close

Вход

Забыли?

вход по аккаунту

?

Highly Resolved Spin-Density Distribution in the Prussian-Blue Precursors Cs2K[Fe(CN)6] and Cs2K[Mn(CN)6].

код для вставкиСкачать
Angewandte
Chemie
Spin-Density Distribution
Highly Resolved Spin-Density Distribution in the
Prussian-Blue Precursors Cs2K[Fe(CN)6] and
Cs2K[Mn(CN)6]**
Frank H. Khler* and Rodrigue Lescouzec
Understanding the properties of magnetic materials at the
atomic level is essential for the design of new materials. This is
particularly true for the chemistry-oriented area of moleculederived magnetic materials, because the macroscopic magnetic properties are the result of interactions between the spin
sources of the molecular building blocks. So far, the most
versatile and successful building blocks are paramagnetic
transition-metal complexes, and there is general agreement
that the magnetic interactions in the materials derived from
them is mediated by bridging ligands, that is, by superexchange.
The ions [Fe(CN)6]3 (S = 1/2) and [Mn(CN)6]3 (S = 1)
are prominent building blocks of magnetic materials. They
feature in Prussian-blue-type magnets,[1] whose general formula is AnM’x[M(CN)6]y(H2O)z (M, M’ = transition metal;
A = optional alkali ion) and whose transition temperature to
bulk magnetism can be adjusted by varying M and M’.[1c] Also,
both ions have recently been used for the assembly of
multiple-spin clusters,[2] ferromagnetic coordination polymers,[3] and “hybrid materials” exhibiting, for instance,
chirality or conducting properties in addition to magnetism.[4]
For all these compounds a fundamental question is: How do
the unpaired electrons get from [M(CN)6]3 to the adjacent
building blocks? The answer can be given based on the
fraction of an unpaired electron, also known as spin density, at
the atoms of the bridging CN ligand. This report demonstrates
that the spin delocalization within [Fe(CN)6]3 and
[Mn(CN)6]3 can be studied in great detail by using 13C and
15
N solid-state NMR spectroscopy.
The magic angle spinning (MAS) NMR spectra of microcrystalline samples of isotopically enriched [Fe(CN)6]3 and
[Mn(CN)6]3 are reproduced in Figures 1 and 2, respectively.
In them the 13C NMR signal patterns are shifted up to
8900 ppm, whereas the 15N NMR signal patterns appear up
to 2200 ppm at the opposite frequency side. Herzfeld–Berger
analysis[5] of the spinning sideband patterns gave the isotropic
signal shifts at the absolute temperature T, diso
T , which are
collected in Table 1. The NMR results are related to the spin
densities of the respective nucleus N, 1(N) (in atomic units),
through Equation (1).[6]
[*] Prof. Dr. F. H. K5hler, Dr. R. Lescou7zec
Chemie Department, Technische Universit9t M:nchen
85747 Garching (Germany)
Fax: (+ 49) 89-2891-3762
E-mail: f.h.koehler@lrz.tu-muenchen.de
[**] This work was supported by the DFG priority program “Molecular
Magnetism”.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2004, 43, 2571 –2573
Figure 1. MAS NMR spectra of Cs2K[Fe(CN)6] at 326.4 K. Rotational
frequencies of 15 kHz and 5 kHz for 13C and 15N, respectively. In each
pair of spectra the upper one has been simulated with the data given
in Table 1. The axial and equatorial isotropic signals are labeled with &
and *, respectively.
Figure 2. MAS NMR spectra of Cs2K[Mn(CN)6] at 326.4 K. Rotational
frequencies of 15 kHz and 8 kHz for 13C and 15N, respectively. In each
pair of spectra the upper one has been simulated with the data given
in Table 1. The (pseudo)axial and (pseudo)equatorial isotropic signals
are labeled with & and * (plus ~), respectively.
1ðNÞ ¼
9kTa30
dconðNÞ
m0 g2avb2eðSþ1Þ T
ð1Þ
Here k is the Boltzmann constant, m0 is the vacuum
permeability, gav is the mean electron g factor, be is the Bohr
magneton, a0 is the Bohr radius, S is the electron spin
DOI: 10.1002/anie.200453726
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2571
Communications
and CN bonding electrons
through spin at the M and N
atoms leads to negative spin den1 (axial)
sity at the C atoms, which overcompensates the small positive
13
diso
(
C)
3269
3295
7760
7709
7709
T
spin density resulting from the
intensity/width
1/1300
2/1300
1/2400
1/4000
1/4000
13
delocalization shown in Figure 3.
