Utilization of phosphinoamide ligands in homobimetallic Fe and Mn complexes: The effect of disparate coordination environments on metal-metal interactions and magnetic and redox properties

Article abstract of DOI:10.1021/ic300776y

A series of homobimetallic phosphinoamide-bridged diiron and dimanganese complexes in which the two metals maintain different coordination environments have been synthesized. Systematic variation of the steric and electronic properties of the phosphinoamide phosphorus and nitrogen substituents leads to structurally different complexes. Reaction of [ⁱPrNKPPh₂] (1) with MCl₂ (M = Mn, Fe) affords the phosphinoamide-bridged bimetallic complexes [Mn(ⁱPrNPPh₂)₃Mn( ⁱPrNPPh₂)] (3) and [Fe(ⁱPrNPPh ₂)₃Fe(ⁱPrNPPh₂)] (4). Complexes 3 and 4 are iso-structural, with one metal center preferentially binding to the three amide ligands in a trigonal planar arrangement while the second metal center is ligated by three phosphine donors. A fourth phosphinoamide ligand caps the tetrahedral coordination sphere of the phosphine-ligated metal center. M?ssbauer spectroscopy of complex 4 suggests that the metals in these complexes are best described as Fe^II centers. In contrast, treatment of MnCl₂ or FeI₂ with [MesNKPⁱPr₂] (2) leads to the formation of the halide-bridged species [(THF)Mn(μ-Cl) (MesNPⁱPr₂)₂Mn(MesNPⁱPr ₂)] (5) and [(THF)Fe(μ-I)(MesNPⁱPr₂) ₂FeI (7), respectively. Utilization of FeCl₂ in place of FeI₂, however, leads exclusively to the C₃-symmetric complex [Fe(MesNPⁱPr₂)₃FeCl] (6), structurally similar to 4 but with a halide bound to the phosphine-ligated Fe center. The M?ssbauer spectrum of 6 is also consistent with high spin Fe^II centers. Thus, in the case of the [ⁱPrNPPh₂]^- and [MesNPⁱPr₂]^- ligands, zwitterionic complexes with the two metals in disparate coordination environments are preferentially formed. In the case of the more electron-rich ligand [ ⁱPrNPⁱPr₂]^-, complexes with a 2:1 mixed donor ligand arrangement, in which one of the ligand arms has reversed orientation relative to the previous examples, are formed exclusively when [ⁱPrNLiPⁱPr₂] (generated in situ) is treated with MCl₂ (M = Mn, Fe): (THF)₃LiCl[Mn(N ⁱPrPⁱPr₂)₂(PⁱPr ₂NⁱPr)MnCl] (8) and [Fe(NⁱPrP ⁱPr₂)₂(PⁱPr₂N ⁱPr)FeCl] (9). Bimetallic complexes 3-9 have been structurally characterized using X-ray crystallography, revealing Fe-Fe interatomic distances indicative of metal-metal bonding in complexes 6 and 9 (and perhaps 4, to a lesser extent). All of the complexes appear to adopt high spin electron configurations, and magnetic measurements indicate significant antiferromagnetic interactions in Mn₂ complexes 5 and 8 and no discernible magnetic superexchange in Fe₂ complex 4. The redox behavior of complexes 3-9 has also been investigated using cyclic voltammetry, and theoretical investigations (DFT) were performed to gain insight into the metal-metal interactions in these unique asymmetric complexes.

Full text of DOI:10.1021/ic300776y

Article
pubs.acs.org/IC
Utilization of Phosphinoamide Ligands in Homobimetallic Fe and Mn
Complexes: The Effect of Disparate Coordination Environments on
Metal−Metal Interactions and Magnetic and Redox Properties
Subramaniam Kuppuswamy,† Mark W. Bezpalko,† Tamara M. Powers,§ Mark M. Turnbull,‡
Bruce M. Foxman,† and Christine M. Thomas*,†
†Department of Chemistry, Brandeis University, 415 South Street, Waltham, Massachusetts 02454, United States
‡Carlson School of Chemistry and Biochemistry, Clark University, 950 Main Street, Worcester, Massachusetts 01610, United States
§Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02139, United States
S
* Supporting Information
ABSTRACT: A series of homobimetallic phosphinoamide-
bridged diiron and dimanganese complexes in which the two
metals maintain different coordination environments have been
synthesized. Systematic variation of the steric and electronic
properties of the phosphinoamide phosphorus and nitrogen
substituents leads to structurally different complexes. Reaction of
[iPrNKPPh2] (1) with MCl2 (M = Mn, Fe) affords the
phosphinoamide-bridged bimetallic complexes [Mn-
(iPrNPPh2)3Mn(iPrNPPh2)] (3) and [Fe(iPrNPPh2)3Fe-
(iPrNPPh2)] (4). Complexes 3 and 4 are iso-structural, with one
metal center preferentially binding to the three amide ligands in a
trigonal planar arrangement while the second metal center is
ligated by three phosphine donors. A fourth phosphinoamide
ligand caps the tetrahedral coordination sphere of the phosphine-ligated metal center. Mossbauer spectroscopy of complex 4
̈
suggests that the metals in these complexes are best described as FeII centers. In contrast, treatment of MnCl2 or FeI2 with
[MesNKPiPr2] (2) leads to the formation of the halide-bridged species [(THF)Mn(μ-Cl)(MesNPiPr2)2Mn(MesNPiPr2)] (5)
and [(THF)Fe(μ-I)(MesNPiPr2)2FeI (7), respectively. Utilization of FeCl2 in place of FeI2, however, leads exclusively to the C3-
symmetric complex [Fe(MesNPiPr2)3FeCl] (6), structurally similar to 4 but with a halide bound to the phosphine-ligated Fe
center. The Mossbauer spectrum of 6 is also consistent with high spin FeII centers. Thus, in the case of the [iPrNPPh ]− and
̈
2
[MesNPiPr2]− ligands, zwitterionic complexes with the two metals in disparate coordination environments are preferentially
formed. In the case of the more electron-rich ligand [iPrNPiPr2]−, complexes with a 2:1 mixed donor ligand arrangement, in
which one of the ligand arms has reversed orientation relative to the previous examples, are formed exclusively when
[iPrNLiPiPr2] (generated in situ) is treated with MCl2 (M = Mn, Fe): (THF)3LiCl[Mn(NiPrPiPr2)2(PiPr2NiPr)MnCl] (8) and
[Fe(NiPrPiPr2)2(PiPr2NiPr)FeCl] (9). Bimetallic complexes 3−9 have been structurally characterized using X-ray
crystallography, revealing Fe−Fe interatomic distances indicative of metal−metal bonding in complexes 6 and 9 (and perhaps
4, to a lesser extent). All of the complexes appear to adopt high spin electron configurations, and magnetic measurements
indicate significant antiferromagnetic interactions in Mn2 complexes 5 and 8 and no discernible magnetic superexchange in Fe2
complex 4. The redox behavior of complexes 3−9 has also been investigated using cyclic voltammetry, and theoretical
investigations (DFT) were performed to gain insight into the metal−metal interactions in these unique asymmetric complexes.
redox chemistry and unique magnetic properties,4,6,7 and in
INTRODUCTION
■
some cases provides an intriguing model of complex enzymatic
active sites.8 Dinuclear active sites featuring two late transition
metals are also prevalent in metalloenzymes, including carbon
monoxide dehydrogenases,9,10 acetyl-coenzymeA synthase,9,10
heme-copper oxidases,11 dimanganese enzymes,12 methane
monooxgyenases,13,14 and hydrogenases.15,16 Metal−metal
interactions in symmetric homobimetallic complexes of second
Multimetallic transition metal complexes have received
considerable attention in recent years owing to their ability to
facilitate multielectron redox processes, their potential
applications in catalysis, and their parallels to enzymatic active
sites.1 Of particular importance is the electronic communication
between multiple metal centers and the resulting effects on
electronic structure. Recent contributions in this area include
elegant multidentate ligand designs capable of supporting well-
defined trinuclear complexes.2−5 The interplay between the
metal centers in these compounds gives rise to multielectron
Received: April 14, 2012
Published: July 17, 2012
© 2012 American Chemical Society
8225
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
and third row transition metals have been investigated
extensively, most notably in the 4-fold symmetric metal−
metal multiple bonding regime.17−19 Metal−metal bonds in
high spin dinuclear complexes of the first transition series,
however, remain quite rare and are understood to a far lesser
extent.
Many groups have been interested in heterobimetallic
combinations of early and late transition metal fragments, as
the unusually polar metal−metal interactions and different
Lewis acid/base properties of the two component metal centers
are likely to lead to new and interesting reactivity patterns.20−24
Our group, in particular, has recently focused on such early/late
metal combinations linked by phosphinoamide ligands, utilizing
the hard/soft acid/base properties of the phosphine and amide
donors to preferentially coordinate late and early metals,
respectively.25 The metal−metal interactions in early/late
heterobimetallic complexes of this type have been shown to
have a drastic effect on redox potentials and lead to interesting
reactivity toward small molecule substrates (e.g., alkyl halides,
H2, CO2).26−30
Using what we have learned from phosphinoamide-linked
early/late heterobimetallic systems, we have begun to
investigate homodinuclear systems in which the metals are in
coordination environments that differ substantially in terms of
hard (amide) and soft (phosphine) donor functionalities. Little
research has been done in the area of homobimetallic
complexes featuring metals in disparate coordination environ-
ments. While a large number of dinuclear late transition metal
complexes have been reported, these are typically formed using
bridging ligands that present similar (hard/hard or soft/soft)
donors to each metal and, thus, lead to metals in similar
electronic environments.31,32 An interesting class of polynuclear
metal clusters composed of metal−metal interactions between
one hard open-shell metal and a second soft metal typically
supported by carbonyls has recently emerged in the
literature.33,34 Theoretical investigations of these “xenophilic”
complexes have suggested that the disparate coordination
environments on two appended late transition metals may lead
to unusual electronic properties and magnetic behavior.35 In
particular, we are curious whether different electronic environ-
ments about two similar metal centers are enough to promote
dative metal−metal interactions. A recent theory posed by
Lindahl suggests that such metal−metal interactions may play a
role in the catalytic mechanism of a number of bimetallic
enzyme active sites.36 Moreover, differentiating the two metal
centers in bimetallic compounds via their coordination spheres
may prove a useful method for promoting two-electron mixed
valence.32
spectroscopic methods.43 The electronic structure of a similar
4-fold symmetric formamidinate complex Fe2(DPhF)4 has
also been investigated computationally by Berry and co-
workers.45 In contrast to iron systems, high spin polynuclear
manganese complexes tend to exhibit antiferromagnetic
coupling and weaker metal−metal interactions.2,46−50
44
To date, all of the reported non-organometallic metal−metal
bonded first row transition metal complexes in the literature are
supported by symmetric ligand sets that present identical
donors to both Fe centers, providing a high degree of
delocalization. Herein, we present the first examples of high
spin diiron and dimanganese complexes in which the metal sites
are coordinated by distinctly different ligand sets. Phosphinoa-
mide ligands of the type [R2PNR′]− are utilized, as an
extension of our earlier work with heterobimetallic com-
pounds.25,51 In contrast to other P,N-binucleating ligands in the
literature that tend to have softer N-donors such as amines or
pyridines, phosphinoamides present one hard and one soft
ligand donor, leading to metal centers whose coordination
spheres are substantially different.31 In the context of
homobimetallic complexes, phosphinoamide ligands have
been reported to stabilize homodinuclear NiII and PdII
complexes,52 as well as CrII complexes with activities toward
ethylene oligomerization.53 We now extend this chemistry to
MnII and FeII, utilizing three different phosphinoamide linkers,
and analyze the complex electronic and magnetic properties of
these new compounds.
