 (6.14); N, 4.47 (4.40). EI mass spectrum, m/z 313 [M⁺].

(E)-2-(2-t-Butylphenylimino)acenaphthylen-1(2H)-one (5). An

ethanol (80 mL) solution of acenaphthenequinone (2.00 g,

11.0 mmol) was treated with 1 mL of formic acid, followed

by dropwise addition (over approx. 5 h) of a solution of 2-

t-butylaniline (1.3 mL, 8.3 mmol) in 50 mL of ethanol. The

resulting mixture was stirred at 60 ^◦C overnight. Unreacted

acenaphthenequinone was removed by ﬁltration and the ﬁltrate

was evaporated to dryness. The orange residue was dissolved in

diethylether and ﬁltered. The resulting solution was reduced in

Synthesis of metal complexes

[(1a)FeCl₂]. A solution containing 0.10 mL (0.91 mmol) of

N,N-dimethylethylenediamine in 50 mL of anhydrous butanol

was added (via cannula) to 0.333 g (0.91 mmol) of 1 and 0.115 g

(0.91 mmol) of FeCl₂. The initial orange color changed to green

and, after stirring overnight under argon at 55 ^◦C, a black–green

solid deposited, which was ﬁltered, washed with 4 mL of THF, and

◦

volume to the saturation point and cooled to −10 C overnight,

yielding 0.588 g (22.7%) of orange solid. ¹H NMR analysis showed

it to be a 15 : 1 mixture of monoimine : diimine; recrystallization

1

from diethylether did not signiﬁcantly alter this ratio. H NMR

dried to yield 0.411 g (80.3%) of grey–green product. H NMR

This journal is

The Royal Society of Chemistry 2007

Dalton Trans., 2007, 2547–2562 | 2557

©

View Article Online

(400 MHz, MeCN-d₃): d = 29.66, 25.56, 25.39, 24.74, 20.57,

18.69, 7.01, 6.44, 6.28, −0.69, −7.92. FTIR (KBr): 2899, 2876,

1573, 1457, 1288, 1024, 827, 778, 734 cm⁻¹. UV-vis (CH₃CN)

[k_max, nm (e, M⁻¹cm⁻¹)] 796 (3600). Anal. Calc. (Found) for

C₂₈H₂₅N₃SCl₂Fe: C, 59.81 (60.32); H, 4.48 (4.74); N, 7.47 (7.20).

l_eff= 4.7 l_B.

did not reveal any discernable transition in the visible region. Anal.

Calc. (Found) for C₂₆H₂₁N₃Cl₂Mn: C, 62.29 (62.15); H, 4.22 (4.60);

N, 8.38 (7.91). l_eff= 5.1 l_B.

[(2c)FeCl₂]. A solution containing 0.11 mL (0.91 mmol) of 2-

(aminoethyl)pyridine in 50 mL of anhydrous butanol was added

(via cannula) to 0.260 g (0.91 mmol) of 2 and 0.115 g (0.91 mmol)

of FeCl₂. The initially orange solution turned dark green and

deposited a green solid within 2 h. After stirring overnight under

argon at 55 ^◦C the solid was ﬁltered, washed with 6 mL of THF,

[(2a)FeCl₂]. A solution containing 0.10 mL (0.91 mmol) of

N,N-dimethylethylenediamine in 50 mL of anhydrous butanol was

added (via cannula) to 0.260 g (0.91 mmol) of 2 and 0.115 g

(0.91 mmol) of FeCl₂. The initially orange solution turned dark

green within 20 min and deposited a microcrystalline solid after

stirring overnight under argon at 55 ^◦C. The solid was ﬁltered,

washed with 6 mL of THF, and dried to yield 0.354 g (80.7%) of

dark green product. ¹H NMR (400 MHz, MeCN-d₃): d = 26.18,

26.01, 20.88, 19.44, 16.35, 6.1, 5.44, 2.50, −0.93, −5.60. FTIR

(KBr): 2904, 1643, 1463, 1287, 1025, 915, 827, 783, 762 cm⁻¹.

UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 776 (3860). Anal. Calc.

(Found) for C₂₄H₂₅N₃Cl₂Fe: C, 59.78 (59.80); H, 5.23 (5.29); N,

8.71 (8.72). l_eff= 4.9 l_B.

1

and dried to yield 0.269 g (57.2%) of green product. H NMR

(400 MHz, MeCN-d₃): d = 44.2, 22.72, 20.0, 18.14, 17.50, 16.32,

15.86, 3.75, 2.96, 2.8, −2.72. FTIR (KBr): 2953, 2904, 1655, 1594,

1484, 1441, 1282, 1046, 832, 773 cm⁻¹. UV-vis (CH₃CN) [k_max, nm

(e, M⁻¹cm⁻¹)] 730 (2500). Anal. Calc. (Found) for C₂₇H₂₃N₃Cl₂Fe:

C, 62.82 (62.58); H, 4.49 (4.42); N, 8.14 (8.13). l_eff= 5.7 l_B.

[(2d)FeCl₂]. A solution containing 0.10 mL (0.76 mmol) of

4-(2-aminoethyl)morpholine in 50 mL of anhydrous butanol was

added (via cannula) to 0.217 g (0.76 mmol) of 2 and 0.096 g

(0.76 mmol) of FeCl₂. Within 2 h, green solid precipitated out of

solution. After stirring overnight under argon at 55 ^◦C, the solid

was ﬁltered, washed with 4 mL of THF, and dried to yield 0.283 g

(71.0%) of green solid. ¹H NMR (400 MHz, MeCN-d₃): d = 26.02,

18.13, 17.76, 17.50, 6.04, 5.49, 5.16, 4.34, 3.40, −1.71. FTIR (KBr):

2959, 2915, 2871, 1671, 1600, 1277, 1118, 1041, 915, 832, 773 cm⁻¹.

UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 736 (2200). Anal. Calc.

(Found) for C₂₆H₂₇N₃OCl₂Fe: C, 59.57 (59.59); H, 5.19 (5.16); N,

8.01 (7.93).

[(2a)MnCl₂]. A solution containing 0.10 mL (0.91 mmol) of

N,N-dimethylethylenediamine in 50 mL of anhydrous butanol was

added (via cannula) to 0.260 g (0.91 mmol) of 2 and 0.115 g

(0.91 mmol) of MnCl₂. After overnight heating under argon at

55 ^◦C, a tan solid precipitated. This solid was ﬁltered, washed

with 6 mL of toluene, and dried to yield 0.324 g (74.0%) of brown

product. The sample was not sufﬁciently soluble in acetonitrile for

NMR analysis. FTIR (KBr): 2888, 1693, 1649, 1594, 1468, 932,

844, 794, 773 cm⁻¹. UV-vis (CH₃CN) spectroscopy did not reveal

any prominent transitions in the visible region. We were unable to

obtain satisfactory carbon analysis on this manganese derivative,

however an X-ray crystal structure determination conﬁrmed its

formulation as the tridentate complex. Anal. Calc. (Found) for

C₂₄H₂₅N₃Cl₂Mn: C, 59.89 (54.37); H, 5.24 (5.19); N, 8.73 (8.81).

l_eff= 5.5 l_B.

[(2e)FeCl₂]. A solution containing 0.10 mL (0.77 mmol) 2-

phenoxyethylamine in 50 mL of anhydrous butanol was added

(via cannula) to 0.220 g (0.77 mmol) of 2 and 0.098 g (0.77 mmol)

of FeCl₂. Immediately upon addition the solution became a dark

green. After stirring overnight under argon at 55 ^◦C, a green

colored solid was ﬁltered, washed with 4 mL of THF, and dried

1

to yield 0.144 g (35.1%) of green product. H NMR (400 MHz,

[(2b)FeCl₂]. A solution containing 0.10 mL (1.0 mmol) of

2-(aminomethyl)pyridine in 50 mL of anhydrous butanol was

added (via cannula) to 0.285 g (1.0 mmol) of 2 and 0.127 g

(1.0 mmol) of FeCl₂. The initially orange solution turned dark

brown within 20 min and deposited a dark solid after stirring

overnight under argon at 55 ^◦C. The solid was ﬁltered, washed with

6 mL of THF, and dried to yield 0.242 g (48.2%) of product. The

sample was not sufﬁciently soluble in acetonitrile for reliable NMR

characterization. FTIR (KBr): 3058, 2954, 2913, 1646, 1605, 1466,

1286, 829, 767 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)]

504 (1500). This compound analyzed consistently low for carbon.

Anal. Calc. (Found) for C₂₆H₂₁N₃Cl₂Fe: C, 62.18 (59.17); H, 4.21

(4.07); N, 8.37 (8.77). l_eff= 4.6 l_B.

MeCN-d₃): d = 12.06, 11.26, 10.17, 8.63, 7.20, 6.78, 5.73, 5.20,

−0.36, −5.78, −11.93. FTIR (KBr): 3038, 2920, 1598, 1501, 1245,

1051, 781, 698 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 648

(430). Anal. Calc. (Found) for C₂₈H₂₄N₂OCl₂Fe: C, 63.30 (63.49);

H, 4.55 (4.55); N, 5.27 (5.36).

[(2g)FeCl₂]. A solution of 0.211 g (0.92 mmol) of 2-

(diphenylphosphino)ethylamine in 40 mL of anhydrous butanol

was added via cannula to 0.263 g (0.92 mmol) of 2 and 0.117 g

(0.92 mmol) of anhydrous FeCl₂. The solution was stirred

overnight under argon at 55 ^◦C. The solid that formed was ﬁltered,

washed with a small amount of THF, and dried to give 0.380 g

1

(66%) of dark green product. H NMR (400 MHz, MeCN-d₃):

d = 39.7, 16.44, 14.15, 13.74, 12.9, 12.5, 8.0, 7.43, 7.1, 6.2, 5.5,

−1.42, −6.3. FTIR (KBr): 2953, 2910, 1655, 1605, 1490, 1441,

838, 789, 739, 701 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)]

752 (1300). Anal. Calc. (Found) for C₃₄H₂₉N₂PCl₂Fe: C, 65.51

(65.07); H, 4.69 (4.68); N, 4.49 (4.49). l_eff= 5.0 l_B.

[(2b)MnCl₂]. A solution containing 0.16 mL (1.6 mmol) of 2-

(aminomethyl)pyridine in 50 mL of anhydrous butanol was added

(via cannula) to 0.450 g (1.58 mmol) of 2 and 0.198 g (1.58 mmol)

of MnCl₂. After overnight heating under argon at 55 ^◦C, a brown

solid precipitated. This solid was ﬁltered, washed with 6 mL

of THF, and dried to yield 0.450 g (56.8%) of brown product.

