Angewandte
Chemie
À
trum Karlsruhe, 76344 Eggenstein-Leopoldshafen, Germany
well as the absence of Ti Ti multiple (p) bonding and,
(fax: (+ 49)7247-808-666; e-mail: crysdata@fiz-karlsruhe.de),
on quoting the depository numbers CSD-413504 and CSD-
413505.
À
furthermore, the avoidance of unfavorable Ti I antibonding
interactions stabilize the cluster anion.
[9] W. Klemm, L. Grimm, Z. Anorg. Allg. Chem. 1942, 249, 198.
[10] W. R. Robinson, J. Solid State Chem. 1974, 9, 255.
Received: January 12, 2004 [Z53739]
[11] Theoretical calculations: Semi-empirical extended Hückel cal-
culations (EH) of the whole solid K4[{Ti4O}I12] were performed
with the program package CAESAR [J. Ren, W. Liang, M.-H.
Whangbo, CAESAR, PrimeColor Software Inc., Raleigh, NC,
USA, 1998] with the following parameters: Hij [eV] (z1): O: 2s
À32.29999 (2.275), 2p À14.8 (2.275), I: 5s À18.0 (2.679), 5p À12.7
(2.322) K: 4s À4.34 (0.874), 4p À2.73 (0.874); Ti: 4s À8.97
(1.075), 4p À5.44 (0.675) for Ti-3d a double zeta-function was
used Hij [eV], z1, coefficient 1, z2, coefficient 2: À10.81, 4.55,
0.4206, 1.4, 0.7839. The program package CACAO [C. Mealli,
D. M. Proserpio, A. Iienco, CACAO, A package of programs for
Molecular Orbital Analysis, Instituto per lo Studio della Stereo-
chimica ed Energetic dei Composti di Coordinazione, Florence
(Italy), D. M. Proserpio, J. Chem. Educ. 1990, 67, 399 – 402] was
used for extended Hückel calculations of the electronic structure
and interpretation of bonding for an isolated [{Ti4O}I12]4À cluster.
Mulliken overlap populations [R. S. Mulliken, J. Chem. Phys.
1955, 23, 1397] are given for selected interactions. For graphical
display of the cluster valence orbitals Hyperchem [HyperChem,
Hypercube Inc., Gainesville, Florida, USA, 2001] was used.
Density functional studies were performed with the Stuttgarter
LMTO program [R. W. Tank, O. Jepsen, A. Burckhardt, O. K.
Andersen, TB-LMTO-ASA Program, Vers. 4.7, Max-Planck-
Institut für Festkörperforschung, Stuttgart, Germany, 1998]
based on the linear-muffin-tin approximation using the local
density functional LDA as the exchange-correlation functional
and the atomic sphere approximation ASA [H. L. Shriver, The
LMTO Method, Springer, Berlin, 1984; O. Jepsen, M. Snob,
O. K. Andersen, Linearized Band-Structure Methods in Elec-
tronic Band-Structure and its Applications, Springer Lecture
Note, Springer, Berlin, 1987; O. K. Anderson, O. Jepsen, Phys.
Rev. Lett. 1984, 53, 2571]. Interstitial spheres are introduced to
achieve space filling. The ASA radii as well as the positions and
radii of additional empty spheres are calculated automatically.
Reciprocal-space integrations are carried out using the tetrahe-
dron method [O. K. Andersen, O. Jepsen, Solid State Commun.
1971, 9, 1763; P. Blöchl, O. Jepsen, O. K. Andersen,Phys. Rev. B
1994, 34, 16223] The basis set of short ranged, atom-centered
TB-LMTOs contained for O 3s, 2p, 3d, for I 6 s, 5p, 5d, and 4f, for
K 4s, 4p, 3d, and 4f and for Ti 4s, 4p, and 3d. For O the 3s and 3d,
for I the 6s, 5d, and 4f, for K the 4p, 3d, and 4f and for Ti 4s, 4p,
and 3d orbitals were treated with the downfolding technique
[W. R. L. Lambrecht, O. K. Andersen, Phys. Rev. B 1986, 34,
2439; O. Jepsen, O. K. Andersen, Z. Phys. B 1995, 97, 35; G.
Krier, O. Jepsen, O. K. Andersen, Max-Planck-Institut für
Festkörperforschung, Stuttgart, Germany, unpublished] The
crystal orbital Hamiltonian population (COHP) method is
used for bond analysis [R. Dronskowski, P. Blöchl, J. Phys.
Chem. 1993, 97, 8617]. Note that the values are negative for
bonding and positive for antibonding interactions, the reverse of
the signs used in crystal orbital overlap population (COOP; [S.
