chim. 1967, 470; G . T. Rogers, R. S. Shadbolt, and T. L . V . Ulbricht, J. chem. SOC.(London) C 1969, 203. [la] J . Pliml and M . Prystas, Advances in Heterocyclic Chemistry 8 , 115.  T . Nishimura, B. Shimizu, and I . Iwai, Chem. pharmac. Bull. (Japan) 11, 1470 (1963); E . Wittenburg, Z. Chem. 4 , 303 (1964).  H . Vorbriiggen and P. Strehlke, Angew. Chem. 81, 997 (1969); Angew. Chem. internat. Edit. 8, 976 (1969). 141 E. Wittenburg, Chem. Ber. 101, 1095 (1968). CNDO Calculation of the Inductive Effect of the Methyl Group By Herbert Kollmar and Harry 0 . Smith [*I MO calculations o n hydrocarbons and carbonium ions predict 11 ~ - a1n increase in the positive charge at the a position by methyl substituents, a prediction apparently confirmed by 1 3 C - N M R [31. This finding contradicts the usual assumption that the methyl group exerts an electron-releasing inductive effect. Various CNDO calculations [41 by our own modified technique [51 have correctly reproducedproperties of hydrocarbons and carbonium ionsr61. We have also been able to demonstrate the energetic stabilization of a positive charge by ormethyl substituents and, in particular, can show that, in terms of MO theory, the fact that methyl renders the or position positive is actually responsible for this stabilization. Table 1 lists the relative heats of formation of a number of carbonium ions; the energies were calculated for complete minimization of the total energy with respect to the geometric parameters (bond lengths and angles). Table I . require negligible activation energy. Our CNDO calculations show structure (2a) to be about 10 kcal/mole more stable than ( l a ) . conversion of ( l a ) into (2a) entailing no activation energy. The same process in the case of the 2-methylated ethyl cation (the n-propyl cation ( I b l ) leads to i-propyl cation ( 3 b ) . According to our calculations, the energy decreases monotonically during the transition: (3b) is 25 kcal/mole more stable than ( l b ) ; thus the nonclassical structure ( 2 6 ) is here less stable than the classical i-propyl cation. This state of affairs is reflected in the fact that the two C-H-3 bonds vary in strength in ( Z b ) , where a n H atom i s situated above the middle of the C-1-C-2 bond: the H atom is drawn towards C-1. This asymmetry is caused by the methyl substituent o n c-2. In the CNDO formalism the total energy of a molecule can be allotted to bonds and atoms 141: E = 2 EA + A The term EABis a measure of the strength of the chemical bond A-B "1, and can be further resolved according to physical principles. It thus becomes clear (Table 2) that the bond 2-3 is weaker than 1-3 because the positive H atom 3 (charge + 0.205) is more strongly repelled by the more positive of the two C atoms, i.e. C-2. This is less obvious from the purely electrostatic bonding components ES than from the resonance energy contributions (ER = one-electron resonance energy; E K = electron exchange energy 171). However, more detailed analysis shows that the contributions of ER and E K to the bonds 1-3 and 2-3 differ only because the positive charge o n C-2 is greater than that o n C-1. Table 2. Energy components of the bonds 1-3and2-3in(Zb). Alkylcarbonium ions: Stability and charge distribution. 2-3 Excess charge on the x-C atom - 2 0 24 -13 - 3 0 25 19 41 [a] AH for the reaction: R+ 55 +0.225 -0.009 '0.190 -1-0.250 C-1 -0.049 C-2 +0.066 +0.273 + C3H8 + n-C3H?+ i- RH (kcal/mole). The tabulated data illustrate the reasonable energetic stabilization of the cation with increasing methyl substitution. The positive charge on the central C atom increases by about 0.02 of a unit charge per methyl group. The significance of the positive charge in t h e a position for the energetic stabilization of carbonium ions will be demonstrated in the present communication. Let us consider the 2,l-hydrogen shift in the classical ethyl cation ( l a ) which proceeds via the nonclassical bridged ion ( 2 a ) . The three-center bond in (2a) can be formed by steady admixture of the empty pz orbital on C-1 in ( l a ) with