RESULTS AND DISCUSSION
■
Synthesis and Characterization of 1−9. Salt metathesis
reactions are a convenient route for the preparation of late
transition metal amides under mild conditions. Treatment of
i
the phosphinoamines PrNHPPh2 and MesNHPiPr2 with 1.02
equiv of KH at room temperature in tetrahydrofuran (THF)
affords [iPrNKPPh2] (1) and [MesNKPiPr2] (2) as air and
moisture sensitive yellow or colorless crystalline solids,
respectively (Scheme 1). Characteristic 31P NMR resonances
Scheme 1
Several high spin dinuclear Fe complexes have been reported,
and the electronic structure and metal−metal bonding within
these compounds appears highly variable.37−41 Notably, the
high spin nature of the Fe centers in these complexes results in
unusually high spin states (as high as S = 4) in combination
with strong metal−metal interactions. The 3-fold symmetric
“trigonal lantern” complex Fe2(DPhF)3 (DPhF = diphenylfor-
mamidinate) presents an archetypical example of metal−metal
multiple bonding in a high spin complex, with an extremely
short Fe−Fe bond distance (2.2318(8) Å) and an S = 7/2
ground state.40 Cotton and co-workers provided a preliminary
explanation for the nature of the metal−metal bonding in
Fe2(DPhF)3,42 and Lu and co-workers recently revisited this
rare example of a ferromagnetically coupled metal−metal
bonded compound using advanced computational and
centered at 46.2 ppm (1) and 72.3 ppm (2) are observed
significantly downfield compared to the protonated ligands (35
and 57 ppm, respectively). The dimeric structures of 1 and 2
are inferred based on analogy to the structurally characterized
lithium phosphinoamides previously reported.54
Potassium amides 1 and 2 are excellent precursors for the
preparation of late transition metal phosphinoamide complexes:
Salt methathesis of 1 with MCl2 (M = Mn or Fe) at ambient
temperature leads to the formation of [Mn(iPrNPPh2)3Mn-
(iPrNPPh2)] (3) and [Fe(iPrNPPh2)3Fe(iPrNPPh2)] (4) as
yellow and purple crystalline solids, respectively (Scheme 2).
1
The H NMR of complex 3 features two broad unresolvable
resonances, as is typical for high spin Mn complexes, but the
analytical purity of 3 is nonetheless confirmed by combustion
8226
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
yield. Conversely, the reaction between [MesNKPiPr2] (2) and
FeI2 affords [(THF)Fe(μ-I)(MesNPiPr2)2FeI] (7) as a brown
crystalline solid in 50% yield. The resonances in the H NMR
Scheme 2
1
spectrum of 6 and 7 appear over a 50 ppm chemical shift
window, but are nonetheless interpretable. The 1H NMR
spectrum of 6 shows 6 very broad resonances between +44 and
+1 ppm, indicative of a C3-symmetric ligand environment
without an additional terminal phosphinoamide ligand. Eleven
broad paramagnetic resonances are observed between +59 and
−16 ppm in the 1H NMR of complex 7, corresponding to a Cs-
symmetric molecule with two signals for the coordinated THF
ligand.
analysis. On the other hand, the analogous iron complex 4 has
an informative 1H NMR spectrum that exhibits 10 para-
magnetically shifted broad resonances between +51 and −31
ppm, indicative of two inequivalent phosphinoamide environ-
ments. On the basis of the number of 1H NMR resonances and
their relative intensities, complex 4 was tentatively assigned as
having three bridging phosphinoamide ligands, with one
additional terminal phosphinoamide (later confirmed crystallo-
graphically, vide infra).
Attempts to isolate alkali metal salts of the electron rich
i
phosphinoamine PrNHPiPr2 failed under various reaction
conditions owing to the solubilizing isopropyl substituents.
Therefore, the lithium amide [iPrNLiPiPr2] was generated in
i
situ via treatment of an equimolar ratio of PrNHPiPr2 with
nBuLi. Subsequent addition of MnCl2 in THF affords an amber
red crystalline solid identified as (THF)3LiCl[Mn-
(NiPrPiPr2)2(PiPr2NiPr)MnCl] (8) in 56% yield (Scheme 4).
Investigating a phosphinoamide ligand set with different
steric and electronic properties led to structurally different
homobimetallic species. Metathesis of [MesNKPiPr2] (2) and
MnCl2 results in formation of [(THF)Mn(μ-Cl)-
(MesNPiPr2)2Mn(MesNPiPr2)] (5) as a reddish yellow
crystalline solid in 40% yield (Scheme 3). Again, the 1H
NMR spectrum of high spin Mn complex 5 is uninformative,
exhibiting only two broad and featureless resonances. The exact
formulation and connectivity of 5 could only be determined
crystallographically (vide infra). In the case of Fe, treatment of
[MesNKPiPr2] (2) with FeCl2 provides [Fe(MesNPiPr2)3FeCl]
(6) as an analytically pure pale yellow crystalline solid in 70%
Scheme 4
Scheme 3
1
Once again, the H NMR spectrum of 8 shows two broad
featureless peaks that are not readily assigned. Complex
[Fe(NiPrPiPr2)2(PiPr2NiPr)FeCl] (9) is synthesized under the
identical reaction conditions of 8, using FeCl2 in place of
MnCl2. The 1H NMR spectrum of 9 exhibits nine broad
paramagnetic resonances that span the range of +85 to −39
ppm, attributed to two inequivalent phosphinoamide environ-
ments. It was later determined crystallographically that this
inequivalence is due to one of the phosphinoamide ligands
coordinating in the opposite direction from the other two in
complexes 8 and 9 (vide infra).
Molecular Structures of 3−9. As predicted, X-ray
crystallographic studies reveal that complexes 3 and 4 are
isomorphous and adopt unusual structures in which the
bimetallic core consists of three symmetric six-membered
M2N2P2 rings (Figure 1, Supporting Information). The
coordination environment about the amide-bound Mn2 center
in 3 is trigonal planar (ΣN−Mn−N 356.99°) and the distorted
tetrahedral coordination sphere of Mn1 is completed by a
terminal phosphinoamide ligand. It is clear from the molecular
structure that the MnII ions adopt different coordination
8227
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
ions in 3 is significantly longer than the sum of the covalent
radii (3.1436(3) Å vs 2.354 Å),56,57 suggesting little or no
bonding interaction between the metal ions.
An X-ray crystallographic study shows that single crystals of 4
are isomorphous with those of 3; however, there are subtle
differences in their geometrical parameters (Figure 1). Most
notably, the Fe−N and Fe−P distances in 4 are substantially
shorter than the Mn−N and Mn−P distances in complex 3.
This phenomenon can be attributed to the higher spin, S, of the
MnII centers resulting in poor ligand field stabilization.
Interestingly, only two tris(phosphine) iron amide complexes
have been reported in the literature, the zwitterionic FeII
tris(phosphino)borates PhB(CH2PPh2)3Fe(NHtolyl) and PhB-
(CH2PiPr2)3FeNPh2.58,59 The Fe1−P distances and Fe1−N4
distances in 4 (avg Fe−P = 2.49 Å; Fe1−N4 = 1.944(2) Å) are
consistent with these two complexes (e.g., avg Fe−P = 2.42 Å
and Fe−N = 1.913(2) Å in PhB(CH2PiPr2)3FeNPh2).58 In
addition, the Fe2−N distances in 4 (avg = 1.95 Å) are similar to
those in the previously reported FeII complex Fe-
(iPrNPPh2)3Cu(Ph2PNHiPr) (avg = 1.97 Å).51 Thus, bond
metrics are consistent with both Fe centers remaining in a 2+
oxidation state. As with 3, the interatomic distance between the
Fe centers in 4 is significantly longer than the sum of the
covalent radii (2.8684(6) Å vs 2.346 Å),56,57 indicative of weak
or absent metal−metal bonding interactions.
A single crystal X-ray structure determination of 5 reveals
two bridging phosphinoamide ligands, a μ2-Cl, and terminally
bound phosphinoamide and THF ligands on either Mn center
(Figure 2). The geometry at both four-coordinate Mn centers is
nearly tetrahedral (avg angles: 109.5° and 108.4°, respectively),
with the largest deviation being the N1−Mn1−N2 angle
(129.68(6)°), which is expanded as a result of the steric bulk of
the mesityl amide substituents. In contrast to 3, the amide-
bound Mn center in 5 has only two amide-donors, resulting in a
more electronically unsaturated metal and necessitating binding
of an additional THF solvent molecule. As seen in 3 and 4 the
pendant phosphine donor bound to the terminal amide ligand
is only weakly interacting with Mn2 (Mn2−P3 = 3.0933(6) Å).
The solid state molecular structure of 6 reveals a C3
symmetric dimer consisting of three symmetrical bridging
phosphinoamides and a chloride at the terminal position on the
phosphine-bound Fe center (Figure 2). Similar to 4, the
geometries at the two Fe centers are best described as trigonal
planar about Fe1 (ΣN−Fe−N 359.89°) and pseudotetrahedral
Figure 1. Displacement ellipsoid (30%) representations of 4 (complex
3 is isomorphous, see Supporting Information). Hydrogen atoms have
been omitted for clarity. Selected bond lengths (Å) and angles (deg):
4: Fe1−Fe2, 2.8684(6); Fe1−P1, 2.4593(8); Fe1−P2, 2.4791(8);
Fe1−P3, 2.5355(8), Fe1−P4, 2.8129(9); Fe1−N4, 1.944(2); Fe2−N1,
1.946(2); Fe2−N2, 1.953(2); Fe2−N3, 1.960(2); N1−Fe2−N2,
121.16(9); N1−Fe2−N3, 118.17(9); N2−Fe2−N3, 118.17(9), Fe1−
N4−P4, 102.05(12). 3: Mn1−Mn2, 3.1436(3); Mn1−P1, 2.6093(4);
Mn1−P2, 2.5843(4); Mn1−P3, 2.6424(4); Mn1−P4, 2.8384(4);
Mn1−N4, 1.9967(13); Mn2−N1, 2.0369(13); Mn2−N2,
2.0304(12); Mn2−N3, 2.0376(12); N1−Mn2−N2, 118.74(5); N1−
Mn2−N3, 117.11(5); N2−Mn2−N3, 116.25(5); Mn1−N4−P4,
101.26(6).