Wijeyesekera, R. Hoffmann, Organometallics 1984, 3, 949])
diagrams in the semi-empirical Hückel treatment. This discrep-
ancy emerges because, to obtain the COOPthe DOS gets
multiplied by the overlap population while for weighting the
DOS in case of the COHPthe corresponding element of the
Hamiltonian is used.
Keywords: cluster compounds · electronic structure · iodine ·
titanium
.
[1] T. Gloger, Dissertation, Universität zu Köln, 1998.
[2] A. Lachgar, D. S. Dudis, J. D. Corbett, Inorg. Chem. 1998, 37,
2242.
[3] B. Krebs, G. Henkel, Z. Anorg. Allg. Chem. 1981, 474, 149.
[4] J. Zhang, R.-I. Qi, J. D. Corbett, Inorg. Chem. 1991, 30, 4794.
[5] D. J. Hinz, G. Meyer, T. Dedecke, W. Urland, Angew. Chem.
1995, 107, 97; Angew. Chem. Int. Ed. Engl. 1995, 34, 71.
[6] D. J. Hinz, G. Meyer, J. Chem. Soc. Chem. Commun. 1994, 125.
[7] Synthesis: KI (Aldrich, 99%) was dried under vacuum at 473 K.
TiI4 was prepared from titanium (chips, Chempur, 99.6%) and
iodine (Riedel de Häen, 99.8%), according to the procedure
described in ref. [9]. TiO2 was used as obtained (Kronos Titan).
All storage and manipulation of all starting materials and
products were handled under dry box conditions (MBraun,
Garching). The reaction was carried out in a sealed tantalum
container, jacketed with a silica tube under argon at a temper-
ature of 798 K. The compound formed as black shiny crystals of
good quality. X-ray powder diffraction measurements at room
temperature revealed the product to consist of only one phase,
K4[{Ti4O}I12]. No phase transition was observed by X-ray
powder diffraction in the temperature range between 293 K
and 10 K. Indexed X-ray powder diffractograms of K4[{Ti4O}I12]
were measured at room temperature on a transmission powder
diffraction system, Guinier Image Plate Camera G670 (Huber,
Rimsting), using monochromatic MoKa radiation (cell parame-
ters at 298 K refined from powder data: a = 1373.0(3) pm, c =
807.1(2) pm). Low temperature X-ray powder diffractograms
were measured on a Guinier diffractometer G645 (Huber,
Rimsting, Germany), using monochromatic CuKa radiation, (cell
parameters at 150 K refined from powder data: a =
1369.1(4) pm, c = 808.8(3) pm).
[8] Crystal data and structure refinement: A suitable single crystal
(0.2 0.2 0.1 mm) was mounted in a glass capillary. Intensity
data were collected on an IPDS diffractometer (Stoe, Darm-
stadt) at 170 K and 293 K. 170 K: Tetragonal, space group I4/m;
a = 1364.5(1) pm, c = 801.51(7) pm, V= 1.4923(2) nm3; Z = 2;
1calcd = 4.199 gcmÀ3; 2.11 < q < 29.978; IPDS II, MoKa radiation
(l = 71.073 pm); T= 170(2) K; F(000) = 1616; m = 13.152 mmÀ1
;
13964 reflections were measured, of which 1169 are unique, 1104
are considered as observed. R1 = 0.0196 and wR2 = 0.0426 for
[I0 > 2s(I0)]. 293 K: Tetragonal, space group I4/m; a =
1373.9(2) pm, c = 810.1(2) pm, V= 1.5293(5) nm3; Z = 2; 1calcd
=
4.097 gcmÀ3; 2.92 < q < 25.988; IPDS I, MoKa radiation (l =
71.073 pm); T= 293(2) K; F(000) = 1616; m = 13.658 mmÀ1
;
6080 reflections were measured, of which 804 are unique, 536
are considered as observed. R1 = 0.0261 and wR2 = 0.0385 for
[I0 > 2s(I0)]. The data were processed with the program systems
SHELX-97 [G. M. Sheldrick, SHELX-97, Universität Göttin-
gen, 1997]. Scattering factors were taken from the International
Tables for Crystallography, Volume C [A. J. C. Wilson, Interna-
tional Tables for Crystallography, Vol. C, Kluwer Dordrecht, the
Netherlands, 1995]. Numerical absorption correction after
crystal shape optimization was performed using the programs
XRED and XSHAPE [Stoe, XRED 1.01 and XSHAPE 1.01,
Darmstadt, 1996]. Further details on the crystal structure
investigations may be obtained from the Fachinformationszen-
[12] M. Kriener, V. Kataev, A. Freimuth, unpublished results.
Angew. Chem. Int. Ed. 2004, 43, 3183 –3185
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3185