Synthesis and Characterization of (smif)2Mn (n = 0, M = V, Cr, Mn, Fe,Co, Ni, Ru; N = +1, M = Cr, Mn, Co, Rh, Ir; Smif =1,3-di-(2-pyridyl)-2-azaallyl)

Article

pubs.acs.org/IC

Synthesis and Characterization of (smif)₂Mⁿ(n = 0, M = V, Cr, Mn, Fe,

Co, Ni, Ru; n = +1, M = Cr, Mn, Co, Rh, Ir; smif =1,3-di-(2-pyridyl)-2-

azaallyl)

Brenda A. Frazier,^†Erika R. Bartholomew,^†Peter T. Wolczanski,*^,†Serena DeBeer,^†,⊥

Mitk’El Santiago-Berrios,^†Hector D. Abruna,^†Emil B. Lobkovsky,^†Suzanne C. Bart,^‡Susanne Mossin,^‡

̃

Karsten Meyer,^‡and Thomas R. Cundari^§

^†Department of Chemistry & Chemical Biology, Baker Laboratory, Cornell University, Ithaca, New York 14853, United States

^‡Department of Chemistry & Pharmacy, University of Erlangen-Nuremberg, Egerlandstrasse 1, D-91058 Erlangen, Germany

^§Department of Chemistry, Center for Advanced Scientific Computing and Modeling (CASCaM), University of North Texas,

Box 305070, Denton, Texas 76203-5070, United States

S

* Supporting Information

ABSTRACT: A series of Werner complexes featuring the

tridentate ligand smif, that is, 1,3-di-(2-pyridyl)-2-azaallyl, have

been prepared. Syntheses of (smif)₂M (1-M; M = Cr, Fe) were

accomplished via treatment of M(NSiMe₃)₂(THF)_n(M = Cr,

n = 2; Fe, n = 1) with 2 equiv of (smif)H (1,3-di-(2-pyridyl)-2-

azapropene); ortho-methylated (^oMesmif)₂Fe (2-Fe) and

(^oMe₂smif)₂Fe (3-Fe) were similarly prepared. Metatheses of

MX₂variants with 2 equiv of Li(smif) or Na(smif) generated 1-M (M = Cr, Mn, Fe, Co, Ni, Zn, Ru). Metathesis of VCl₃(THF)₃

with 2 Li(smif) with a reducing equiv of Na/Hg present afforded 1-V, while 2 Na(smif) and IrCl₃(THF)₃in the presence of

NaBPh₄gave [(smif)₂Ir]BPh₄(1⁺-Ir). Electrochemical experiments led to the oxidation of 1-M (M = Cr, Mn, Co) by AgOTf to

produce [(smif)₂M]OTf (1⁺-M), and treatment of Rh₂(O₂CCF₃)₄with 4 equiv Na(smif) and 2 AgOTf gave 1⁺-Rh.

Characterizations by NMR, EPR, and UV−vis spectroscopies, SQUID magnetometry, X-ray crystallography, and DFT

calculations are presented. Intraligand (IL) transitions derived from promotion of electrons from the unique CNC^nb

(nonbonding) orbitals of the smif backbone to ligand π*-type orbitals are intense (ε ≈ 10 000−60 000 M⁻¹cm⁻¹), dominate

the UV−visible spectra, and give crystals a metallic-looking appearance. High energy K-edge spectroscopy was used to show that

the smif in 1-Cr is redox noninnocent, and its electron configuration is best described as (smif(−))(smif(2−))Cr(III); an

unusual S = 1 EPR spectrum (X-band) was obtained for 1-Cr.

occurred while conducting a metalation reaction with Cr{N-

(TMS)₂}₂THF₂.^41,42As Scheme 1 indicates, a C−N bond was

cleaved in the process to yield forest green {1,3-di-(2-pyridyl)-

2-azaallyl}CrN(TMS)₂, abbreviated as (smif)CrN(TMS)₂(smif =

1,3-di-(2-pyridyl)-2-azaallyl). In the proposed mechanism of its

formation,⁴³a chromium amide functions as a base, thereby

prompting an alternative synthesis from (smif)H, the

protonated azaallyl 1,3-di-(2-pyridyl)-2-azapropene, and Cr{N-

(TMS)₂}₂THF₂. A related degradation of a di-((2-pyridyl)-

CH₂)amide on Zn led to the isolation of the first smif-complex,

(smif)₂Zn, according to Westerhausen and Kniefel.⁴⁴

Several features of (smif)CrN(TMS)₂were intriguing,

especially the optical density of the species, which manifested

two intraligand bands of great intensity at 674 nm (ε ≈ 15 000

M⁻¹cm⁻¹) and 396 nm (ε ≈ 27 000 M⁻¹cm⁻¹). The azaallyl

unit is also isoelectronic with popular N-heterocyclic carbene

ligands,⁴⁵⁻⁴⁷as Figure 1 illustrates. Most importantly, due to its

monoanionic charge and tridentate capacity, the possibility of

INTRODUCTION

■

The beginnings of inorganic coordination chemistry can be

traced to the laboratories of Alfred Werner,¹whose archived

collection of cobalt complexes at the University of Zurich can still

be viewed today with an appropriate request.²For a significant

part of the 20th century and continuing into the 21st century,

investigations into amines and N-donor heterocycles played a

dominant role in the field of coordination chemistry, with more

attention given to chelates such as bipyridines,^3,4terpyridines,⁵⁻¹⁷

pyrazolyl borates,¹⁸⁻²⁰2-picolyl-amines,²¹⁻²⁷dipyridylimines and

pyridinediimines,²⁸⁻³²and bis-(2-pyridylcarbonyl)aminates,

among countless others.³³⁻⁴⁰The majority of these compounds

contain cationic complex ions, and while first row transition metal

(TM) species are prominent, there are plenty of examples

containing second and third row TM elements. Even today, one

can hardly open a journal with inorganic content and not see a

contribution describing the synthesis, characterization or

application of a Werner-type coordination compound.

During the course of investigating tetradentate chelate

ligands as applied to first row TM elements, a degradation

Received: February 22, 2011

Published: November 17, 2011

12414

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Scheme 1

Figure 1. N-Heterocyclic carbenes are isoelectronic with the azaallyl portion of smif.

Scheme 2

synthesizing Werner-type coordination complexes that were

neutral and amenable to study in nonaqueous solvents was

appealing. Herein we report the syntheses, characterizations

and initial calculations of pseudo-octahedral MN₆complexes

with the composition (smif)₂Mⁿ; a few of the derivatives (M =

Fe, Co, Co⁺, Ni) were previously communicated,⁴⁸and all the

complexes described were included in a patent application.⁴⁹

bis-amide precursors M{N(TMS)₂}₂THF_n(M = Cr, n = 2;^41,42

M = Fe, n = 1)⁵²acted as bases when treated with smifH to

afford (smif)₂M (1-M; M = Cr, Fe) concomitant with 2 equiv

HN(TMS)₂. This method was also used to prepare di- and

tetra-ortho-methylated iron derivatives (^oMesmif)₂Fe (2-Fe)

and (^oMe₂smif)₂Fe (3-Fe). While the metal-amide precursor

route was considered the cleanest, the requirement of

synthesizing various bis-amides added a step that was

circumvented by standard metathetical procedures with simple

metal halides. Typically, Li(smif) was prepared in situ from

smifH and LiN(TMS)₂at −78 °C in THF (Scheme 2), and the

metal halides^53,54were added and allowed to stir for 16−36 h

to afford several (smif)₂M (1-M, M = Mn, Fe, Co, Ni, Zn) in

reasonable to very good yield. The organometallic precursor

RESULTS

■

Syntheses. 1. (smif)H and Derivatives. Condensation

of 2-pyridinecarboxaldehyde with 2-(aminomethyl)pyridine

afforded 1,3-di-(2-pyridyl)-2-azapropene in >90% yield according

to a literature procedure.⁵⁰As shown in Scheme 2, the method

was adopted for two methylated species derived from the

o-methylated pyridine carboxaldehyde and o-methylated

2-(aminomethyl)pyridine. One provided a smifH with a methyl

55

(COD)RuCl₂and 2 equiv of Na(smif) were used to prepare

(smif)₂Ru in similar fashion. Utilization of Rh₂(O₂CCX₃)₄(X =

H, F)⁵⁶in metathesis reactions with M′(smif) (M′ = Li, Na)

failed to elicit the desired Rh(II) complex, and PdX₂and PtX₂

sources also failed to produce the corresponding bis-smif

derivatives, although ample chemistry, especially for Pd, was

noted. While the proposed 19 and 20 e⁻compounds might not

be expected to be stable, the potential for smif to accommodate

additional electrons as a redox-active ligand (vide infra)

prompted the attempts. Scheme 3 also provides the colors of

the (smif)₂M (1-M) derivatives and their complementary solid

state appearances, which were often gold-bronze (or gold

o

group in a single ortho position, that is, MesmifH, and the

other led to a di-o-methylated smif, ^oMe₂smifH. Deprotonation

of smifH was effected with strong bases, and utilization of

51

LiN(TMS)₂and NaN(TMS)₂afforded gold crystals of

Li(smif) and Na(smif) in excellent yields (≥90%). The alkali

metal anions were deep magenta in solution, with bands at 583

(ε ≈ 18 000 M⁻¹cm⁻¹) and 420 nm (ε ≈ 7000 M⁻¹cm⁻¹)

measured for Li(smif) in benzene.

2. (smif)₂M (M = TM). Neutral (smif)₂M derivatives were

prepared via three routes that are illustrated in Scheme 3. The

12415

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Scheme 3

Figure 2. Summary of electrochemical potentials determined from cyclic voltammetry (typically 1 V/s) in THF with ∼1 mM 1-M, 0.1 M TBAP as

supporting electrolyte, Pt wire working electrode, and Ag wire as a pseudoreference electrode (approximate to Ag⁰/Ag⁺(0.1 M), which is 0.49 V vs

SCE; 0.73 V vs NHE). Reversible potentials are shown with a bold line, and irreversible potentials are labeled; origins of irreversible potentials (when

identified) are indicated by the red connectivity and arrows.

58

depending on crystal size) and indicative of reflectivity because

of a large absorption in the red region of the visible spectrum.

Completion of the entire series of first row TMs required

synthesis of titanium, vanadium, and copper derivatives. While

the preparation of other smif-containing titanium species has

been realized, the work will be reported in a separate article.⁵⁷

Various attempts at preparing (smif)₂Cu failed, and it is

suspected that the azaallyl bridge of the complex is too reactive

with regard to CC-bond forming processes. Finally, since

appropriate V(II) precursors are not common, the combination

of 2 equiv Li(smif), 1 equiv of reducing agent (i.e., Na/Hg),

and VCl₃(THF)₃enabled the synthesis of (smif)₂V in 81%

yield.

3. Electrochemistry. Li(smif) and a select group of 1-M

(M = Cr, Mn, Fe, Co, Ni) were subjected to cyclic voltammetry

to help determine whether redox processes could be used to

synthesize related cations or anions. Note that the neutral

character and air-sensitivity of the species limited analyses to

THF solutions, in part because acetonitrile proved to react in

certain cases. Figure 2 shows a chart summarizing the assay that

reveals possible noninnocent redox activity of the smif at high

negative potentials (∼ −1.6 V vs Ag⁰/Ag⁺) observed for both

12416

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Scheme 4

Scheme 5

56

Li(smif) and 1-M. In general, the electrochemistry of these first

row elements is rich, yet the number of irreversible processes,

especially those originating from apparent smif reduction,

suggests that the reactivity of the system might hamper the

isolation of some ions. There is also no clear trend for

oxidation, but reversible potentials ≤0.0 V (Ag⁰/Ag⁺at ∼0.49 V

vs SCE) suggested that cations could be prepared from mild

oxidants, so reactions with AgOTf were explored.

Treatment of Rh₂(O₂CCF₃)₄with 2 equiv of Na(smif) in

toluene in the presence of AgOTf afforded [(smif)₂Rh]OTf

(1⁺-Rh) in 53% yield, despite the fact that AgOTf oxidized

Na(smif) quite rapidly in a separate experiment. It is plausible

that the relative insolubilities of the reagents in toluene result

in preferential oxidation of smif-containing Rh species over

Na(smif), ultimately leading to the Rh(III) product. A related

effort to prepare [(smif)₂Cu]⁺was unsuccessful. The correspond-

ing Ir(III) compound, [(smif)₂Ir]BPh₄(1⁺-Ir), was prepared via

4. Cations [(smif)₂M]⁺. The redox chemistry observed

electrochemically was synthetically scaled with modest success,

as indicated in Scheme 4. AgOTf oxidations of (smif)₂M (1-M;

M = Cr, Mn, Co) in THF produced the corresponding cations

[(smif)₂M]OTf (1⁺-M; M = Cr, Mn, Co) in excellent yields

(75−81%). The ready preparation of 1⁺-Mn was surprising

given the lack of a reversible metal-based couple; however, an

oxidation wave at 0.59 V (vs Ag⁰/Ag⁺) was present, and served

as the origin of three irreversible reduction waves. Both

(smif)₂V (1-V) and 1-Ni yielded intractable brown-black solids

when exposed to AgOTf, despite the observation of a reversible

couple at −0.68 V (anodic sweep) for the latter. It is plausible

that a Ni(III) species was stable on the time scale of the

electrochemical experiment (1 V/s), but underwent chemical

degradation during the time of a chemical oxidation.

59

metathesis of IrCl₃(THT)₃with Na(smif) in the presence of

NaBPh₄.

Iron derivatives (smif)₂Fe (1-Fe) and (^oMe₂smif)₂Fe (3-Fe)

were not successfully oxidized with Ag(I) or other mild

oxidants with the necessary potential, and Scheme 5 provides

some indication of the fate of the former. Oxidation of 1-Fe led

to a mixture of four products, all containing spectral signatures

of smifH; two were identified by independent syntheses as the

cation [(smifH)(smif)Fe]OTf (4⁺-Fe) and the dication

[(smifH)₂Fe](OTF)₂(5²⁺-Fe). The addition of 2 equiv

smifH to FeX₂(X = Br, OTf) led to the known dication

[(smifH)₂Fe]X₂(5²⁺-Fe),²⁸as did direct diprotonation of 1-Fe

with 2 equiv of HBF₄. Dication 5²⁺-Fe and 1-Fe comproportio-

nated to give 4⁺-Fe in an NMR tube scale experiment in

CD₃CN. If oxidation of 1-Fe occurred at the ligand, as

predicted by calculations (vide infra), it is possible that H-atom

abstraction events led to the products containing smifH.

Since neutral second and third row (smif)₂M species could

not be directly synthesized, with the exception of 1-Ru, d⁶

cations were directly prepared for M⁺= Rh(III) and Ir(III).

12417

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

5. Chemical Reductions. All attempts to prepare

[(smif)₂M]⁻(1⁻-M) from the neutral precursors failed, even

though varied reaction conditions and a host of reagents were

employed. While the electrochemical measurements suggested

that smif could harbor an additional electron for all of the

complexes, 1⁻-M must not be stable in the reduction medium

or on the time scale of the chemical reduction. Cyclic

voltammetry on [(smif)₂Rh]OTf (1⁺-Rh) in THF revealed

quasi-reversible waves at ∼ −1.2 V and ∼ −1.6 V and

irreversible oxidations at 0.92 and 1.38 V. Attempts to

chemically reduce 1⁺-Rh to form a Rh(II) species that would

have a formal smif dianion failed and evidence of free smif

anion was observed optically. Related measurements on

[(smif)₂Ir]BPh₄(1⁺-Ir) in THF showed quasi-reversible

waves at −1.18 V and ∼ −1.6 V, an irreversible reduction at

−0.96 V and an irreversible oxidation at 1.28 V among several

minor waves. Again, chemical reductions failed concomitant

with the appearance of free smif anion.

ligand HOMO and subsequent reactivity of the ligand radical

thus created, providing a rationale for products derived from

H-atom abstraction (Scheme 5).

Allowed transitions from the CNC^nborbitals to an e set of

ligand π*orbitals at −1.20 eV transfer charge from the CNC-

backbone to the pyridines of the smif, incurring a large electric

dipole change (“red” IL bands). A second set of intraligand

(IL) transitions from the CNC^nborbitals to a group of ligand

π* orbitals at −0.14 to −0.19 eV also transfers charge from the

backbone to the pyridines (“blue” IL bands). Large intensities

are expected from IL bands of this type, and this is

experimentally observed. The energy differences between filled

orbitals and virtual (empty) orbitals should not be construed as

accurate, but the relative energies of filled vs filled orbitals, etc.,

are reasonably accurate,^61,62hence the IL bands are expected to

be roughly 1.0 eV apart.

2. Comparison of (smif)₂M (1-M; M = V, Cr, Mn, Fe,

Co, Ni). Because of spin polarization, unrestricted DFT calculations

of the “open-shell” systems, that is, (smif)₂M (1-M; M = V, Cr,

Mn, Co, Ni), can generate alpha and beta spins at different

energies and with differing orbital compositions despite

population of congeneric spatial orbitals. In Figure 4, energies

of the alpha and beta spins of the same orbital parentage have

been “averaged” to produce truncated molecular orbital

diagrams that can be better visually assessed. This approx-

imation is not without its problems, but there are some clear

predictions that can be tested with experiments. In the cases of

1-Cr, 1-Mn, and 1-Co, additional electrons are calculated to

reside in smif π* orbitals, rendering these metals M(III) with

the (smif)₂ligands carrying a total of 3− charge. The total spin

state is ambiguous in these cases. For example, the calculations

Calculations. 1. (smif)₂Fe (1-Fe). As a guide toward

understanding the electronic structures and the unusual optical

phenomena affiliated with (smif)₂M (1-M; M = V, Cr, Mn, Fe,

Co, Ni), density functional (DFT) calculations were conducted.