The sample was not sufﬁciently soluble in acetonitrile for NMR

analysis. FTIR (KBr): 3058, 2920, 1688, 1646, 1591, 1473, 1438,

1362, 1293, 1044, 836, 781 cm⁻¹. UV-vis (CH₃CN) spectroscopy

[(2g)CoCl₂]. A solution containing 0.242 g (1.05 mmol) 2-

(diphenylphosphino)ethylamine in 40 mL of anhydrous butanol

was added (via cannula) to 0.301 g (0.96 mmol) of 2 and 0.137 g

(1.05 mmol) of CoCl₂. After stirring overnight under argon at

2558 | Dalton Trans., 2007, 2547–2562

This journal is

The Royal Society of Chemistry 2007

©

View Article Online

55 ^◦C, the solution was cooled and the resulting solid was ﬁltered,

washed with diethylether, and dried to yield 0.363 g (55.1%) of

rusty-brown product. The sample was not sufﬁciently soluble in

acetonitrile for NMR analysis. FTIR (KBr): 2910, 1666, 1627,

1584, 1484, 1436, 838, 778, 740, 696 cm⁻¹. UV-vis spectroscopy

did not reveal any discernable transition in the visible region.

Anal. Calc. (Found) for C₃₄H₂₉N₂PCl₂Co: C, 65.19 (64.49); H,

4.67 (4.73); N, 4.47 (4.41). l_eff= 4.4 l_B.

808 (2000). This compound analyzed satisfactorily as the THF

monosolvate. Anal. Calc. (Found) for C₃₄H₃₇N₃OCl₂Fe: C, 64.78

(64.48); H, 5.92 (6.09); N, 6.67 (6.61). l_eff= 5.0 l_B.

[(3c)FeCl₂]. A solution containing 0.11 mL (0.91 mmol) of 2-

(aminoethyl)pyridine in 50 mL of anhydrous butanol was added

(via cannula) to 0.310 g (0.91 mmol) of 3 and 0.115 g (0.91 mmol)

of FeCl₂. The initially orange solution turned dark brown and

deposited a green solid after stirring overnight under argon at

55 ^◦C. The solid was ﬁltered, washed with 6 mL of THF, and dried

[(2h)FeCl₂]. A solution containing 0.229 g (0.94 mmol) 3-

(diphenylphosphino)-1-propylamine in 50 mL of anhydrous bu-

tanol was added (via cannula) to 0.268 g (0.94 mmol) of 2 and

0.119 g (0.94 mmol) of FeCl₂. After stirring overnight under argon

1

to yield 0.290 g (55.7%) of green microcrystalline product. H

NMR (400 MHz, MeCN-d₃): d = 45.5, 26.25, 21.70, 21.0, 18.10,

5.85, 5.50, 3.46, 3.00, −1.38, −3.23. FTIR (KBr): 2920, 1653,

1598, 1438, 1313, 1182, 829, 766 cm⁻¹. UV-vis (CH₃CN) [k_max, nm

(e, M⁻¹cm⁻¹)] 720 (2400). Anal. Calc. (Found) for C₃₁H₃₁N₃Cl₂Fe:

C, 65.05 (64.81); H, 5.46 (5.55); N, 7.34 (7.30). l_eff= 5.3 l_B.

◦

at 55 C, the solution was reduced in volume to 10 mL and the

resulting solid was ﬁltered, washed with 4 mL of THF, and dried

to yield 0.406 g (67.7%) of aqua-colored product. The compound

was not sufﬁciently soluble in acetonitrile for reliable NMR

characterization. FTIR (KBr): 2950, 1642, 1594, 1474, 1436, 832,

789, 745, 701 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 714

(1100). This compound analyzed as the monohydrate. Anal. Calc.

(Found) for C₃₅H₃₃N₂OPCl₂Fe: C, 64.14 (63.02); H, 5.07 (4.90); N,

4.27 (3.91); O, 2.44 (2.99).

[(3d)FeCl₂]. A solution containing 0.10 mL (0.76 mmol) 4-

(2-aminoethyl)morpholine in 50 mL of anhydrous butanol was

added (via cannula) to 0.260 g (0.76 mmol) of 3 and 0.096 g

(0.76 mmol) of FeCl₂. The initial orange color changed to green

◦

and, after stirring overnight under argon at 55 C, a green solid

deposited, which was ﬁltered, washed with 4 mL of THF, and dried

1

[(2i)FeCl₂]. A solution containing 0.202 g (0.928 mmol) of 2-

(4-chlorophenylthio)ethylamine in 10 mL of anhydrous butanol

was added (via cannula) to 0.265 g (0.928 mmol) of 2 and 0.118 g

(0.928 mmol) of FeCl₂in 40 mL of anhydrous butanol. The initially

orange solution turned dark green and deposited a green solid

within 2 h. After overnight stirring under argon at 55 ^◦C, the green

solid was ﬁltered, washed with 6 mL of THF, and dried to yield

0.203 g (37.5%) of product. ¹H NMR (400 MHz, MeCN-d₃): d =

11.79, 11.39, 9.93, 9.50, 7.24, 4.88, −3.03, −11.80. FTIR (KBr):

2939, 2915, 1665, 1633, 1584, 1479, 1101, 833, 811, 789, 773 cm⁻¹.

UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 656 (340). Anal. Calc.

(Found) for C₅₆H₄₆N₄S₂Cl₆Fe₂: C, 57.80 (57.58); H, 3.98 (4.03); N,

4.82 (4.68). l_eff= 4.4 l_Bper Fe.

