Synthesis and reaction of [{HC(CMeNAr)2}Mn]2 (Ar = 2,6-iPr2C6H3): The complex containing three-coordinate manganese(I) with a Mn-Mn bond exhibiting unusual magnetic properties and electronic structure

Article abstract of DOI:10.1021/ja042269e

This paper reports on the synthesis, X-ray structure, magnetic properties, and DFT calculations of [{HC(CMeNAr)₂)}Mn]₂ (Ar = 2,6-iPr₂C₆H₃) (2), the first complex with three-coordinate manganese(I). Re

Full text of DOI:10.1021/ja042269e

Published on Web 06/01/2005
Synthesis and Reaction of [{HC(CMeNAr)2}Mn]2 (Ar )
2,6-iPr2C6H3): The Complex Containing Three-Coordinate
Manganese(I) with a Mn-Mn Bond Exhibiting Unusual
Magnetic Properties and Electronic Structure
Jianfang Chai,† Hongping Zhu,† A. Claudia Stu¨ckl,† Herbert W. Roesky,*,†
Jo¨rg Magull,† Alessandro Bencini,‡ Andrea Caneschi,‡ and Dante Gatteschi‡
Contribution from the Institut fu¨r Anorganische Chemie der UniVersita¨t Go¨ttingen,
Tammannstrasse 4, D-37077 Go¨ttingen, Germany, and Dipartimento di Chimica and INSTM
Research Unit, Polo Scientifico dell’UniVersita` di Firenze, Via della Lastruccia 3, I-50019,
Sesto Fiorentino, Italy
Received December 23, 2004; E-mail: hroesky@gwdg.de
Abstract: This paper reports on the synthesis, X-ray structure, magnetic properties, and DFT calculations
of [{HC(CMeNAr)2}Mn]2 (Ar ) 2,6-iPr2C6H3) (2), the first complex with three-coordinate manganese(I).
Reduction of the iodide [{HC(CMeNAr)2}Mn(µ-I)]2 (1) with Na/K in toluene afforded 2 as dark-red crystals.
2+
The molecule of 2 contains a Mn2 core with a Mn-Mn bond. The magnetic investigations show a rare
example of a high-spin manganese(I) complex with an antiferromagnetic interaction between the two Mn(I)
centers. The DFT calculations indicate a strong s-s interaction of the two Mn(I) ions with the open shell
configuration (3d54s1). This suggests that the magnetic behavior of 2 could be correctly described as the
5
coupling between two S1 ) S2 ) /2 spin centers. The Mn-Mn bond energy is estimated at 44 kcal mol-1
by first principle calculations with the B3LYP functional. The further oxidative reaction of 2 with KMnO4 or
O2 resulted in the formation of manganese(III) oxide [{HC(CMeNAr)2}Mn(µ-O)]2 (3). Compound 3 shows
an antiferromagnetic coupling between the two oxo-bridged manganese(III) centers by magnetic measure-
ments.
Introduction
with manganese(I) (or lower oxidation states) are known to be
important due to their participation in photochemistry.5 In the
Transition metal complexes containing low-valent and low-
coordinate metal centers are rare. This is partly due to the
difficulty associated with such metal ions acquiring the stable
16 or 18 valence electrons in the metal orbitals. The employment
of sterically hindered organic ligands can somewhat kinetically
stabilize transition metal centers in low coordination; however,
the reactivity studies of these compounds are not well investi-
gated. Recent publications have shown that the unusual orbital
structure of these complexes resulting from the low-coordinate
metal centers would render their reactivity very interesting.1 It
has been reported that reduction of the three-coordinate iron-
(II) chloride {HC(CtBuNAr)2}FeCl (Ar ) 2,6-iPr2C6H3) leads
to the formation of the iron(I) dinitrogen complex [{HC-
(CtBuNAr)2}FeN]2. Upon further reduction of this compound
with potassium, [K{HC(CtBuNAr)2}FeN]2 was obtained con-
taining a [FeNNFe]0 core.2 Other transition metal compounds
with low-valent and low-coordinate metal centers have been
synthesized, for example, for Co, Ni, and Cu,1,3 and the
reactivity studies of Cu(I) compounds are notable.4 Complexes
course of our investigation, the great majority of manganese(I)
(and lower oxidation states) complexes are reported for carbonyl
compounds. Complexes with low-valent manganese incorporat-
ing noncarbonyl ligands are not abundant,6 although the
chemistry of these species is quite interesting.7 Correspondingly,
a few complexes with three-coordinate, even two-coordinate,
manganese centers are known; however, these compounds have
oxidation states of +2 or higher in metal centers.8 In this regard,
(3) (a) Holland, P. L.; Tolman, W. B. J. Am. Chem. Soc. 1999, 121, 7270-
7271. (b) Holland, P. L.; Tolman, W. B. J. Am. Chem. Soc. 2000, 122,
6331-6332. (c) Randall, D. W.; George, S. D.; Holland, P. L.; Hedman,
B.; Hodgson, K. O.; Tolman, W. B.; Solomon, E. I. J. Am. Chem. Soc.