Figure 3 shows a truncated MO diagram for D_2d1-Fe, chosen

2

of (smif)₂Mn (1-Mn) show one electron in d_{x −y}and one in

smif π*, with a total spin of S = 3/2, but computationally the

S = 5/2 solution is very close in energy, hence the latter spin

state was chosen for the diagram in compliance with

experiment. That is not the least of the problems with 1-Mn,

as the calculation yields half-filled orbitals lower than filled, and

a clear discontinuity in the trends of dπ- and dσ-orbitals relative

to 1-Cr and 1-Fe. Multireference calculations suggest that this

is actually a conventional HS Mn(II) species with a weak-field

d⁵configuration above the CNC^nborbitals in energy.⁶³

Bis-smif vanadium 1-V is predicted to be S = 3/2, but the

HOMO of the system has considerable ligand character (∼50%

smif π*), whereas the HOMO of 1-Cr is clearly smif π* and

contains an electron that is antiferromagnetically (AF) coupled

to a Cr(III) S = 3/2 center, resulting in a total spin system of

S = 1. The Co complex 1-Co is similarly portrayed by the

calculations, with its odd electron residing in a smif π* orbital,

and 1-Ni is calculated to be a standard Ni(II) species with both

2

d_{x −y}and d_zsingly occupied. It was with these calculations in

hand that complementary physical inorganic investigations of

1-M were conducted, but the inconsistencies prompted higher

level computational approaches.⁶³

Figure 3. Truncated molecular orbital diagram of (smif)₂Fe (1-Fe)

showing origin of dominant intraligand bands in red and blue regions

of the UV−vis spectrum.

General Structural Features of (smif)₂Mⁿ(1ⁿ-M). Table 1

lists selected data acquisition and refinement details pertaining

to (smif)₂M (1-M; M = V, Cr, Mn, Fe, Co, Ni) and

[(smif)₂M]OTf (1⁺-M; M = Cr, Co). Curiously, all of the

neutral compounds aside from D_2d1-Fe crystallized with two

molecules in the asymmetric unit, and often a solvent molecule

was present; in many cases SQUEEZE was applied during

refinement. Since the bite angle formed by the azaalyl nitrogen

(N_aza) and each pyridine nitrogen (N_py) is ∼80°, there is

considerable space for modest deviations from D_2dsymmetry,

because the electron−electron correlation factors do not

complicate this representation since it is a closed-shell system.

The HOMO and HOMO-1 orbitals are two linear combina-

tions (b₁and a₂) of the azaallyl nonbonding orbital containing a

node at nitrogen and opposing phase p-orbitals at the two

carbons (CNC^nb). Roughly 0.9 eV below this set are the e

(d_xz, d_yz) and b₁(d_xy) orbitals of this low spin d⁶complex that

are nearly pure d in character. Perhaps the failed oxidation

attempts on 1-Fe resulted from removal of an electron from the

12418

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

2

Figure 4. Truncated molecular orbital diagrams for (smif)₂M (1-M; M = V, Cr, Mn, Fe, Co, Ni); for comparison to octahedral systems, d_{x −y}placed

on the bond axes. All orbital energies should be considered approximate; for the open shell cases alpha- and β-spins of related orbital parentage were

2

averaged to generate the filled orbital. Dashed red lines correlate the nonbonding d_xy, d_xz, and d_yzset of orbitals and the d_{x −y}and d_zset of sigma-

antibonding orbitals. The CNC^nborbitals have essentially the same energies independent of metal. Ligand orbitals above the CNC^nbpair are

primarily pyridine Lπ*, and ligand orbitals below are essentially pyridine π-bonding in character.

and Figure 5 illustrates the types of distortions expected. An

axial or propeller twist about the N_aza−M−N_a^′_zaaxis removes

the mirror planes and renders the resulting structure D₂, and if

pyridines of opposing smif ligands approach one another in

addition to the twist, the symmetry is further lowered to C₂. If

one smif is pulled away from the metal while the N_aza-M-N′_aza

angle remains 180°, this axial elongation results in C_2v

symmetry, but if the smif is canted such that /N_aza-M-N′_azais

no longer 180°, only a mirror plane remains (C_s).

Figure 7, as previously described,⁴⁸whose δ of 0.30(1) mm/s,

modest ΔE_Qof 0.62(1) mm/s, and sharp line width of 0.25(1)

mm/s may be construed as typical for low spin Fe(II) complexes

with relatively symmetric electron density. Single o-Me substitution

of the smif ligand, that is, (^oMesmif)₂Fe (2-Fe), generates a rather

modest change in the Mossbauer spectrum, with a greater ΔE

̈

Q

of 0.99 mm/s and an increased line width indicating slightly

greater asymmetry in the electron density about the Fe(II)

center. A notable change occurs in (^oMe₂smif)₂Fe (3-Fe),

whose isomer shift (δ = 1.04 mm/s) is now more consistent

with a high spin Fe(II) configuration, with a ΔE_Qof 2.18 mm/s

that is also typical of an S = 2 iron center.⁶⁴

Pertinent structural parameters for [(smif)₂M]ⁿ(n = 0, 1-M;

n = 1, 1⁺-M) are given in Table 2, where they are listed as

average values when statistically appropriate. All of the

compounds are roughly D_2d, with very subtle deviations that

render all species except (smif)₂Fe (1-Fe) and [(smif)₂Co]OTf

(1⁺-Co) rigorously assigned C₁symmetry. The distortions that

best describe the subtle changes from D_2dare listed in the table,

where the N_aza−M−N_p^′_yangles and visualization often provided

the best means to assess the deviations.

Characterizations. 1. Iron. Figure 6 illustrates (smif)₂Fe

(1-Fe), which has a regular D_2dstructure with no clear

distortions, as expected for a low spin d⁶configuration. The

azaallyl nitrogen−iron distances average 1.9012(14) Å, which is

significantly shorter than the corresponding Fe-py average

distance of 1.9634(12) Å. The azaallyl nitrogens are essentially

linear about the iron (179.11(6)°), and smif possesses a

N_azaFeN_pybite angle of 82.3(2)°, consistent with tight binding

in the low spin complex. Pyridine nitrogens on opposing smif

ligands are 91.0(12)° apart, and N_azaFeN_p^′_yangles average

97.7(5)°. The d(CN) of the azaallyl group is 1.333(3) Å,

consistent with significant double bond character.

1

Five and six resonances are observed in the H and ¹³C{¹H}

NMR spectra of (smif)₂Fe (1-Fe), respectively, with the azaallyl

proton at δ 7.59 and its accompanying carbon at δ 112.19. The

1

expected ten resonances were found in the H NMR spectrum

of (^oMesmif)₂Fe (2-Fe), and while most of the shifts were

reasonable, the azaallyl hydrogens were at δ 11.43 and 12.04

with linewidths of 29 and 46 Hz, respectively; furthermore, the

C³H proton of the unsubstituted pyridine was found at δ 13.31

with a line width of 52 Hz. Together, these broad, downfield

resonances suggest some paramagnetic character in the sample.

The dimethylated smif complex, (^oMe₂smif)₂Fe (3-Fe) is

1

characterized by a H NMR spectrum clearly characteristic of

a paramagnetic species,⁶⁵with shifts ranging from δ −9.64

((CH)₂N, ν_1/2∼ 110 Hz) to δ 167.44 (py-CH, ν_1/2= 53 Hz).

Figure 8 illustrates SQUID magnetic measurements for

(^oMesmif)₂Fe (2-Fe) and (^oMe₂smif)₂Fe (3-Fe) that corrobo-

rate the Mossbauer and NMR spectroscopic details above. The

̈

dimethylated smif derivative 3-Fe has a μ_eff(298 K) of ∼5.5 μ_B

from 50 to 300 K indicative of a high spin S = 2 center with

significant orbital or spin-orbit contributions,^66,67as is plausible

The diamagnetism observed for (smif)₂Fe (1-Fe) is

corroborated by the zero field Mossbauer spectrum shown in

̈

12419

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Table 1. Selected Crystallographic and Refinement Data for (smif)₂M (1-M; M = V, Cr, Mn, Fe, Co, Ni) and [(smif)₂M]OTf

(1⁺-M; M = Cr, Co)

1-V

1-Cr

1⁺-Cr

1-Mn

ab

,

ac

,

d

ac

,

formula

formula wt

space group

Z

C₂₄H₂₀N₆V

443.40

C₂₄H₂₀N₆Cr

444.46

C₂₉H₂₈N₆O₄F₃SCr

665.63

C₂₄H₂₀N₆Mn

447.40

P1

P1 bar

̅

4

̅

4

2

a, Å

8.940(6)

14.245(9)

17.526(9)

94.41(5)

98.06(5)

96.16(4)

2188(2)

1.346

8.9763(6)

14.4378(10)

16.9883(11)

94.086(4)

97.642(4)

97.152(4)

2156.5(3)

1.369

8.7769(4)

12.5575(6)

15.6932(7)

86.978(3)

76.027(3)

80.361(3)

1654.63(13)

1.336

8.9189(11)

14.4448(19)

17.849(2)

94.240(6)

98.287(6)

93.988(5)

2261.8(5)

1.314

b, Å

c, Å

α, deg

β, deg

γ, deg

V, Å³

ρ_calcd, g·cm⁻³

μ, mm⁻¹

temp, K

λ (Å)

0.477

0.554

0.467

0.606

100(2)

173(2)

296(2)

0.97890

0.71073

R indices

R1 = 0.0482

wR2 = 0.1444

R1 = 0.0500

wR2 = 0.1469

1.071

R1 = 0.0484

wR2 = 0.1008

R1 = 0.0816

wR2 = 0.1109

1.034

R1 = 0.0482

wR2 = 0.1135

R1 = 0.0677

wR2 = 0.1208

1.075

R1 = 0.0433

gh

,

[I > 2σ(I)]

R indices

wR2 = 0.0910

R1 = 0.0754

wR2 = 0.0997

1.024

gh

,

(all data)

i

GOF

1-Fe

1-Co

1⁺-Co

1-Ni

a

d

f

formula

C₂₄H₂₀N₆Fe

448.31

C₂₄H₂₀N₆Co

C₂₉H₂₈N₆O₄F₃SCo

672.56

C_25.5H_21.5N₆Ni

470.70

formula wt

space group

Z

451.39

P2₁/n

P1

C2/c

P1

̅

4

̅

4

8

a, Å

8.7442(4)

27.4138(14)

9.2149(4)

90

9.028(6)

14.398(9)

16.882(9)

93.92(5)

98.35(5)

97.35(4)

2145(2)

30.8068(12)

14.4243(5)

18.5224(7)

90

9.0129(8)

14.4774(12)

17.0517(14)

94.368(3)

97.829(3)

97.521(3)

2175.2(3)

1.437

b, Å

c, Å

α, deg

β, deg

γ, deg

V, Å³

ρ_calcd, g·cm⁻³

μ, mm⁻¹

temp, K

113.809(2)

90

126.141(2)

90

2020.93(16)

1.473

6646.9(4)

1.344

1.398

0.771

0.824

0.637

0.918

173(2)

100(2)

173(2)

213(2)

λ (Å)

0.71073

R1 = 0.0424

wR2 = 0.0932

R1 = 0.0586

wR2 = 0.1001

1.041

0.97890

0.71073

R indices

R1 = 0.0474

wR2 = 0.1325

R1 = 0.0488

wR2 = 0.1349

1.030

R1 = 0.0551

wR2 = 0.1424

R1 = 0.0782

wR2 = 0.1543

1.055

R1 = 0.0436

wR2 = 0.0795

R1 = 0.0882

wR2 = 0.0934

1.004

gh

,

[I > 2σ(I)]

R indices

gh

,

(all data)

i

GOF

a

b

c

The asymmetric unit contains two formula units. SQEEZE applied to 1/2 molecule of toluene per asymmetric unit. SQEEZE applied to toluene

d

f

molecule. The asymmetric unit contains one molecule of 1⁺-M and one molecule of THF; SQEEZE was applied to a second THF. The asymmetric

g

h

i

unit contains two molecules of 1-Ni and 1/2 molecule of C₆H₆. R1 = Σ∥F_o| − |F_c∥/Σ|F_o|. wR2 = [Σw(|F_o| − |F_c|)²/ΣwF_o]^1/2. GOF (all data) =

2

[Σw(|F_o| − |F_c|)²/(n − p)]^1/2, n = number of independent reflections, p = number of parameters.

for octahedral Fe(II) centers. The downturn in moment below

50 K is a consequence of zero field splitting (ZFS), a combination

of spin−orbit coupling and low symmetry effects (JulX fit, see

Supporting Information).^66,67In contrast, (^oMesmif)₂Fe (2-Fe)

has a μ_effof 1.22 μ_Bat 298 K that declines to ∼0.8 μ_Bat 30 K.

This steady decrease is consistent with temperature independent

magnetism (TIP) as a significant factor in the paramagnetism of

arises from mixing of a nearby excited state that is not thermally

populated, hence 2-Fe would appear to be on the cusp of

changing spin state, yet possesses an S = 0 ground state.

The aforementioned data suggests that the moderately strong

field imparted by the smif ligands in (smif)₂Fe (1-Fe) is

disturbed by single methylation of the ligand in the ortho-

positions, but the minor steric change does not disrupt the

binding enough to change the spin state. Nonetheless, the

methylation has brought an excited state into proximity from an

energy standpoint. Dimethylation of the smif is enough to

sterically hamper binding and significantly weaken its field

strength, thereby incurring a change to the high spin, S = 2

ground state found for (^oMe₂smif)₂Fe (3-Fe).

2-Fe, and also explains why its NMR and Mossbauer spectra were

̈

only slightly changed from that of (smif)₂Fe (1-Fe); the curve is

greater than that expected from TIP alone, hence a small amount

of paramagnetic impurity (PI, fit as S = 2) is likely present (JulX

fit (D is the zero field splitting parameters): g = 2.00, |D| = 0.729

cm⁻¹, E/D = 0.137, TIP = 286 × 10⁻⁶emu, PI = 2.7%). TIP

12420

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Figure 5. Common structural distortions of [(smif)₂M]ⁿ(n = 0, 1-M; n = 1, 1⁺-M).

The structural changes implicated in the characterizations of

(smif)₂Fe (1-Fe), (^oMesmif)₂Fe (2-Fe) and (^oMe₂smif)₂Fe

(3-Fe) should be corroborated by UV−vis spectroscopy, but

there are complications due to the intense intraligand (IL)

absorptions intrinsic to the smif ligand. A TDDFT calculation

of the spectrum of 1-Fe (see Supporting Information) was

helpful in assessing the bands, although the calculated spectrum

was blue-shifted by 0.24−0.39 eV depending on the feature.

The origins of the major absorptions were predominantly

intraligand (IL) in character, although several contain some

MLCT or LMCT character (Figure 9). The major band in 1-Fe

at 603 nm (ε = 17 800 M⁻¹cm⁻¹) is an IL band in which

charge is transferred from the smif-CNC^nbto smif-π* orbitals,

as are the other major bands at 506 (ε = 19 500 M⁻¹cm⁻¹),

473 (ε = 21 900 M⁻¹cm⁻¹), and 437 nm (ε = 42 000

M⁻¹cm⁻¹). Referring to Figure 3, the lowest energy band is

illustrated by the “red” transition, whereas the latter absorptions

correspond to the “blue” transitions according to orbital origins

given by the TDDFT calculation; some metal to ligand charge

transfer (MLCT) is also included. The d-d band(s) expected

for the distorted octahedral complex are completely dwarfed by

the IL features, but are likely to be in the 550 and 400 nm

regions according the calculations, and Δ_octis estimated to be

∼18 000 cm⁻¹with an accompanying B of ∼470 cm⁻¹from

Tanabe-Sugano fits of calculated bands adjusted for the blue

shift. Both values are consistent with a relatively covalent complex

also possible the 645, 603, and 560 nm bands are the ν_GS= 0 to

ν_ES= 0, 1, and 2 vibrational components, provided the ES

potential energy surface is substantially displaced from the GS.

The difference of ∼1250 cm⁻¹is comparable to ground state IR

absorptions that are assigned to the CNC linkage, and it is

likely that related excited state features are at similar energies.

Related explanations of features observed for mono-smif

compounds have been made.⁴³The band at 263 nm is another

IL band (″black-hashed″ in Figure 3), and the low energy

features at 885 nm and 790 nm are likely to be triplet

components of the IL bands.⁷⁰Their substantial intensities (ε =

2500 M⁻¹cm⁻¹) may be a function of how much the iron

mediates spin−orbit coupling, thereby relaxing the forbidden

character to the transitions.

o-Methylation of the smif in (^oMesmif)₂Fe (2-Fe) causes a

blue shift in the IL bands and lends credence to the possibility

that the “red IL band” reveals a distinct structural change in the

ES, provided the absorptions at 630, 580, and 530 nm are

vibrational components. The spectrum also shows an increase

in intensity of the “red IL band”, and features present in 1-Fe at

∼500 nm are lost, perhaps in accordance with the spectral

changes of the main IL bands. Dimethylation of the smif in high

spin (^oMe₂smif)₂Fe (3-Fe) affords major changes. The “red IL

band” has increased in intensity to 58 000 M⁻¹cm⁻¹while the

major “blue IL band” decreased to 32 000 M⁻¹cm⁻¹. The

origin of the intensity change is not clear, but the change in

profile suggests that the ES pertaining to the “red IL band” is

not as displaced relative to the GS as in 1-Fe and 2-Fe,

providing a reason for the intensity gain. Note that the low

energy features are diminished (850 nm, 1030 nm (ε ≈ 500

M⁻¹cm⁻¹)) in 3-Fe, suggesting that the S = 2 center is less

effective at helping relax the selection rules that permit

observation of spin forbidden IL bands.

implicated by its diamagnetism and Mossbauer parameters.