[(3a)FeCl₂]. A solution containing 0.10 mL (0.91 mmol) of

N,N-dimethylethylenediamine in 75 mL of anhydrous butanol was

added (via cannula) to 0.310 g (0.91 mmol) of 3 and 0.115 g

(0.91 mmol) of FeCl₂. The initially orange solution turned dark

green within 20 min and deposited a microcrystalline solid after

stirring overnight under argon at 55 ^◦C. The solid was ﬁltered,

washed with 6 mL of THF, and dried to yield 0.327 g (66.7%) of

dark green product. ¹H NMR (400 MHz, MeCN-d₃): d = 26.51,

25.15, 24.19, 21.25, 19.86, 7.27, −1.42, −3.62, −5.34. FTIR (KBr):

2959, 2915, 2849, 1649, 1441, 1287, 1145, 827, 778 cm⁻¹. UV-vis

(CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 784 (4000). Anal. Calc. (Found)

for C₂₈H₃₃N₃Cl₂Fe: C, 62.47 (62.36); H, 6.18 (6.20); N, 7.81 (7.83).

l_eff= 4.9 l_B.

to yield 0.267 g (60.4%) of green product. H NMR (400 MHz,

MeCN-d₃): d = 24.47, 17.2, 15.72, 10.2, 7.76, 4.97, 3.64, −1.19,

−2.28. FTIR (KBr): 2948, 2858, 1666, 1598, 1459, 1293, 1113, 829,

781 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 738 (1900).

This compound analyzed satisfactorily with 1.5 H₂O per metal

complex. Anal. Calc. (Found) for C₃₀H₃₈N₃O_2.5Cl₂Fe: C, 59.32

(59.27); H, 6.30 (6.26); N, 6.92 (6.48). l_eff= 5.0 l_B.

[(3f)FeCl₂]. A solution containing 0.10 mL (0.79 mmol) of

N-phenylethylenediamine in 50 mL of anhydrous butanol was

added (via cannula) to 0.269 g (0.79 mmol) of 3 and 0.100 g

(0.79 mmol) of FeCl₂. After overnight heating under argon at

◦

55 C, a dark microcrystalline solid precipitated. This solid was

ﬁltered, washed with 6 mL of THF, and dried to yield 0.153 g

1

(32.9%) of product. H NMR (400 MHz, MeCN-d₃): d = 15.79,

10.04, 9.5, 9.0, 6.3, 5.3, 2.99 −0.62, −6.0, −7.4. FTIR (KBr):

2968, 2864, 1598, 1486, 1286, 1210, 946, 829, 781 cm⁻¹. UV-vis

(CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 758 (890). Anal. Calc. (Found)

for C₃₂H₃₃N₃Cl₂Fe: C, 65.55 (64.69); H, 5.67 (5.64); N, 7.17 (7.05).

l_eff= 4.8 l_B.

[(3g)FeCl₂]. A solution containing 0.297 g (1.29 mmol) 2-

(diphenylphosphino)ethylamine in 50 mL of anhydrous butanol

was added (via cannula) to 0.44 g (1.29 mmol) of 3 and 0.164 g

(1.29 mmol) of FeCl₂. After stirring overnight under argon at

55 ^◦C, a microcrystalline solid was ﬁltered, washed with 4 mL of

THF, and dried to yield 0.505 g (57.6%) of aqua-colored product.

¹H NMR (400 MHz, MeCN-d₃): d = 21.5, 15.86, 12.80, 6.68, 4.21,

0.52, −1.41, −2.15, −6.5. FTIR (KBr): 2959, 2860, 1643, 1594,

1430, 1293, 1041, 832, 773, 690 cm⁻¹. UV-vis (CH₃CN) [k_max, nm

(e, M⁻¹cm⁻¹)] 754 (1600). This compound analyzed satisfactorily

as the monohydrate. Anal. Calc. (Found) for C₃₈H₃₉N₂OPCl₂Fe:

C, 65.44 (65.92); H, 5.64 (5.82); N, 4.02 (3.72). l_eff= 4.7 l_B.

[(3b)FeCl₂]. A solution containing 0.10 mL (1 mmol) of 2-

(aminomethyl)pyridine in 50 mL of anhydrous butanol was added

(via cannula) to 0.342 g (0.91 mmol) of 3 and 0.127 g (1 mmol)

of FeCl₂. The initially orange solution turned dark green within

20 min and deposited a dark green solid after stirring overnight

under argon at 55 ^◦C. The solid was ﬁltered, washed with 6 mL of

THF, and dried to yield 0.449 g (80%) of dark green product. The

sample was not sufﬁciently soluble in acetonitrile for reliable NMR

characterization. FTIR (KBr): 2959, 2920, 2860, 1643, 1594,

1298, 1046, 778 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)]

[(3h)FeCl₂]. A solution containing 0.276 g (1.13 mmol) 3-

(diphenylphosphino)-1-propylamine in 50 mL of anhydrous bu-

tanol was added (via cannula) to 0.387 g (1.13 mmol) of 3 and

0.144 g (1.13 mmol) of FeCl₂. After stirring overnight under argon

This journal is

The Royal Society of Chemistry 2007

Dalton Trans., 2007, 2547–2562 | 2559

©

View Article Online

at 55 ^◦C, a dark green solid precipitated. This solid was collected,

washed with diethylether, and dried to yield 0.577 g (73.7%) of

not sufﬁciently soluble in acetonitrile for NMR analysis. FTIR

(KBr): 2964, 2936, 2873, 1627, 1604, 1484, 1445, 1296, 1249, 1160,

835, 780, 767 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 656

(520). Anal. Calc. (Found) for C₂₇H₂₃N₃Cl₂Co: C, 62.44 (62.05);

H, 4.46 (4.47); N, 8.09 (8.01). l_eff= 4.3 l_B.