2000, 122, 11632-11642. (d) Jazdzewski, B. A.; Holland, P. L.; Pink, M.;
Young, V. G., Jr.; Spencer, D. J. E.; Tolman, W. B. Inorg. Chem. 2001,
40, 6097-6107.
(4) (a) Spencer, D. J. E.; Aboelella, N. W.; Reynolds, A. M.; Holland, P. L.;
Tolman, W. B. J. Am. Chem. Soc. 2002, 124, 2108-2109. (b) Aboelella,
N. W.; Lewis, E. A.; Reynolds, A. M.; Brennessel, W. W.; Cramer, C. J.;
Tolman, W. B. J. Am. Chem. Soc. 2002, 124, 10660-10661. (c) Aboelella,
N. W.; Kryatov, S. V.; Gherman, B. F.; Brennessel, W. W.; Young, V. G.,
Jr.; Sarangi, R.; Rybak-Akimova, E. V.; Hodgson, K. O.; Halman, B.;
Soloman, E. I.; Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 2004,
126, 16896-16911. (d) Lewis, E. A.; Tolman, W. B. Chem. ReV. 2004,
104, 1047-1076.
† Universita¨t Go¨ttingen.
(5) Roundhill, D. M. Photochemistry and Photophysics of Metal Complexes;
Plenum Press: New York, 1994.
‡ Universita` di Firenze.
(1) Holland, P. L.; Cundari, T. R.; Perez, L. L.; Eckert, N. A.; Lachicotte, R.
J. J. Am. Chem. Soc. 2002, 124, 14416-14424.
(2) Smith, J. M.; Lachicotte, R. J.; Pittard, K. A.; Cundari, T. R.; Lukat-Rodgers,
G.; Rodgers, K. R.; Holland, P. L. J. Am. Chem. Soc. 2001, 123, 9222-
9223.
(6) (a) Reardon, D.; Aharonian, G.; Gambarotta, S.; Yap, G. P. A. Organo-
metallics 2002, 21, 786-788. (b) Gibson, V. C.; Segal, J. A.; White, A. J.
P.; Williams, D. J. J. Am. Chem. Soc. 2000, 122, 7120-7121. (c) Bourget-
Merle, L.; Lappert, M. F.; Severn, J. R. Chem. ReV. 2002, 102, 3031-
3066.
9
10.1021/ja042269e CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 9201-9206
9201

A R T I C L E S
Chai et al.
Scheme 1. Preparation of Compound 2
following the recent success that low-valent and/or low-
coordinate metal centers are stabilized by the sterically bulky
â-diketiminato ligands, it is of great interest to assemble
complexes with low-valent and low-coordinate manganese by
using such ligands and, meanwhile, to explore their reactivity.
Herein, we report on the synthesis and X-ray structure of [{HC-
(CMeNAr)2}Mn]2 (2), the first example of a three-coordinate
manganese(I) compound. The magnetic properties and electronic
structure (by DFT calculations) of 2 are also investigated, as
well as its oxidative reaction to manganese(III) oxide [{HC-
(CMeNAr)2}Mn(µ-O)]2 (3).
Figure 1. ORTEP drawing for complex 2 (30% probability ellipsoids).
Table 1. Crystallographic Data for Complexes 2 and 3
2
3
formula
Fw
C
58H82Mn2N4
C
58H82Mn2N4O2
945.16
977.16
Results and Discussion
T (K)
133(2)
133(2)
cryst. syst.
space group
a (Å)
b (Å)
c (Å)
monoclinic
P2(1)/c
14.4602(9)
14.2263(6)
26.795(2)
94.999(6)
5491.1(6)
4
1.143
0.498
2032
3.24 to 49.62
38085
monoclinic
C2/c
Synthesis, Structure, and Reaction. The reduction of the
bis(µ-iodo)dimanganese(II) complex [{HC(CMeNAr)2}Mn(µ-
I)]2 (1)9 with a Na/K alloy in toluene at room temperature affords
[{HC(CMeNAr)2}Mn]2 (2) as dark-red crystals in about 15%
yield. Compound 2 is highly air-sensitive, and decomposition
occurs, even when it is kept in solid state at argon atmosphere.
It is thermally stable and melts at 154-156 °C. The most intense
peak in the EI mass spectrum of 2 appeared at m/z 472,
corresponding to half of the molecule. The signal at m/z 944
(5%) was assigned to the molecular ion [M+]. Single crystals
of 2 suitable for X-ray structural analysis were obtained by
keeping the toluene solution of 2 at 4 °C for 1 week.