̈

There are two potential ″red″ IL bands according to Figure 3

(a₂b₁→ a₂b₁e¹, A₁→¹E; a₂b₁→ a₂b₁e¹, A₁→¹E), but

the TDDFT calculation shows only one band. The band at 560

nm (ε = 15 600 M⁻¹cm⁻¹) is not accounted for by the TDDFT

and may be the ν_GS= 0 to ν_ES= 1 vibrational component

affiliated with the 603 nm (ν_GS= 0 to ν_ES= 0) band.^68,69It is

2

1

2

1

2

1

12421

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

12422

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Figure 6. A molecular view of D_2d(smif)₂Fe (1-Fe, a.), and one of

(smif)₂Mn (1-Mn, b.) that illustrates a C_sdistortion.

2. Vanadium. The structure of (smif)₂V (1-V) has a

significant C_sdistortion (with slight C₂twist) observed in the

N_azaVN_a^′_zaangle of 172.1(4)° and varied N_azaVN_p^′_yangles

(95.08(10)-110.10(10)°). The d(VN_aza) distance of 2.058(18) Å

is 0.15 Å longer than those of 1-Fe, consistent with the greater

covalent radius of vanadium coupled with its low valent V(II)

center. The d(VN_py) distances of 2.118(7) Å are similarly

longer, and the N_azaVN_p^′_ybite angle is diminished to 77.3(2) as

the smif is farther off the metal.

Figure 8. SQUID data (1 T) for (^oMesmif)₂Fe (2-Fe) and (^oMe₂smif)₂Fe

(3-Fe); μ_eff(298K) = 1.22 μ_Bfor 2-Fe and μ_eff(298K) = 5.47 μ_Bfor 3-Fe.

(smif)₂Fe (1-Fe) is diamagnetic.

A quartet ground state was calculated for (smif)₂V (1-V), and

the EPR spectrum shown in Figure 10, taken at 6 K in toluene

glass, was consistent with a rhombic S = 3/2 spin system, albeit

with a small E/D of 0.05(2). An 8-line hyperfine coupling to

⁵¹V (I = 7/2, 99.76%) of ∼150 MHz was found for the

forbidden transition (M_s= −3/2 → M_s= 3/2). Figure 11

illustrates a plot of SQUID magnetometry data for 1-V that

shows a μ_effof 3.76 μ_Bat 300 K which remains fairly constant

until 50 K, when the effects of ZFS are evident; the data are

readily fit with the EPR parameters (See Supporting

Information). Evans’ method solution studies⁷¹afforded a μ_eff

of 3.2 μ_B, and while both solid state and solution values are

somewhat lower than the spin-only value of 3.87 μ_B, some

attenuation due to spin−orbit coupling is normal.^66,67

Figure 9. UV−vis spectra for (smif)₂Fe (1-Fe, green), (^oMesmif)₂Fe

(2-Fe, red), and (^oMe₂smif)₂Fe (3-Fe, blue) in pentane.

The UV−vis spectrum of (smif)₂V (1-V) is shown in Figure

12 with the spectra of the iron and chromium derivatives, and

while its greatest absorptions range from ∼8000−12 000 M⁻¹

cm⁻¹, they are clearly diminished with respect to iron, and

significant changes are apparent. The visible region of the

spectrum is essentially covered from its UV limit to ∼600 nm

́

by a melange of bands that are presumably IL or CT in origin.

From the calculations in Figure 3, MLCT bands are expected at

a lower energy than the lowest IL band, yet the spectrum is

Figure 7. Zero field Mossbauer spectra of (smif) Fe (1-Fe), (^oMesmif) Fe (2-Fe) and (^oMe smif) Fe (3-Fe) taken at 80 K. Spectra of 3-Fe contain

̈

2

variable amounts (∼10−20%) of an unidentified impurity (δ = 1.02 mm/s, ΔE_Q= 0.65 mm/s, Γ_FWHM= 0.50 mm/s) that was a component of the fit.

12423

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Figure 12. UV−vis spectra of (smif)₂M (1-M; M = V, Cr, Fe) in

pentane.

176.45(8)°) characterizes a secondary C_sdistortion, which is

also evident in the N_azaCrN′_pyand N_pyCrN′_pyangles.

Solution studies (Evans’ method)⁷¹and SQUID magneto-

metry data for (smif)₂Cr (1-Cr) gave μ_effof 2.67 μ_Bat 300 K,

consistent with the expected S = 1 ground state for a Cr(II)

derivative. In particular, the temperature dependence of μ_effwas

remarkably flat to 10 K, and the slight upturn at 5 K (1 T,

Figure 11) was not evident in data acquired at 3 and 5 T; it may

be due to a weak intermolecular interaction (JulX fit θ = 0.393 K).

In D_2dsymmetry the d_xy(b₂) and d_xz,d_yz(e) orbitals are

intrinsically different, and ZFS from the expected (b₂)²(e)²GS

configuration mixing with appropriate excited states should be

modest, but substantial structural deviations were observed in

the solid state. The |D| ≪ 1 cm⁻¹was thus quite unexpected,

and perhaps indicative of an unusual electronic configuration.

EPR spectra were obtained for the S = 1 system, and since

the |D| values for triplet systems often do not permit observation

of signals on standard X-band spectrometers (e.g., for 9 GHz;

systems with |D| > 0.30 cm⁻¹not observable), care was taken to

ensure that the samples were not compromised. Five sets of

variable temperature EPR data were obtained on three different

samples, and all were identical. Furthermore, temperature-

dependent spectral changes were verified for two data sets (i.e.,

reproducible spectra upon cooling−heating−cooling, etc.), and a

spin-counting experiment showed the signal to be 65% (relative

to CuSO₄) of that expected, a reasonable value. Comparative

spectra of [(smif)₂Cr]OTf (1⁺-Cr) ruled out the possibility of an

impurity due to a simple one electron oxidation.

Figure 10. EPR spectrum (dχ″/dB) of (smif)₂V (1-V) at 6 K, showing

g = 5.67 (forbidden line, I(V) = 7/2, A ≈ 150 MHz (5.00 × 10⁻³

cm⁻¹), 3.82 and 1.92 lines. Simulation of the spectrum was performed

with S = 3/2: g_x= g_y= 1.92, g_z= 1.91; |D| = 2.6 cm⁻¹; E/D = 0.05.

The signals containing hyperfines centered at g = 2.0 are likely because

of a trace V(IV) impurity.

Figure 11. SQUID magnetometry data (1 T) for (smif)₂M (1-M; M = V,

Cr, Mn, Co, Ni) and [(smif)₂M]OTf (1⁺-M; M = Cr, Mn); sight lines are

not data fits. For JulX fits, see Supporting Information.

Spectra of (smif)₂Cr (1-Cr) taken in toluene glass at 10 K

and in solution at 296 K are given in Figure 13. Simulation of

the low temperature spectrum was achieved with “D” =

0.00435 cm⁻¹(E/D = 0.276), and a fit obtained at 296 K

possessed similarly small values (“D” = 0.0029 cm⁻¹, E/D = 0.1);

the parameters were successfully used to fit the SQUID data.

Half field signals are often diagnostic of S = 1 systems, but in

the limit of D = 0, the intensity of the forbidden transition

diminishes, and in this instance small signals in the g ≈ 4 region

were more consistent with common trace Fe impurities.

Furthermore, “D” should not be observed in fluid solution, as

the anisotropic part should average out in solution; note that

some of the anisotropy is gone in fluid solution, and the spectra

were still fit because the “D” in this case is a phenomenological

one. In the absence of evidence pertaining to impurities, the

reproducible, phenomenological “D” may signify a system

where a ligand S = 1/2 component is mapped onto a Cr(III)

S = 3/2 environment, that is, an S = 3/2 antiferromagnetically

surprisingly devoid of features from ∼600 nm to broad, lower

intensity bands (ε ≈ 2,000 M⁻¹cm⁻¹) at 870 and 950 nm.

Similar absorptions appear in most other (smif)₂M (1-M), and

these appear to be “spin-forbidden” transitions affiliated with

the IL bands. Bands at 468 and 397 nm can be considered

plausible “blue IL” components according to Figure 3, but the

apparent blue shift of the “red IL” band is puzzling, unless the

ES of 1-V is displaced significantly from its GS geometry.

3. Chromium. The two (smif)₂Cr (1-Cr) molecules in the

asymmetric unit are quite similar, with both exhibiting C_2v

distortions as the major deviation from D_2d. Axial nitrogen

distances of 1.992(2) and 2.026(2) Å are significantly longer

than their counterparts of 1.9481(19) and 1.932(2) Å,

respectively, yet both axial distances are shorter than the

pyridine-nitrogen distances of 2.035(12) Å. A subtle lean of one

smif relative to the other (∠N_azaCrN_aza= 175.81(18)°,

12424

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Figure 13. Observed and simulated (red) EPR spectra (dχ″/dB) in toluene of (smif)₂Cr (1-Cr) taken at 20 K in a toluene matrix (a, simulation

parameters: S = 1, g_x= 1.979, g_y= 1.985, g_z= 2.007, “D” = 0.00435 cm⁻¹, E/D = 0.276) and in solution at 296 K (b, simulation parameters: S = 1,

g

_iso= 1.9885, “D” = 0.0029 cm⁻¹, E/D = 0.1); in both cases 9.5% of ⁵³Cr (I = 3/2, A_iso= 60 MHz (2.0 × 10⁻³cm⁻¹) was taken into account in the

simulations. The spectral changes were shown to be reversible. The observed “D” is a phenomenological observable (see text and Supporting

Information).

coupled to an S = 1/2 system does not necessarily have the

spectral signature of an S = 1 system. Unfortunately, the

expertise necessary to fully understand and vet the spin-

Hamiltonian for this system is lacking in this group, but

hopefully this phenomenological view of the system will interest

others. The data is nonetheless consistent with a ground state

that is (smif(−))(smif(2−))Cr(III), in line with K-edge X-ray

absorption spectroscopic measurements described below.

The UV−vis spectrum (Figure 12) of (smif)₂Cr (1-Cr) is

fairly similar to the iron derivative from ∼650 nm to high

energies, with the “red IL” band spanning the region from 530

to 650 nm (632 nm, ε ≈ 12 900 M⁻¹cm⁻¹), and a “blue IL”

band at 394 nm (ε ≈ 25 000 M⁻¹cm⁻¹); a similar feature at

489 nm (ε ≈ 16 300 M⁻¹cm⁻¹) is likely to be an MLCT band

or another “blue IL” band. Below 650 nm is a relatively flat

region (ε ≈ 6 000 M⁻¹cm⁻¹) extending to ∼850 nm that is

unique to the set of (smif)₂M (1-M) in this study. If 1-Cr is

reconfigured as (smif(−))(smif(2−))Cr(III), the absorptions

in this region may be assigned as ligand to metal charge transfer

(LMCT) bands, but the lack of related absorptions in (smif)₂V

(1-V) suggests an alternative. Redox noninnocent ligands⁷²⁻⁸¹

with radical character often exhibit ligand−ligand charge

transfer (LLCT) or ligand−ligand intervalence charge transfer

(IVCT) transitions,⁷²⁻⁷⁶and the bands in 1-Cr may fall in the

latter category, although their modest intensities belie such an

assignment. A flat region of ε ≈ 1000 M⁻¹cm⁻¹extends from

∼900−1150 nm that presumably encompasses triplet features

affiliated with the IL bands.

similarities are consistent with the low spin character attributed

to 1-Cr, but may also be indicative of noninnocent behavior of

the smif ligands.

For comparison, the cation [(smif)₂Cr]OTf (1⁺-Cr) was also

examined via SQUID magnetometry and EPR spectroscopy. As

Figure 12 illustrates, 1⁺-Cr manifests a μ_effof 3.76 μ_Bat 300 K,

and a modest amount of ZFS revealed at temperatures less than

25 K (JulX g = 2.050, D = −0.003 cm⁻¹, E/D = 0, TIP = 1768 ×

10⁻⁶emu, θ = −0.415 K). Some attenuation of the spin only value

of 3.87 μ_Bis common for Cr(III) in octahedral complexes,^66,67

and it is not unreasonable for it to be slightly more pronounced in

D_2d. Its EPR spectrum is rather featureless (see Supporting

Information), but consistent with a symmetric S = 3/2 center.

The UV−vis spectrum of [(smif)₂Cr]OTf (1⁺-Cr) is

illustrated in Figure 14 along with the other cations in the

The X-ray crystal structure of [(smif)₂Cr]OTf (1⁺-Cr)

revealed the cation to be much more symmetric than (smif)₂Cr

(1-Cr), with only a modest C₂distortion revealed by slightly

varying N_azaCrN_p^′_yangles (97.02(9)−102.13(9)°), and an

N_azaCrN_a^′_zaangle of 176.72(9)°. The d(CrN_aza) distances that

average 1.994(3) Å are actually slightly longer than the average

distances affiliated with 1-Cr (1.975(42) Å), perhaps because of

diminished overlaps for Cr(III), but the real surprise is how

similar they are. The chromium pyridine distances are identical

at 2.035(2) Å. In high spin Cr(II)/Cr(III) comparisons, the

chromous species is often 0.1−0.2 Å larger in d(CrL). These

Figure 14. UV−vis spectra of cations [(smif)₂Cr]OTf (1⁺-Cr),

[(smif)₂Mn]OTf (1⁺-Mn), and [(smif)₂Co]OTf (1⁺-Co) in THF.

system, and while it is significantly different than (smif)₂Cr (1-Cr)

in terms of intensity, the main IL bands have a distinct

correspondence to the neutral species. The “red IL” band at

626 nm (ε ≈ 22 000 M⁻¹cm⁻¹) is the most intense, and it is

slightly blue-shifted from 1-Cr, with a modest shoulder at ∼583

nm that could be a second red IL band or a vibrational

component. The band shape and lack of an obvious vibrational

12425

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Figure 15. (a) Normalized Cr K-edge spectra (10 K) for (smif)₂Cr (1-Cr) and [(smif)₂Cr]OTf (1⁺-Cr). (b) Expanded Cr K-pre-edge region

2

showing 1s → 3d_xz/3d_yzand higher energy 1s → 3d_{x −y}transitions.

progression is consistent with the low amount of distortion in

1⁺-Cr relative to its IL excited state. A “blue IL” band is clearly

evident at 385 nm (ε ≈ 13 200 M⁻¹cm⁻¹), but the band at

491 nm (ε ≈ 12 900 M⁻¹cm⁻¹) may be an MLCT feature or

another IL band. A low energy feature at 741 nm is again likely

to be a triplet absorption associated with an IL band.

and little ZFS evident (JulX g = 2.03, D = 1.00 cm⁻¹, E/D =

0.250). A broad featured EPR spectrum (see Supporting

Information) of 1-Mn obtained at 8 K was consistent with an

S = 5/2 center; at 296 K a broad, featureless resonance was observed

at g_iso= 2.02.

The UV−vis spectrum (Figure 16) of (smif)₂Mn features the

“red IL band” at 588 nm with the largest molar absorptivity

In view of the structural similarities in (smif)₂Cr (1-Cr) and

[(smif)₂Cr]OTf (1⁺-Cr), the unusually small D found in the

EPR spectra and SQUID data of (smif)₂Cr (1-Cr), and the

correspondence in band maxima of their respective UV−vis

spectra, it is prudent to recognize that calculations suggested

1-Cr is best formulated as (smif(−))(smif(2−))Cr(III), that is,

the smif ligand is noninnocent⁷²⁻⁸¹in its ability to

accommodate excess charge. High-energy spectroscopies, in

particular K-edge spectroscopy, have proven to be useful in

discerning the effective charge at metal centers when

appropriate models are available.⁸²⁻⁸⁴As Figure 15 reveals,

comparison of the normalized Cr K-edge spectra for 1-Cr and

1⁺-Cr indicates a similar effective nuclear charge for both

species, indicative of a ligand-based oxidation process

pertaining to removal of an electron from smif(2−). Both

2

compounds exhibit 1s → 3d_xz/3d_yzand 1s → 3d_{x −y}transitions

(confirmed by calculations) at roughly the same energies, and

the crude Δ_octof 17 500−19 500 cm⁻¹obtained from their

difference is a reasonable field strength commensurate with the

structural parameters. As suggested by the calculations, the

ground state of 1-Cr appears to be a Cr(III) S = 3/2 center

antiferromagnetically coupled to a smif S = 1/2 center.

4. Manganese. Given the likelihood of (smif)₂Mn (1-Mn)

being high spin, its distorted structure was not surprising, since

longer bond lengths affiliated with weaker Mn-smif binding and

the modest bite-angle of smif enable deviations from D_2dto be

energetically feasible. Figure 6b illustrates the C_sdistortion in

1-Mn, and a secondary C₂twist is also present. The N_aza−Mn

distances in the two independent molecules ranged from

2.186(3) to 2.220(3) Å and were accompanied by N_aza−Mn−

N_a^′_zaangles of 169.22(10)° and 166.51(9)°, while the Mn−

N_pydistances were the longest of the neutral derivatives at

2.235(7) Å. The scale of the distortions is most evident in the

∼19° spread in N_azaMnN_p^′_yangles.