1

dark green product. H NMR (400 MHz, MeCN-d₃): d = 26.9,

21.9, 19.1, 17.0, 11.6, 5.8, −1.16, −3.22, −4.71, −7.4. FTIR (KBr):

3052, 2959, 2860, 1643, 1468, 1430, 1288, 788, 734, 701 cm⁻¹. UV-

vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 714 (2400). Anal. Calc.

(Found) for C₃₉H₃₉N₂PCl₂Fe: C, 67.55 (67.43); H, 5.67 (6.00); N,

4.04 (3.83). l_eff= 5.2 l_B.

[(7b)MnCl₂]. A solution containing 0.15 mL (1.5 mmol) of

2-(aminomethyl)pyridine in 50 mL of anhydrous butanol was

added (via cannula) to 0.450 g (1.44 mmol) of 7 and 0.181 g

(1.44 mmol) of MnCl₂. After overnight heating under argon at

55 ^◦C, a yellow–orange solid precipitated. This solid was ﬁltered,

washed with 6 mL of THF, and dried to yield 0.305 g (56.8%)

of product. The sample was not sufﬁciently soluble in acetonitrile

for NMR analysis. FTIR (KBr): 2962, 1673, 1584, 1293, 1051,

1016, 781 cm⁻¹. UV-vis (CH₃CN) spectroscopy did not reveal any

prominent transitions in the visible region. Anal. Calc. (Found) for

C₂₈H₂₅N₃Cl₂Mn: C, 63.53 (63.34); H, 4.76 (5.09); N, 7.94 (7.55).

l_eff= 5.3 l_B.

[(3i)CoCl₂]. A solution containing 0.318 g (1.46 mmol) of 2-

(4-chlorophenylthio)ethylamine in 10 mL of anhydrous butanol

was added (via cannula) to 0.501 g (1.46 mmol) of 3 and 0.191 g

(1.46 mmol) of CoCl₂in 40 mL of anhydrous butanol. After

stirring overnight under argon at 55 ^◦C, the solution was taken to

dryness and the resulting solid was washed with diethylether and

dried to yield 0.727 g (79.5%) of dark green product. The sample

was not sufﬁciently soluble in acetonitrile for NMR analysis.

FTIR (KBr): 2970, 1632, 1589, 1485, 1299, 1096, 1014, 838,

778 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 672 (390).

This compound is presumably the chloride-bridged dimer, by

analogy to iron derivative [(2i)FeCl₂]. Anal. Calc. (Found) for

C₆₄H₆₂N₄S₂Cl₆Co₂: C, 59.96 (59.54); H, 4.88 (4.91); N, 4.37 (4.42).

l_eff= 4.4 l_Bper Co.

[(7g)FeCl₂]. A solution containing 0.206 g (0.90 mmol) 2-

(diphenylphosphino)ethylamine in 40 mL of anhydrous butanol

was added (via cannula) to 0.281 g (0.90 mmol) of 7 and 0.114 g

(0.90 mmol) of FeCl₂. After stirring overnight under argon at

55 ^◦C, the solution was reduced in volume and ﬁltered. The

residue was dissolved in THF, ﬁltered and precipitated by adding

diethylether. The solid was collected and dried to yield 0.307 g

(52.4%) of dark green product. FTIR (KBr): 2964, 2926, 2866,

[(3j)FeCl₂]. A solution containing 0.189 g (1.04 mmol) of

2-(3,5-dimethylphenylthio)ethylamine in 50 mL of anhydrous

butanol was added (via cannula) to 0.356 g (1.04 mmol) of 3 and

0.132 _◦g (1.04 mmol) of FeCl₂. After stirring overnight under argon

at 55 C, volatiles were removed in vacuo and the resulting solid

was ﬁltered, washed with 6 mL of diethylether, and dried to yield

1594, 1430, 1145, 838, 784, 696 cm⁻¹. UV-vis (CH₃CN) [k_max

,

nm (e, M⁻¹cm⁻¹)] 752 (850). This compound gave satisfactory

elemental analysis as the monohydrate. Anal. Calc. (Found) for

C₃₆H₃₅N₂OPCl₂Fe: C, 64.59 (64.20); H, 5.27 (5.16); N, 4.18 (3.96);

O, 2.39 (2.69).

1

0.320 g (48.7%) of green–brown product. H NMR (400 MHz,

MeCN-d₃): d = 13.80, 10.71, 10.41, 8.18, 7.28, 4.86, 2.50, −0.16,

−9.72. FTIR (KBr): 2766, 2925, 2888, 1582, 1488, 1464, 1286,

1054, 832, 781 cm⁻¹. UV-vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)]

444sh (3180). The compound analyzed as the monohydrate. Anal.

Calc. (Found) for C₃₄H₃₆N₂SCl₂Fe: C, 62.87 (60.85); H, 5.90 (5.79);

N, 4.31 (4.32); O, 2.46 (1.92).