Crystallographic data are given in Table 1 and selected bond
lengths and angles in Table 2. The molecular structure of 2
15.4921(12)
16.3958(9)
21.3688(17)
90.802(6)
5427.3(7)
4
1.196
0.509
2096
3.62 to 49.54
18243
â (°)
V (Å3)
Z
d
calcd (g cm-3
)
µ (mm-1
F(000)
)
2θ range (°)
No. of rflns collected
No. of ind rflns
9401 (R(int) ) 0.1276) 4641 (R(int) ) 0.0637)
No. of data/restraints/params 9401/0/597
4641/0/308
1.024
0.0421, 0.0873
0.0664, 0.0962
0.325 to -0.257
goodness of fit, F2
R1, wR2 [I > 2σ(I)]
R1, wR2 (all data)
0.953
0.0414, 0.0781
0.0716, 0.0825
0.496 to -0.567
largest diff peak (e Å-3
)
2+
reveals a dimer containing a central Mn2 core with a Mn-
Mn bond (Figure 1). Each Mn atom is three-coordinate by two
N atoms of the L ligand and one adjacent Mn atom. The sum
of the bond angles around each Mn averages to 359.4°,
indicating the trigonal planar coordination geometry of the
respective manganese center. The two Mn-Mn atoms are each
terminally chelated by L ligands, forming two six-membered
C3N2Mn rings in approximately an orthogonal array (dihedral
angle, 80.6°). In the C3N2Mn ring, the C3N2 part is planar and
a Mn atom is significantly out of this plane (av 0.53 Å). The
Mn-N bond lengths (av 2.10 Å) are slightly longer than those
observed in 1 (av 2.07 Å), and the N-Mn-N angle (89.82°)
in 2 is more acute than that in 1 (av 94.59°).9 The Mn-Mn
bond length in 2 (2.72 Å) is significantly shorter than that in
the Mn(0) compound Mn2(CO)10 (2.90 Å)10 and can be
compared with that in the Mn(I) complex Mn2(CO)7(µ-S2) (2.67
Å),11 whereas similar Mn-Mn separations are found in the
alkyl-bridged Mn(II) compounds [(PhCMe2CH2)Mn(µ-CH2-
Table 2. Selected Bond Lengths (Å) and Bond Angles (°) for
Compounds 2 and 3
Compound 2
Mn(1)-N(1)
Mn(1)-N(2)
Mn(1)-Mn(2)
Mn(2)-N(3)
Mn(2)-N(4)
2.097(2)
2.101(2)
2.721(1)
2.116(2)
2.094(2)
N(1)-Mn(1)-N(2)
N(1)-Mn(1)-Mn(2)
N(2)-Mn(1)-Mn(2)
N(3)-Mn(2)-N(4)
N(3)-Mn(2)-Mn(1)
89.72(9)
139.96(6)
129.67(6)
89.92(8)
141.26(6)
Compound 3
Mn(1)-N(1)
Mn(1)-N(2)
Mn(1)-O(1)
Mn(1)-O(1A)
Mn(1)-Mn(1A)
2.013(2)
2.013(2)
1.809(2)
1.827(2)
2.659(1)
N(1)-Mn(1)-N(2)
N(1)-Mn(1)-O(1)
N(2)-Mn(1)-O(1)
O(1)-Mn(1)-O(1A)
N(1)-Mn(1)-O(1A)
90.28(8)
96.64(8)
157.90(9)
85.99(7)
156.26(8)
CMe2Ph)]2 (2.72 Å)12 and [{HC(CMeNAr)2}Mn(µ-Me)]2 (2.81
Å).3b However, these Mn-Mn bond distances are shorter than
2+
those observed for the “naked” Mn2 (3.4 Å) and Mn2 (3.06
Å) species.14
(7) Treichel, P. M. In ComprehensiVe Organometallic Chemistry; Wilkinson,
G., Stone, F. G. A., Abel, E. W., Eds.; Elsevier: Oxford, UK, 1982.
(8) (a) Ellison, J. J.; Power, P. P.; Shoner, S. C. J. Am. Chem. Soc. 1989, 111,
8044-8046. (b) Wehmschulte, R. J.; Power, P. P. Organometallics 1995,
14, 3264-3267. (c) Panda, A.; Stender, M.; Wright, R. J.; Olmstead, M.
M.; Klavins, P.; Power, P. P. Inorg. Chem. 2002, 41, 3909-3916. (d) Chai,
J.; Zhu, H.; Fan, H.; Roesky, H. W.; Magull, J. Organometallics 2004, 23,
1177-1179.
(9) Chai, J.; Zhu, H.; Most, K.; Roesky, H. W.; Vidovic, D.; Schmidt, H.-G.;
Noltemeyer, M. Eur. J. Inorg. Chem. 2003, 4332-4337.