environmentsMn2 binds three negatively charged amides
while Mn1 is bound by three neutral phosphine donors and
one terminal amideresulting in an unusual zwitterionic
homobimetallic complex. In addition to the terminal amide
ligand bound to Mn1, the solid state structure suggests an
additional weaker interaction with the pendant phosphine
(Mn1−P4 = 2.8384(4) Å). The terminal Mn1−N distance is
0.04 Å shorter than that of Mn2 with the bridging
phosphinoamides, likely a result of additional amide π-donation
to the more electronically unsaturated Mn1. The Mn−N
distances are comparable to those in the previously reported
heterobimetallic Mn/Cu complex supported by this ligand
system, Mn(iPrNPPh2)3Cu(Ph2PNHiPr),51 and the average
Mn−P distance in 3 (2.6120(4) Å) is comparable to that in the
few examples of tris(phosphine)-supported pseudotetrahedral
MnII complexes.55 The interatomic distance between the Mn
Figure 2. Displacement ellipsoid (30%) representations of 5, 6, and 7. Hydrogen atoms have been omitted for clarity. Selected bond lengths (Å) and
angles (deg): 5: Mn1−Mn2, 3.0033(4); Mn1−N1, 2.0404(14); Mn1−N2, 2.0397(15); Mn1−O1, 2.1184(13); Mn1−Cl1, 2.4462(5); Mn2−Cl1,
2.5242(5); Mn2−P1, 2.6110(5); Mn2−P2, 2.6205(5); Mn2−P3, 3.0933(6); Mn2−N3, 2.0114(15), Mn2−N3−P3, 112.85(8). 6: Fe1−Fe2,
2.5855(4); Fe1−N1, 2.0201(15); Fe1−N2, 1.9999(15); Fe1−N3, 1.9959(16), Fe2−P1, 2.4497(6); Fe2−P2, 2.4383(6); Fe2−P3, 2.4546(6); Fe2−
Cl1, 2.2630(5); N1−Fe1−N2, 121.56(6); N1−Fe1−N3, 118.25(6); N2−Fe1−N3, 120.08(7). 7: Fe1−Fe2, 2.8134(3); Fe1−N1, 1.9808(12); Fe1−
N2, 1.9542(12); Fe1−O1, 2.0796(11); Fe1−I1, 2.8128(3); Fe2−P1, 2.4702(5); Fe2−P2, 2.4523(4); Fe2−I1, 2.7004(3); Fe2−I2, 2.5729(3); N1−
Fe1−N2, 134.89(5); N1−Fe1−O1, 103.44(5); N1−Fe1−I1, 104.19(4); N2−Fe1−O1, 100.15(5); N2−Fe1−I1, 105.25(4); O1−Fe1−I1, 106.31(3).
8228
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
Figure 3. Displacement ellipsoid (30%) representations of 8 and 9. For clarity, the [(THF)3Li]+ cation of 8, one of the two independent molecules
in the asymmetric unit of 9, and all hydrogen atoms have been omitted. Selected bond lengths (Å): For 8: Mn1−Mn2, 3.2464(9); Mn1−N2,
2.079(3); Mn1−N3, 2.071(2); Mn1−P1, 2.624(1); Mn1−Cl1, 2.439(1); Mn2−P2, 2.574(1); Mn2−P3, 2.582(1); Mn2−N1, 2.053(2); Mn2−Cl2,
2.347(1). For 9: Fe1−Fe2, 2.6112(7); Fe1−N1, 1.938(3); Fe1−N2, 1.944(3); Fe1−P3, 2.444(1); Fe2−P1, 2.486(1); Fe2−P2, 2.498(1); Fe2−N3,
1.967(3); Fe2−Cl1, 2.3096(9).
a
Table 1. Mossbauer Parameters for Complexes 4, 6, 7, and 9
̈
FeN
FeP
complex
δ (mm/s)
|ΔEQ| (mm/s)
δ (mm/s)
|ΔEQ| (mm/s)
4
6
7
9
0.57
0.69
0.64
0.60
0.65
2.17
2.63
2.77
1.80
2.05
0.51
0.63
0.83
0.60
1.67
1.16
1.27
1.80
[Fe(iPrNPPh2)3Cu2(iPrNPPh2)]51
PhB(CH2PiPr2)3FeCl72
0.58
0.55
1.65
1.75
72
PhB(CH2PiPr2)3FeNR2
a
The parameters for similar monometallic complexes are also provided for comparison.
about Fe2 (avg. bond angle = 109.3°). Like 3 and 4, the ligand
orientation in complex 6 renders the complex zwitterionic.
Unlike the aforementioned complexes, however, the inter-
metallic distance in complex 6 (2.5855(4) Å) is short enough
to indicate a metal−metal interaction. This can be rationalized
by the fact that the Cl− is donating far less electron density than
the apical phosphinoamide in 4. The Fe−Fe distance in the
symmetrically bridged tetragonal lantern complex Fe2(DPhF)4
(2.462(2) Å44,45) is significantly shorter than those in 4 or 6.
This phenomenon can be tentatively attributed to the two
disparate coordination environments of the two Fe centers in 4
and 6 and the resulting differences in orbital energies, leading
to poorer overlap between the metal d orbitals. The immediate
coordination sphere of Fe2 in complex 6 bears resemblance to
the tris(phosphino)borate FeIICl systems reported by Peters
and co-workers,60,61 however, the P−Fe2−P angles in 4 and 6
are ∼101−105° while the tris(phosphine)borate complexes
have contracted P−Fe−P angles between 90° and 95°.
An X-ray structure determination of the iodide derivative 7
reveals a geometry consisting of two bridging phosphinoamides
and a bridging iodide, with a terminal iodide on Fe2 and a
bound THF molecule on Fe1 completing the coordination
spheres at each end of the bimetallic framework (Figure 2).
The bridging iodide is asymmetrically located between the two
Fe atoms (Fe1−I1 = 2.8128(3) Å, and Fe2−I1 = 2.7004(3) Å),
likely a consequence of the different electronic environments at
the two Fe centers. Unlike complex 6, the Fe−Fe distance in 7
(2.8134(3) Å) precludes any significant metal−metal inter-
action.
A single crystal X-ray diffraction study of the [NiPrPiPr2]−-
ligated Mn complex 8 reveals an asymmetric triply bridged
scaffold. Unlike the aforementioned complexes 3−7, the three-
bridging phosphinoamide ligands in 8 orient in two different
directions, with two amides bound to Mn1 and the third amide
bonded to Mn2 (Figure 3). Mn2 completes its coordination
sphere with a terminal chloride ligand, while a LiCl(THF)3 unit
bonds to Mn1 (see Supporting Information). While the
zwitterionic complex 3 maintains a trigonal planar geometry
at the tris(amide)-bound Mn center, the bis(amide) bound Mn
center in 8 is more electron-deficient and binds tightly to a LiCl
salt. Notably, repeated attempts to remove this LiCl salt via
extraction into nonpolar solvents were unsuccessful. Similar to
3 and 5, there is no indication of a Mn−Mn interaction, as the
Mn centers in 8 are separated by 3.2464(9) Å.
The solid state molecular structure of complex 9 contains
two crystallographically independent molecules with similar
geometrical parameters in the unit cell (Figure 3, Supporting
Information). The composition of complex 9 is identical to that
of 8, except for the absence of a coordinated LiCl salt. The
geometry at Fe1 is distorted trigonal planar (ΣL‑Fe‑L 359.26°),
with the angle between the amide donors larger than either P−
Fe−N angle (N1−Fe1−N2 = 137.17(12)°). At first glance, this
distortion seems unusual since the amide donors should be less
sterically encumbered than the phosphines. However, examin-
ing the opposite side of the molecule, one can see that this
8229
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
expansion of the N1−Fe-1−N2 angle is due to steric repulsion
between the diisopropylphosphine substituents bound to Fe2.
Interestingly, a short intermetallic distance (2.6112(7) Å) is
observed in 9, suggesting a metal−metal interaction.
trigonal planar tris(amido) Fe center in 4 on the basis of similar
quadrupole splittings. Other three-coordinate high spin Fe(II)
complexes reported in the literature have similar isomer shifts
(∼0.6 mm/s to ∼0.8 mm/s).68−71 The second quadrupole
doublet centered at δ = 0.51 mm/s (|ΔEQ| = 1.67 mm/s) can,
therefore, be attributed to the tetrahedral tris(phosphine)-
supported amido-FeII center. The parameters associated with
this Fe center are similar to those reported for Peters’
tris(phosphine)borate FeII-X complexes (X = Cl, δ = 0.58
mm/s, |ΔEQ| = 1.65 mm/s; X = NR2, δ = 0.55 mm/s, |ΔEQ| =
1.75 mm/s), particularly given the inherent differences (weak
interactions between the pendant phosphine and other Fe
center) between the former complexes and complex 4.72
Four coordinate Mn/Fe centers with two phosphines, one
amide, and one chloride in monometallic complexes are
relatively rare. A tripodal amido-bis(phosphine) Fe complex
was reported by Peters and co-workers in 2006,62 and a similar
Mn complex was reported by Arnold et al. in 2008.63 The bond
distances about Fe2 in 9 are similar to those in the amido-
bis(phosphine) system,62 and in Mn complex 8, which lacks a
metal−metal bond, both the bond distances and bond angles
associated with Mn2 are similar to those in the tripodal amido-
bis(phosphine) complex [tBuNSiMe2N(CH2CH2PiPr2)2]-
MnCl.63 Pincer type complexes with similar amide bis-
(phosphine) coordination environments have also been
reported.64−66
Zero-field 57Fe Mossbauer spectra of the [iPr PNMes]−-
̈
2
substituted complexes 6 and 7 are also composed of two
distinct quadrupole doublets, consistent with two Fe centers in
disparate coordination environments (Figure 5). The Mossba-
̈
̈
Mossbauer Spectroscopy. To further examine the
uer spectrum of complex 6 is adequately fitted by two
quadrupole doublets centered at δ = 0.63 mm/s (|ΔEQ| = 1.16
mm/s) and δ = 0.69 mm/s (|ΔEQ| = 2.63 mm/s), indicative of
high spin FeII.67 While the intensity ratio should be 1:1 based
on the stoichiometry of the complex, the best fit was obtained
with a 38% contribution of the former signal and a 62%
contribution of the latter doublet. We attribute these deviations
from 50% to a minor impurity in the sample, likely due to
oxidation during sample preparation.73 Again, based on
comparisons to the aforementioned trigonal planar tris(amido)
Fe/Cu2 trimetallic and tris(phosphino)borate FeII-X com-
plexes,51,72 the doublet with larger quadrupole splitting
centered at δ = 0.69 is attributed to the tris(amido) Fe center
while the doublet at δ = 0.63 mm/s with smaller quadrupole
splitting is assigned as the tris(phosphine)-ligated Fe center.
In contrast to the above compounds, zero-field 57Fe
electronic structure and coordination environment of diiron
complexes 4, 6, 7, and 9, these complexes were investigated by
zero-field 57Fe Mossbauer spectroscopy, and isomer shift and
̈
quadrupole splitting data is presented in Table 1. The
Mossbauer spectrum of 4 reveals a broad quadrupole doublet
̈
that is best modeled by two overlapping signals of equal
intensity (Figure 4), as expected for an asymmetric diiron
Mossbauer spectroscopy of 9 reveals a single quadrupole
̈
doublet centered at δ = 0.60 mm/s (|ΔEQ| = 1.80 mm/s, Figure
6). The quadrupole splitting is, again, consistent with high spin
FeII, but the two Fe centers are electronically similar enough to
preclude distinction of the two Fe centers by Mossbauer
̈
spectroscopy. Coincidental overlap of the Mossbauer signals for
̈
the two Fe centers in 9 is consistent with the more similar
coordination environment of the two Fe centers in comparison
to 6 or 4.