Figure 16. UV−vis spectra of [(smif)₂M] (1-M; M = Mn, Co, Ni, Zn)

in pentane.

observed (ε ≈ 60 000 M⁻¹cm⁻¹), a shoulder at 557 nm (ε ≈

39 000 M⁻¹cm⁻¹) that is either a vibrational component or

another IL band, and a single “blue IL band” at 400 nm (ε ≈

25 000 M⁻¹cm⁻¹). Lower energy shoulders off the latter band

at 442 (ε ≈ 13 000 M⁻¹cm⁻¹) and 457 (ε ≈ 8000 M⁻¹cm⁻¹)

are likely to be additional IL absorptions or charge-transfer

bands. Note that the situation has reversed from Figure 12;

now the lower energy IL band is the most intense, and it is

blue-shifted from where the ν_GS= 0 to ν_ES= 0 components

were proposed for 1-Cr and 1-Fe. There are also smaller

features at ∼635 and ∼750 nm that are common to 1-M where

M = Zn, Co, and Ni. These are likely to be triplet absorptions

affiliated with the IL transitions with significant singlet

character admixed.

The UV−vis features of 1-Mn are remarkably similar to the

broader IL absorptions of [(smif)₂Mn]OTf (1⁺-Mn, Figure 14),

which are observed at 570 (ε ≈ 26 000 M⁻¹cm⁻¹) and 399 nm

(smif)₂Mn (1-Mn) is an S = 5/2 molecule, as the SQUID

magnetometry data in Figure 11 indicates. The μ_effat 300 K is

5.73 μ_B, very near the expected 5.9 μ_Bfor a spin-only system,

12426

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

(ε ≈ 9,700 M⁻¹cm⁻¹). While no X-ray crystal structure of

1⁺-Mn was obtained, SQUID magnetometry revealed a μ_effof

5.43 for the cation (Figure 11) augmented by a slight amount

of TIP, but with a noticeably greater amount of ZFS. In D_2d

2

symmetry, the d_z(a₁) and d_{x ‑y}(b₁) orbitals are intrinsically

different, hence a significant ZFS is expected for a high spin d⁴

configuration. In addition, the μ_effvalue is high for an S = 2

center (μ(spin only) = 4.9 μ_B) that should have no orbital

contribution. A Guoy balance measurement afforded a μ_effof

5.0 μ_B, which is certainly closer to the spin-only moment, yet

still high. Spin−orbit contributions from triplet states with

significant orbital angular momentum can be the origin of

upward deviations from the spin-only value for Mn(III).

5. Cobalt. Two molecules of (smif)₂Co (1-Co) were again

present in the asymmetric unit, but in this case the metric

parameters of the two were statistically distinct. In molecule

one, two significantly different d(CoN_aza) of 1.946(3) and

1.888(3) Å are accompanied by two d(CoN_py) sets that average

2.184(13) Å and 1.971(13) Å, respectively. The resulting C_2v

distortion is augmented by a C_sdistortion evident in the

N_azaCoN_a^′_zaangle of 177.30(11)° and N_aza−Co−N′_pyangles that

vary from 95 to 101°. Molecule two has roughly identical

CoN_azadistances of 1.945(3) and 1.939(3) Å with accompany-

ing CoN_pyaverage distances of 2.105(16) and 2.053(5) Å,

respectively. While the remaining core angles are similar to

molecule one, its distortion is best construed as C₂.

Figure 17. EPR spectrum (dχ″/dB) of rhombic S = 1/2 (smif)₂Co

(1-Co, I = 7/2) and fit (red): g_x= 2.01, A_x(Co) = 140 MHz (4.67 ×

10⁻³cm⁻¹); g_y= 2.14, A_y(Co) = 200 MHz (6.67 × 10⁻³cm⁻¹); g_z=

2.21, A_z(Co) = 200 MHz (6.67 × 10⁻³cm⁻¹). The fit includes a 0.1%

impurity modeled as an isotropic species with g = 2.055, A(Co) =

50 MHz (1.67 × 10⁻³cm⁻¹), and a line width 1/7 of the others.

In stark contrast to the structural vagaries of (smif)₂Co (1-Co),

the corresponding diamagnetic cation [(smif)₂Co]OTf (1⁺-Co)

has a regular D_2dstructure with shorter bond distances attri-

butable to low spin Co(III): d(CoN_aza) = 1.8768(11) Å and

d(CoN_py) = 1.9252(19) Å. The N_azaCoN_a^′_zaangle approaches

linearity at 179.05(10)°, and the average N_a^′_zaCoN^′_pyand

N_pyCoN^′_pyangles of 96.1(7)° and 90.7(19)°, respectively, are a

testament to the regularity of the structure.

The μ_effof 2.8 μ_Bobtained for (smif)₂Co (1-Co) by

Evans’ measurements (6 trials) at 293 K was confirmed by

Gouy balance (2.8 μ _Bat 294 K), prompting further

investigation. As implied by the room temperature measure-

ments, spin crossover behavior was observed in SQUID

magnetometry data, as revealed by a decline in the μ_efffrom

3.19 μ_Bat 300 K to 1.75 μ_Bat 10 K.⁸⁵The Evans’⁷¹and Gouy

balance values are consistent with a rough 1:1 mixture of S =

3/2 and S = 1/2 species at room temperature; at temperatures

<100 K, there is little high spin Co(II) remaining.

Figure 17 illustrates the EPR spectrum of (smif)₂Co (1-Co)

at 30 K, where virtually all of the Co is in the S = 1/2

configuration. The temperatures at which the S = 3/2 species

has a reasonable concentration typically lead to broadened

signals due to increased relaxation and they are often not

observed; no signals were observed in toluene solution at 296 K.

While S = 1/2 species can be often seen at room temperature, in

this instance exchange with the S = 3/2 species may be rapid,

and other relaxation mechanisms are also common. The cobalt

hyperfine couplings in this rhombic system are quite uniform,

ranging from 48 to 66 G (140−200 MHz), and differ from

related coordination compounds such as [(terpy)₂Co]²⁺(Co

hyperfine) in which the coupling is disparate.⁸⁵

Despite the spectroscopic investigations, the origin of the

structural discrepancy between the two molecules in the

asymmetric unit is still uncertain, and there were different

scenarios that could explain the data. The crystal structure

reported above was obtained at 100 K using the Cornell High

Energy Synchrotron Source (CHESS), but a second set was

also obtained at 173 K on a conventional diffractometer. While

the data in the latter collection resulted in a lesser quality

solution, the metric parameters for the two essentially matched,

and the two molecules in the asymmetric unit were still

statistically different. Neither molecule could be attributed to an

S = 3/2 species, as both sets of bond distances were more

consistent with a low spin Co species, and calculated high spin

versions gave metric parameters that were well off the experi-

mental values.

While it is conceivable that “crystal packing” effects

determine the different geometries in the asymmetric unit, it

seems very unlikely that such modest interactions can change

bond distances to the degree that is observed. Furthermore,

inspection of the asymmetric unit failed to uncover a significant

interatomic interaction capable of inducing a significant struc-

tural change. The remaining scenario regarding the independ-

ent geometries is the possibility of several S = 1/2 ground states

of nearly the same energy. Although this has been probed and

verified by high-level calculations,⁶³one additional experiment

was conducted to limit the scope of plausible S = 1/2 states,

since initial calculations suggested an electronic configuration

of (d_xy)²(d_xz,d_yz)⁴(CNC^nb)⁴(smif-π*)¹(Figure 4), that is, a low-

spin Co(III) center and an additional electron in a π* orbital of

the smif ligand.

Metal K-edge X-ray absorption spectroscopic measure-

ments⁸²⁻⁸⁴were utilized to assign the orbital parentage of the

odd electron in S = 1/2 (smif)₂Co (1-Co) through comparison

to [(smif)₂Co]OTf (1⁺-Co). As Figure 18 illustrates, a significant

shift (∼2 eV) in the leading edge to higher energies is observed

when 1-Co is oxidized to 1⁺-Co. In addition, the 1s to 3d pre-

edge features are also shifted by ∼1.5 eV, consistent with a metal-

based, rather than ligand-based, oxidation process. The data

clearly portray the electronic configuration from Figure 4 as

incorrect, and 1-Co is thus considered a Co(II) complex. As a

consequence, the two distinct geometries evident in the crystal

12427

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Figure 18. (a) Comparison of the Co K-edge spectra (10 K) for (smif)₂Co (1-Co) and [(smif)₂Co]OTf (1⁺-Co) that indicates an increase of ∼2 eV

in the effective nuclear charge on Co upon oxidation. (b) Expansion of the pre-edge region, showing that the ligand field has increased by ∼1.5 eV

upon oxidation.

structure of 1-Co are likely to be two S = 1/2 Co(II) species

whose energies are close.

spectral measurements are consistent with a standard d⁸Ni(II)

center.

If the conventional electronic assessment of (smif)₂Co (1-Co)

as Co(II) is correct, its UV−vis spectrum should be

significantly different than that of the corresponding cation,

[(smif)₂Co]OTf (1⁺-Co). From views of Figures 14 and 16,

this is correct, although each spectrum is dominated by the IL

features common to all bis-smif derivatives. The “red IL”

absorption in 1-Co has a maximum at 555 nm (ε ≈ 29 000 M⁻¹

cm⁻¹) with a low energy shoulder at ∼610 nm and a high

energy shoulder at ∼515 nm that may be construed as

vibrational components. Very low energy features at ∼666 nm

and ∼724 nm are likely to be triplet states that accompany

the intense IL transitions. There are additional complications

due to the possibility of LMCT bands, and the band shape of

the 555 nm absorption is distorted enough to encourage

speculation. A single “blue IL” band is observed at 400 nm (ε ≈

22 100 M⁻¹cm⁻¹) and it has shoulders consistent with

additional IL or CT absorptions. The cation 1⁺-Co has a

broad, featureless band at 591 nm (ε ≈ 23 000 M⁻¹cm⁻¹)

assigned to the “red IL” band, and two less intense “blue IL”

absorptions at 385 (ε ≈ 13 000 M⁻¹cm⁻¹) and 320 nm (ε ≈

12 000 M⁻¹cm⁻¹) with a low energy shoulder on the former.

6. Nickel. In (smif)₂Ni (1-Ni), the d(NiN_aza) and d(NiN_py)

distances of 2.019(5) and 2.093(9) Å are elongated relative to

Co because of the single occupation of two sigma-antibonding

orbitals, according to Figure 4. Only high spin d⁵(smif)₂Mn

(1-Mn) has longer distances among the first row metals whose

covalent radii are roughly the same. The remaining parameters

reveal subtle C₂and C_sdistortions to the two molecules in the

asymmetric unit.

7. Zinc. (smif)₂Zn (1-Zn) was prepared by Westerhausen²⁵

and structurally characterized, thus the examination of

diamagnetic 1-Zn was limited to standard NMR character-

ization and UV−vis spectroscopy. The “red IL” band at 566 nm

in the spectrum of 1-Zn is again dominant (ε ≈ 24 000 M⁻¹

cm⁻¹), and possesses a high energy shoulder at 537 nm. The

single “blue IL” band is at 396 nm (ε ≈ 8700 M⁻¹cm⁻¹) with a

series of low energy shoulders consistent with a vibrational

progression of ∼1200 cm⁻¹similar to those previously

observed. The low-lying absorption at ∼635 nm may be

attributed to a triplet corresponding to an IL band.

8. Ruthenium, Rhodium, and Iridium. The two second row

complexes, (smif)₂Ru (1-Ru) and [(smif)₂Rh]OTf (1⁺-Rh),

and the third row [(smif)₂Ir]BPh₄(1⁺-Ir) are all low spin,

diamagnetic d⁶derivatives that were not structurally charac-

1

terized. Each possesses five H NMR spectral resonances and

six ¹³C{¹H} NMR signals indicative of D_2dsymmetry in

solution, and the usual intense UV−vis spectral bands

associated with the smif IL transitions (Figure 19). The

(smif)₂Ni (1-Ni) is “EPR silent”, indicative of a ZFS large

enough to obviate X-band observation. SQUID magnetometry

(Figure 11) indicates a μ_effof 2.81 μ_Bat 300 K and a modest

decline below 10 K because of ZFS. The UV−vis spectrum of

1-Ni consists of dominant “red IL” band at 571 (ε ≈ 50 000

M⁻¹cm⁻¹) with a shoulder at ∼540 nm (ε ≈ 31 500 M⁻¹

cm⁻¹), and a lesser “blue IL” absorption at 398 nm (ε ≈ 19 000

M⁻¹cm⁻¹) with a low energy shoulder at ∼425 nm (ε ≈ 12 000

M⁻¹cm⁻¹). A low-lying absorption at 652 nm (ε ≈ 2300 M⁻¹

cm⁻¹) is consistent with triplet excitation affiliated with an IL

band, as has been observed in several previous cases. All

Figure 19. UV−vis spectra of (smif)₂Ru (1-Ru), [(smif)₂Rh]OTf (1⁺-Rh),

and [(smif)₂Ir]BPh₄(1⁺-Ir).

spectrum of 1-Ru reveals the “red IL band” at 566 nm (ε ≈ 10 000

M⁻¹cm⁻¹) as a broad featureless absorbance flanked by a lesser

shoulder at ∼650 nm (ε ≈ 3600 M⁻¹cm⁻¹) and a band at 740 nm

12428

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

(ε ≈ 1000 M⁻¹cm⁻¹). The latter two absorptions may be triplets

affiliated the IL transitions, most likely the intense “blue IL”

transition at 417 nm (ε ≈ 33 000 M⁻¹cm⁻¹). A band at ∼505 nm

(ε ≈ 10 500 M⁻¹cm⁻¹) is likely to be another IL component or an

MLCT transition.

sphere highly susceptible to the minor structural perturbations.

There is no indication from the SQUID and EPR data that the

molecule has significant distortions due to electronic factors.

The bis-smif nickel derivative, 1-Ni, also has only small distortions,

3

1

2

consistent with its B₂ground state (i.e., (d_z) (d_{x −y}) ), and the

core distances are long, as expected for pseudo-octahedral Ni(II).

The cobalt derivative (smif)₂Co (1-Co), whose structure at

100 K represents a low spin d⁷configuration according to

K-edge spectroscopy, is considerably distorted from D_2d

symmetry in each statistically different, independent molecule.

In principle, the ²B₂state (D_2d) should reflect an axial

The lowest energy IL bands, that is, the “red-IL bands”, are

red-shifted in [(smif)₂Rh]OTf (1⁺-Rh), and [(smif)₂Ir]BPh₄

(1⁺-Ir) relative to (smif)₂Ru (1-Ru), and they are the most

intense, a change also seen when (smif)₂Fe (1-Fe) is compared

to (smif)₂Co (1-Co) and [(smif)₂Co]OTf (1⁺-Co). For 1⁺-Rh,

the 618 nm (ε ≈ 30 000 M⁻¹cm⁻¹) absorption has a shoulder

at 580 nm (ε ≈ 17 600 M⁻¹cm⁻¹), and these are significantly

more intense than the “blue IL band” at 385 nm (ε ≈ 12 400

M⁻¹cm⁻¹) with its accompanying shoulder at 425 nm (ε ≈

7300 M⁻¹cm⁻¹). The typical IL features are attenuated in 1⁺-Ir,

whose most prominent absorption occurs at 658 nm (ε ≈ 14

700 M⁻¹cm⁻¹), with apparent vibrational components at 623

nm (ε ≈ 12 900 M⁻¹cm⁻¹) and 570 nm (ε ≈ 5800 M⁻¹cm⁻¹).

Its “blue IL band” has an apparent vibrational progression

witnessed as similarly intense (ε ≈ 5000 M⁻¹cm⁻¹)

absorptions at 405, 386, 366, and 346 nm.

2

compression/equatorial elongation, and vice versa for a A₁

state, but such standard distortions are not compatible with

tridentate ligands. One molecule of 1-Co shows a clear

asymmetry (C_2v) in the binding of the two smif ligands, with

one roughly 0.06 Å closer to the cobalt at the N_azaposition and

0.21 Å closer at the pyridine-N positions. The distortion of the

second molecule is not so clearly seen, but a C₂twist is the

main change.

The most interesting GS configurations belong to the

vanadium and chromium derivatives, (smif)₂V (1-V) and

(smif)₂Cr (1-Cr), respectively. A C_sdistortion is found for

1-V, and the N_aza−V−N^′_azaangle of 172.1(4)° is the most severe

cant of a smif seen aside from those found for 1-Mn. If the

compound was a true D_2dd³V(II) system (i.e., ⁴B₁), no orbital

impetus for such a distortion would be expected. The EPR

spectrum (Figure 10) also reveals a modest but significant

distortion in the S = 3/2 system, and SQUID magnetometry

(Figure 11) indicates a large ZFS not expected for a d³

configuration (cf [(smif)₂Cr]OTf (1⁺-Cr)). The “d_yzorbital”

in the calculation of 1-V is the HOMO in the system, and its

composition is greatly mixed with a ligand π* component. It is

possible the system should be considered (smif(−))-

(smif(2−))V(III), that is, the smif ligand is redox noninnocent,

but this would require a ferromagnetic coupling of the smif

electron with the remaining d²core, whereas the opposite is

seen for 1-Cr. When an electron is promoted into a ligand, the

spatial separation from the remaining valence electrons causes a

decrease in coulomb or exchange energy affiliated with every

pairwise exchange. For a pseudo-octahedral d³case, “redox

noninnocence” would actually cause a loss in exchange energy

(stabilization) that would need to be compensated by the

stability of the d²π*¹configuration. If the calculations in Figure

3 are reasonable, the smif π* orbitals are still energetically far

from the “t_2g” set, and it may not be feasible for the transfer to

occur. Instead the HOMO is highly mixed, and the compound

behaves as a standard S = 3/2 system, albeit with the modest

asymmetry indicated by a large ZFS, which may be derived

from the a significant ligand component to the GS

configuration. Unfortunately, K-edge measurements, which

could provide greater insight into the GS composition, were

not obtained on 1-V. The distances of the core are consistent

with either V(II) or V(III) since the electron in question with

either occupy a π^bd-orbital or a ligand π*-orbital and the latter

would not be expected to impart a noticeable bond length

change from the former.