(10) Bianchi, R.; Gervasio, G.; Marabello, D. Inorg. Chem. 2000, 39, 2360-
2366 and references therein.
In view of the rich chemistry of low-valent Mn-Mn
complexes, such as Mn2(CO)10 and Mn2(CO)7(µ-S2),10,11 it is
of great interest to explore the reactivity of 2. Here, we examined
(11) (a) Adams, R. D.; Kwon, O.-S.; Smith, M. D. Inorg. Chem. 2001, 40, 5322-
5323. (b) Adams, R. D.; Kwon, O.-S.; Smith, M. D. Inorg. Chem. 2002,
41, 5525-5529.
(12) Andersen, R. A.; Carmona-Guzman, E.; Gibson, J. F.; Wilkinson, G. J.
Chem. Soc., Dalton Trans. 1976, 2204-2211.
9
9202 J. AM. CHEM. SOC. VOL. 127, NO. 25, 2005

Synthesis and Reaction of [HC(CMeNAr)2Mn]2
A R T I C L E S
Scheme 2. Formation of 3 by Oxidation of 2
its reactivity toward the oxidative reagents KMnO4 and molec-
ular O2 in an attempt to approach manganese oxides, which are
believed to be involved in many natural processes such as water
oxidation and O2 evolution by means of the photosystem II.14
After stirring a mixture of 2 and excess KMnO4 in toluene at
room temperature for 2 days, a dimeric manganese oxide [{HC-
(CMeNAr)2}Mn(µ-O)]2 (3) was smoothly afforded as red
crystals in high yield. The alternative reaction of 2 with predried
dioxygen gave the same result. Compound 3 is thermally stable.
It melts at 213-215 °C, and the EI-MS spectrum gives the most
intense peak at m/z 976 for the molecular ion of 3.The molecular
structure of 3 is shown in Figure 2. Crystallographic data and
selected bond lengths and angles are given in Tables 1 and 2,
respectively. This compound contains two peripheral six-
membered C3N2Mn rings and one central four-membered Mn2O2
ring. The dihedral angle between the six- and the four-membered
rings is 39.9°. Each manganese center is tetrahedrally coordi-
nated. The Mn-N distances (av 2.01 Å) are a slightly shorter
than those in 2, while the N-Mn-N angle (90.28(8)°) is close
to that in 2. The Mn-O distances in 3 average to 1.82 Å and
fall in the normal range (1.79-1.86 Å).15 The Mn-Mn distance
in 3 is 2.66 Å and is the shortest separation among those bis-
(µ-oxo)dimanganese(III) centers. It is a little shorter compared
to that in 2, indicating considerable contacts within the Mn2O2
core.
Figure 2. ORTEP drawing for complex 3 (50% probability ellipsoids).
Figure 3. Temperature dependence of the magnetic susceptibility of 2.
Magnetic Properties and Electronic Structure of 2. The
temperature dependence of the magnetic susceptibility of 2 in
the temperature range of 2-290 K is shown in Figure 3. The
room temperature effective magnetic moment (µeff ) 3.93 µB)
suggests that the manganese(I) ions are in a high spin state
95 GHz, respectively, showed a number of signals disappearing
at temperatures lower than 10 K (Figures S2a and S2b, in the
Supporting Information) and are indicative of an antiferromag-
netic behavior, contrary to the ferromagnetism observed for
7
derived either from the free ion (3d54s1) configuration, S, or
2+
Mn2 isolated in rare gas matrixes.17
from the 3d6 electronic configuration, 5D. However, it is much
smaller than that expected for two noninteracting manganese-
(I) centers, µeff ) 9.8 µB for 7S and µeff ) 6.9 µB for 5D states,
respectively. This may indicate that an antiferromagnetic
interaction between the metal centers with Curie temperature
(TC) higher than 300 K is operative, as confirmed by the fact
that ø smoothly decreases in the temperature range from 300
to 20 K without going through a maximum. Below 10 K, the ø
increases versus T, again, indicating the presence of a small
paramagnetic impurity. As a matter of fact, 2 was found to be
highly air-sensitive, and although particular care was taken in
the preparation of the sample, some decomposition was likely
to occur. In the temperature range of 10-20 K, the ø versus T
curve reaches a minimum significantly different from zero. This
implies that a sizable temperature-independent paramagnetism
is present together with some paramagnetic impurity. EPR
spectra recorded with freshly filtered single crystals at 9.4 and
Direct information for the value of the spins Si of the
manganese(I) ions in 2 cannot be obtained since we are not
able to measure susceptibility values at temperatures higher than
TC. However, it is worth mentioning that the ø versus T curve
could be reproduced with any value of 3 g Si g 2, as shown
by the calculations. The magnetic data alone are not sufficient
enough to unambiguously determine the exact spin of the
interacting centers; therefore, quantum chemical calculations of
the electronic structure of 2 using density functional theory
(DFT)18 were performed as described in the Experimental
Section.