Magnetism. To better examine the effects of metal−metal
interactions in complexes 3−9 on their magnetic properties,
magnetic susceptibility measurements were carried out in both
solution (via Evans’ method) and the solid state (via a
superconconducting quantum interference device, SQUID). In
general, the magnetic behavior of the diiron and dimanganese
complexes is different and will be discussed separately.
The solution magnetic moment of 4 at room temperature
(μeff = 9.26 μB) is consistent with two high spin S = 2 Fe
centers (μspin only = 9.80 μB). To probe the magnetic behavior of
4 at low temperature, the solid state magnetic susceptibility was
measured using SQUID magnetometry. A plot of χT as a
function of temperature (per mole of Fe in 4) is shown in
Figure 7. The value of χT is virtually constant from room
temperature down to near 20 K, below which a small decrease
(∼20%) is observed. The solid line shows a fit of the data to a
dimer model with g = 2.43(1) and a single ion anisotropy (D)
of −3.4(3) cm−1, with J = 0 K, and the spin Hamiltonian used
to fit the data is shown in eq 1 (where D is D/kT). Although
the data could be fit well using only a single-ion anisotropy
term, it is likely that it could be fit similarly with a very weak
Figure 4. Zero-field 57Fe Mossbauer spectrum of a solid sample of 4 at
110 K (black) fitted as a combination of quadrupole doublets (red) at
55% (δ = 0.57 mm/s, |ΔEQ| = 2.17 mm/s, blue) and 45%(δ = 0.51
mm/s, |ΔEQ| = 1.67 mm/s, green) intensity.
̈
complex. At first glance, the immediate coordination sphere of
each Fe center in complex 4 might suggest a mixed valent FeIII/
FeI configuration; however, the isomer shifts revealed by fitting
of the Mossbauer spectrum are consistent with two high spin
̈
FeII centers (δ = 0.57 mm/s, 0.51 mm/s) linked together in a
zwitterionic arrangement.67 Comparing the Mossbauer spec-
̈
trum of 4 with that of a similar phosphinoamide-linked Fe/Cu2
trimetallic complex, [Fe(iPrNPPh2)3Cu2(iPrNPPh2)] (δ = 0.65
mm/s, |ΔEQ| = 2.05 mm/s),51 the quadrupole doublet centered
at δ = 0.57 mm/s (|ΔEQ| = 2.17 mm/s) can be assigned as the
8230
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
Figure 6. Zero-field 57Fe Mossbauer spectrum of solid samples of 9 at
̈
110 K (black) fitted as a single quadrupole doublet (red, δ = 0.60 mm/
s, |ΔEQ| = 1.80 mm/s).
Figure 5. Zero-field 57Fe Mossbauer spectra of solid samples of 6 and
̈
7 at 110 K (black) fitted as a combination of quadrupole doublets
(red). For complex 6, the signal is best fitted with a combination of
two signals at 38% (blue; δ = 0.63 mm/s, |ΔEQ| = 1.16 mm/s) and
62% (green; δ = 0.69 mm/s, |ΔEQ| = 2.63 mm/s) intensity. The
deviation of these values from 50% is likely attributed to a minor
impurity in the sample. The spectrum of complex 7 is adequately
modeled by two signals of roughly equal intensity (blue: 51%, δ = 0.83
mm/s, |ΔEQ| = 1.27 mm/s; green, 49%, δ = 0.64 mm/s, |ΔEQ| = 2.77
mm/s).
Figure 7. Variable temperature magnetic susceptibility of 4, reported
per mole Fe (rather than per mole complex). Solid line represents a fit
to the data as described in the text.
⎛
(6eD + 4e−2D
)
⎛
⎞
⎜
⎜
⎝
χ = 0.375(g2)
⎜
⎝
⎟
⎠
e2D + 2eD + 2e−2D
2(9eD − 7eD − 2e−2D
)
⎞
2
⎟
⎟
⎠
D
+
/9(T − θ)
e2D + 2eD + 2e−2D
(1)
antiferromagnetic exchange. Thus, in the case of 4, it appears
that measurable magnetic superexchange is not present within
the complex in either the solid state or solution. Notably, the
formamidinate-linked FeIIFeI and FeIIFeII systems originally
reported by Cotton and co-workers display ferromagnetic
behavior, perhaps as a result of significantly shorter metal−
metal distances.40−44
In contrast, the dimanganese complexes display magnetic
behavior indicative of significant antiferromagnetic exchange
within the complexes. Complex 3 has a magnetic moment at
room temperature in solution (μeff = 10.33 μB) that is slightly
below the spin-only value of 11.83 μB expected for two S = 5/2
ions. Solid state SQUID measurements suggest antiferromag-
8231
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
Figure 8. Variable temperature magnetic susceptibility of 5 and 8, with contributions for a small (4−8%) paramagnetic impurity subtracted out. χ is
reported per mole of Mn.
netic coupling between the two Mn(II) centers. However, the
magnetic behavior of 3 is dominated by the presence of a
paramagnetic impurity to the extent that no meaningful
interpretation is presented here (see Supporting Informa-
tion).74 The Mn(II) centers in complexes 5 and 8 are also
antiferromagetically coupled, this time with a solution magnetic
moment significantly lower than the expected spin-only value
for two S = 5/2 ions (6.91 μB and 6.87 μB, respectively).75
Further analysis of these results are presented below.
rapid decrease at lower temperatures as expected for an
antiferromagnetic bimetallic species. Attempts to fit the data to
an S = 5/2 (per Mn) bimetallic model were unsuccessful (likely
due to the error introduced by removal of the paramagnetic
impurity contribution which accounted for ∼50% of the
observed susceptibility at 40 K), but do suggest a J value in the
range −5 to −8 K.
The antiferromagnetic behavior of complex 5 is, perhaps,
expected given the presence of a halide bridging the two Mn
centers. However, similar magnetic susceptibility results were
observed for complex 8, which lacks a bridging halide. The M
vs H data at 1.8 K is similarly lower than anticipated for an
uncoupled high spin Mn(II) bimetallic system, but shows a
slightly larger percentage paramagnetic impurity (the saturation
magnetization is approaching 5000 emu/mol at 5 kOe,
suggesting an 8% impurity, see Supporting Information).74
The magnetic susceptibility data again shows only a monotonic
increase with decreasing temperature (see Supporting In-
formation). However, subtraction of an 8% paramagnetic
impurity results in a maximum near 40 K, similar to that
observed for 5, again indicative of antiferromagnetic
interactions in the bulk material (Figure 8).74 The lack of the
chloride bridge in 8, relative to 5, suggests that the magnetic
superexchange is propagated through the phosphinoamide M-
P-N-M bridge.
It is particularly interesting that this phosphinoamide-
mediated magnetic superexchange pathway is operative in the
Mn(II) bimetallic complexes, but completely absent in the case
of the Fe(II) analogues. Since the metal−metal distances in the
diiron complexes are significantly shorter than those in the
dimanganese complexes (vide supra), it can be concluded that
the observed antiferromagnetic coupling in the Mn2 complexes
must occur through the bridging phosphinoamide ligands. The
question as to why the phosphinoamide-mediated magnetic
superexchange mechanism is operative only in the case of
manganese is intriguing, and may relate to better angular
overlap between the metal and ligand orbitals once metal−
metal bonding interactions are disrupted.
Magnetization data as a function of field was collected on the
halide-bridged complex 5 from 0 kOe to 50 kOe at 1.8 K.
Several data points were also collected as the field was reduced
to zero to check for hysteresis effects; none were observed (see
Supporting Information). Even at 5 kOe, the magnetization of
the sample has clearly not reached an appreciable percentage of
the saturation magnetization. For a dimanganese complex, such
as 5, the saturation magnetization should approach 60,000
emu/mol, but even at 5 kOe the moment has yet to reach 2000
emu/mol (see Supporting Information). This suggests that
either there are significant antiferromagnetic interactions in the
sample, or that the bulk of the sample is in a singlet ground
state at low temperature and the moment observed at 1.8 K is
due to the presence of a small amount of a paramagnetic
impurity.74 The latter interpretation is borne out by the
susceptibility data. Magnetic susceptibility data were collected
as a function of temperature in a 1 kOe applied field from 1.8 to
310 K. An initial plot of χ vs T reveals that the susceptibility
rises continuously as temperature decreases to a maximum near
0.10 emu/mol-Oe at 1.8 K, suggestive of a paramagnetic system
(see Supporting Information). However, if this were strictly a
paramagnetic dimanganese complex with no interactions, the
value of χ at 1.8 K should be 4.89 emu/mol-Oe, nearly 50 times
the observed value. Again, this is suggestive of an
antiferromagnetic material, but one where the observed
susceptibility is dominated by a paramagnetic impurity.
Estimating the percentage impurity to be ∼4% from the M vs
H plot, the contribution of the paramagnetic impurity was
subtracted from the susceptibility data, and the result is shown
in Figure 8, with χ reported per mole of Mn rather than per
mole of 5. The data show a maximum in χ near 45 K and then a
Cyclic Voltammetry. The redox properties of the triply
bridged complexes 3, 4, 6, and 8−9 were investigated using
8232
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
cyclic voltammetry (CV). In general, the diiron complexes were
found to undergo both oxidative and reductive events, while the
dimanganese complexes only displayed oxidative behavior
within the measurable THF solvent window. Measured redox
potentials are tabulated in Table 2. The CV of dimanganese
features, although in this case interpretation of the CV is not as
straightforward on account of very broad features (see
Supporting Information). Comparing the oxidation potential
of the Fe/Cu complex Fe(iPrNPPh2)3Cu(Ph2PNHiPr) we
recently reported (Epa = −0.60 V)51 to those of 4, it can be
concluded that more facile first oxidation of 4 involves the
tris(amide)-ligated iron center, while the second oxidation
occurs at the tris(phosphine)-ligated Fe center. Since no
reductive features were observed in the CV of the Fe/Cu
complex Fe(iPrNPPh2)3Cu(Ph2PNHiPr), it can also be
concluded that the reduction of 4 at −1.98 V is centered on
the tris(phosphine)ligated Fe center.51 This is also consistent
with the facile reduction chemistry of other tris(phosphine)Fe
complexes in the literature.61,77 While assigning redox events to
each metal individually appears reasonable in this case, it should
be noted that this explanation is an oversimplification since a
significant amount of d-orbital mixing occurs between the two
metals (vide infra).
Table 2. Redox Potentials of 3, 4, 8, and 9 (vs Fc/Fc+)
a
Measured by Cyclic Voltammetry
redox potential
III/II
III/II
II/I
II/I
complex
MN
MP
MP
MN
b
c
c
c
c
c
c
3
4
8
9
−0.58 V
−0.74 V
−0.44 V
−1.02 V
0.05 V
−0.24 V
−0.23 V
b
d
d
d
−1.98 V
−2.15 V
−2.64 V
−3.03 V
a
b
0.4 M [nBu3N][PF6] in THF; scan rate 100 mV/s. E1/2 for a
reversible process. Epa for an irreversible oxidation. Epc for an
c
d
irreversible reduction.