DISCUSSION

■

Syntheses of (smif)₂M and (smif)₂M⁺. An unusual series

of neutral “Werner complexes” has been prepared by using the

1,3-di-(2-pyridyl)-2-azaallyl ligand, coined “smif”. Straightforward

metatheses of transition metal salts with (smif)M (M =

Li, Na), and the addition of 1,3-di-(2-pyridyl)-2-azapropene,

(smif)H, to appropriate metal amides led to the desired (smif)₂M

(1-M, M = Cr, Mn, Fe, Co, Ni, Ru) and [(smif)₂Ir]BPh₄(1⁺-Ir)

complexes in modest to excellent yields. An additional reducing

equiv was used to prepare (smif)₂V (1-V) from VCl₃(THF)₃,

and, surprisingly, the oxidant AgOTf was compatible with the

smif anion during the course of its reaction with Rh₂(O₂CCF₃)₄

to afford [(smif)₂Rh]OTf (1⁺-Rh). Presumably this is a fortunate

circumstance of relative reagent solubilities and relative rates.

Using electrochemical measurements as a guide, mild Ag⁺

oxidations cleanly provided [(smif)₂M]OTf (1⁺-M; M = Cr,

Mn, Co). For those oxidations that failed despite favorable

electrochemical indications, it is likely that reactions of the

CNC-backbone of the smif ligand,^48,57perhaps due to ligand-

based oxidations,⁸⁶lead to degradation of cations on a chemical

time scale that is slower than the sweep rates used in the CV

experiments.

Electronic Factors Influencing (smif)₂M and (smif)₂M⁺

Structures. A number of the (smif)₂M (1-M, M = V, Cr, Mn,

Fe, Co, Ni) and [(smif)₂M]OTf (1⁺-M; M = Cr, Co)

complexes have been structurally characterized and details

have been given above. In general, deviations from D_2d

symmetry may be construed as arising from their d-counts and

the <180° bite angle (N_py−M−N_p^′_y) intrinsic to the smif ligand.

As compared to rigorously octahedral species, the electronic

asymmetry inherent to 6-coordinate D_2dcomplexes manifests

itself in EPR spectra and SQUID measurements, but these

contributions are minor.

The most symmetric species are the low spin d⁶complexes

[(smif)₂Co]OTf (1⁺-Co) and (smif)₂Fe (1-Fe), and the d³

chromium cation, [(smif)₂Cr]OTf (1⁺-Cr), which all have short

d(M−N_aza) and d(M−N_py) consistent with significant ligand

field stabilization energies. The most distorted of the remaining

cases is (smif)₂Mn (1-Mn). Its high spin d⁵configuration leads

to long metal−ligand distances and weaker bonding, and the

expected lack of covalency for Mn(II) renders the coordination

In the case of (smif)₂Cr (1-Cr), K-edge spectroscopy

suggests that its GS configuration is (smif(−))(smif(2−))Cr-

(III), or d³Cr(III) antiferromagnetically coupled to a smif π*

electron, that is, d³π*¹. Here it is important to recognize that

there are three configurations of interest: high- and low-spin d⁴

and d³π*¹. It is plausible that the high and low spin forms of

Cr(II) are likely to be energetically similar, given the above

12429

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

Figure 20. Simplified decomposition of the CNC^nbsmif backbone “allyl” orbital (p orbital components A and B; electrons 1 and 2), ignoring

contributions from pyridine π-orbitals.

assessments of the smif field strength. The resulting redox

noninnocent d³π*¹configuration represents a situation that

relieves the sigma-antibonding character of a HS (high spin) d⁴

GS, or one that relieves the coulomb interaction of a LS d⁴GS

by spatially separating the “paired electrons”. In addition, the

pairwise exchange energies in the d³core of the Cr(III) center

will be more favorable as the 3d orbitals contract. Evidence for

the contraction is seen in the d(Cr−N), which compares

favorably with those of [(smif)₂Cr]OTf (1⁺-Cr). The difference

between the d(Cr−N) of (bipy)₃Cr²⁺and (bipy)₃Cr³⁺is about

0.10 Å,^3,87and it is possible that the former is actually

(bipy)₂(bipy(−))Cr³⁺by the same reasoning. Consistent with

this argument is the ∼10⁹M⁻¹s⁻¹self-exchange rate for

(bipy)₃Cr^2+/3+, which is only sensible if the chromous center is

low spin or contains a redox “noninnocent” bipy ligand.⁸⁸

The EPR spectrum of (smif)₂Cr (1-Cr) is shown in Figure 13

and simulated as a “very low |D|”, S = 1 system. It is unusual to

find observable S = 1 systems, and this particular case may be

justified by rationalizing the chromium to be a highly

symmetric Cr(III) center AF-coupled to a smif radical dianion.

It is conceivable that in related Cr(II) systems with potentially

redox active ligands, EPR spectra may provide a signature for

redox active ligands antiferromagnetically coupled to metals. Fast

electron transfer rates, such as the self-exchange rates, may also

be indicative of electrons in redox “noninnocent” ligands.

UV−vis Spectroscopy. The individual descriptions of the

(smif)₂M^0/+species are familiar in the sense that most of their

spectroscopic and magnetic behavior typifies Werner-type

coordination compounds, with modest changes intrinsic to

the bis-smif framework. The individual species have already

been discussed above.

1⁺-M) have considerable anionic character localized on the

CNC^nbportion of the orbitals, and transfer of this charge to the

pyridines in the IL excited states renders a large change in

dipole moment. The resulting intensities of the IL bands dwarf

those of the MLCT bands common to virtually all other

Werner complexes.³⁻⁴⁰Betley’s recent examination of dipyrro-

methane and related ligands^89,90revealed extraordinary UV−vis

absorption intensities due to related intraligand transitions, and

it appears that smif shares some of their features. Figure 20

illlustrates a simple view of the CNC^nborbital, where its carbon

backbone p-orbital components can be shown to possess either

ionic or covalent “diradical” character, according to the

expansion of the orbital component of its wave function. In a

future submission, the C−C bond-forming chemistry originat-

ing from the CNC^nbbackbone will be elucidated.⁵⁷

Because of the nonbonding character of the CNC^nbsmif

orbitals (Figure 3), they are energetically near the d-orbitals for

all the first row metals. For (smif)₂M (1-M, M = V, Cr, and

likely Mn), they are slightly below the “t_2g” set, for 1-Fe they

are slightly above the “t_2g” orbitals, and for 1-M (M = Co, Ni),

they are amid the pseudo-octahedral field. For all 1-M

calculated, the CNC^nborbitals are not substantially perturbed

by changes in M, nor are the corresponding smif π*-orbitals,

hence the relative consistency of the “red IL” (λ_max≈ 590(30)

nm and “blue IL” (λ _max≈ 420(40)) band maxima.

Unfortunately, the intensity of these bands obscures much of

the other features of interest, and makes locating d-d bands

implausible.

The tremendous intensities of the IL bands permit ready

identification in all of the UV−vis spectra pertaining to

(smif)₂M (1-M) aside from 1-V, and the “red IL” bands often

feature an apparent vibrational progression, typically around

∼1100−1200 cm⁻¹.⁴³IR spectra of 1-M manifest several

absorptions in this region, and it is likely that the relevant

excited states feature similar vibrations that are appropriately

coupled. In some instances, similar progressions are observed

for the “blue IL” transitions, although the vibrations appear at

slightly higher frequencies. There is a noticeable change in the

relative intensities of the IL bands upon moving from 1-Zn to

1-V. For 1-Zn to 1-Co, the “red IL” band is significantly more

intense than its blue partner, whereas for 1-Fe, (o-Mesmif)₂Fe

The major difference in bis-smif derivatives, relative to

simpler N-donor complexes, stems from the unique electronic

features of the smif anion, and its HOMO, the CNC^nb

backbone orbital with opposing phases on the carbons adjacent

to the nitrogen that contains a nodal plane (Figure 20).

Transitions arising from the promotion of electrons in these

orbitals (Figure 3; linear combinations a₂and b₁) to smif π*

orbitals located mostly on the pyridines are the origin of the

“red IL” and “blue IL” UV−vis bands in each bis-smif

compound. The HOMOs of [(smif)₂M]ⁿ(n = 0, 1-M; n = 1,

12430

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

toluene-d₈were dried over sodium, vacuum transferred and stored

over activated 4 Å molecular sieves. THF-d₈was dried over sodium

and vacuum transferred from sodium benzophenone ketyl prior to use.

VCl₃(THF)₃,⁵⁸CrCl₂(THF),⁴¹Cr{N(TMS)₂}(THF)₂,⁴²

FeBr₂(THF)₂,⁵³Fe{N(TMS)₂}₂(THF),⁵²NiCl₂(DME),⁵⁴sodium

bis(trimethylsilyl)amide,⁵¹and 1,3-di-(2-pyridyl)-2-azapropene

(smifH)⁵⁰were prepared according to literature procedures. Lithium

bis(trimethylsilyl)amide was purchased from Aldrich and recrystallized

from hexanes prior to use. All other chemicals were commercially

available and used as received.

(2-Fe), and 1-Cr, the situation subtly reverses. The breadth of

these absorptions, whether due to vibrational progressions or

multiple IL transitions, hampers the determination of the

overall relative intensities. Interestingly, the high spin 1-Mn and

(o-Me₂smif)₂Fe (3-Fe) cases clearly have “red IL” bands that

are significantly more intense than their “blue IL” counterparts.

In the cations [(smif)₂M]⁺(1⁺-M, M = Cr, Mn, Co, Rh, Ir), the

“red IL” band is clearly dominant, even for the second row

species, yet the neutral 1-Ru has a similar overall spectrum as its

iron congener.

The majority of the complexes exhibit weaker features

between 650 and 1100 nm (the spectra are devoid of

absorptions to 1700 nm). These are formulated as singlet →

triplet absorptions whose intensity may be dependent on how

well the metal helps the “intensity stealing”, that is, how

effective the metal aids in mixing singlet character into the

triplet wave functions of the excited states. Calculations on the

lowest lying triplet state of (smif)⁻place it ∼1.3 eV above the

singlet. Using this as a guide,⁹¹the weak features are likely to

be triplet components of the “blue IL” bands, since they lie

∼1.0−1.7 eV below the 380−440 nm region.

The similarity of the IL and plausible MLCT features in

UV−vis spectra of (smif)₂Cr (1-Cr) and [(smif)₂Cr]OTf

(1⁺-Cr) provided a corroboration of the proposed electronic

configuration of 1-Cr as (smif(−))(smif(2−))Cr(III), the

redox noninnocence of smif. However, the spectrum of 1-Cr

shows features too intense to be S → T absorptions in the

region from 675−850 nm (ε ≈ 6,000 M⁻¹cm⁻¹). The existence

of a redox “noninnocent” ligand containing an electron in a π*-

orbital is often detected via the appearance of an intervalence

charge transfer (IVCT) band,⁷²⁻⁷⁶which can be vibrationally

broadened.⁷⁴A low energy, that is, in the red or near IR, IVCT

band is featured in systems having one normal ligand and one

possessing radical character, and is often very intense, with

extinction coefficients as high as 60 000 M⁻¹cm⁻¹. In many of

these cases, the π-systems responsible for the transition are

roughly coplanar, as in square planar complexes. In (smif)₂Cr,

the π-systems of the smif ligands are essentially orthogonal. If

these bands are of the IVCT type and not MLCT transitions,

perhaps the orthogonality of the smif π*-orbitals are

attenuating the intensities.

NMR spectra were obtained using an INOVA 400 and 500 MHz

spectrometers. Chemical shifts are reported relative to benzene-d₆(¹H

δ 7.16; ¹³C{¹H} δ 128.39), toluene-d₈(¹H δ 2.09; ¹³C{¹H} δ 20.4),

and THF-d₈(¹H δ 3.58; ¹³C{¹H} δ 67.57). Infrared spectra were

recorded on a Nicolet Avatar 370 DTGX spectrophotometer

interfaced to an IBM PC (OMNIC software). UV−vis spectra were

obtained on a Shimadzu UV-2102 interfaced to an IBM PC (UV

Probe software). Solution magnetic measurements were conducted via

Evans’ method in toluene-d₈.⁷¹Solid state magnetic measurements

were performed using a Johnson Matthey magnetic susceptibility

balance calibrated with HgCo(SCN)₄. Elemental analyses were

performed at the University of Erlangen-Nuremberg and Robertson

Analytical (New Jersey).

Procedures. 1. ^oMesmifH. To a suspension of anhydrous

MgSO₄(12.421 g, 103.19 mmol) in 40 mL of CH₂Cl₂was added 6-

methyl-2-pyridinecarboxaldehyde (2.500 g, 20.64 mmol) followed by

the slow addition of 2-(aminomethyl)pyridine (2.232 g, 20.64 mmol).

The suspension stirred at 23 °C for 3 h. The reaction mixture was

filtered and washed with CH₂Cl₂. The solvent was removed under

1

vacuum to yield a pale yellow liquid (4.25 g, 97%). H NMR (C₆D₆,

400 MHz): δ 2.39 (s, CH₃, 3 H), 4.92 (s, CH₂, 2 H), 6.60 (d, py^Me

C⁵H, 1 H, J = 7.6 Hz), 6.61 (t, py-C⁵H, 1 H, J = 5.3 Hz), 7.04 (t, py^Me

-

C⁴H, 1 H, J = 7.6 Hz), 7.08 (td, py-C⁴H, 1 H, J = 7.5, 1.6 Hz), 7.20 (d,

py-C³H, 1 H, J = 7.8 Hz), 8.02 (d, py^Me-C³H, 1 H, J = 7.8 Hz), 8.48

(d, py-C⁶H, 1 H, J = 4.8 Hz), 8.61 (s, im-CH, 1 H). ¹³C{¹H} NMR

(C₆D₆, 100 MHz): δ 24.62 (CH₃), 67.24 (CH₂), 118.48 (py^im-C³H),

122.16 (py-C³H), 122.53 (py^im-C⁵H), 124.39 (py-C⁵H), 136.36 (py-

C⁴H), 137.70 (py^im-C⁴H), 149.95 (py-C⁶H), 154.25 (py^im-C⁶), 158.54

(py^im-C²), 160.22 (py-C²), 164.97 (im-CH).

2. ^oMe₂smifH. To a suspension of anhydrous MgSO₄(5.036 g,

41.82 mmol) in 16 mL of CH₂Cl₂was added 6-methyl-2-

pyridinecarboxaldehyde (1.014 g, 8.37 mmol) followed by the slow

addition of 6-methyl-2-pyridylmethylamine (1.022 g, 8.37 mmol). The

yellow suspension stirred at 23 °C for 3 h. The reaction mixture was

filtered and washed with CH₂Cl₂. The solvent was removed under

1

vacuum to yield a pale yellow solid (1.73 g, 92%). H NMR (C₆D₆,

500 MHz): δ 2.38 (s, py-CH₃, 3 H), 2.39 (s, py^im-CH₃, 3 H), 4.93 (s,

CH₂, 2 H), 6.61 (d, py-C³H, py-C⁴H, 2 H, J = 6.5 Hz), 7.06 (t, py^im-

C⁴H, 1 H, J = 7.5 Hz), 7.10 (d, py-C⁵H, py^im-C⁵H, 2 H, J = 7 Hz),

8.02 (d, py^im-C³H, 1 H, J = 8 Hz), 8.62 (s, CH, 1 H). ¹³C{¹H} NMR

(C₆D₆, 125 MHz): δ 24.62 (py-CH₃), 24.84 (py^im-CH₃), 67.45 (CH₂),

118.47 (py-C³H), 119.56 (py^im-C³H), 121.60 (py-C⁵H), 124.36 (py^im-

C⁵H), 136.70 (py-C⁴H), 136.78 (py^im-C⁴H), 155.38 (py^im-C²), 158.46

(py-C⁶), 158.50 (py^im-C⁶), 159.46 (py-C²), 164.79 (im-CH).

CONCLUSIONS

■

In contrast to typical cationic coordination compounds, neutral,

hydrocarbon soluble, “Werner complexes” [(smif)₂M]ⁿ(n = 0,

1-M, M = V, Cr, Mn, Fe, Co, Ni, Zn, Ru; n = 1, 1⁺-M, M = Cr,

Mn, Co, Rh, Ir; smif =1,3-di-(2-pyridyl)-2-azaallyl) and the

related azaallyl species (^oMesmif)₂Fe (2-Fe) and (^oMe₂smif)₂Fe

(3-Fe) have been prepared. While similar in field strength to

terpy and other tridentate N-donors, the smif complexes are

distinguished by extremely intense intraligand aborptions in

blue and red regions of UV−vis spectrum. In one instance, the

smif ligand has been shown to be redox “noninnocent”, and the

singlet diradical character of the ligand HOMO suggests

potential reactivity at the ligand CNC-backbone.^48,49,57

3. Li(smif). To a solution of lithium bis(trimethylsilyl)amide (1.273 g,

7.60 mmol) in 50 mL THF was slowly added a solution of smifH

(1.500 g, 7.60 mmol) in 50 mL THF at −78 °C under argon. The

solution immediately turned magenta and was stirred at −78 °C for

2 h. After the mixture was stirred at 23 °C for 2 h, the volatiles were

removed in vacuo. The solid was triturated with Et₂O and filtered.