It is well-known that the calculation of the magnetic structure
of polynuclear transition metal complexes is a delicate quantum
chemical problem challenging computational and theoretical
chemists for many years.19 This is because the computed
energies fall in the range of kT and are, therefore, strongly
influenced by small effects, such as the electron correlation.
(13) Selected samples: (a) Seder, T. A.; Church, S. P.; Weitz, E. J. Am. Chem.
Soc. 1986, 108, 1084-1086. (b) Seder, T. A.; Church, S. P.; Weitz, E. J.
Am. Chem. Soc. 1986, 108, 7518-7524.
(16) (a) Kitajima, N.; Singh, U. P.; Amagai, H.; Osawa, M.; Morooka, Y. J.
Am. Chem. Soc. 1991, 113, 7757-7758. (b) Goodson, P. A.; Oki, A. R.;
Glerup, J.; Hodgson, D. J. J. Am. Chem. Soc. 1990, 112, 6248-6254. (c)
Goodson, P. A.; Hodgson, D. J. Inorg. Chem. 1989, 28, 3606-3608.
(17) Van Zee, R. J.; Weltner, W., Jr. J. Chem. Phys. 1988, 89, 4444-4446.
(18) ReViews of Modern Quantum Chemistry; Sen, K. D., Ed.; World Scien-
tific: Singapore, 2002.
(14) Nayak, S. K.; Rao, B. K.; Jena, P. J. Phys.: Condens. Matter 1998, 10,
10863-10878.
(15) (a) Hoganson, C. W.; Babcock, G. T. Science 1997, 277, 1953-1956. (b)
Stubbe, J.; van der Donk, W. A. Chem. ReV. 1998, 98, 705-762 and
references therein.
9
J. AM. CHEM. SOC. VOL. 127, NO. 25, 2005 9203

A R T I C L E S
Chai et al.
The description of the lowest spin state of the systems, in
particular, requires the use of open-shell multiconfigurational
wave functions that can be correctly handled only with
multireference configuration interaction approaches.20 These
approaches become, however, unusable for large molecular
systems, due to the extremely large number of configurations
required, and formalisms have been developed to approximately
include these effects into DFT,21 which is a single determinant-
based theory. These efforts result in DFT being the most widely
used tool for the interpretation of magnetic properties due to
its high accuracy/CPU cost ratio. The use of the B3LYP
functional and the broken symmetry formalism was found to
provide results that are generally, at least, in semiquantitative
agreement with the experimental data. Furthermore, the magnetic
behavior of molecules is strongly influenced by modelization
of the chemical systems,22 which is unavoidable in multicon-
figurational calculations in order to reach a reasonable compu-
tational time. Again, within DFT, it is possible to use the real
chemical system in the calculations of the magnetic coupling
constants.
Figure 4. The s bond orbital of 2 (left: the R spin orbital; right: the â
spin orbital).
Table 3. The Mn-Mn Natural Bond Orbitals
spin
occupancy
%
Mn(1)
%
Mn(2)
R
â
0.965
0.965
56.5
43.6
4s0.964p0.033d0.04
4s0.914p0.023d0.09
43.5
56.4
4s0.904p0.023d0.10
4s0.964p0.033d0.04
densities are 4.91, with opposite sign on each manganese atom.
The Mn-Mn NBOs are computed for R and â spins with 0.96
occupancy. Their composition is shown in Table 3. The Mn-
Mn bond is, therefore, mainly formed by the 4s orbitals of the
manganese atoms with a small contribution of 3d electrons. A
crude estimate of the Mn-Mn bonding energy can be obtained
by subtracting from the total energy of the molecule those of
the fragments. This procedure gave a bonding energy of 44 kcal
mol-1 (B3LYP), which can compare with that calculated for
Single-point calculations were performed using the molecular
coordination obtained by the X-ray single-crystal structure to
compute the magnetic exchange coupling constant (J) and to
characterize the bonding interactions. The Heisenberg-Dirac-
vanVleck spin Hamiltonian in the form H ) J(S1‚S2) was used,
with a positive J value determining a singlet ground state. The
magnetic exchange coupling constant was computed by using
the broken symmetry (BS) approach.19 The BS calculations were
performed imposing mirror symmetry to the R and â electrons
on the two Mn atoms, and the resulting wave function has,
therefore, MS ) 0. Due to the mirror symmetry imposed on the
spin densities, the BS molecular orbital can be spatially localized
onto different parts of the molecule and is assumed to be a good
description of the natural magnetic orbitals, with the extent of
the localization depending on the overlap between the orbitals.19d
In the SOMO region, the BS wave function of 2 is characterized
by a series of occupied 3d spin-orbitals almost centered onto
the manganese atoms and rather well localized on the two
different halves of the molecule according to their R/â oc-
cupancy (R orbitals being localized on the Mn(1) atom). The
HOMO, on the contrary, which lies above the SOMO’s, is
delocalized onto the two metal centers, as shown in Figure 4.