Similar to dimanganese complex 3, the CV of complex 8 has
oxidative features, although in this case they are both
irreversible (see Supporting Information). The shift in
potentials of these two oxidations (Epa = −0.44 V and −0.23
V) with respect to 3one to more positive potential and one
to more negative potentialreflect the asymmetric coordina-
tion environments of the two Mn centers in this complex.
Diiron complex 9, again, displays more complex redox behavior
in its cyclic voltammogram, with a reversible oxidation at −1.01
V and several irreversible reductive features at Epc = −2.15 V
and −3.03 V (Figure 9). The negative shift in oxidation
potential with respect to 4 can be attributed to the more
complex 3 displayed two oxidative events,76 including a
reversible oxidation at E1/2 = −0.58 V and an irreversible
oxidation at Epa = 0.05 V (Figure 9). In this fully high spin
system, these oxidations must correspond to removal of
electrons from metal−metal and metal−ligand antibonding
orbitals.
In contrast, the CV of the analogous diiron complex 4
displays much richer redox properties with two quasi-reversible
oxidative events at Epa = −0.74 V and Epa = −0.24 V, as well as
two reductive events (the additional oxidative feature that
appears at slightly more negative potential was shown to be an
artifact that is not present in the absence of scans to negative
potentials, see Supporting Information). The first reduction
appears at E1/2 = −1.98 V and is fully reversible, while the
i
electron-donating Pr substituents on the phosphine donors,
also leading to more negative reduction potentials.
Theoretical Investigations. To gain further insight into
the metal−metal interactions in these homobimetallic com-
plexes, and to assess the effects of inequivalent coordination
spheres on the orbital interactions between two otherwise
identical metal centers, a computational investigation was
undertaken using density functional theory (DFT, Gaussian
second reductive feature is an irreversible reduction at Epc
=
−2.64 V (Figure 9). In this case, reduction involves the addition
of electrons to metal−metal bonding orbitals with little metal−
ligand antibonding character. The CV of MesNPiPr2-linked
diiron complex 6 also displays both oxidative and reductive
Figure 9. Cyclic voltammograms of 3, 4, and 9 (∼2 mM analyte in 0.4 M [nBu4N][PF6] in THF, scan rate: 100 mV/s). Potentials are reported
versus the ferrocene/ferrocenium redox couple. Since reductive features were absent in the case of complex 3, only the oxidative portion of the cyclic
voltammogram is shown for clarity (a full CV is reported in the Supporting Information).
8233
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
09). Starting from crystallographically derived coordinates, the
geometries of 4, 6, and 9 were optimized to a minimum. The
frontier molecular orbital diagram of 4 is complicated by both
significant ligand-based contributions and the reduced
symmetry resulting from the fourth phosphinoamide ligand
bound to the apical position of the tris(phosphine)Fe center
(see Supporting Information). However, significant d orbital
mixing can be seen in the calculated frontier MO diagram and it
is clear from the computational results that the unpaired
electrons reside almost equally on both Fe centers (Mulliken
spin densities: FeN = 3.77; FeP = 3.45, Table 3).
two SOMOs shown in Figure 10 that appear to be Fe−Fe σ* in
character, the difference between them is that one is π-bonding
with respect to the amide-based p orbitals while the other is
Fe−N π-antibonding. The lowest energy, doubly occupied
orbital shown in Figure 10 is clearly an Fe−Fe σ bond formed
2
via overlap of two dz orbitals. While most of the frontier MOs
contain contributions from both metal centers, minimal π- and
δ-bonding contributions are present as a result of the disparate
energies of the d orbitals of the two Fe centers with different
ligand donor sets. Moreover, as previously reported for other
high spin diiron complexes, since metal−metal bonding and
antibonding orbitals are occupied to some extent, the maximum
bond order even in a symmetric complex would be 1.40−45
While the calculated frontier MO diagram of 9 is less
complicated by ligand contributions based on the absence of
aryl substituents, the C3-symmetric ligand framework is
disrupted as the two metals now have 2:1 combinations of
donor ligands (Figure 11). As a result of the mixed donor
environments in 9, the unpaired spin density resides almost
equally on the two Fe centers (FeN2P: 3.42; FeNP2: 3.48, Table
3). As in 6, more than 10 d-orbital-derived MOs are shown in
Figure 11 as a result of substantial orbital contributions to the
SOMOs. An Fe−Fe σ-bonding orbital is again seen as one of
the doubly occupied frontier MOs, and in this case several
orbital combinations with the proper symmetry for π-overlap
are visible.
Table 3. Calculated Wiberg Bond Indices (WBIs) and
Mulliken spin densities on the two Fe centers in compexes 4,
6, and 9
Mulliken spin density
complex
WBI
FeP
FeN
4
6
0.34
0.42
0.43
3.45
3.35
3.48
3.77
3.71
3.42
a
9
a
FeN here refers to the Fe center with two coordinated amides and
one phosphine donor while FeP refers to the Fe center with two
coordinated phosphines and one amide.
To gain more quantitative insight into the metal−metal
interactions in 4, 6, and 9, natural bond orbital (NBO)
calculations were performed, and the resulting Fe−Fe NBOs
are shown in Figure 12. For complexes 4 and 6 only σ-bonding
NBOs were found. In both cases, the NBOs had stronger
contributions from the tris(amido)Fe center, but this was more
The calculated frontier molecular orbital diagram of the more
symmetric halide-bound complex 6 is more easily interpreted
(Figure 10). First, it should be noted that more than 10 largely
metal-based molecular orbitals are shown in Figure 10. This is a
result of substantial ligand contributions to several of the singly
occupied molecular orbitals (SOMOs). For example, there are
Figure 10. Calculated frontier molecular orbital diagram of complex 6 (BP86/LANL2TZ(f)/6-311+G(d)/D95 V).
8234
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
Figure 11. Calculated frontier molecular orbital diagram of complex 9 (BP86/LANL2TZ(f)/6-311+G(d)/D95 V).
Figure 12. Pictorial representations of the Fe−Fe NBOs of 4, 6, and 9. Composition of Fe−Fe NBO of 4: 32.3% FeP (75.8% d, 16.6% p, 7.6% s);
67.7% FeN (91.2% d, 7.9% s). Composition of Fe−Fe NBO of 6: 45.3% FeP (69.6% d, 16.6% p, 13.8% s); 54.7% FeN (93.0% d, 5.9% s). Composition
of Fe−Fe NBOs of 9: σ - 50.9% FeP (68.1% d, 30.7% s); 49.1% FeN (48.2% d, 41.0% p, 10.8% s). π - 52.2% FeP (86.1% d, 9.8% p, 4.1% s); 47.8% FeN
(84.2% d, 14.5% p).
pronounced in the Fe−Fe NBO of complex 4, likely a result of
competing interactions with the additional phosphinoamide
ligand. Since one metal center contributes substantially more
electron density to the metal−metal bond, this interaction can
be seen as dative in nature, with the electron-rich phosphine-
ligated Fe center donating electron density to the amide-ligated
Fe center. The dative character in this interaction, however, is
certainly less pronounced than in a complex composed of two
different metal centers, such as an early/late heterobimetallic
complex. The tris(amido)Fe contribution to the Fe−Fe σ-
bonding NBO in both cases is largely composed of d character,
while the tris(phosphine)Fe atom’s contribution to bonding
contains substantial s and p character as well. The Wiberg bond
index (WBI) calculated for the Fe−Fe bond in complex 4 is
0.34 (compared to ∼0.6 for the Fe−P dative bonds), suggestive
of a bond order significantly lower than 1 (Table 3). This is
consistent with the rather long Fe−Fe distance (2.8684(6) Å)
observed in the solid state structure of 4. For the Fe−Fe bond
in 6, the calculated WBI is slightly higher (0.42), in accordance
with the shorter Fe−Fe distance observed for 6 (2.5855(4) Å).
The origin of this difference is, again, competing interactions
with the apical phosphinoamide donor in 4.
In the case of complex 9, both σ- and π-bonding NBOs were
found. In both of these orbitals, the contributions from each Fe
center to the metal−metal bonding NBOs were essentially
equivalent as a result of more similar coordination environ-
ments in this complex. As expected, the π-bonding NBO is
largely composed of d orbital character; however, the Fe−Fe σ-
bonding NBO has substantial s character (30.7%) from the
bis(phosphine)-ligated Fe center and substantial p character
8235
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
Table 4. Crystallographic Data and Refinement Parameters for 3, 4, 5 and 6
3·0.5 Et2O
4
5
6·pentane
chemical formula
fw
C62H73Mn2N4O0.5P4
1116.04
120
C60H68Fe2N4P4
1081.81
120
C49H83Cl1Mn2N3O1P3
968.45
C50H87Cl1Fe2N3P3
970.33
T (K)
120
120
λ (Å)
0.71073
12.6984(5)
44.4400(13)
11.7591(3)
90
0.71073
12.6193(9)
44.242(3)
11.7062(7)
90
0.71073
8.6509(4)
26.9140(11)
21.7536(9)
90
0.71073
a (Å)
12.3024(4)
14.4580(4)
16.5435(5)
101.105(2)
91.933(2)
112.297(1)
2653.02(14)
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
space group
Z
116.967(1)
90
117.149(3)
90
91.192(2)
90
5914.3(3)
P21/c
5815.6(7)
P21/c
5063.8(4)
P21/c
P1
̅
4
4
4
2
Dcalc (g/cm3)
μ (cm−1
1.253
1.234
1.270
1.215
)
5.77
6.49
6.84
7.22
a
R1 (I > 2σ(I)), wR2 (all)
Rint
0.0778, 0.0829
0.034
0.0476, 0.0959
0.082
0.0336, 0.0836
0.053
0.0410, 0.1198
0.044
Nref(all), Nref(I > 2σ(I))
17214, 14249
12677, 8253
12814, 9411
15469, 11542
a
2
2
R1 = ∑||Fo| − |Fo||/∑|Fo|; wR2 = {∑[w(Fo − Fc2)2]/∑[w(Fo )2]}1/2
.
Table 5. Crystallographic Data and Refinement Parameters for 7, 8 and 9
7
8
9
chemical formula
fw
C34H58Fe2I2N2O1P2
938.30
C39H88.6Cl2LiMn2N3O3P3
928.36
C27H63Cl1Fe2N3P3
669.88
T (K)
120
120
120
λ (Å)
0.71073
0.71073
0.71073
a (Å)
9.3067(6)
10.8588(8)
21.1880(15)
104.534(3)
93.880(3)
106.983(3)
1959.2(2)
11.866(2)
12.897(3)
17.631(4)
90.101(10)
93.261(11)
107.517(10)
2568.4(9)
11.3057(4)
17.9804(7)
18.6204(7)
101.656(2)
91.054(2)
107.739(2)
3517.9(2)
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
space group
Z
P1
̅
2
P1
̅
2
P1
̅
4, Z′ = 2
1.265
Dcalc (g/cm3)
μ (cm−1
1.590
1.200
)
24.25
7.23
10.57
a
R1 (I > 2σ(I)), wR2 (all)
Rint
0.0195, 0.0465
0.027
0.0550, 0.1444
0.051
0.0472, 0.1191
0.070
Nref(all), Nref(I > 2σ(I))
11290, 9938
13083, 9279
13823, 9411
a
2
2
R1 = ∑||Fo| − |Fo||/∑|Fo|; wR2 = {∑[w(Fo − Fc2)2]/∑[w(Fo )2]}1/2
.