the orbital of the C-2-H-3 bond. The transition should therefore EAB A>B E ER -0.303 -0.286 EK -o.ioo Es +0.082 -0.262 -0.271 -0.082 +0.092 Thus a relationship is established between the charge shift induced by the methyl group and the energetic stabilization of a positive charge by a n or-methyl substituent: The CH3 group increases the positive charge in the a position and is, in turn, bonded particularly strongly to a positive center. In alkylcarbonium ions, the methyl groups donate negative charge t o the empty p orbital of the central C atom (hyperconjugation). This effect is overcompensated by electron withdrawal from the o bonds by the methyl grouprll. In methyl substituted x systems, the appearance of a positive charge in the a position is accompanied by the occurrence of a corresponding negative charge in the p position. A methyl substituent therefore has a polarizing effect on a i c system [21. This is exemplified by propene [21, toluene, and also by the ion (2b) (see Table 1). Received: March 16, 1970 IZ 198 IE] German version: Angew. Chem. 82,444 (1970) [*I Dr. H. Kollmar Battelle-Institut e.V. 6 Frankfurt 90, Postfach 900160 (Germany) - B (a), R = H;(b)>R = CH, -< ,. R+ (3) 462 H. 0.Smith Max-Planck-Institut fur Medizinische Forschung 69 Heidelberg, Jahnstrasse 29 (Germany) [l] R . Hoflmann, J. chem. Physics 40, 2480 (1964); N . S. Isaacs, Tetrahedron 25, 3555 (1969).  J. A . Pople and M . Gordon, J. Amer. chem. SOC.89, 4253 (1957); M. 5 . Newton and W . N . Lipscomb, ibid. 89, 4261 (1 967).  G. A . Olah and A . M . White, J. Amer. chem. SOC.91, 5802 (1969).  J . A . Pople, 5.P. Santry, and G. A . Segal, J. chem. Physics 43, S . 129 (1965). Angew. Chem. internat. Edit. 1 Vol. 9 (1970)/ No. 6 H . Fischer and H . Kollmar, Theoret. chim. Acta 13, 213 (1969). 161 H. Kollmar and H . 0 . Smith, Chem. Physics Letters 5 , 7  (1970), and further literature cited there.  H . Fischer and H . Kollmar, Theoret. chim. Acta 16, 163 (1970). Crystal Structure of TeCl, :Presence of Tetramers in Solid Chalcogen(1v) Halides By Bruno Buss and Bernt Krebs [*I O n the basis of X-rayf11 and spectroscopic studies[21 of possible structures of Group VIB halides in the solid state and the steric effect of the lone pairs of chalcogen atoms, two structural models have been proposed for SeC14, TeC14, and TeBr4: one model is ionic and contains E1X;X- (the cations having C3,, symmetry) the other one being covalent and based o n E& (CzV;trigonal bipyramid) by analogy with the gaseous state. No unequivocal evidence in favor of either structure has hitherto been adduced. Relevant knowledge has been limited to the structure of TeF4[31. We have determined the structure of solid tellurium(1v) chloride by a complete X-ray analysis of a twinned crystal. TeC14 forms monoclinic crystals of space group C2/c having a = 17.076 (8), b = 10.404 ( 5 ) , c = 15.252 (8) A; p = 116.82'; Z = 16; dX-ray = 2.959, dexp 3.01 g cm-3. The lattice constants are in good agreement with theknown values [lb.ldl. The structure was determined o n the basis of about 3000 independent reflections (recorded with a four circle diffractometer and reduced to a single crystal). Refinement (isotropic) has so far reached R = 0.09. The solid consists of isolated Te4C116 units (Figure) with a cubane-like structure and having only van der Waals contacts (Cl . . . C1) with one another. Each of the T e atoms appears at the apex of a n equilateral trigonal pyramid (mean TeCl distance 2.31 A; 5 for individual values 0.006 A), three terminal C1 atoms forming the base. The coordination of the Te atom is made up to a n octahedron by three much more remote bridging C1 atoms (mean TeCl distance 2.93 A), the Te being displaced from the center parallel to the C3 axis. The corners of the cubane-like skeleton are occupied alternately by Te and bridging C1 atoms. Mean bond angles 85.1 for are 94.8 for Cl(termina1)-Te-Cl(terminal), Cl(bridge)-Te-Cl(bridge), and 94.7 for Te-Cl(bridge)-Te. The Te4C116 unit (exact symmetry C,) has almost Td symmetry. While the shorter Te-CI bonds are