dcon
(
C)
3491
3431
8482
8231
8231
298
0.02583
0.02537
0.04705
0.04566
0.04566
1(13C)
In Cs2K[Fe(CN)6] the spin density
[c]
4022
3998
9223
9194
9194
dpara
298; xx
at
the six C atoms corresponds to
[c]
dpara
4096
4204
9532
9612
9612
298; yy
()15.3 % of the unpaired elecpara [c]
3067
3069
7188
7059
7059
d298; zz
tron, while ()27.8 % is found for
15
Cs2K[Mn(CN)6] (see Table 1). The
dcon
(
N)
727
704
896
992
1026
298
ratio of the spin densities is close to
intensity/width
1/400
2/400
1/300
1/300
1/300
15
dcon
(
N)
873.1
857.1
1039.2
1189.2
1226.5
what one expects for S = 1/2 and
298
1(15N)
0.00646
0.00639
0.00576
0.00660
0.00680
S = 1 species, respectively. Surpris[c]
dpara
370
333
883
756
727
298; xx
ingly, at the N atoms the spin
[c]
231
261
413
450
422
dpara
298; yy
density is virtually the same for
para [c]
2063
1978
1922
2289
2481
d298; zz
both compounds (3.8 % and 3.9 %,
[a] T = 326.4 K, d in ppm, width in Hz. [b] 1 in au. [c] j dzzdiso j j dyydiso j j dxxdiso j .
respectively), giving rise to speculations whether in Prussian blues
the efficiency of spin transfer
beyond the N atoms varies or not.
We
are
not
aware
of
any
experimental method providing
quantum number, and dcon
(N)is
the
Fermi
contact
shift
of
T
this sort of information. A polarized neutron diffraction study
nucleus N at the temperature T.
of Cs2K[Fe(CN)6][10] gave good results for the spin density at
The resolution of the spin densities at both the C and the
N atoms of the cyano groups is unusually high. It is possible to
iron, but it was less precise for the C and N atoms, claiming
distinguish differences of 5 A 105 au, which corresponds to a
()5 % and 6 % of an unpaired electron, respectively. So it is
no wonder that those numbers differ from ours. Early static
signal shift difference of 7 ppm in the MAS NMR spectrum of
solid-state 13C and 14N NMR studies gave only broad and
Cs2K[Fe(CN)6] in Figure 1. Therefore, in this compound even
a small distortion of the spin-density distribution toward axial
featureless signals,[11] and the 13C NMR signals Takeda et al.[12]
and equatorial ligand sites can be detected. For
obtained recently from static samples of cobalt and mangaCs2K[Mn(CN)6] the symmetry is still lower. This follows
nese hexacyanoferrates were very broad as well.
The analysis of the signal patterns in Figures 1 and 2
from the different spin densities at the N atoms (6.60 A 103 au
revealed considerable signal shift anisotropies, which pointed
and 6.80 A 103 au, respectively) of what would be equatorial
to nonspherical distributions of the spin densities about the C
CN ligands at lower resolution. Corresponding spin densities
and N atoms. Spectrum simulation showed that the spin
at the C atoms could not be resolved, although the greater
distribution about the these atoms has axial symmetry for
width of the 13C NMR signals of the equatorial CN groups
[Fe(CN)6]3 and nearly so for [Mn(CN)6]3. This applies to all
points to two overlapping signals (Table 1). More generally,
comparison of the spectra in Figures 1 and 2 shows that the
nonequivalent CN groups, while the individual values differ.
para
resolution of the 15N data is better than that of the 13C data.
Therefore, the principal shift values at 298 K, dpara
298; xx, d298; yy,
para
One would expect that the distortion of the spin-density
and d298; zz, given in Table 1 are due to the unpaired electrons.
distribution corresponds to the crystallographic nonequivaThere is a striking difference between the shift anisotropies:
para
para
lence of the CN groups of [M(CN)6]3. However, for
For the C atoms j dpara
298; zz j < j d298; xx j ffi j d298; yy j holds, whereas
3
[7]
para
para
[Fe(CN)6] conflicting results have been reported, while
for the N atoms we find j d298; zz j > j d298; xx j ffi j dpara
298; yy j . If the
for [Mn(CN)6]3 only axial symmetry has
shift anisotropies are visually represented by spheroids, this
means that they would be oblate and prolate at C and N,
been mentioned.[7d, 8] It follows that for
respectively. We suggest that the orientation of the spheroids
the title compounds the resolution of the
relative to the MCN axis is as shown in Figure 4. This
NMR data is better than that of the
would correspond more closely to the expectation that the
diffraction data.
induction of spin at the C atoms
All of the 13C and 15N NMR signals
involves polarization of electrons in sappear at low and high frequencies
type orbitals parallel to the MCN
(Figure 1 and 2), respectively, and hence
Figure 3. p-Type
orbital contribuaxis, while direct spin transfer to the N
the spin density at the C atoms is
tions of a MCN
Figure 4. Sketch of
atoms involves p-type orbitals perpennegative, while it is positive at the N
fragment qualitathe anisotropic spin
dicular to that axis.
atoms. This proves that the spin delocaltively representing
distribution about
In conclusion, by using solid-state
ization proceeds through p-type orbitals
the transfer of posthe C and N atoms
NMR spectroscopy, we have deter(Figure 3) with large and small contribuitive spin density
of the FeCN fragmined quantitatively small spin densitions of the N and C atomic orbitals,
from the metal to
ment as derived
ties on the CN ligands of [Fe(CN)6]3
respectively.[9] Polarization of the MC
the CN ligand.
from the NMR data.