(41.0%) from the bis(amide)-ligated Fe center. Despite both σ
and π contributions to bonding, the WBI calculated for 9 is
nearly identical to that of 6 (0.43), which is in agreement with
their very similar Fe−Fe distances (2.6112(7) Å for 9).
dimanganese complexes, whereas the C3-symmetric diiron
complexes had Fe−Fe distances indicative of some degree of
orbital overlap. Computational studies confirmed the presence
of weak metal−metal bonding in these high spin systems.
Interestingly, magnetic studies revealed that while there were
negligible magnetic interactions between the metal centers in
the diiron complexes, all of the dimanganese complexes
exhibited antiferromagnetic superexchange. Examination of
the redox behavior of 3−9 using cyclic voltammetry reveals
that all of the bimetallic complexes are prone to undergo two
sequential one-electron oxidations at mild potentials, and the
diiron complexes 4 and 9 also have readily accessible reductive
processes. Future studies will focus on systematic investigations
of these redox processes and the further reactivity of complexes
3−9.
CONCLUSION
■
In summary, we have successfully isolated and structurally
characterized a series of homobimetallic manganese and iron
phosphinoamides complexes. The complexes formed via this
one-pot synthetic route varied with the identity of the metal
halide precursor and the phosphorus and nitrogen substituents.
Structural and magnetic characterization of bimetallic com-
plexes 3−9 revealed that the metal centers remain in a high
spin MII state in these complexes, and Mossbauer spectroscopy
̈
of the diiron derivatives confirms the assignment of high spin
Fe(II) for both metal centers. The solid state molecular
structures revealed relatively long intermetallic distances in the
8236
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
electron correlation functional84 (BP86). For open shell systems,
unrestricted wave functions were used in energy calculations. A mixed-
basis set was employed, using the LANL2TZ(f) triple-ζ basis set with
effective core potentials for iron,85−87 Gaussian09’s internal 6-
311+G(d) for atoms bonded directly to the metal centers (nitrogen
and phosphorus), and Gaussian09’s internal LANL2DZ basis set
(equivalent to D95 V88) for carbon and hydrogen. Starting with
crystallographically determined geometries as a starting point, when
available, the geometries were optimized to a minimum, followed by
analytical frequency calculations to confirm that no imaginary
frequencies were present. NBO analysis was performed using NBO
3.1,89 as implemented by Gaussian09. Pictorial representations of
molecular orbitals were generated using Jimp 2.90−92
EXPERIMENTAL SECTION
■
General Considerations. Unless specified otherwise, all manip-
ulations were performed under an inert atmosphere using standard
Schlenk or glovebox techniques. Glassware was oven-dried before use.
Benzene, pentane, diethyl ether, tetrahydrofuran, and toluene were
dried using a Glass Contours drying column. All solvents were stored
over 3 Å molecular sieves. Benzene-d6 and toluene-d8 (Cambridge
Isotopes) were degassed via repeated freeze−pump−thaw cycles, and
dried over 3 Å molecular sieves. THF-d8 was dried over CaH2,
vacuum-transferred, and degassed via repeated freeze−pump−thaw
cycles. Ph2PNHiPr,54,78 (iPr)2PNHMes,26 and (iPr)2PNHiPr,26 were
synthesized using literature procedures. Anhydrous MnCl2, FeCl2, and
FeI2 were purchased from Strem Chemicals and used after 12 h drying
at 100 °C/1 Torr. NMR spectra were recorded at ambient
temperature unless otherwise stated on a Varian Inova 400 MHz
i
[iPrNKPPh2]2 (1). A solution of PrNHPPh2 (243 mg, 1.00 mmol)
in THF (3 mL) was cooled to −32 °C and added to KH (41 mg, 1.0
mmol) in THF (2 mL). The resulting mixture was stirred at room
temperature for 2 h at which point the remaining KH was removed via
filtration through Celite. All volatiles were removed in vacuo, and the
crude product was washed with pentane (3 × 5 mL) to obtain 1 as an
analytically pure pale yellow crystalline solid (250 mg, 89%). 1H NMR
(400 MHz, THF-d8): δ 7.13 (dd, JHP = 1.2 Hz, JHH = 5.2 Hz, 4H, Ar-
H), 6.80 (t, JHH = 7.2 Hz, 4H, Ar-H), 6.69 (t, JHH = 7.2 Hz, 2H, Ar-H),
instrument. Chemical shifts are reported in δ (ppm). For H and 13C
1
NMR spectra the solvent resonance was referenced as an internal
standard, and for 31P{1H} NMR spectra the 85% H3PO4 resonance
was referenced as an external standard. IR spectra were recorded on a
Varian 640-IR spectrometer controlled by Resolutions Pro software.
UV−vis spectra were recorded on a Cary 50 UV−vis spectropho-
tometer using Cary WinUV software. Elemental analyses were
performed at Complete Analysis Laboratory Inc., Parsippany, NJ.
Electrochemistry. Cyclic voltammetry measurements were carried
out in a glovebox under a dinitrogen atmosphere in a one-
compartment cell using a CH Instruments electrochemical analyzer.
A glassy carbon electrode and platinum wire were used as the working
and auxiliary electrodes, respectively. The reference electrode was Ag/
AgNO3 in THF. Solutions (THF) of electrolyte (0.40 M [nBu4N]-
[PF6]) and analyte (2 mM) were also prepared in the glovebox. All
potentials are reported versus the ferrocene/ferrocenium couple.
X-ray Structure Determinations. All operations were performed
on a Bruker-Nonius Kappa Apex2 diffractometer, using graphite-
monochromated MoKα radiation. All diffractometer manipulations,
including data collection, integration, scaling, and absorption
corrections were carried out using the Bruker Apex2 software.79
Preliminary cell constants were obtained from three sets of 12 frames.
Crystallographic parameters are provided in Tables 4 and 5 and further
experimental crystallographic details are described for each compound
in the Supporting Information.
3.08 (dsept, JHP = 20.8 Hz, JHH = 6.0 Hz, 1H, iPr−CH), 0.54 (d, JHH
=
i
6.0 Hz, 6H, Pr−CH3). 13C{1H} NMR (100.53 MHz, THF-d8): δ
154.3 (d, JCP = 40.9 Hz), 132.8 (d, JCP = 10.8 Hz), 127.7 (d, JCP = 3.6
i
i
Hz), 125.8 (s), 51.2 (s, Pr-CH), 30.8 (d, JCP = 8.5 Hz, Pr-CH3).
31P{1H} NMR (161.84 MHz, THF-d8): δ 46.4. Anal. Calcd for
C15H17KNP: C, 64.03; H, 6.09; N, 4.98. Found: C, 63.91; H, 6.07; N,
4.89.
[MesNKPiPr2]2 (2). A solution of MesNHPiPr2 (251 mg, 1.00
mmol) in THF (3 mL) was cooled to −32 °C and added to KH (41
mg, 1.0 mmol) in THF (2 mL). The resulting mixture was stirred at
room temperature for 2 h, at which point the remaining KH was
removed via filtration through Celite. All volatiles were removed in
vacuo, and the crude product was washed with pentane (3 × 5 mL) to
obtain 2 as an analytically pure colorless crystalline solid (260 mg,
1
92%). H NMR (400 MHz, THF-d8): δ 6.43 (s, 2H, Ar-H), 2.25 (s,
6H, Mes-CH3), 1.99 (s, 3H, Mes-CH3), 1.30 (sept, JHH = 6.8 Hz, 2H,
iPr−CH), 1.04 (overlapped doublets, JHH = 7.6 Hz, JHH = 8.0 Hz, 6H,
iPr−CH3), 0.89 (d, JHH = 7.2 Hz, 3H, iPr−CH3), 0.86 (d, JHH = 7.2 Hz,
3H, iPr−CH3). 13C{1H} NMR (100.53 MHz, THF-d8): δ 160.8, 130.5,
129.8, 119.1, 30.7 (d, JCP = 80.2 Hz, iPr-CH), 26.5 (s, Mes-CH3), 23.1
(d, JCP = 42.3 Hz, iPr-CH), 21.0 (s, Mes-CH3), 20.5 (d, JCP = 19.8 Hz,
Mossbauer Spectroscopy. Iron-57 Mossbauer spectra were
̈
measured on a constant acceleration spectrometer (SEE Co,
Minneapolis, MN) with a Janis SVT-100 cryostat. Isomer shifts are
quoted relative to α-Fe foil (<25 μm thick) at room temperature. The
Fe foil standard spectrum has linewidths Γ (fwhm) of 0.292 and 0.326
mm/s for the doublets within the 4 mm/s window when measured
outside the cryostat at room temperature. Samples of 4, 6, 7, and 9
were prepared using approximately 30 mg of sample suspended in
paratone-N oil. Data were analyzed using a package written by E. R.
King and modified by E. V. Eames in Igor Pro (Wavemetrics) using a
simple model consisting of Lorentzian lineshapes with optional
asymmetry.
Magnetic Measurements. Solution magnetic moments were
measured using Evans’ method.80,81 Crystalline samples of 4, 5, and 8
were loaded into gel caps in a dinitrogen-filled glovebox and then
sealed with Eicosane wax. The samples were transported in sealed vials
under nitrogen, and the waxed capsules were taken directly from the
vial, placed in a mounting straw, and the straw loaded into a Quantum
Design MPMS measurement system. The sample chamber was
immediately evacuated, and filled with He gas. Magnetic susceptibility
data were collected as a function of temperature in a 1 kOe applied
field from 1.8 to 310 K. Magnetization as a function of field was also
collected from 0 kOe to 5 kOe at 1.8 K. Several data points were also
collected as the field was reduced to zero to check for hysteresis
effects; none were observed (see Supporting Information).
i
iPr-CH3), 18.7 (d, JCP = 9.9 Hz, Pr-CH3). 31P{1H} NMR (161.84
MHz, THF-d8): δ 72.3. Anal. Calcd for C15H25KNP: C, 62.25; H, 8.71;
N, 4.84. Found: C, 62.21; H, 8.73; N, 4.79.
[Mn(iPrNPPh2)3Mn(iPrNPPh2)] (3). A solution of MnCl2 (63 mg,
0.50 mmol) in THF (3 mL) was cooled to −32 °C and to this a THF
i
solution (3 mL) of PrNKPPh2 (281 mg, 1.00 mmol) was added
dropwise over a period of 5−10 min. The reaction mixture was
gradually warmed to room temperature and continuously stirred for 12
h. The volatiles were subsequently removed in vacuo, and the
remaining crude reddish brown materials were extracted with toluene
(3 × 2 mL) and filtered through Celite to remove KCl and other
insoluble byproducts. The volatiles were removed from the filtrate
under vacuum to obtain analytically pure 3 as a yellow crystalline solid
(230 mg, 83%). X-ray quality single crystals were grown by careful
layering of pentane onto a concentrated THF solution of 3 at room
temperature. UV−vis (THF) λmax, nm (ε, M−1 cm−1): 495 (517), 611
(219). Evans’ method (THF-d8): 10.38 μB. Anal. Calcd for
C60H68N4P4Mn2: C, 66.79; H, 6.35; N, 5.19. Found: C, 66.70; H,
6.37; N, 5.21.