The R and â components have almost the same spatial extent,
which shows the formation of a metal-metal σ bond. The
Mulliken population analysis assigns a spin population (spin
density) of 4.9 to each Mn center. Transforming the computed
molecular orbitals into natural bond orbitals (NBO)23,24 allowed
a natural population analysis (NPA) that assigned to each
Mn2(CO)10 (38 or 35 kcal mol-1 25
) and is much lower than that
found for the H2 molecule (104 kcal mol-1).26 The binding
energies computed for Mn2 (82 kcal mol-1) and Mn22+ (70 kcal
mol-1) are also larger than that computed for 2.
The above description of the electronic structure of 2 suggests
that its magnetic behavior could be properly described as the
5
coupling between two S1 ) S2 ) /2 spin centers, and the 4s
electrons are involved in forming a metal-metal single bond.
The magnetic electrons are mainly localized in the metal 3d
atomic orbitals. This result could not be easily anticipated from
simple considerations based on the electronic structure of the
free manganese(I) ion, as usually done for rationalizing the
magnetic behavior of transition metal dimers,25 due to the
“quenching” of the magnetic contribution of the 4s electrons
when the Mn-Mn bond is formed. The exchange coupling
constant (J) can thus be conveniently computed from the energy
difference between the S ) 10 high spin state and the BS state
as 110 and 160 cm-1 using the B3LYP and BP functionals,
respectively.
The best fitting of the magnetic data of 2 was obtained with
the procedure described in the Experimental Section. The
presence of some S ) 5/2 paramagnetic impurity was considered
in order to reproduce the low-temperature tail of the ø versus T
manganese center the electronic configuration (4s1.023d5.144p0.01
)
with a natural charge of 0.83, in agreement with the formal
oxidation state (+1) of the manganese ion. The natural spin
(24) We mention here that the description of bond properties and other molecular
properties based on MO pictures and population analysis is subject to several
limitations due to the arbitrariness in the choice of the basis functions and
of the scheme for partitioning the electronic density in atomic regions. For
a critical comparison of some population analyses, one can refer to: Guerra,
C. F.; Handgraaf, J.-W.; Baerends, E. J.; Bickelhaupt, F. M. J. Comput.
Chem. 2003, 25, 189-210. It must also be stressed that population analysis
is currently used in interpreting polarized neutron diffraction experiments
with semiquantitative agreement with the experimental data (see for
example: Gillon, B.; Mathoniere, C.; Ruiz, E.; Alvarez, S.; Cousson, A.;
Rajendiran, T. M.; Kahn, O. J. Am. Chem. Soc. 2002, 124, 14433-14441.
(25) Rosa, A.; Ehlers, A.; Baerends, E. J.; Snijders, J. G.; te Velde, G. J. Phys.
Chem. 1996, 100, 5690-5696.
(19) (a) Noodleman, L.; Norman, J. G., Jr. J. Chem. Phys. 1979, 70, 4903-
4906. (b) Noodleman, L. J. Chem. Phys. 1981, 74, 5737-5743. (c)
Noodleman, L.; Davidson, E. R. Chem. Phys. 1986, 109, 131-143. (d)
Ciofini, I.; Daul, C. A.; Bencini, A. Modeling Molecular Magnetism. In
Recent AdVances in DFT Methods; Barone, V., Bencini, A., Fantucci, P.,
Eds.; World Scientific: Singapore, 2002.
(20) Miralles, J.; Castell, O.; Caballol, R.; Malrieu, J. P. Chem. Phys. 1993,
172, 33.
(21) Illas, F.; Moreira, I.; De, P. R.; de Graaf, C.; Barone, V. Theor. Chem.
Acc. 2000, 104, 265.
(22) Adamo, C.; Barone, V.; Bencini, A.; Totti, F.; Ciofini, I. Inorg. Chem.
1999, 38, 1996-2004.
(23) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88, 899-926.
(26) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry, 5th ed.; John
Wiley & Sons Inc.: New York, 1988.
9
9204 J. AM. CHEM. SOC. VOL. 127, NO. 25, 2005

Synthesis and Reaction of [HC(CMeNAr)2Mn]2
A R T I C L E S
filtration, the solution was concentrated to ca. 10 mL. Dark-red crystals
were obtained after 1 week at 4 °C. Yield: 0.07 g, 15%. Mp 154-156
°C. Anal. Calcd for C58H82Mn2N4 (945.16): C, 73.65; H, 8.68; N, 5.93.