Table 1: Solid-state NMR data[a] and spin densities[b] of Prussian-blue precursors.
Cs2K[Fe(CN)6]
Site of cyano group
axial
equatorial
2572
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Cs2K[Mn(CN)6]
Site of cyano group
2 (equatorial)
3 (equatorial)
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 2571 –2573
Angewandte
Chemie
and [Mn(CN)6]3. Because CN ligands are responsible for the
interaction in magnetic materials derived from these complexes, it is enlightening that the spin distribution in different
directions of the crystal lattice can be distinguished as well.
Finally, the shape of the spin distribution about the C and N
atoms can be derived and related to the spin delocalization
mechanisms engaged. It is expected that in the same way
many related compounds having suitable electron spin
relaxation times can be investigated in great detail.
[5]
[6]
[7]
Experimental Section
Samples of Cs2K[Fe(CN)6] and Cs2K[Mn(CN)6] were prepared
according to the literature[7d, 13] by using K 13CN and KC 15N, each
99 % enriched. The powders used for packing the 4-mm ZrO2 rotors
in a glove box were obtained from large single crystals. Nickelocene
was added to the powders for measuring the temperature inside the
rotor.[14] The spectra were recorded with a Bruker Avance 300
spectrometer. For spectrum analysis and simulation the respective
programs HBA and Wsolids[15] were used. The isotropic signal shifts
at the measuring temperature T, diso
T , were determined relative to
external adamantane and ammonium nitrate. The diso
T values and the
anisotropic signal shifts were converted to paramagnetic shifts at
298 K by subtracting the respective signal shifts of Cs2K[Co(CN)6]
and by considering the Curie law, dpara / 1/T. For obtaining the contact
para
values (see
shifts, dcon
298 , the dipolar shifts were subtracted from the d
the Supporting Information for details). The contact shifts were
converted to the spin densities, 1, by using Equation (1).
[8]
[9]
[10]
[11]
[12]
[13]
Received: January 12, 2004 [Z53726]
.
Keywords: cyanides · magnetic properties · NMR spectroscopy ·
spin-density distribution
[14]
[15]
F. M. Romero, E. Rusanov, H. Stoeckli-Evans, Inorg. Chem.
2002, 41, 4615 – 4617.
J. Herzfeld, A. E. Berger, J. Chem. Phys. 1980, 73, 6021 – 6030.
F. H. KOhler in Magnetism: Molecules to Materials (Eds.: J. S.
Miller, M. Drillon), Wiley-VCH, Weinheim, 2001, p. 379 – 430.
a) S. R. Fletcher, T. C. Gibb, J. Chem. Soc. Dalton Trans. 1977,
309 – 316; b) F. Herren, A. Ludi, P. Fischer, Acta Crystallogr.
Sect. B 1979, 35, 3129 – 3130; c) C. A. Daul, P. Day, B. N. Figgis,
H. U. GPdel, F. Herren, A. Ludi, P. A. Reynolds, Proc. R. Soc.
London Ser. A 1988, 419, 205 – 219; d) B. Ziegler, R. Haegele, D.
Babel, Z. Naturforsch. A 1989, 44, 896 – 902; e) B. N. Figgis, E. S.
Kucharski, J. M. Raynes, P. A. Reynolds, J. Chem. Soc. Dalton
Trans. 1990, 3597 – 3604; f) S. Peschel, B. Ziegler, M. Schwarten,
D. Babel, Z. Anorg. Allg. Chem. 2000, 626, 1561 – 1566.
B. N. Figgis, A. N. Sobolev, E. S. Kucharski, V. Broughton, Acta
Crystallogr. Sect. C 2000, 56, 735 – 737.
J. Cano, E. Ruiz, S. Alvarez, M. Verdaguer, Comments Inorg.
Chem. 1998, 20, 27 – 56.
P. Day, C. D. Delfs, B. N. Figgis, P. A. Reynolds F. Tasset, Mol.
Phys. 1993, 78, 769 – 780.
a) D. G. Davies, R. J. Kurland, J. Chem. Phys. 1967, 46, 388 – 390;
b) B. R. McGarvey, J. Perlman, J. Magn. Reson. 1969, 1, 178 –
184.
a) S. Takeda, Y. Umehara, H. Hara, G. Maruta, Mol. Cryst. Liq.