X-Ray crystallography

Single crystals were mounted in thin-walled glass capillaries and

transferred to a Bruker-Nonius MACH3S X-ray diffractometer

for data collections at either 25 ^◦C (1, complexes (3a)FeCl₂,

(2c)FeCl₂, (1a)FeCl₂, (2i)FeCl₂(MeCN), and [(2i)FeCl₂]₂) or

−100 ^◦C (complexes (2d)FeCl₂, (2a)MnCl₂, (3g)FeCl₂, and

[(4g)FeCl₂]. A solution of 0.235 g (1.02 mmol) of 2-

(diphenylphosphino)ethylamine in 10 mL of anhydrous butanol

was added via cannula to 0.380 g (1.02 mmol) of 4 and 0.130 g

(1.02 mmol) of anhydrous FeCl₂followed by 40 mL of anhydrous

butanol. The green solution was stirred overnight under argon

at 55 ^◦C and reduced in volume to about 10 mL. The solid

that precipitated was ﬁltered, washed with a small amount of

diethylether, and dried to give 0.352 g (48.8%) of dark green

product. ¹H NMR (400 MHz, MeCN-d₃): d = 39.7, 16.44, 14.15,

13.74, 12.9, 12.5, 8.0, 7.43, 7.1, 6.2, 5.5, −1.42, −6.3. FTIR (KBr):

2970, 2866, 1611, 1490, 1436, 1293, 838, 784, 745, 707 cm⁻¹. UV-

vis (CH₃CN) [k_max, nm (e, M⁻¹cm⁻¹)] 732 (740). This compound

afforded satisfactory elemental analysis as the monohydrate. Anal.

Calc. (Found) for C₄₀H₄₃N₂OPCl₂Fe: C, 66.22 (65.80); H, 5.97

(5.77); N, 3.86 (3.60).

˚

(3j)FeCl₂) using graphite monochromated Mo Ka(k = 0.71073 A)

radiation. Unit cell constants were determined from a least squares

reﬁnement of the setting angles of 25 intense, high angle reﬂections.

Intensity data were collected using the x/2h scan technique to a

maximum 2h valueof50–55^◦. Absorption corrections were applied

based on azimuthal scans of several reﬂections for each sample as

required. The data were corrected for Lorentz and polarization

effects and converted to structure factors using the Crystal-

Structure software package.³⁰Space groups were determined

based on systematic absences and intensity statistics. Successful

direct-methods solutions were calculated for each compound

using the SHELXTL suite of programs.³¹Any non-hydrogen

atoms not identiﬁed from the initial E-map were located after

several cycles of structure expansion and full matrix least squares

reﬁnement on F². Hydrogen atoms were added geometrically. All

non-hydrogen atoms were reﬁned with anisotropic displacement

parameters, while hydrogen atoms were reﬁned using a riding

model with group isotropic displacement parameters. Relevant

[(6c)CoCl₂]. A solution containing 0.10 mL (0.85 mmol) of

2-(aminoethyl)pyridine in 40 mL of anhydrous butanol was added

(via cannula) to 0.243 g (0.85 mmol) of 6 and 0.110 g (0.85 mmol) of

CoCl₂. A dark solid deposited after stirring overnight under argon

at 55 ^◦C. The solid was ﬁltered, washed with 6 mL of diethylether,

and dried to yield 0.195 g (44.2%) of product. The compound was

2560 | Dalton Trans., 2007, 2547–2562

This journal is

The Royal Society of Chemistry 2007

©

View Article Online

crystallographic information for the compounds is summarized in

Tables 1 and 3.

Acknowledgements

We thank Chevron Phillips Chemical Company for their generous

support of this work at the University of Wisconsin-Eau Claire

and for permission to publish these results. In addition, we thank

the donors of the Petroleum Research Fund, administered by the

American Chemical Society (Grant 37885-GB3 to M. J. C.), Re-

search Corporation (Cottrell College Science Award to M. J. C.),

the National Science Foundation (CHE-0078746 to J. A. H.) and

the University of Wisconsin-Eau Claire for ﬁnancial support.

Instrumentation for infrared analyses (CHE-0216058) and NMR

spectroscopic measurements (CHE-0521019) was obtained with

the support of the National Science Foundation MRI grant

program. We also thank Mr Daniel Bates for assistance in

acquiring UV-vis and IR characterization data for this study.

In complex (2c)FeCl₂, the sample was found as expected

with two crystallographically independent yet chemically similar

complexes occupying the asymmetric unit in addition to a

single acetonitrile solvate (resulting in the fractional elemental

compositions found in the empirical formula). The sample was

robust throughout the majority of the room temperature data

collection. However, within the last 2–3 h, the sample experienced

signiﬁcant (<60%) decay due to rapid desolvation. No attempt

was made to scale the data for this decay, as the reﬂections

being measured at that time were principally weak, high h data.

However, as a result of the sample decay, azimuthal scan data

for a semi-empirical absorption correction could not be collected.

Thus, the data for (2c)FeCl₂were not treated for either decay or

absorption.

For complex (2d)FeCl₂, doubling of the unit cell dimensions

yields cell parameters that are close to a C-centered monoclin_◦ic

References

˚

cell (with a = 20.56 A, b = 20.27 A, c = 12.32 A, a = 89.9 ,

1 (a) L. K. Johnson, C. M. Killian and M. Brookhart, J. Am. Chem. Soc.,

1995, 117, 6414–6415; (b) C. M. Killian, D. J. Tempel, L. K. Johnson

and M. Brookhart, J. Am. Chem. Soc., 1996, 118, 11664–11665; (c) S. A.