Found: C, 73.62; H, 8.67; N, 5.74. EI-MS: m/z (%) 944 (5) [M]+,
472 (100) [1/2M]+. IR (KBr, Nujol mull, cm-1): ν˜ 1698 (w), 1654
(w), 1624 (w), 1577 (w), 1555 (w), 1261 (s), 1092 (s), 1019 (s), 937
(w), 866 (w), 799 (s), 762 (w), 722 (w), 667 (w), 614 (w), 568 (w).
curve. The best fitting parameters are J ) 109.8(2) and F )
0.031(4)% (F being the percentage of paramagnetic impurity
assumed to have a spin S ) S1). The Zeeman g factor and the
temperature-independent paramagnetism were kept constant at
g ) 2.00 and TIP ) 830 × 10-6 cm3 mol-1 27
. The agreement
factor R (R ) ∑i|øi° - øic|/∑iøi°) was 0.88%. The computed J
value is in good agreement with the experimental one, and with
the B3LYP results in better quantitative agreement as often
found in the literature. Equivalent fits of the magnetic data were
obtained for S1 ) S2 ) 3 (J ) 109.9(2), F ) 0.023(4)%, TIP )
830 × 10-6 cm3 mol-1, R ) 0.88%) and S1 ) S2 ) 2 (J )
109.7(2), F ) 0.045(4)%, TIP ) 830 × 10-6 cm3 mol-1, R )
0.95%).
[{HC(CMeNAr)2}Mn(µ-O)]2 (3). Route a: KMnO4 (0.5 g, 3.2
mmol) was added to a solution of 2 (0.2 g, 0.2 mmol) in toluene (20
mL) at room temperature. The mixture was stirred at room temperature
for 2 days. Unreacted KMnO4 was removed by filtration. The solvent
was removed in vacuum, and the residue was extracted with diethyl
ether. Red crystals were obtained after 4 days at 4 °C. Yield: 0.14 g,
72%. Mp 213-215 °C. Anal. Calcd for C58H82Mn2N4O2 (977.16): C,
71.23; H, 5.94; N, 5.73. Found: C, 70.92; H, 5.67; N, 5.54. EI-MS:
m/z (%) 976 (100) [M]+. IR (KBr, Nujol mull, cm-1): ν˜ 1659 (w),
1623 (w), 1592 (w), 1552 (w), 1528 (w), 1261 (s), 1094 (s), 1025 (s),
936 (w), 919 (w), 842 (w), 801 (s), 761 (w), 720 (w), 699 (w), 668
(w), 607 (w), 514 (w), 467 (w). Route b: Dry O2 was introduced into
a solution of 2 (0.2 g, 0.2 mmol) in toluene (20 mL) at -78 °C instead
of N2. The solution was kept at this temperature for 2 h and slowly
warmed to room temperature and stirred for 12 h. All volatiles were
removed in vacuum, and the residue was extracted with diethyl ether.
Red crystals were obtained after 7 days at 4 °C. Yield: 0.12 g, 63%.
Magnetic Measurement of 3. The magnetic data of 3 were
recorded in the temperature range of 4-300 K. The magnetic
behavior is indicative of an antiferromagnetic coupling between
the two manganese(III) centers, S1 ) S2 ) 2 in a pseudo-
tetrahedral chromophore, mediated by the two oxo-bridges. The
µeff (4.7 µB) at room temperature is much lower than the spin-
only value expected for two noninteracting S ) 2 spins (µeff
)
6.9 µB). The µeff steadily decreases to ∼0 µB at about 24 K
(Figure S1). This behavior is consistent with the geometry and
the type of bridging atoms. The temperature dependence of the
magnetic susceptibility was fitted as described in the Experi-
mental Section. The best fit values are J ) 91.5(5), g ) 1.99-
(2), and R ) 0.98%.
X-ray Crystallography. Crystallographic data of complexes 2 and
3 were collected on a Stoe IPDS II-array detector system with graphite-
monochromated Mo KR radiation (λ ) 0.71073 Å). The structures were
solved by direct methods (SHELXS-97) and refined against F2 using
SHELXL-97.28 All non-hydrogen atoms were located by difference
Fourier synthesis and refined anisotropically. The hydrogen atoms were
included at geometrically calculated positions and refined using a riding
model.
Conclusions
In summary, we have prepared [{HC(CMeNAr)2}Mn]2 (Ar
) 2,6-iPr2C6H3) (2), the first complex with three-coordinate
manganese(I), by the Na/K alloy reduction of [{HC(CMeNAr)2}-
Computational Details. All of the calculations were performed in
the framework of the density functional theory (DFT) with the
NWChem4.6.29 Single-point calculations of the experimental X-ray
structure were performed using the B3LYP functional.30 Calculations
with the Becke-Perdew (BP) functional31,32 were also done for
comparison. All electron double-ú basis set proposed by Ahlrichs was
applied to all of the atoms except for Mn, which was treated with the
Ahlrichs triple-ú basis set.33
2+
Mn(µ-I)]2 (1). Compound 2 has a Mn2 core with a Mn-Mn
bond. The reaction of 2 with KMnO4 or O2 affords a Mn2O2
core compound [{HC(CMeNAr)2}Mn(µ-O)]2 (3). The latter
oxidation with O2 of the two Mn(I) centers involves a homolytic
cleavage of the Mn-Mn bond, indicating the interesting
reactivity of 2 for further investigations.