ideal (covalent) single bonds, insofar as the bond length corresponds to the sum of the covalent radii, ionic bonding plays a dominant role in the much weaker Te-CI bridging bonds. Thus the polar resonance structure can be fairly accurately described as an - array of TeC1; ions, of approximate symmetry C3,,, and C1ions. TeCI; ions have also been detected in TeCI;AlCI,[2a, 2e1, TeCl;AsF;[2e,51, and TeCl;SbCI;[2el. The shortest Te . . . Te distances in the tetramer are 4.30 A, thus ruling out Te-Te bonding. The C1 atoms are approximately close packed (C1 . . . CI distances 3.38 t o 4.21 A), T e filling 1 , 4 of the octahedral sites (cf. ref. [lcl). Interestingly, the Te-CI bond lengths in the TeCl; groups of TeC14 (2.31 A) are much shorter than those found for TeIv-CI in, e.g., (CH3)zTeCIz (2.51 A) [GI, TeCIi- (2.56 A) [71, o r tran~-TeC14(SC[N(CH3)2]2)2(2.53 A) [81 (cf. also ref. [99. A knowledge of the structure of solid TeC14 permits resolution of apparent contradictions in the properties of the compound, such as its electrical conductivity in the liquid state and its solubility in nonpolar solvents, e.g. benzene. A benzene solution of TeC14 is reported to contain trimers or tetramers[2b,101; this can readily be deduced from the structure. Under certain conditions, however, the possibility of an equilibrium with the monomer cannot be discounted [2c, *e,Zf, 101. Moreover, the vibrational spectra can now be unequivocally assigned, the twinning interpreted, and characteristic data such as the dipole moment, the high boiling point, the nuclear quadrupole resonance spectrum [ I 11, etc. explained. SeC14 and TeBr4 are isotypic with TeC14 (cf. also ref. rlc, Id]). Received: March 18, 1970 [ Z 202 I € ] German version: Angew. C h e m . 82, 446 (1970) [*I Stud.-Ref. B. Buss and Priv.-Doz. Dr. B. Krebs Anorganisch-Chemisches Institut der Universitat 34 Gottingen, Hospitalstrasse 8-9 (Germany) [I ] a) W. R . Blackmore, S . C . Abrahams, and J. Kalnajs, Acta crystallogr. 9, 295 (1956); b) A . W . Cordes, R . F . Kruh, E. K . Gordon, and M . K . Kemp, ibid. 17, 756 (1964); c) C . B . Shoemaker and S . C . Abrahams, ibid. 18,296 (1965); d) P. Khodadad, P . Laruelle, and J . Flahaut, C. R. hebd. SBances Acad. Sci. 259, 794 (1964).  a) H . Gerding and H . Houtgraaf, Recueil Trav. chim. PaysBas 73, 737, 759 (1954); b) N . N . Greenwood, B. P . Straughan, and A . E. Wilson, J. chem. SOC.(London) A 1966, 1479; 1968, 2209; c) D . M . Adams and P. J . Lock, ibid. 1967, 145; d) G . C . Hayward and P . J . Hendra, ibid. 1967, 643; e) I . R . Beattie and H . Chudzynska, ibid. 1967, 984; f) N . Katsaros and J. W . George, Inorg. chim. Acta 3, 165 (1969).  A . J . Edwards and F. I . Hewaidy, J. chem. SOC.(London) A 1968, 2977.  A similar structure has been described for trimethylplatinum(rv)-chloride: R . E. Rundle and J . H . Sturdivant, J. Amer. chem. SOC.69, 1561 (1947).  W . Sawodny and K . Dehnicke, Z. anorg. allg. Chem. 349, 169 (1967).  G. D . Christofferson, R . A . Sparks, and J . D . McCullough, Acta crystallogr. 11, 782 (1958).  A . C . Hazell, Acta chem. scand. 20, 165 (1966). 181 S . Husebye and J. W . George, Inorg. Chem. 8, 313 (1969). 191 0. Foss and S . Husebye, Acta chem. scand. 20, 132 (1966); 0. Foss, K . Johnsen, K . Maartmann-Moe, and K . Maroy, ibid. 20, 113 (1966). [lo] V. S . Yarkovleva and B. P . Troitskii, Chem. Abstr. 54, 117991 (1960); K . J . Wynne and P . S . Pearson, Inorg. Chem. 9, 106 (1970). [ll] A . Schmitt and W. Zeil, Z . Naturforsch. 18a, 428 (1963). 8,8-Dichloro-1,2,3,4,5,6,7,8heptathiatellur(1v)-ocaneI* *I By Johannes Weiss and Manfred Pupp [*I Figure: Te4Cl16 structural unit in solid tellurium(rv) chloride Reaction of TeC14 with polysulfanes affords a compound having the composition C12TeS7. We have been able t o isolate this product as orange crystals that are stable in air and insoluble in the usual solvents; they are, however, slightly soluble in CS2 (ca. 2 g/l). O n heating, a black coloration is observed at 95 OC (m.p. 110-112°C). 1 Vol. 9 (1970)1 No. 6 463 Angew. Chem. internat. Edit.