Svejda, L. K. Johnson and M. Brookhart, J. Am. Chem. Soc., 1999, 121,

10634–10635; (d) L. H. Shultz and M. Brookhart, Organometallics,

2001, 20, 3975–3982.

2 (a) L. K. Johnson, S. Mecking and M. Brookhart, J. Am. Chem. Soc.,

1996, 118, 267–268; (b) D. J. Tempel, L. K. Johnson, R. L. Huff, P. S.

White and M. Brookhart, J. Am. Chem. Soc., 2000, 122, 6686–6700.

3 (a) G. J. P. Britovsek, V. C. Gibson, B. S. Kimberley, P. J. Maddox,

S. J. McTavish, G. A. Solan, A. J. P. White and D. J. Williams, Chem.

Commun., 1998, 849–850; (b) G. J. P. Britovsek, M. Bruce, V. C. Gibson,

B. S. Kimberley, P. J. Maddox, S. Mastroianni, S. J. McTavish, C.

Redshaw, G. A. Solan, S. Stro¨mberg, A. J. P. White and D. J. Williams,

J. Am. Chem. Soc., 1999, 121, 8728–8740; (c) G. J. P. Britovsek, S.

Mastroianni, G. A. Solan, S. P. D. Baugh, C. Redshaw, V. C. Gibson,

A. J. P. White, D. J. Williams and M. R. J. Elsegood, Chem.–Eur. J.,

2000, 6, 2221–2231.

b = 107.1^◦, c = 90.8^◦); however, the Laue symmetry of the

¯

data is clearly 1 rather than 2/m. Moreover, R(merge) for the

monoclinic cell is more than 0.6. Assignment of the Laue class

as triclinic is supported by the successful solution and reﬁnement

¯

of the structure in the triclinic space group P1. Two independent

molecules of (2d)FeCl₂are present in the asymmetric unit, with

metrical differences that well exceed the uncertainties in the

calculated bond lengths and angles.

In the case of (1a)FeCl₂, nearly all of the peaks in the ﬁnal

difference map are located within channels that are arrayed around

¯

3 axes that traverse the unit cell. It is highly likely that fractional,

disordered solvent is present within these channels. However,

attempts to model this solvent with CH₂Cl₂or pentane (from

which the sample was crystallized) was not successful. The diffuse

scattering of the low-occupancy, disordered solvent is responsible

for the large residuals reported for this complex. In support of

4 (a) B. L. Small, M. Brookhart and A. M. A. Bennett, J. Am. Chem.

Soc., 1998, 120, 4049–4050; (b) B. L. Small and M. Brookhart, J. Am.

Chem. Soc., 1998, 120, 7143–7144; (c) A. M. A. Bennett, Chemtech,

1999, 29, 24.

5 Y. Chen, R. Chen, C. Qian, X. Dong and J. Sun, Organometallics, 2003,

22, 4312–4321.

this notion, the SOLV/SQUEEZE utility in the PLATON suite

3

˚

of programs identiﬁed a solvent accessible volume of 1484 A

6 Y. Chen, C. Qian and J. Sun, Organometallics, 2003, 22, 1231–1236.

7 M. E. Bluhm, C. Folli and M. Do¨ring, J. Mol. Catal. A: Chem., 2004,

212, 13–18.

8 C. Bianchini, G. Mantovani, A. Meli, F. Migliacci, F. Zanobini, F.

Laschi and A. Sommazzi, Eur. J. Inorg. Chem., 2003, 1620–1631.

9 C. Bianchini, G. Giambastiani, I. T. Guerrero, A. Meli, E. Passaglia

and T. Gragnoli, Organometallics, 2004, 23, 6087–6089.

10 W.-H. Sun, X. Tang, T. Gao, B. Wu, W. Zhang and H. Ma,

Organometallics, 2004, 23, 5037–5047.

11 T. M. Smit, A. K. Tomov, V. C. Gibson, A. J. P. White and D. J. Williams,

Inorg. Chem., 2004, 43, 6511–6512.

12 G. J. P. Britovsek, V. C. Gibson, B. S. Kimberley, S. Mastroianni, C.

Redshaw, G. A. Solan, A. J. P. White and D. J. Williams, J. Chem. Soc.,

Dalton Trans., 2001, 1639–1644.

13 G. J. P. Britovsek, V. C. Gibson, O. D. Hoarau, S. K. Spitzmesser, A. J. P.

White and D. J. Williams, Inorg. Chem., 2003, 42, 3454–3465.

14 M. Gasperini, F. Ragaini, E. Gazzola, A. Caselli and P. Macchi, Dalton

Trans., 2004, 3376–3382.

15 We have also reported the synthesis of asymmetric PBI ligands and

monoimine acetylpyridine (NNO) ligands for use in chromium-based

oleﬁn polymerization catalysis. B. L. Small, M. J. Carney, D. M.

Holman, C. E. O’Rourke and J. A. Halfen, Macromolecules, 2004,

37, 4375–4386.