1

Li(smif) was isolated as a metallic gold solid (1.389 g, 90%). H NMR

(C₆D₆, 400 MHz): δ 5.98 (t, py-C⁵H, 1 H, J = 8 Hz), 6.50 (d, py-C³H,

1 H, J = 8 Hz), 6.84 (t, py-C⁴H, 1 H, J = 8 Hz), 7.16 (s, CH, 1 H),

7.66 (d, py-C⁶H, 1 H, J = 4 Hz). ¹³C{¹H} NMR (C₆D₆, 100 MHz): δ

113.20 (CH), 117.95 (py-C³H), 118.65 (py-C⁵H), 136.18 (py-C⁴H),

148.90 (py-C⁶H), 159.44 (py-C²).

EXPERIMENTAL SECTION

■

General Considerations. All manipulations were performed

using either glovebox or high vacuum line techniques. All glassware

was oven-dried. THF and ether were distilled under nitrogen from

purple sodium benzophenone ketyl and vacuum transferred from the

same prior to use. Hydrocarbon solvents were treated in the same

manner with the addition of 1−2 mL/L tetraglyme. Benzene-d₆and

4. Na(smif). To a solution of sodium bis(trimethylsilyl)amide

(1.395 g, 7.60 mmol) in 50 mL THF was slowly added a solution of

smifH (1.500 g, 7.60 mmol) in 50 mL THF at −78 °C under argon.

The solution immediately turned magenta and was stirred at −78 °C

12431

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

for 2 h. After it was stirred at 23 °C for 2 h, the volatiles were removed

in vacuo. The solid was triturated with Et₂O (3 × 15 mL) prior to

filtering. Na(smif) was isolated as a metallic gold solid (1.602 g, 96%).

¹H NMR (C₆D₆, 400 MHz): δ 6.19 (t, py-C⁵H, 1 H, J = 5.6 Hz), 6.55

(d, py-C³H, 1 H, J = 8 Hz), 6.97 (t, py-C⁴H, 1 H, J = 7.2 Hz), 7.04 (s,

CH, 1 H), 7.72 (d, py-C⁶H, 1 H, J = 4 Hz). ¹³C{¹H} NMR (C₆D₆,

100 MHz): δ 112.19 (CH), 115.70 (py-C³H), 119.05 (py-C⁵H),

135.62 (py-C⁴H), 149.81 (py-C⁶H), 160.23 (py-C²).

5. (smif)₂V (1-V). To a 50 mL 3-neck flask charged with lithium

bis(trimethylsilyl)amide (0.170 g, 1.02 mmol) and 0.95% Na/Hg

(1.288 g, 0.53 mmol) was vacuum transferred 10 mL THF at −78 °C.

A solution of smifH (0.200 g, 1.01 mmol) in THF (8 mL) was slowly

added to the flask via a dropping funnel under argon. The solution

immediately turned magenta and stirred at −78 °C for 3 h prior to the

addition of VCl₃(THF)₃(0.189 g, 0.51 mmol). The reaction mixture,

which turned cherry red after slowly warming to 23 °C and stirring for

12 h, was degassed and filtered. The volatiles were removed in vacuo,

and the microcrystalline solid was triturated and filtered in Et₂O to

yield 0.185 g of 1-V (81%). EA attempts failed for this extremely air

sensitive material. μ_eff(SQUID, 300 K) = 3.76 μ_B.

6. (smif)₂Cr (1-Cr). To a solution of Cr{N(SiMe₃)₂}₂(THF)₂

(0.425 g, 0.82 mmol) in 8 mL of Et₂O was slowly added a solution

of smifH (0.325 g, 1.65 mmol) in 10 mL Et₂O at 23 °C. The solution

immediately became dark emerald green. The reaction was degassed,

warmed to 23 °C, and stirred for 12 h while dark green crystals

precipitated from solution. The reaction was concentrated, and the

green suspension was filtered to yield 0.288 g of crystalline 1-Cr

(79%). ¹H NMR (C₆D₆, 400 MHz): δ −103.60 (ν_1/2≈ 1700 Hz, py-

CH, 1 H), −22.67 (ν_1/2≈ 1900 Hz, py-CH, 1 H), −19.83 (ν_1/2≈ 200

Hz, py-CH, 1 H), 19.35 (ν _1/2≈ 130 Hz, CH, 1 H), 22.08 (ν _1/2≈ 100

Hz, py-CH, 1 H). Anal. Calcd. H₂₀C₂₄N₆Cr: C, 64.86; H, 4.54; N,

18.91. Found: C, 63.81; H, 4.21; N, 17.43 (extreme air sensitivity

hampered EA). μ_eff(SQUID, 300 K) = 2.67 μ_B.

smifH (0.250 g, 1.27 mmol) in Et₂O (15 mL) at 23 °C. The solution

immediately changed from pale green to deep forest green. The

reaction was degassed and warmed to 23 °C. Black-metallic purple

crystals began to precipitate from solution after stirring for 30 min.

The reaction mixture was stirred for an addition 9.5 h. The volatiles

were removed, and the solid was triturated and filtered in Et₂O to yield

black-metallic purple crystals of 1-Fe (0.229 g, 80%). b. A solution of

smifH (5.00 g, 25.35 mmol) in 100 mL of THF was added dropwise to

a solution of lithium bis(trimethylsilyl)amide (4.242 g, 25.35 mmol) in

50 mL of THF at −78 °C under argon. The solution turned magenta

and was stirred at −78 °C for 3 h prior to the addition of

FeBr₂(THF)₂(4.561 g, 12.67 mmol). After stirring at 23 °C for 16 h, a

purple crystalline solid precipitated from the forest green solution. The

volatiles were removed in vacuo, and the residue was dissolved in

toluene and filtered. Toluene was removed, and the solid was

triturated with Et₂O and filtered to yield black-metallic purple crystals

1

of 1-Fe (2.980 g, 52%). H NMR (C₆D₆, 400 MHz): δ 5.73 (t, py-

C⁵H, 1 H, J = 5.9 Hz), 6.11 (d, py-C³H, 1 H, J = 7.9 Hz), 6.38 (t, py-

C⁴H, 1 H, J = 7.8 Hz), 7.59 (s, CH, 1 H), 7.66 (d, py-C⁶H, 1 H, J =

5.2 Hz). ¹³C{¹H} NMR (C₆D₆, 100 MHz): δ 112.19 (CH), 115.64

(py-C³H), 118.34 (py-C⁵H), 134.68 (py-C⁴H), 151.81 (py-C⁶H),

165.65 (py-C²). Anal. Calcd. H₂₀C₂₄N₆Fe: C, 64.30; H, 4.50; N, 18.75.

Found: C, 63.76; H, 4.64; N, 17.69.

11. (^oMesmif)₂Fe (2-Fe). To a solution of Fe{N(SiMe₃)₂}₂(THF)

(0.500 g, 1.11 mmol) in 15 mL of Et₂O was slowly added a solution of

^oMesmifH (0.471 g, 2.22 mmol) in Et₂O (10 mL) at 23 °C. The

solution immediately changed from pale green to a brilliant blue. The

reaction was degassed and warmed to 23 °C. Purple crystals began to

precipitate from the deep blue solution while stirring for 20 h. The

volatiles were removed, and the solid was triturated and filtered in

Et₂O to yield purple crystals of 2-Fe (0.311 g, 59%). ¹H NMR (C₆D₆,

400 MHz): δ 2.04 (s, py^Me-CH₃, 3 H), 6.44 (t, py^Me-C⁴H, 1 H, J = 6.8

Hz), 6.51 (t, py-C⁵H, 1 H, J = 6.8 Hz), 6.83 (d, py^Me-C⁵H, 1 H, J = 6.8

Hz), 6.97 (br s, py^Me-C³H, py-C⁴H, 2 H), 7.59 (d, py-C⁶H, 1 H, J = 6

Hz), 11.43 (ν_1/2≈ 29 Hz, CH, 1 H), 12.04 (ν_1/2≈ 46 Hz, CH, 1 H),

13.31 (ν_1/2≈ 52 Hz, py-C³H, 1 H). ¹³C{¹H} NMR (C₆D₆, 100

MHz): δ 25.93 (py^Me-CH₃), 100.75 (py^Me-C³H), 103.26 (py^Me-C⁵H),

107.54 (py-C³H), 112.22 (py-C⁵H), 114.11 (py^Me-C⁴H), 119.50

(CH), 123.66 (py-C⁴H), 132.93 (CH), 136.73 (py-C⁶H), 149.10

(py^Me-C⁶H), 160.93 (py^Me-C²), 170.83 (py-C²). Anal. Calcd

H₂₄C₂₆N₆Fe: C, 65.56; H, 5.08; N, 17.64. Found: C, 65.58; H, 5.25;

N, 17.17. μ_eff(SQUID, 5 K) = 0.5 μ_Bto μ_eff(SQUID, 300 K) = 1.22 μ_B.

12. (^oMe₂smif)₂Fe (3-Fe). To a solution of Fe{N(SiMe₃)₂}₂(THF)

(0.747 g, 1.66 mmol) in 12 mL Et₂O was slowly added a solution of

^oMe₂smifH (0.750 g, 3.33 mmol) in Et₂O (10 mL) at 23 °C. The

solution immediately changed from pale green to deep forest green.

The reaction was degassed and warmed to 23 °C. Gold-bronze crystals

began to precipitate from the deep cobalt blue solution after stirring

for 30 min. The reaction mixture was stirred for an addition 15.5 h.

The volatiles were removed, and the solid was triturated and filtered in

7. [(smif)₂Cr](OTf) (1⁺-Cr). To a 25 mL round-bottom flask charged

with (smif)₂Cr (0.300 g, 0.67 mmol) and AgOTf (0.173 g, 0.67 mmol)

was vacuum transferred 8 mL Et₂O at −78 °C, and the reaction

mixture became green within 5 min. The flask warmed slowly to 23 °C

and was stirred for 2 d while a dark green solid precipitated from the

pale blue solution. The volatiles were removed in vacuo.

Recrystallization of the dark green solid from THF at 80 °C under

a blanket of argon for 16 h led to the formation of metallic red crystals

1

of 1⁺-Cr (0.309 g, 75%). H NMR (C₆D₆, 400 MHz): δ −12.19 (ν_1/2

≈ 600 Hz, py-CH, 1 H), −3.95 (ν_1/2≈ 600 Hz, py-CH, 1 H). μ_eff

(Gouy balance, 295K) = 3.6 μ_B; μ_eff(SQUID, 300 K) = 3.76 μ_B.

8. (smif)₂Mn (1-Mn). To a solution of lithium bis(trimethylsilyl)-

amide (0.425 g, 2.54 mmol) in 15 mL THF at −78 °C was added

dropwise a solution of smifH (0.500 g, 2.53 mmol) in 10 mL of THF

under argon. The solution immediately turned magenta and stirred at

−78 °C for 2 h prior to the addition of MnCl₂(0.160 g, 1.27 mmol).

The reaction mixture became deep purple after stirring at 23 °C for 36

h. The volatiles were removed in vacuo, and the solid was dissolved

and filtered in toluene. Toluene was removed, and the solid was

triturated and filtered in Et₂O to isolate metallic gold crystals of 1-Mn

1

Et₂O to yield gold-bronze crystals of 3-Fe (0.712 g, 85%). H NMR

(C₆D₆, 400 MHz): δ −9.64 (υ_1/2≈ 110 Hz, CH, 1 H), 7.44 (ν_1/2

≈

17 Hz, CH₃, 3 H), 36.73 (ν_1/2≈ 20 Hz, py-CH, 1 H), 52.87 (ν_1/2

≈

1

(0.410 g, 72%). H NMR (C₆D₆, 400 MHz): δ −13.52 (ν_1/2≈ 1200

15 Hz, py-CH, 1 H), 167.44 (ν _1/2≈ 53 Hz, py-CH, 1 H). Anal. Calcd

H₂₈C₂₈N₆Fe: C, 66.67; H, 5.60; N, 16.66. Found: C, 66.54; H, 5.47; N,

16.19. μ_eff(SQUID, 300 K) = 5.47 μ_B.

Hz, py-CH, 1 H), 48.08 (ν_1/2≈ 4100 Hz, py-CH, 1 H). Anal. Calcd.

H₂₀C₂₄N₆Mn: C, 64.43; H, 4.51; N, 18.78. Found: C, 64.21; H, 4.40;

N, 18.52. μ_eff(SQUID, 300K) = 5.73 μ_B.

13. (smif)₂Co (1-Co). To a solution of lithium bis(trimethylsilyl)-

amide (0.425 g, 2.54 mmol) in 15 mL THF at −78 °C was added

dropwise a solution of smifH (0.500 g, 2.53 mmol) in 10 mL THF

under argon. The reaction solution immediately turned magenta and

was stirred at −78 °C for an additional 2 h prior to the addition of

CoCl₂(0.165 g, 1.27 mmol). After stirring at 23 °C for 36 h, the

solution had darkened to a deep purple-magenta. The volatiles were

removed, and the residue was dissolved and filtered in toluene.

Toluene was removed in vacuo, and the solid was triturated with Et₂O

9. [(smif)₂Mn](OTf) (1⁺-Mn). To a 100 mL round-bottom flask

charged with (smif)₂Mn (1-Mn, 0.700 g, 1.56 mmol) and AgOTf

(0.402 g, 1.56 mmol) was vacuum transferred 50 mL THF at −78 °C.

The dark magenta-purple solution slowly warmed to 23 °C and

darkened to a deeper purple. After stirring at 23 °C for 1.5 d, the

volatiles were removed in vacuo resulting in a red-bronze metallic solid

which was filtered in toluene and THF. Filtrates were concentrated,

cooled to 23 °C, and filtered to yield metallic red-bronze microcrystals

of 1⁺-Mn (0.728 g, 78%). Anal. Calcd H₂₀C₂₅N₆O₃F₃SMn: C, 50.34;

H, 3.38; N, 16.66; S, 5.38. Found: C, 50.18; H, 5.50; N, 12.75; S, 5.56.

μ_eff(Gouy balance, 295K) = 4.45 μ_B; μ_eff(SQUID, 300 K) = 5.43 μ_B.

10. (smif)₂Fe (1-Fe). a. To a solution of Fe{N(SiMe₃)₂}₂(THF)

(0.284 g, 0.63 mmol) in 15 mL Et₂O was slowly added a solution of

1

and filtered to yield metallic gold crystals of 1-Co (0.501 g, 87%). H

NMR (C₆D₆, 400 MHz): δ 10.06 (ν_1/2≈ 50 Hz, CH, 1 H), 37.63

(ν_1/2≈ 70 Hz, py-CH, 1 H), 39.90 (ν _1/2≈ 80 Hz, py-CH, 1 H), 85.19

(ν_1/2≈ 140 Hz, py-CH, 1 H), 108.94 (ν _1/2≈ 480 Hz, py-CH, 1 H).

Anal. Calcd (for (smif)₂Co·(C₇H₈)_0.5) H₂₄C_27.5N₆Co: C, 66.40; H,

12432

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

4.86; N, 16.89. Found: C, 65.92, 64.99; H, 4.68, 4.47; N, 17.23, 16.92.

μ_eff(SQUID, 10 K) = 1.75 μ_Band μ_eff(SQUID, 300 K) = 3.19 μ_B.

14. [(smif)₂Co](OTf) (1⁺-Co). To a 10 mL round-bottom flask

charged with 0.200 g (0.44 mmol) (smif)₂Co and 0.114 g (0.44 mmol)

AgOTf was vacuum transferred 8 mL THF at −78 °C. The reaction

mixture changed from deep purple to cobalt blue within 5 min and

slowly warmed to 23 °C. After it was stirred at 23 °C for 12 h, a

magenta solid precipitated from solution. The volatiles were removed

in vacuo. Recrystallization of the magenta solid in THF at 80 °C under

a blanket of argon for 16 h led to the formation of metallic red crystals

of 1⁺-Co (0.215 g, 81%). ¹H NMR (THF-d₈, 400 MHz): δ 6.55 (t, py-

C⁵H, 1 H, J = 6.4 Hz), 6.85 (d, py-C³H, 1 H, J = 8 Hz), 7.24 (s, CH, 1

H), 7.28 (t, py-C⁴H, 1 H, J = 7.2 Hz), 7.59 (d, py-C⁶H, 1 H, J = 6.0

Hz). ¹³C{¹H} NMR (THF-d₈, 100 MHz): δ 117.47 (CH), 118.23 (py-

C³H), 119.54 (py-C⁵H), 120.36 (py-C⁴H), 139.24 (py-C⁶H), 148.56

(py-C²).

15. (smif)₂Ni (1-Ni). A solution of lithium bis(trimethylsilyl)amide

(0.425 g, 2.54 mmol) in 15 mL THF under argon at −78 °C was

slowly treated with a solution of smifH (0.500 g, 2.53 mmol) in THF

(10 mL) . The solution instantly turned magenta and was stirred at

−78 °C for 2 h prior to the addition of NiCl₂(dme) (0.278 g, 1.27

mmol). After stirring at 23 °C for 36 h, the volatiles were removed in

vacuo from the magenta reaction mixture. The solid was dissolved and

filtered in toluene. Toluene was removed, and the solid was triturated

and filtered in Et₂O to yield metallic gold crystals of (smif)₂Ni (0.385 g,

67%). ¹H NMR (C₆D₆, 400 MHz): δ 9.40 (ν_1/2≈ 170 Hz, CH, 1 H),

51.75 (ν_1/2≈ 400 Hz, py-CH, 1 H), 57.01 (ν _1/2≈ 470 Hz, py-CH,

1 H), 140.85 (ν_1/2≈ 3300 Hz, py-CH, 1 H), 248.32 (ν _1/2≈ 6200 Hz,

py-CH, 1 H). Anal. Calcd. (for (smif)₂Ni·(C₆H₆)_0.5) H₂₃C₂₇N₆Ni: C,

66.15; H, 4.73; N, 17.14. Found: C, 65.52; H, 4.61; N, 17.13. μ_eff

(SQUID, 300 K) = 2.81 μ_B.