Population analysis based on natural atomic orbitals (NAO) and
natural bond orbitals (NBO)23 was performed using the NBO 5.0
program34 implemented in NWChem.
Experimental Section
General Considerations. All reactions were performed using
standard Schlenk and drybox techniques. Solvents were appropriately
dried and distilled under dinitrogen prior to use. Elemental analyses
were performed by the Analytisches Labor des Instituts fu¨r Anorga-
nische Chemie der Universita¨t Go¨ttingen. Mass spectra were obtained
on a Finnigan Mat 8230. IR spectra were recorded on a Bio-Rad Digilab
FTS-7 spectrometer as Nujol mulls between KBr plates. Magnetic
measurements were performed by using a Cryogenic S600 SQUID
magnetometer operating with an applied field of 1 Torr. The sample
was prepared in a glovebox by grinding freshly filtered crystals. The
powder was wrapped in a Teflon tape placed in a special sample holder
and placed inside the SQUID without any contact with air. Magnetic
susceptibilities were corrected for diamagnetism by using Pascal’s
constants. [{HC(CMeNAr)2}Mn(µ-I)]2 (1) was synthesized as previ-
ously reported in the literature.9
Calculations of the exchange coupling constant were performed using
the broken symmetry formalism19 with the working equation, J ) 2/25
[E(HS) - E(BS)], where E(HS) and E(BS) are the SCF energies of
the high spin (S ) 5) and broken symmetry (MS ) 0) states,
respectively. The ø versus T curve was fitted using the standard
procedure35 by minimizing the squares’ sum of the differences between
computed and measured molar susceptibilities with the MINUIT
program package.36 The best fit values are reported in the text.
(28) (a) Sheldrick, G. M. Acta Crystallogr. A 1990, 46, 467-473. (b) Sheldrick,
G. M. SHELXL-97, Program for Crystal Structure Refinement; Universita¨t
Go¨ttingen: Go¨ttingen, Germany, 1997.
(29) High-performance computational chemistry group, NWChem: A Compu-
tational Chemistry Package for Parallel Computers, version 4.6, Pacific
Northwest National Laboratory: Richland, WA, 2004.
(30) Becke, A. D. J. Chem. Phys. 1990, 98, 5648-5652.
[{HC(CMeNAr)2}Mn]2 (2). At room temperature, a suspension of
1 (0.60 g, 0.5 mmol) in toluene (30 mL) was added to a Na/K alloy
(Na 0.01 g, 0.5 mmol; K 0.04 g, 1 mmol). The mixture was stirred at
room temperature for 4 days, and a red solution was obtained. After
(31) Becke, A. D. Phys. ReV. A 1988, 38, 3098-3100.
(32) Perdew, J. P. Phys. ReV. B 1986, 33, 8822-8824.
(33) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571-2577.
(34) NBO 5.0. Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J.
E.; Bohmann, J. A.; Morales, C. M.; Weinhold, F. Theoretical Chemistry
Institute, University of Wisconsin: Madison, WI, 2001; http://www.chem-
.wisc.edu/-nbo5.
(27) Due to the high variance-covariance between F and TIP (0.87), we
performed a series of calculations for various values of TIP between 600
× 10-6 and 900 × 10-6 cm3 mol-1. The value shown in the text gave the
better agreement.
(35) (a) Kahn, O. Molecular Magnetism; VCH Publishers: New York, 1993.
(b) O’Connor, C. J. Prog. Inorg. Chem. 1982, 29, 203-283.
(36) James, F. MINUIT, version 94.1; CERN Program Library; CERN Geneva,
Switzerland.
9
J. AM. CHEM. SOC. VOL. 127, NO. 25, 2005 9205

A R T I C L E S
Chai et al.
Acknowledgment. We are grateful to the Fonds der Che-
mischen Industrie, and the Go¨ttinger Akademie der Wissen-
schaften for support of this work. FIRB and PRIN funds from
Italian MIUR are gratefully acknowledged.
temperature dependence of the magnetic susceptibility of 3
(Figure S1, PDF). The electronic spectra (EPR) of 2 (Figures
S2a and S2b, PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
Supporting Information Available: Details of the single-
crystal X-ray structure determination of 2 and 3 (CIF files). The
JA042269E
9
9206 J. AM. CHEM. SOC. VOL. 127, NO. 25, 2005