[Fe(iPrNPPh2)3Fe(iPrNPPh2)] (4). A solution of FeCl2 (63 mg,
0.50 mmol) in THF (3 mL) was cooled to −32 °C and to this a THF
i
solution (3 mL) of PrNKPPh2 (281 mg, 1.00 mmol) was added
Computational Details. All calculations were performed using
Gaussian09-E.0182 for the Linux operating system. Density functional
theory calculations were carried out using a combination of Becke’s
1988 gradient-corrected exchange functional83 and Perdew’s 1986
dropwise over the period of 5 min. The reaction mixture was gradually
warmed to room temperature and continuously stirred for 12 h. The
volatiles were subsequently removed in vacuo, and the crude purple-
brown materials were extracted with Et2O (3 × 2 mL) and filtered
8237
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
i
through Celite. Cooling the concentrated Et2O solution to −32 °C to
[Fe(NiPrPiPr2)2(PiPr2NiPr)FeCl] (9). A solution of PrNHPiPr2
(175 mg, 1.00 mmol) in THF (3 mL) was cooled to −32 °C and
to this nBuLi (0.63 mL, 1.6 M in hexanes, 1.0 mmol) was added
dropwise over 10 min. The resulting pale yellow solution was warmed
1
afforded analytically pure purple blocks of 4 (160 mg, 59.3%). H
NMR (400 MHz, C6D6): δ 53.7, 31.2, 12.9, 11.5, 10.2, 3.3, 1.1, −3.2,
−20.5, −30.9 (bs). UV−vis (C6H6) λmax, nm (ε, M−1 cm−1): 541
(323). Evans’ method (C6D6): 9.26 μB. Anal. Calcd for
C60H68N4P4Fe2: C, 66.68; H, 6.34; N, 5.18. Found: C, 66.78; H,
6.25; N, 5.10.
i
to room temperature and stirred for 2 h to form PrNLiPPh2 in situ.
The resulting mixture was added to a THF solution (3 mL) of FeCl2
(85 mg, 0.67 mmol) at −32 °C, and the reaction mixture was gradually
warmed to room temperature and continuously stirred for 12 h.
During this time period, the solution became homogeneous and
brown. All volatiles were subsequently removed in vacuo, and the
crude red materials were extracted with ether (3 × 2 mL) and filtered
through Celite. The ether solution was again dried in vacuo to obtain
analytically pure 9 as a yellow crystalline solid (150 mg, 67.3%).1H
NMR (400 MHz, C6D6): δ 85.4, 75.2, 64.4, 40.5, 29.9, 4.2, 1.8, −4.7,
−38.5 (bs). UV−vis (C6H6) λmax, nm (ε, M−1 cm−1): 470 (615).
Evans’ method (C6D6): 8.95 μB. Anal. Calcd for C27H63ClN3P3Fe2: C,
48.41; H, 9.48; N, 6.27. Found: C, 48.33; H, 9.36; N, 6.14.
[(THF)Mn(μ-Cl)(MesNPiPr2)3Mn(MesNPiPr2)] (5). A solution of
MnCl2 (84 mg, 0.67 mmol) in THF (3 mL) was cooled to −32 °C
and to this a THF solution (3 mL) of MesNKPiPr2 (289 mg, 1.00
mmol) was added dropwise over the period of 5 min. The resulting
mixture was stirred at room temperature for 12 h, at which time the
insoluble materials were removed via filtration through Celite. The
volatiles were subsequently removed in vacuo, and the crude product
was extracted with toluene and filtered through Celite to remove KCl
and other insoluble byproducts. Volatiles were removed from the
filtrate under vacuum to obtain analytically pure 5 as a red-orange
crystalline solid (230 mg, 71%). X-ray quality single crystals were
grown from a THF/pentane mixture of 5 at room temperature. UV−
vis (THF) λmax, nm (ε, M−1 cm−1): 500 (383), 634 (253), 1059 (400).
Evans’ method (THF-d8 ): 6.91 μB . Anal. Calcd for
C49H83ClN3OP3Mn2: C, 60.77; H, 8.64; N, 4.34. Found: C, 60.73;
H, 8.53; N, 4.29.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional spectral, cyclic voltammetry, magnetic and crystallo-
graphic data for complexes 1−9 (in CIF format). This material
is available free of charge via the Internet at http://pubs.acs.org.
[Fe(MesNPiPr2)3FeCl] (6). A solution of FeCl2 (85 mg, 0.67
mmol) in THF (3 mL) was cooled to −32 °C and to this a THF
solution (3 mL) of MesNKPiPr2 (289 mg, 1.00 mmol) was added
dropwise over a period of 5 min. The resulting mixture was stirred at
room temperature for 12 h, at which point the insoluble materials were
removed via filtration. The volatiles were subsequently removed in
vacuo, and the crude product was extracted with toluene and filtered
through Celite to remove KCl and other insoluble byproducts.
Volatiles were removed from the filtrate under vacuum to obtain
analytically pure 6 as a pale yellow crystalline solid (170 mg, 57%). 1H
NMR (400 MHz, C6D6): δ 43.3, 32.6, 29.8, 26.7, 16.6, 3.3. UV−vis
(THF) λmax, nm (ε, M−1 cm−1): 490 (102), 668 (46). Evans’ method
(THF-d8): 6.80 μB. Anal. Calcd for C45H75ClN3P3Fe2: C, 60.18; H,
8.42; N, 4.68. Found: C, 59.94; H, 8.37; N, 4.53.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: thomasc@brandeis.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This material is based upon work supported by the Department
of Energy under Award No. DE-SC0004019. C.M.T. is grateful
for a 2011 Sloan Research Fellowship. The authors also thank
Brandeis University for initially funding this project, Prof.
Theodore Betley for access to a Mossbauer spectroscopy
facility, and Prof. Christopher Landee for insightful discussions
related to magnetic data.
[(THF)Fe(μ-I)(MesNPiPr2)2FeI] (7). A solution of FeI2 (310 mg,
1.00 mmol) in THF (3 mL) was cooled to −32 °C and to this a cold
THF solution (3 mL) of MesNKPiPr2 (289 mg, 1.00 mmol) was
added dropwise over a period of 5 min. The resulting mixture was
stirred at room temperature for 12 h, at which point the insoluble
materials were removed via filtration through Celite. The volatiles were
subsequently removed in vacuo, and the crude product was extracted
with toluene and filtered through Celite to remove KI and other
insoluble byproducts. The volatiles were removed from the filtrate
under vacuum to obtain analytically pure 7 as a brown crystalline solid
̈
REFERENCES
■
(1) Adams, R. D.; Cotton, F. A. Catalysis by Di- and Polynuclear Metal
Cluster Complexes; Wiley-VCH: New York, NY, 1998.
(2) Tsui, E. Y.; Kanady, J. S.; Day, M. W.; Agapie, T. Chem. Commun.
2011, 47, 4189−4191.
(3) Zhao, Q.; Harris, T. D.; Betley, T. A. J. Am. Chem. Soc. 2011, 133,
8293−8306.
1
(340 mg, 74%). H NMR (400 MHz, C6D6): δ 58.7, 53.7, 40.9, 31.5,
30.8, 30.0, 3.3, 1.3, 1.1, −0.5, −15.3 (bs). UV−vis (C6H6) λmax, nm (ε,
M−1 cm−1): 511 (819). Evans’ method (C6D6): 8.53 μB. Anal. Calcd
for C34H58N2I2OP2Fe2: C, 43.52; H, 6.23; N, 2.99. Found: C, 43.48;
H, 6.32; N, 3.05.
(4) Zhao, Q.; Betley, T. A. Angew. Chem., Int. Ed. 2011, 50, 709−712.
(5) Powers, T. M.; Fout, A. R.; Zheng, S.-L.; Betley, T. A. J. Am.
Chem. Soc. 2011, 133, 3336−3338.
(THF)3LiCl[Mn(NiPrPiPr2)2(PiPr2NiPr)MnCl] (8). A solution of
iPrNHPiPr2 (175 mg, 1.00 mmol) in THF (3 mL) was cooled to −32
(6) Eames, E. V.; Harris, T. D.; Betley, T. A. Chem. Sci. 2012, 3, 407−
n
°C and to this BuLi (0.63 mL, 1.6 M in hexanes, 1.0 mmol) was
415.
added dropwise over 10 min. The resulting pale yellow solution was
(7) Harris, T. D.; Zhao, Q.; Sanchez, R. H.; Betley, T. A. Chem.
Commun. 2011, 47, 6344−6346.
(8) Kanady, J. S.; Tsui, E. Y.; Day, M. W.; Agapie, T. Science 2011,
333, 733−736.
(9) Ragsdale, S. W. J. Inorg. Biochem. 2007, 101, 1657−1666.
(10) Lindahl, P. A.; Graham, D. E. Acetyl-coenzyme A Synthases and
Nickel-Containing Carbon Monoxide Dehydrogenases. In Nickel and
Its Surprising Impact in Nature; John Wiley & Sons, Ltd: Hoboken, NJ,
2007; pp 357−415.
(11) Kim, E.; Chufan, E. E.; Kamaraj, K.; Karlin, K. D. Chem. Rev.
2004, 104, 1077−1134.
(12) Dismukes, G. C. Chem. Rev. 1996, 96, 2909−2926.
(13) Friedle, S.; Reisner, E.; Lippard, S. J. Chem. Soc. Rev. 2010, 39,
2768−2779.
i
warmed to room temperature and stirred for 2 h to form PrNLiPPh2
in situ. The resulting mixture was added to a THF solution (3 mL) of
MnCl2 (84 mg, 0.67 mmol) at −32 °C, and the reaction mixture was
gradually warmed to room temperature and continuously stirred for 12
h. During this time period, the solution became homogeneous and
turned red-orange. All volatiles were subsequently removed in vacuo,
and the crude materials were dissolved in THF and layered with
pentane at room temperature. Red-orange blocks of 8 were obtained
after 12 h (176 mg, 56%). UV−vis (THF) λmax, nm (ε, M−1 cm−1):
500 (440), 642 (400), 1063 (340). Evans’ method (THF-d8): 6.87 μB.
Because of both extreme air and moisture sensitivity and the labile
solvent molecules on the Li ion, repeated combustion analysis data of
8 was unsatisfactory.
8238
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
(14) Merkx, M.; Kopp, D. A.; Sazinsky, M. H.; Blazyk, J. L.; Muller,
(53) Thapa, I.; Gambarotta, S.; Korobkov, I.; Duchateau, R.;
Kulangara, S. V.; Chevalier, R. Organometallics 2010, 29, 4080−4089.
(54) Poetschke, N.; Nieger, M.; Khan, M. A.; Niecke, E.; Ashby, M.
T. Inorg. Chem. 1997, 36, 4087−4093.
̈
J.; Lippard, S. J. Angew. Chem., Int. Ed. 2001, 40, 2782−2807.
(15) Tard, C.; Pickett, C. J. Chem. Rev. 2009, 109, 2245−2274.
(16) Fontecilla-Camps, J. C.; Volbeda, A.; Cavazza, C.; Nicolet, Y.
Chem. Rev. 2007, 107, 4273−4303.
(17) Cotton, F. A.; Murillo, C. A.; Walton, R. A. Multiple Bonds
Between Metal Atoms; Springer Science and Business Media, Inc.: New
York, 2005.