1 H, J = 7.7 Hz), 7.60 (d, py-C⁶H, 1 H, J = 6.0 Hz). ¹³C{¹H} NMR

(THF-d₈, 125 MHz): δ 116.08 (CH), 138.08 (py-C³H), 140.52 (py-

C⁵H), 149.47 (py-C⁴H), 163.36 (py-C⁶H), 169.14 (py-C²). Anal.

Calcd. H₄₀C₄₈N₆BRh: C, 63.78; H, 4.46; N, 9.30. Found: C, 62.22,

65.03; H, 4.84, 4.82; N, 8.02, 8.14.

EPR Spectroscopy. Solution and frozen glass EPR spectra were

recorded on a JEOL continuous wave spectrometer, JES-FA200

equipped with an X-band Gunn oscillator bridge, a cylindrical mode

cavity, and a helium cryostat. For all samples, a modulation frequency

of 100 kHz and a time constant of 0.1 s were employed. Frequencies

were close to 9.0 GHz and all spectra were obtained on freshly

prepared solutions (1−10 mM in toluene) in quartz tubes with J.

Young valves and were checked carefully for reproducibility.

Background spectra were obtained on clean solvents at the same

measurement conditions. Spectral simulations were performed using

the programs W95EPR by Prof. Dr. Frank Neese⁹²and ESRSIM by

Prof. Dr. Høgni Weihe, University of Copenhagen, Denmark. The

fittings were performed by the “chi by eye” approach. Collinear g and

A tensors were used, and deviations from isotropic parameters in

spectra of frozen glasses were only used when clearly justified.

Magnetic Susceptibility Measurements. Magnetic susceptibil-

ity measurements of crystalline powdered samples (10−30 mg) were

performed on a Quantum Design MPMS-5 SQUID magnetometer at

10 kOe (1 T) between 5 and 300 K for all samples. All sample

preparations and manipulations were performed under an inert

atmosphere because of the air sensitivity of the samples. The samples

were measured in gelatin capsules, and the diamagnetic contribution

from the sample container was subtracted from the experimental data.

Pascal’s constants⁶⁷were used to subtract diamagnetic contributions,

yielding paramagnetic susceptibilities. The program julX written by E.

Bill was used for (elements of) the simulation and analysis of magnetic

susceptibiltiy data.⁹³

16. (smif)₂Ru (1-Ru). To a small bomb reactor charged with

Na(smif) (0.400 g, 1.82 mmol) and (COD)RuCl₂(0.256 g, 0.946

mmol) was vacuum transferred 15 mL THF at −78 °C. After warming

to 23 °C, the bomb was heated in a 60 °C oil bath for 2 d as the

magenta solution became dark green with dark purple solids. The

reaction mixture was filtered cold in THF, and all volatiles were

removed in vacuo. The resulting dark purple, metallic solid was

Mossbauer Spectroscopy. ⁵⁷Fe Mossbauer spectra were recorded

̈

on a WissEl Mossbauer spectrometer (MRG-500) at 77 K in constant

̈

acceleration mode. ⁵⁷Co/Rh was used as the radiation source.

WinNormos for Igor Pro software has been used for the quantitative

evaluation of the spectral parameters (least-squares fitting to

Lorentzian peaks). The minimum experimental line widths were

0.20 mms⁻¹. The temperature of the samples was controlled by an

1

washed with pentane, and 0.276 g (smif)₂Ru were isolated (61%). H

NMR (C₆D₆, 400 MHz): δ 5.64 (t, py-C⁵H, 1 H, J = 6.4 Hz), 6.07 (d,

py-C³H, 1 H, J = 8.3 Hz), 6.81 (s, CH, 1 H), 6.31 (t, py-C⁴H, 1 H, J =

7.6 Hz), 7.80 (d, py-C⁶H, 1 H, J = 5.1 Hz). ¹³C{¹H} NMR (C₆D₆, 125

MHz): δ 113.66 (CH), 113.73 (py-C³H), 115.03 (py-C⁵H), 134.77

(py-C⁴H), 151.24 (py-C ⁶H), 167.54 (py-C ²). Anal. Calcd.

H₂₄C₂₀N₆Ru: C, 58.41; H, 4.08; N, 17.03. Found: C, 58.48; H, 4.22;

N, 13.28.

̈

MBBC-HE0106 MOSSBAUER He/N₂cryostat within an accuracy of

±0.3 K. Isomer shifts were determined relative to α-iron at 298 K.

XAS Spectroscopy. XAS data were measured at the Stanford

Synchrotron Radiation Lightsource using focused beamline 9−3,

under ring conditions of 3 GeV and 60−100 mA. A Si(220) double-

crystal monochromator was used for energy selection and a Rh-coated

mirror (set to an energy cutoff of 9 keV) was utilized in combination

with 30% detuning for rejection of higher harmonics. All samples were

prepared as dilutions in BN and measured as transmission spectra.

Sample were maintained at 10 K using an Oxford continuous flow. To

check for reproducibility, 2−3 scans were measured for all samples.

The energy was calibrated from Cr and Co foil spectra, with the first

inflection set to 5989.0 and 7709.5 eV, respectively. A step size of

0.11 eV was used over the edge region. Data were averaged, and a smooth

background was removed from all spectra by fitting a polynomial to the

pre-edge region and subtracting this polynomial from the entire spectrum.

Normalization of the data was accomplished by fitting a flattened

polynomial or straight line to the postedge region and normalizing the

edge jump to 1.0.

17. [(smif)₂Rh][OTf] (1⁺-Rh). To a small bomb reactor charged with

Na(smif) (0.072 g, 0.328 mmol), AgOTf (0.042 g, 0.163 mmol) and

Rh₂(TFA)₄(0.054 g, 0.082 mmol) was vacuum transferred 5 mL

toluene at −78 °C. After it was warmed to 23 °C, the solution turned

from magenta to purple and was placed in a 100 °C oil bath for 1 d.

The bright blue reaction mixture was filtered and washed with toluene.

All volatiles were removed in vacuo leaving a bright red metallic solid

1

1⁺-Rh (0.028 g, 53%). H NMR (THF-d₈, 400 MHz): δ 6.51 (t, py-

C⁵H, 1 H, J = 6.4 Hz), 6.87 (d, py-C³H, 1 H, J = 8.1 Hz), 6.85 (s, CH,

1 H), 7.29 (t, py-C⁴H, 1 H, J = 7.5 Hz), 7.77 (d, py-C⁶H, 1 H, J = 5.6

Hz). ¹³C{¹H} NMR (THF-d₈, 125 MHz): δ 114.72 (CH), 118.20 (py-

C³H), 138.70 (py-C⁵H), 148.20 (py-C⁴H), 148.29 (py-C⁶H), 165.22

(py-C²). Anal. Calcd. H₂₀C₂₅N₆O₃F₃SRh: C, 46.60; H, 3.13; N, 13.04.

Found: C, 44.39, 44.89; H, 4.95, 3.66; N, 8.87, 9.28.

Computational Methods. B3LYP⁹⁴⁻⁹⁸geometry optimization

utilized the Gaussian03 suite of programs; the 6-31G(d) basis set was

employed. Tests with the larger 6-311+G(d) basis set did not reveal

significant differences in the optimized geometries. No symmetry

constraints were employed in geometry optimization. Where

applicable, geometry optimizations were started from both a pseudo-

D_2dstructure (akin to crystal structure of (smif)₂Fe (1-Fe)) and/or

a highly Jahn−Teller distorted starting geometry (e.g., (smif)₂Co

(1-Co)). Calculation of the energy Hessian was performed to confirm

species as minima on their respective potential energy surfaces at this

level of theory. All plausible spin multiplicities were investigated for

the different M(smif)₂complexes. Modeling of open-shell species with

18. [(smif)₂Ir][BPh₄] (1⁺-Ir). To a small bomb reactor charged with

Na(smif) (0.195 g, 0.889 mmol), NaBPh₄(0.152 g, 0.444 mmol) and

IrCl₃(THT)₃(0.250 g, 0.444 mmol) was vacuum transferred 5 mL

THF at −78 °C. Upon warming to 23 °C, the magenta solution

quickly turned navy blue. The bomb was placed in a 70 °C oil bath for

2 d after which the solution was turquoise. The reaction mixture was

filtered and washed in THF. All volatiles were removed in vacuo

leaving dark purple metallic solid [(smif)₂Ir][BPh₄] (0.200 g, 50%).

¹H NMR (THF-d₈, 400 MHz): δ 6.36 (t, py-C⁵H, 1 H, J = 6.7 Hz),

6.53 (d, py-C³H, 1 H, J = 8.3 Hz), 6.30 (s, CH, 1 H), 7.03 (t, py-C⁴H,

12433

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

density functional theory employed unrestricted Kohn−Sham

methods.

of argon and slowing cooling to room temperature. A total of 28 729

reflections were collected with 6,783 determined to be symmetry

independent (R_int= 0.0622), and 4,835 were greater than 2σ(I). A

semiempirical absorption correction from equivalents was applied,

Electrochemistry. All electrochemical experiments were done in a

glovebox. Solutions of ∼1 mM of the desired complex were prepared

in THF containing 0.1 M TBAP. The electrochemical experiments

were performed using a platinum electrode as the working electrode, a

silver wire as a pseudoreference electrode⁹⁹and a platinum foil as the

counter electrode. A BAS-27 W potentiostat was used to perform the

experiments, and data were digitally recorded using WinDaq Serial

Acquisition software (DATAQ Instruments).

Single Crystal X-ray Diffraction Studies. Upon isolation, the

crystals, except those of 1-V, were covered in polyisobutenes and

placed under a 173 K N₂stream on the goniometer head of a Siemens

P4 SMART CCD area detector (graphite-monochromated MoK_α

radiation, λ = 0.71073 Å). The structures were solved by direct

methods (SHELXS). All non-hydrogen atoms were refined anisotropi-

cally unless stated, and hydrogen atoms were treated as idealized

contributions (Riding model).

and the refinement utilized w⁻¹= σ²(F_o) + (0.0846p)²+ 0.0000p,

2

where p = ((F_o+ 2F_c²)/3).

8. (smif)₂Ni (1-Ni). A metallic gold plate (0.60 × 0.20 × 0.03 mm)

was obtained from the slow evaporation of benzene at 23 °C. A total of

25 347 reflections were collected with 5,631 determined to be

symmetry independent (R_int= 0.0809), and 3,712 were greater than

2σ(I). A semiempirical absorption correction from equivalents was

applied, and the refinement utilized w⁻¹= σ²(F_o) + (0.0331p)²+ 0.8315p,

2

where p = ((F_o+ 2F_c²)/3).

ASSOCIATED CONTENT

* Supporting Information

■

S

CIF files for 1-M (M = V, Cr, Mn) and 1⁺-Cr (those for 1-M

(M = Fe, Co, Ni) and 1⁺-Co can be found in the Supporting

Information of ref 48) and additional spectroscopic details and

experimental considerations, including JulX fits of all SQUID

data. This material is available free of charge via the Internet at

http://pubs.acs.org.

1. (smif)₂V (1-V). A metallic gold needle (0.30 × 0.02 × 0.01 mm)

was obtained from toluene at 23 °C. It was covered with polyiso-

butenes, placed in a goniometer at MacCHESS station A1, cooled to

100 K, and subjected to the beamline (λ = 0.97890 Å, Si monochro-

mator). A 360° sweep of data was collected in 2° ψ-scans. A total of

4,195 reflections were collected with 4,195 determined to be symmetry

independent (R_int= 0.0000), and 3,918 were greater than 2σ(I). Data

reduction was conducted using HKL200 software,¹⁰⁰and the structure

was solved by direct methods and refined (F²) using full matrix least-

squares techniques and SHELXTL¹⁰¹software. A semiempirical ab-

sorption correction from equivalents was applied, and the refinement

AUTHOR INFORMATION

Corresponding Author

*Fax: 607 255 4173. E-mail: ptw2@cornell.edu.

■

Present Address

utilized w⁻¹= σ²(F_o) + (0.1134p)²+ 1.3469p, where p = ((F_o

+

2

^⊥Max-Planck Institut fur Bioanorganische Chemie, Stiftstr. 34-

̈

2F_c²)/3).

36, D-45470 Mulheim an der Ruhr, Germany.

̈

2. (smif)₂Cr (1-Cr). A dark green plate (0.40 × 0.15 × 0.02 mm)

was obtained from the slow evaporation of toluene at 23 °C. A total of

36,536 reflections were collected with 7,312 determined to be

symmetry independent (R_int= 0.0760), and 5,019 were greater than

2σ(I). A semiempirical absorption correction from equivalents was

ACKNOWLEDGMENTS

■

P.T.W. thanks the NSF (CHE-0718030) and Cornell

University, and TRC the DOE (DE-FG02-03ER15387) for

financial support. S.D. thanks Cornell University for funding,

and we thank Prof. Frank Neese for helpful discussions and for

aid in fitting EPR spectra, and Dr. Kyle Lancaster for calculating

K-edge spectra. We also thank Valerie A. Williams and Dr.

Carsten Milsmann for aid in fitting of SQUID data, and Dr.

Marat Khusniyarov for additional EPR spectra. Portions of

this research were carried out at the Stanford Synchrotron

Radiation Lightsource, a Directorate of SLAC National

Accelerator Laboratory and an Office of Science User Facility

operated for the U.S. Department of Energy Office of Science

by Stanford University. The SSRL Structural Molecular Biology

Program is supported by the DOE Office of Biological and

Environmental Research, and by the National Institutes of

Health, National Center for Research Resources, Biomedical

Technology Program (P41RR001209).

applied, and the refinement utilized w⁻¹= σ²(F_o) + (0.0489p)²+

2

0.0000p, where p = ((F_o+ 2F_c²)/3).

3. [(smif)₂Cr]OTf (1⁺-Cr). A metallic red plate (0.30 × 0.10 × 0.03

mm) was obtained from a solution of tetrahydrofuran at −40 °C. A

total of 23,529 reflections were collected with 5,637 determined to be

symmetry independent (R_int= 0.0540), and 4,220 were greater than

2σ(I). A semiempirical absorption correction from equivalents was

applied, and the refinement utilized w⁻¹= σ²(F_o) + (0.0583p)²+

2

0.0687p, where p = ((F_o+ 2F_c²)/3).

4. (smif)₂Mn (1-Mn). A metallic gold plate (0.60 × 0.20 × 0.03

mm) was obtained from the slow evaporation of toluene at 23 °C. A

total of 30,750 reflections were collected with 6,863 determined to be

symmetry independent (R_int= 0.0622), and 4492 were greater than

2σ(I). A semiempirical absorption correction from equivalents was

applied, and the refinement utilized w⁻¹= σ²(F_o) + (0.0447p)²+

2

0.0000p, where p = ((F_o+ 2F_c²)/3).

5. (smif)₂Fe (1-Fe). A black-metallic purple block (0.45 × 0.30 ×

0.20 mm) was obtained from the slow evaporation of benzene at

23 °C. A total of 25,212 reflections were collected with 5007

determined to be symmetry independent (R_int= 0.0497), and 3994

were greater than 2σ(I). A semiempirical absorption correction from

REFERENCES

■

(1) Kauffman, G. B. Alfred WernerFounder of Coordination

Chemistry; Springer-Verlag: Berlin, 1966.

equivalents was applied, and the refinement utilized w⁻¹= σ²(F_o) +

2

(2) Spingler, B.; Scanavy-Grigorieff, M.; Werner, A.; Berke, H.;

Lippard, S. J. Inorg. Chem. 2001, 40, 1065−1066.

(3) Kaes, C.; Katz, A.; Hosseini, M. W. Chem. Rev. 2000, 100, 3553−

3590.

(4) Constable, E. C. Adv. Inorg. Chem. 1989, 34, 1−63.

(5) Constable, E. C. Adv. Inorg. Chem. 1986, 30, 69−121.

(6) Wadman, S. H.; van der Geer, E. P. L.; Havenith, R. W. A.;

Gebbink, R. J. M.; van Klink, G. P. M.; van Koten, G. J. Organomet.

Chem. 2008, 693, 3188−3190.

(7) Ward, M. D.; McCleverty, J. A.; Jeffery, J. C. Coord. Chem. Rev.

2001, 222, 251−272.

(8) Ziessel, R. Coord. Chem. Rev. 2001, 216, 195−223.

(9) Dobson, J. C.; Taube, H. Inorg. Chem. 1968, 7, 254−261.

(0.0422p)²+ 0.9777p, where p = ((F_o+ 2F_c²)/3).

2

6. (smif)₂Co (1-Co). A metallic gold thin plate (0.25 × 0.20 × 0.01

mm) was obtained after heating a solution of toluene at 80 °C for 8 h

in a sealed tube under a blanket of argon and slowing cooling to room

temperature. A total of 3,587 reflections were collected with 3,587

determined to be symmetry independent (R_int= 0.0000), and 3,406

were greater than 2σ(I). A semiempirical absorption correction from

equivalents was applied, and the refinement utilized w⁻¹= σ²(F_o²) +

(0.1045p)²+ 1.8144p, where p = ((F_o+ 2F_c²)/3).