Cryst. Sci. Technol. Sect. A 2002, 379, 253 – 258; b) H. Ishiyama,
G. Maruta, T. Kobayashi, S. Takeda, Polyhedron 2003, 22, 1981 –
1987.
a) P. N. Hawker, M. V. Twigg in Comprehensive Coordination
Chemistry, Vol 4. (Eds.: G. Wilkinson, R. D. Gillard, J. A.
McCleverty), Pergamon, Oxford, 1987, p. 1204; b) M. K. Basu,
M. N. Das, J. Chem. Soc. A 1968, 2182 – 2185; c) V. Gadet, PhD
thesis, UniversitQ de Paris VI, 1992, p. 257; d) J. A. Lower, W. C.
Fernelius, T. Moller, V. O. Brecht, Inorg. Synth. 1946, 2, 213 –
214.
a) F. H. KOhler, X. Xie, Magn. Reson. Chem. 1997, 35, 487 – 492;
b) H. Heise, F. H. KOhler, X. Xie, J. Magn. Reson. 2001, 150,
198 – 206.
K. Eichele, R. E. Wasylishen, HBA 1.4 and Wsolids, Dalhousie
University, Canada, 2001 and 1998.
[1] a) W.-D. Griebler, D. Babel, Z. Naturforsch. B 1982, 37, 832 –
837; b) D. Babel, Comments Inorg. Chem. 1986, 5, 285 – 320;
c) M. Verdaguer, A. Bleuzen, V. Marvaud, J. Vaissermann, M.
Seuleiman, C. Desplanches, A. Scuiller, C. Train, R. Garde, G.
Gelly, C. Lomenech, I. Rosenman, P. Veillet, C. Cartier, F.
Villain, Coord. Chem. Rev. 1999, 190–192, 1023 – 1047.
[2] a) H. Iyasaka, N. Matsumoto, H. Okazawa, N. Re, E. Gallo, C.
Floriani, J. Am. Chem. Soc. 1996, 118, 981 – 994; b) K. van Langenbach, D. C. R. Hockless, B. Moubaraki, K. S. Murray, Synth.
Met. 2001, 122, 573 – 580; c) C. P. Berlinguette, D. Vaughn, C.
Canada-Vilalta, J. R. Galan-Mascaros, K. Dunbar, Angew.
Chem. 2003, 115, 1561 – 1564; Angew. Chem. Int. Ed. 2003, 42,
1523 – 1526; d) X. Shen, B. Li, J. Zou, H. Hu, Z. Xu, J. Mol.
Struct. 2003, 657, 325 – 331; e) S. Su, J.-K. Tang, Z.-Q. Liu, D.-Z.
Liao, Z.-H. Jiang, S.-P. Yan, P. Cheng, Inorg. Chem. Commun.
2003, 6, 1109 – 1112.
[3] H.-Z. Kou, S. Gao, W.-M. Bu, D.-Z. Liao, B.-Q. Ma, Z.-H. Jiang,
G.-P. Yan, Y.-G. Fan, G.-L. Wang, J. Chem. Soc. Dalton Trans.
1999, 2477 – 2480; b) H.-Z. Kou, W.-M. Bu, S. Gao, D.-Z. Liao,
Z.-H. Jiang, S.-P. Yan, Y.-G. Fan, G.-L. Wang, J. Chem. Soc.
Dalton Trans. 2000, 2996 – 3000; c) W. Sun, L. Liang, J. Weng, B.
He, Z. Shen, Mater. Chem. Phys. 2003, 78, 676 – 683.
[4] a) M. Clemente-Leon, E. Coronado, J. R. Galan-Mascaros, C.
Gimenez-Saiz, C. J. Gomez-Garcia, E. Ribera, J. Vidal-Gancedo, C. Rovira, E. Canadell, V. Laukhin, Inorg. Chem. 2001, 40,
3526 – 3533; b) M. Clemente-Leon, E. Coronado, J. R. GalanMascaros, C. Gimenez-Saiz, C. J. Gomez-Garcia, J. M. Fabre,
G. A. Mousdis, G. C. Papavassilion, J. Solid State Chem. 2002,
168, 616 – 625; c) E. Coronado, C. J. Gomez-Garcia, A. Nuez,
Angew. Chem. Int. Ed. 2004, 43, 2571 –2573
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2573
Документ
Категория
Без категории
Просмотров
1
Размер файла
116 Кб
Теги
spina, resolved, distributions, cs2k, blue, prussia, precursors, density, highly
1/--страниц
Пожаловаться на содержимое документа