(55) Lu, C. C.; Peters, J. C. Inorg. Chem. 2006, 45, 8597−8607.
(56) Pauling, L. The Nature of the Chemical Bond, 3rd ed.; Cornell
University Press: Ithaca, NY, 1960.
(57) For completeness we include the van der Waals radii of Mn
(2.05 A) and Fe (2.04 A): Hu., S.-Z.; Zhou, Z.-H.; Robertson, B. E. Z.
Kristallogr. 2009, 224, 375−383.
(58) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 4538−
4539.
(18) Collman, J. P.; Boulatov, R. Angew. Chem., Int. Ed. 2002, 41,
3948−3961.
(19) Chisholm, M. H. Acc. Chem. Res. 1990, 23, 419−425.
(20) Gade, L. H. Angew. Chem., Int. Ed. 2000, 39, 2658−2678.
(21) Wheatley, N.; Kalck, P. Chem. Rev. 1999, 99, 3379−3420.
(22) Stephan, D. W. Coord. Chem. Rev. 1989, 95, 41−107.
(23) Bullock, R. M.; Casey, C. P. Acc. Chem. Res. 1987, 20, 167−173.
(24) Cooper, B. G.; Napoline, J. W.; Thomas, C. M. Catal. Rev. - Sci.
Eng. 2012, 54, 1−40.
(59) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 6252−
6254.
(60) Brown, S. D.; Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2002,
125, 322−323.
(61) Betley, T. A.; Peters, J. C. Inorg. Chem. 2003, 42, 5074−5084.
(62) Whited, M. T.; Rivard, E.; Peters, J. C. Chem. Commun. 2006,
1613−1615.
(25) Thomas, C. M. Comments Inorg. Chem. 2011, 32, 14−38.
(26) Greenwood, B. P.; Forman, S. I.; Rowe, G. T.; Chen, C.-H.;
Foxman, B. M.; Thomas, C. M. Inorg. Chem. 2009, 48, 6251−6260.
(27) Greenwood, B. P.; Rowe, G. T.; Chen, C.-H.; Foxman, B. M.;
Thomas, C. M. J. Am. Chem. Soc. 2010, 132, 44−45.
(28) Krogman, J. P.; Foxman, B. M.; Thomas, C. M. J. Am. Chem. Soc.
2011, 133, 14582−14585.
(29) Thomas, C. M.; Napoline, J. W.; Rowe, G. T.; Foxman, B. M.
Chem. Commun. 2010, 46, 5790−5792.
(30) Zhou, W.; Napoline, J. W.; Thomas, C. M. Eur. J. Inorg. Chem.
2011, 2011, 2029−2033.
(63) Chomitz, W. A.; Mickenberg, S. F.; Arnold, J. Inorg. Chem. 2007,
47, 373−380.
(64) Adhikari, D.; Basuli, F.; Fan, H.; Huffman, J. C.; Pink, M.;
Mindiola, D. J. Inorg. Chem. 2008, 47, 4439−4441.
(65) Buschhorn, D.; Pink, M.; Fan, H.; Caulton, K. G. Inorg. Chem.
2008, 47, 5129−5135.
(66) Fryzuk, M. D.; Leznoff, D. B.; Ma, E. S. F.; Rettig, S. J.; Young,
V. G. Organometallics 1998, 17, 2313−2323.
(67) Schunemann, V.; Paulsen, H. Mossbauer Spectroscopy. In
Applications of Physical Methods to Inorganic and Bioinorganic
Chemistry; Scott, R. A., Lukehart, C. M., Eds.; John Wiley & Sons,
Ltd.: Hoboken, NJ, 2007; pp 241−269.
(68) Evans, D. J.; Hughes, D. L.; Silver, J. Inorg. Chem. 1997, 36,
747−748.
(69) Sanakis, Y.; Power, P. P.; Stubna, A.; Munck, E. Inorg. Chem.
2002, 41, 2690−2696.
(70) MacDonnell, F. M.; Ruhlandt-Senge, K.; Ellison, J. J.; Holm, R.
H.; Power, P. P. Inorg. Chem. 1995, 34, 1815−1822.
(71) Andres, H.; Bominaar, E. L.; Smith, J. M.; Eckert, N. A.;
Holland, P. L.; Munck, E. J. Am. Chem. Soc. 2002, 124, 3012−3025.
(72) Hendrich, M. P.; Gunderson, W.; Behan, R. K.; Green, M. T.;
Mehn, M. P.; Betley, T. A.; Lu, C. C.; Peters, J. C. Proc. Natl. Acad. Sci.
U.S.A. 2006, 103, 17107−17112.
(73) Repeated measurements were very similar. Alternative fits that
either included this impurity or attempted to fit to doublets of 50%
intensity were unsatisfactory. See Supporting Information for more
details.
(31) Simona, M. Coord. Chem. Rev. 2009, 253, 1793−1832.
(32) Rosenthal, J.; Bachman, J.; Dempsey, J. L.; Esswein, A. J.; Gray,
T. G.; Hodgkiss, J. M.; Manke, D. R.; Luckett, T. D.; Pistorio, B. J.;
Veige, A. S.; Nocera, D. G. Coord. Chem. Rev. 2005, 249, 1316−1326.
(33) Gade, L. H. Angew. Chem., Int. Ed. Engl. 1996, 35, 2089−2090.
(34) Whittlesey, B. R. Coord. Chem. Rev. 2000, 206−207, 395−418.
(35) Xu, Z.; Lin, Z. Chem.Eur. J. 1998, 4, 28−32.
(36) Lindahl, P. A. J. Inorg. Biochem. 2012, 106, 172−178.
(37) Nguyen, T.; Merrill, W. A.; Ni, C.; Lei, H.; Fettinger, J. C.; Ellis,
B. D.; Long, G. J.; Brynda, M.; Power, P. P. Angew. Chem., Int. Ed.
2008, 47, 9115−9117.
(38) Hess, C. R.; Weyhermuller, T.; Bill, E.; Wieghardt, K. Angew.
̈
Chem., Int. Ed. 2009, 48, 3703−3706.
(39) Klose, A.; Solari, E.; Floriani, C.; Chiesi-Villa, A.; Rizzoli, C.; Re,
N. J. Am. Chem. Soc. 1994, 116, 9123−9135.
(40) Cotton, F. A.; Daniels, L. M.; Falvello, L. R.; Matonic, J. H.;
Murillo, C. A. Inorg. Chim. Acta 1997, 256, 269−275.
(41) Cotton, F. A.; Daniels, L. M.; Matonic, J. H.; Murillo, C. A.
Inorg. Chim. Acta 1997, 256, 277−282.
(74) All of the compounds reported herein are highly air sensitive,
and were transported and handled outside of an inert atmosphere to
(42) Cotton, F. A.; Feng, X.; Murillo, C. A. Inorg. Chim. Acta 1997,
256, 303−308.
collect Mossbauer and SQUID data. Although precautionary measures
̈
(43) Zall, C. M.; Zherebetskyy, D.; Dzubak, A. L.; Bill, E.; Gagliardi,
L.; Lu, C. C. Inorg. Chem. 2012, 51, 728−736.
were taken to prevent exposure to air, it is likely that decomposition
occurred to some extent, giving rise to the impurities observed using
these techniques.
(44) Cotton, F. A.; Daniels, L. M.; Murillo, C. A. Inorg. Chim. Acta
1994, 224, 5−9.
(75) Repeated measurements gave similar results, indicating that the
solution magnetic behavior of these compounds may differ from that
in the solid state.
(76) The resting potential was measured to be more negative than
−0.6 V, allowing both redox events in the CV of 3 to be assigned as
oxidative in nature.
(45) Timmer, G. H.; Berry, J. F. C.R. Chim. 2012, 15, 192−201.
(46) Fout, A. R.; Zhao, Q.; Xiao, D. J.; Betley, T. A. J. Am. Chem. Soc.
2011, 133, 16750−16753.
(47) Lu, D.-Y.; Yu, J.-S. K.; Kuo, T.-S.; Lee, G.-H.; Wang, Y.; Tsai, Y.-
C. Angew. Chem., Int. Ed. 2011, 50, 7611−7615.
(48) Ashley, A. E.; Cooper, R. T.; Wildgoose, G. G.; Green, J. C.;
O’Hare, D. J. Am. Chem. Soc. 2008, 130, 15662−15677.
(49) Chai, J.; Zhu, H.; Stuckl, A. C.; Roesky, H. W.; Magull, J.;
Bencini, A.; Caneschi, A.; Gatteschi, D. J. Am. Chem. Soc. 2005, 127,
9201−9206.
(77) Thoreson, K. A.; Follett, A. D.; McNeill, K. Inorg. Chem. 2010,
49, 3942−3949.
(78) Sisler, H.; Smith, N. J. Org. Chem. 1961, 26, 611−613.
(79) Apex 2: Version 2 User Manual, M86-E01078; Bruker Analytical
X-ray Systems: Madison, WI, 2006.
(50) Hatnean, J. A.; Raturi, R.; Lefebvre, J.; Leznoff, D. B.; Lawes, G.;
Johnson, S. A. J. Am. Chem. Soc. 2006, 128, 14992−14999.
(51) Kuppuswamy, S.; Cooper, B. G.; Bezpalko, M. W.; Foxman, B.
M.; Powers, T. M.; Thomas, C. M. Inorg. Chem. 2012, 51, 1866−1873.
(52) Fenske, D.; Maczek, B.; Maczek, K. Z. Anorg. Allg. Chem. 1997,
623, 1113−1120.
(80) Evans, D. F. J. Chem. Soc. 1959, 2003−2005.
(81) Sur, S. K. J. Magn. Reson. 1989, 82, 169−173.
(82) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Braone, V.; Mennucci,
B.; Petersson, G. A. et al. Gaussian 09, Revision A. 1; Gaussian, Inc.:
Wallingford, CT, 2009.
8239
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240

Inorganic Chemistry
Article
(83) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100.
(84) Perdew, J. P. Phys. Rev. B 1986, 33, 8822−8824.
(85) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299−310.
(86) Roy, L. E.; Hay, P. J.; Martin, R. L. J. Chem. Theory Comput.
2008, 4, 1029−1031.
(87) Ehlers, A. W.; Bohme, M.; Dapprich, S.; Gobbi, A.; Hollwarth,
A.; Jonas, V.; Kohler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G.
Chem. Phys. Lett. 1993, 208, 111−114.
(88) Dunning, T. H.; Hay, P. J. In Modern Theoretical Chemistry;
Schaefer, H. F., Ed.; Plenum: New York, 1976; Vol. 3, pp 1−28.
(89) Glendening, E. D.; Reed, A. E.; Carpenter, J. E. NBO Version 3.1.
(90) Hall, M. B. F., R. F. Inorg. Chem. 1972, 11, 768−779.
(91) Bursten, B. E.; Jensen, J. R.; Fenske, R. F. J. Chem. Phys. 1978,
68, 3320.
(92) Manson, J.; Webster, C. E.; Perez, L. M.; Hall, M. B. http://
www.chem.tamu.edu/jimp2/index.html.
8240
dx.doi.org/10.1021/ic300776y | Inorg. Chem. 2012, 51, 8225−8240