2

7. [(smif)₂Co]OTf (1⁺-Co). A metallic red-orange rod (0.40 ×

0.15 × 0.10 mm) was obtained after heating a solution of

tetrahydrofuran at 80 °C for 16 h in a sealed tube under a blanket

12434

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

(10) Hughes, M. C.; Macero, D. J. Inorg. Chem. 1976, 15, 2040−

2044.

(45) Jacobsen, H.; Correa, A.; Poater, A.; Costabile, C.; Cavallo, L.

Coord. Chem. Rev. 2009, 253, 687−703.

(11) Hogg, R.; Wilkins, R. C. J. Chem. Soc. 1962, 341−350.

(12) Harris, C. M.; Patil, H. R. H.; Sinn, E. Inorg. Chem. 1969, 8,

101−104.

(46) de Fremont, P.; Marion, N.; Nolan, S. P. Coord. Chem. Rev.

2009, 253, 862−892.

(47) Schuster, O.; Yang, L. R.; Raubenheimer, H. G.; Albrecht, M.

Chem. Rev. 2009, 109, 3445−3478.

(13) Braterman, P. S.; Song, J. I.; Peacock, R. D. Inorg. Chem. 1992,

31, 555−559.

(48) Frazier, B. A.; Wolczanski, P. T.; Lobkovsky, E. B.; Cundari,

T. R. J. Am. Chem. Soc. 2009, 131, 3428−3429.

(14) (a) Kremer, S.; Henke, W.; Reinen, D. Inorg. Chem. 1982, 21,

3013−3022. (b) Schmidt, J. D.; Brey, W. S.; Stoufer, R. C. Inorg. Chem.

1967, 6, 268−271.

(49) Di(1,3-2-pyridyl)-2-azaallyl as Chelate for Metals. CCTEC

D4598−02. U.S. Patent Pending No. 12/704,170, Filed March 17,

2010.

(15) Henke, W.; Reinen, D. Z. Anorg. Allg. Chem. 1977, 436, 187−

200.

(16) Behrens, H.; Brandl, H. Z. Naturforsch., B: Chem. Sci. 1967, 22,

(50) Incarvito, C.; Lam, M.; Rhatigan, B.; Rheingold, A. L.; Qin,

C. J.; Gavrilova, A. L.; Bosnich, B. J. Chem. Soc., Dalton Trans. 2001,

3478−3488.

(51) Clark, D. L.; Sattelberger, A. P. Inorg. Synth. 1997, 31, 307−315.

(52) (a) Olmstead, M. M.; Power, P. P.; Shoner, S. C. Inorg. Chem.

1991, 30, 2547−2551. (b) Andersen, R. A.; Faegri, K.; Green, J. C.;

Haaland, A.; Lappert, M. F.; Leung, W. P.; Rypdal, K. Inorg. Chem.

1988, 27, 1782−1786.

(53) Ittel, S. D.; English, A. D.; Tolman, C. A.; Jesson, J. P. Inorg.

Chim. Acta 1979, 33, 101−106.

(54) Ward, L. G. L. Inorg. Synth. 1971, 13, 154−165.

(55) Kaesz, H. D.; Albers, M. O.; Singleton, E.; Yates, J. E.;

Mccormick, F. B. Inorg. Synth. 1989, 26, 249−258.

(56) Johnson, S. A.; Hunt, H. R.; Neumann, H. M. Inorg. Chem.

1963, 2, 960−962.

(57) Frazier, B. A., Ph.D. Thesis, Cornell University, 2010.

(58) (a) Liu, X.; Ellis, J. E.; Miller, T. D.; Ghalasi, P.; Miller, J. S.

Inorg. Synth. 2004, 34, 96−103. (b) Manzer, L. E. Inorg. Synth. 1982,

21, 135−140.

(59) John, K. D.; Salazar, K. V.; Scott, B. L.; Baker, T. R.;

Sattelberger, A. P. Organometallics 2001, 20, 296−304.

(60) Roth, J. P.; Yoder, J. C.; Won, T.-J.; Mayer, J. M. Science 2001,

294, 2524−2526.

1216.

(17) Behrens, H.; Brandl, H.; Lutz, K. Z. Naturforsch., B: Chem. Sci.

1967, 22, 99−100.

(18) (a) Trofimenko, S. J. Am. Chem. Soc. 1967, 89, 3170−3177.

(b) Jesson, J. P.; Trofimenko, S.; Eaton, D. R. J. Am. Chem. Soc. 1967,

89, 3148−3158. (c) Trofimenko, S. J. Am. Chem. Soc. 1967, 89, 6288−

6294.

(19) (a) Trofimenko, S. Acc. Chem. Res. 1971, 4, 17−22.

(b) Trofimenko, S. Chem. Rev. 1972, 72, 497−509.

(20) Churchill, M. R.; Gold, K.; Maw, C. E. Inorg. Chem. 1970, 9,

1597−1604.

(21) Glerup, J.; Goodson, P. A.; Hodgson, D. J.; Michelsen, K.;

Nielsen, K. M.; Weihe, H. Inorg. Chem. 1992, 31, 4611−4616.

(22) Nelson, S. M.; Rodgers, J. J. Chem. Soc. A 1968, 272−276.

(23) Butcher, R. J.; Addison, A. W. Inorg. Chim. Acta 1989, 158,

211−215.

(24) Davies, C. J.; Solan, G. A.; Fawcett, J. Polyhedron 2004, 23,

3105−3114.

(25) Malassa, A.; Gorls, H.; Buchholz, A.; Plass, W.; Westerhausen,

M. Z. Anorg. Chem. 2006, 632, 2355−2362.

(26) Velusamy, M.; Palaniandavar, M.; Thomas, K. R. J. Polyhedron

1998, 17, 2179−2186.

(61) (a) Poli, R.; Cacelli, I. Eur. J. Inorg. Chem. 2005, 12, 2324−2331.

(b) Petit, A.; Cacelli, I.; Poli, R. Chem.Eur. J. 2006, 12, 813−823.

(62) Zhang, G.; Musgrave, C. B. J. Phys. Chem. A 2007, 111, 1554−

1561.

(63) Hachmann, J.; Frazier, B. A.; Wolczanski, P. T.; Chan, G. K.-L.

Chem. Phys. Chem. 2011. DOI: 10.1002/cphc.201100286.

(27) Palaniandavar, M.; Butcher, R. J.; Addison, A. W. Inorg. Chem.

1996, 35, 467−471.

(28) Lions, F.; Martin, K. V. J. Am. Chem. Soc. 1957, 79, 2733−2738.

(29) Goodwin, H. A.; Lions, F. J. Am. Chem. Soc. 1959, 81, 6415−

6422.

(64) (a) Parish, R. V. NMR, NQR, EPR, and Mossbauer Spectroscopy

(30) Geldard, J. F.; Lions, F. J. Am. Chem. Soc. 1962, 84, 270−282.

(31) Figgins, P. E.; Busch, D. H. J. Am. Chem. Soc. 1960, 82, 820−

824.

̈

in Inorganic Chemistry; Ellis Horwood: West Sussex, England, 1990.

(b) Stevens, J. G. Hyperfine Interact. 1983, 148, 71−84.

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

Paramagnetic Molecules. In Current Methods in Inorganic Chemistry;

Elsevier: London, 2001.

(66) Figgis, B. N.; Hitchman, M. A. Ligand Field Theory and Its

Applications; Wiley-VCH: New York, 2000.

(67) Carlin, R. L. Magnetochemistry; Springer-Verlag: New York,

1986.

(32) Juhas

32, 882−883.

́

z, G.; Hayami, S.; Inoue, K.; Maeda, Y. Chem. Lett. 2003,

́

(33) Marcos, D.; Folgado, J.-V.; Beltran-Porter, D. Polyhedron 1990,

9, 2699−2704.

(34) Wocadlo, S.; Massa, W.; Folgado, J.-V. Inorg. Chim. Acta 1993,

207, 199−206.

́

(35) Marcos, D.; Martinez-Mane, R.; Folgado, J.-V.; Beltran-Porter,

̃

(68) (a) Mack, J.; Stillman, M. J.; Kobayashi, N. Coord. Chem. Rev.

2007, 251, 429−453. (b) Mack, J.; Stillman, M. J. Coord. Chem. Rev.

2001, 219−221, 993−1032.

́

A.; Beltran-Porter, D.; Fuertes, A. Inorg. Chim. Acta 1989, 159, 11−18.

(36) Kajiwara, T.; Sensui, R.; Noguchi, T.; Kamiyama, A.; Ito, T.

Inorg. Chim. Acta 2002, 337, 299−307.

(37) Nonoyama, M.; Yamasaki, K. Inorg. Chim. Acta 1971, 5, 124−

128.

(69) (a) Gudel, H. U.; Zilian, A. Coord. Chem. Rev. 1991, 111,

̈

33−38. (b) Colombo, M. G.; Brunold, T. C.; Riedener, T.; Gudel,

̈

(38) Rowland, J. M.; Olmstead, M. M.; Mascharak, P. K. Inorg. Chem.

H. U.; Fortsch, M.; Burgi, H.-B. Inorg. Chem. 1994, 33, 545−550.

̈

2002, 41, 2754−2760.

(70) Ichino, T.; Villano, S. M.; Gianola, A. J.; Goebbert, D. J.;

Velarde, L.; Sanov, A.; Blanksby, S. J.; Zhou, X.; Hrovat, D. A.; Borden,

W. T.; Lineberger, W. C. Angew. Chem., Int. Ed. 2009, 48, 8509−8511.

(71) (a) Evans, D. F. J. Chem. Soc. 1959, 2003−2005. (b) Schubert,

E. M. J. Chem. Educ. 1992, 69, 62.

(39) Hazra, S.; Naskar, S.; Mishra, D.; Gorelsky, S. I.; Figgie, H. M.;

Sheldrick, W. S.; Chattopadhyay, S. K. Dalton Trans. 2007, 4143−

4148.

(40) Hellyer, R. M.; Larsen, D. S.; Brooker, S. Eur. J. Inorg. Chem.

2009, 1162−1171.

(72) Ray, K.; Petrenko, T.; Wieghardt, K.; Neese, F. Dalton Trans.

(41) Kern, R. J. J. Inorg. Nucl. Chem. 1962, 24, 1105−1109.

(42) Bradley, D. C.; Hursthouse, M. B.; Newing, C. W.; Welch, A. J.

J. Chem. Soc., Chem. Commun. 1972, 567−568.

(43) Frazier, B. A.; Wolczanski, P. T.; Lobkovsky, E. B. Inorg. Chem.

2009, 48, 11576−11585.

2007, 1552−1556.

(73) Ray, K.; DeBeer George, S.; Solomon, E. I.; Wieghardt, K.;

Neese, F. Chem.Eur. J. 2007, 13, 2783−2797.

(74) Ray, K.; Weyhermuller.; Neese, F.; Wieghardt, K. Inorg. Chem.

̈

2007, 44, 5345−5360.

(44) Westerhausen, M.; Kneifel, A. N. Inorg. Chem. Commun. 2004,

7, 763−766.

(75) Petrenko, T.; Ray, K.; Wieghardt, K. E.; Neese, F. J. Am. Chem.

Soc. 2006, 128, 4422−4436.

12435

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Inorganic Chemistry

Article

(76) Herebian, D.; Wieghardt, K. E.; Neese, F. J. Am. Chem. Soc.

2003, 125, 10997−11005.

(98) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.;

Gordon, M. S.; Jensen, J. J.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.;

Su, S.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem.

1993, 14, 1347−1363.

(77) (a) Spikes, G. H.; Bill, E.; Weyhermuller, T.; Wieghardt, K.

̈

Angew. Chem., Int. Ed. 2008, 47, 2973−2977. (b) Lu, C. C.; Bill, E.;

(99) Mills, J. L.; Nelson, R.; Shore, S. G.; Anderson, L. B. Anal.

Chem. 1971, 43, 157−159.

Weyhermuller, T.; Bothe, E.; Wieghardt, K. J. Am. Chem. Soc. 2008,

̈

130, 3181−3197.

(100) Otwinowski, Z.; Minor, W. Macromolecular Crystallography,

Part A. Methods in Enzymology, Vol. 276; Carter, C. W., Sweet, Fr. R. M.,

Eds.; Academic Press: New York; pp 307−326.

(101) Scheldrick, G. M. SHELXTL, version 5.10; Bruker AXS, Inc.:

Madison, WI, 1999.

(78) (a) Bart, S. C.; Chlopek, K.; Bill, E.; Bouwkamp, M. W.;

Lobkovsky, E.; Neese, F.; Wieghardt, K.; Chirik, P. J. J. Am. Chem. Soc.

2006, 128, 13901−13912. (b) Sylvester, K. T.; Chirik, P. J. J. Am.

Chem. Soc. 2009, 131, 8772−8773. (c) Tondreau, A. M.; Milsmann,

C.; Patrick, A. D.; Hoyt, H. M.; Lobkovsky, E.; Wieghardt, K.; Chirik,

P. J. J. Am. Chem. Soc. 2010, 132, 15046−15059. (d) Bowman, A. C.;

NOTE ADDED AFTER ASAP PUBLICATION

Milsmann, C.; Bill, E.; Lobkovsky, E.; Weyhermuller, T.; Wieghardt,

̈

■

K.; Chirik, P. J. Inorg. Chem. 2010, 49, 6110−6123.

Due to a production error, this paper was published on the

Web on November 17, 2011, before the text corrections were

implemented. The corrected version was reposted on

November 18, 2011.

(79) (a) Nguyen, A. I.; Zarkesh, R. A.; Lacy, D. C.; Thorson, M. K.;

Heyduk, A. F. Chem. Sci. 2011, 2, 166−169. (b) Nguyen, A. I.;

Blackmore, K. J.; Carter, S. M.; Zarkesh, R. A.; Heyduk, A. F. J. Am.

Chem. Soc. 2009, 131, 3307−3316. (c) Blackmore, K. J.; Lal, N.; Ziller,

J. W.; Heyduk, A. G. Eur. J. Inorg. Chem. 2009, 735−743.

(80) Stanciu, C.; Jones, M. E.; Fanwick, P. E.; Abu-Omar, M. M.

J. Am. Chem. Soc. 2007, 129, 12400−12401.

(81) (a) Pierpont, C. G. Coord. Chem. Rev. 2001, 219, 415−433.

(b) Pierpont, C. G. Coord. Chem. Rev. 2001, 216, 99−125.

(82) Sarangi, R.; DeBeer George, S.; Jackson Rudd, D.; Szilagyi, R.

K.; Ribas, X.; Rovira, C.; Holm, R. H.; Hedman, B.; Hodgson, K. O.;

Solomon, E. I. J. Am. Chem. Soc. 2007, 129, 2316−2326.

(83) Kapre, R.; Bothe, E.; Weyhermuller, T.; DeBeer George, S.;

̈

Muresan, N.; Wieghardt, K. Inorg. Chem. 2007, 46, 7827−7839.

(84) Banerjee, P.; Sproules, S.; Weyhermuller, T.; DeBeer George,

̈

S.; Wieghardt, K. Inorg. Chem. 2009, 48, 5829−5847.

(85) Krivokapic, I.; Zerara, M.; Daku, M. L.; Vargas, A.; Enachescu,

C.; Ambrus, C.; Treganna-Piggott, P.; Amstutz, N.; Krausz, E.; Hauser, A.

Coord. Chem. Rev. 2007, 251, 364−378.

(86) Volpe, E. C.; Wolczanski, P. T.; Lobkovsky, E. B. Organo-

metallics 2010, 29, 364−377.

(87) Goodwin, K. V.; Pennington, W. T.; Petersen, J. D. Inorg. Chem.

1989, 28, 2016−2018.

(88) Brunschwig, B.; Sutin, N. J. Am. Chem. Soc. 1978, 100, 7568−

7577.

(89) King, E. R.; Betley, T. A. J. Am. Chem. Soc. 2009, 131, 14374−

14380.

(90) Sazama, G. T.; Betley, T. A. Inorg. Chem. 2010, 49, 2512−2524.

(91) Turro, N. J. Modern Molecular Photochemistry; University

Science Books: Sausalito, CA, 1991.

(92) Neese, F.; Zumft, W. G.; Antholine, W. E.; Kroneck, P. M. H.

J. Am. Chem. Soc. 1996, 118, 8692−8699.

(93) http://ewww.mpi-muelheim.mpg.de/bac/logins/bill/julX_en.

php

(94) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.

(95) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.

(96) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.,Jr.; Vreven, T.;

Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;

Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;

Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;

Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,

O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross,

J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,

O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;

Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;

Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;

Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,

J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;

Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;

Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;

Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,

W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian’03, revision

C.02; Gaussian, Inc.: Carnegie, PA.

(97) Nakano, H. J. Chem. Phys. 1993, 99, 7983−7992.

12436

dx.doi.org/10.1021/ic200376f|Inorg. Chem. 2011, 50, 12414−12436

Article Doi

Synthesis and Characterization of (smif)2Mn (n = 0, M = V, Cr, Mn, Fe,Co, Ni, Ru; N = +1, M = Cr, Mn, Co, Rh, Ir; Smif =1,3-di-(2-pyridyl)-2-azaallyl)

DOI: 10.1021/ic200376f

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ic200376f

Full text of DOI:10.1021/ic200376f

Products guided by the article

R&D Labs maybe for 1349903-53-2

Relevant to this article

Hot Product