16 Recently, Ragaini et al. postulated that n-alkyl imines of acenaph-

thenequinone are unstable and decompose via an isomerization process

that releases ring strain in the initially formed alkylimine. The

isomerization involves a sp²–sp³rehybridization of the acenaphthene

ring carbon atom, followed by further reaction with another equivalent

in the unit cell (11.4% of total cell volume), occupied by a total

disordered electron density of 71 e⁻per cell.³²This corresponds

to approximately two pentane or two CH₂Cl₂molecules per

cell (1/9th of a solvate molecule per molecule of complex per

cell). The contributions of these disordered solvate molecules to

F_owere removed using the SQUEEZE utility, and reﬁnement

converged using the modiﬁed structure factors. The contributions

of the removed solvate molecules are not included in the ﬁnal

d

_calcd, F₀₀₀, or the formula weight of the complex. Data collection

for (1a)FeCl₂was not possible at reduced temperature as all of

the specimens examined suffered from cracking or splitting at

−100 ^◦C, presumably the result of a phase change.

Crystals of (2i)FeCl₂(MeCN) were found to be highly solvated.

There is one acetonitrile ligand, one fully occupied acetonitrile

solvate, and a second, half-occupied acetonitrile solvate that

is disordered over an inversion center. Finally, in the case of

(3j)FeCl₂, a half-occupied molecule of THF was found to be

disordered on an inversion center. Reﬁnement of the remaining

structures proceeded normally.

CCDC reference numbers 636867–636876.

For crystallographic data in CIF or other electronic format see

DOI: 10.1039/b702197f

This journal is

The Royal Society of Chemistry 2007

Dalton Trans., 2007, 2547–2562 | 2561

©

View Article Online

of alkylamine to release an alkyl-imine product. Similar monoimine

instability and decomposition processes may account for our inability

to isolate monoimines using donor-modiﬁed alkylamines. F. Ragaini,

M. Gasperini, E. Gallo and P. Macchi, Chem. Commun., 2005, 1031–

1033.

22 The Fe(II)–P distances in six-coordinate [FeCl₂(o-(Ph₂P)₂C₆H₄)₂] are

˚

2.612 A and 2.623 A; J. E. Barclay, A. Hills, D. L. Hughes and G. J.

Leigh, J. Chem. Soc., Dalton Trans., 1988, 2871.

23 The Fe(II)–P distances in four-coordinate [FeBr₂(PEt₃)₂] range from

˚

2.406 A to 2.451 A; B. S. Snyder and R. H. Holm, Inorg. Chem., 1988,

17 The geometric parameter s is deﬁned as s = |b−a|/60, where a and

b are the two trans basal angles in pseudo-square pyramidal geometry

(see ref. 18). In idealized square-planar geometries, a = b, but trigonal

distortions result in a = b.

27, 2339.

24 The Fe(II)–P distances in four-coordinate [Fe(B(CH₂PPh₂)₃)(NH-p-

˚

C₆H₄Me)] are 2.427 A and 2.452 A; S. D. Brown and J. C. Peters,

J. Am. Chem. Soc., 2004, 126, 4538.

18 A. W. Addison, T. N. Rao, J. van Reedijk, J. Rijn and G. C. Vorschoor,

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

25 A. R. Katritzky, Y.-J. Hu, H.-Y. He and S. Mehta, J. Org. Chem., 2001,

66, 5590–5594.

19 M. Di Vaira, S. Midollini and L. Sacconi, Inorg. Chem., 1981, 20,

3430–3435.

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

27 C. Cairns, S. M. Nelson and M. G. B. Drew, J. Chem. Soc., Dalton

Trans., 1981, 1965.

28 R. E. Marsh, Acta Crystallogr., Sect. B: Struct. Sci., 2004, B60, 252.

29 B. L. Small, R. Rios, E. R. Fernandez and M. J. Carney, Organo-

metallics, 2007, 26, 1744–1749.

30 CrystalStructure 3.6.0: Crystal Structure Analysis Package, Rigaku and

Rigaku/MSC 9009, New Trails Dr., The Woodlands, TX 77381 USA,

2000–2004.

5

20 In addition, the Fe(II)–P distances in low spin Fe(g -C₅H₅)-

˚

[N(SiMe₂CH₂PPh₂)₂] range from 2.242 to 2.256 A; M. D. Fryzuk, D. B.

Leznoff, E. S. F. Ma, S. J. Rettig and V. G. Young, Jr., Organometallics,

1998, 17, 2313–2323.

=

21 The Fe(II)–P distances in six-coordinate [FeBr₂(PPh₂CH CHPPh₂)₂]

˚

are 2.636 A and 2.666 AR. Cecconi, M. di Vaira and L. Sacconi,

Cryst. Struct. Commun., 1981, 10, 1523. The Fe(II)–P distances in

˚

=

[FeCl₂(PPh₂CH CHPPh₂)₂] range from 2.576 A to 2.592 A; R.

Cecconi, M. di Vaira, S. Midollini, A. Orlandini and L. Sacconi, Inorg.

Chem., 1981, 20, 3423.

31 SHELXTL-NT V. 6.12, Bruker AXS, Inc., 5465 E. Cheryl Parkway

Madison, WI 53711–5373 USA, 2000.

32 A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7–13.

2562 | Dalton Trans., 2007, 2547–2562

This journal is

The Royal Society of Chemistry 2007

©

Article Doi

Alternatives to pyridinediimine ligands: Syntheses and structures of metal complexes supported by donor-modified α-diimine ligands

DOI: 10.1039/b702197f

Source and publish data:

Authors:

Article abstract of DOI:10.1039/b702197f

Full text of DOI:10.1039/b702197f

Products guided by the article

R&D Labs maybe for 913618-67-4

Relevant to this article

Hot Product