Read 3810 Problem set 7 in 2003.doc text version
CHEMISTRY 3810 Problem Set #7 Warning: cited literature articles have been added to the end of this assigment; you may not wish to print off all the pages. Topic: Chemistry of the Group 17 and 18 Elements. You are responsible for Ch. 12 of Shriver-Atkins 1. Preferably without consulting reference material, write out the halogens and noble gases as they appear in the periodic table, and indicate the trends in (a) physical state (s, l, or g) at room temperature and pressure, (b) electronegativity, (c) hardness of the halide ion, and (d) colour. Describe how the halogens are recovered from their naturally occurring halides and rationalize the approach in terms of standard potentials. Give balanced chemical equations and conditions where appropriate. Sketch the chlor-alkali cell. Show the half-cell reactions and indicate the direction of diffusion of the ions. Give the chemical equation for the unwanted reaction that would occur if OH migrated through the membrane and into the anode compartment. Sketch the form of the vacant sigma* orbital of a dihalogen molecule and describe its role in the Lewis acidity of the dihalogens. Nitrogen trifluoride, NF3, boils at -129°C and is devoid of Lewis basicity. By contrast the lower molar mass compound NH3 boils at -33°C and is well known as a Lewis base. (a) Describe the origins of this very large difference in volatility. (b) Describe the probable origins of the difference in basicity. Based on the analogy between halogens and pseudohalogens: (a) Write the balanced equation for the probable reaction of cyanogen (CN) 2 with aqueous sodium hydroxide. (b) Write the equation for the probable reaction of excess thiocyanate with the oxidizing agent Mn0 2(s) in acidic aqueous solution. (c) Write a plausible structure for trimethylsilyl cyanide. (a) Use the VSEPR model to predict the probable shapes of IF6+ and IF7. (b) Give a plausible chemical equation for the preparation of [IF6][SbF6]. Predict whether each of the following solutes is likely to make liquid BrF3 a stronger Lewis acid or a stronger Lewis base: (a) SbF5, (b) SF6, (c) CsF. Predict whether each of the following compounds is likely to be dangerously explosive in contact with BrF3, and explain your answer: (a) SbF5, (b) CH30H, (c) F2, (d) S2Cl2.
10. The formation of Br 3 from a tetraalkylammonium bromide and Br 2 is only slightly exoergic. Write an equation (or NR for no reaction) for the interaction of [NR4][Br3] with excess I2 in CH2Cl2 solution, and give your reasoning. 11. Explain why CsI3 (s) is stable with respect to the elements but NaI3 (s) is not. 12. Write plausible Lewis structures for (a) ClO2 and (b) I205 and predict their shapes and point groups. 13. (a) Give the formulas and the probable relative acidities of perbromic acid and periodic acid. (b) Which is the more stable? 14. (a) Describe the expected trend in the standard potential of an oxoanion in a solution with decreasing pH (b) Demonstrate this phenomenon by calculating the reduction potential of ClO4 at pH = 7 and comparing it with the tabulated value at pH = 0. 15. With regard to the general influence of pH on the standard potentials of oxoanions, explain why the disproportionation of an oxoanion is often promoted by low pH. 16. Which oxidizing agent reacts more readily in dilute aqueous solution, perchloric acid or periodic acid? Give a mechanistic explanation for the difference. 17. (a) For which of the following anions is disproportionation thermodynamically favourable in acidic solution: ClO, CIO2 , ClO3 and ClO4? (if you do not know the properties of these ions, determine them from a table of standard potentials.) (b) For which of the favourable cases is the reaction rate very low at room temperature? 18. Which of the following compounds present an explosion hazard? (a) NH4ClO4, (b) Mg(ClO4)2, (c) NaCI0 4, (d) [Fe(OH2)6][ClO4]2. Explain your reasoning. 19. Explain why helium is present in low concentration in the atmosphere even though it is the second most abundant element in the universe.
Problem Set #7
20. Which of the noble gases would you choose as (a) the lowest temperature liquid refrigerant, (b) an electric discharge light source requiring a safe gas with the lowest ionization energy, (c) the least expensive inert atmosphere? 22. Give the formula and describe the structure of a noble gas species that is isostructural with (a) ICl 4, (b) IBr 2, (c) BrO3, (d) CIF. 23. (a) Give Lewis formulas and formal charges for ClO, Br 2, and XeF+. (b) Are these species isolobal? (c) Describe the chemical similarities as judged by their reactions with nucleophiles. (d) Rationalize the trends in electrophilicity and basicity. 21. By means of balanced chemical equations and a statement of conditions, describe a suitable synthesis of (a) xenon difluoride, (b) xenon hexafluoride, (c) xenon trioxide. 24. (a) Give a Lewis structure for XeF7. (b) Speculate on its possible structures by using the VSEPR model and analogy with other xenon 25. Given the bond lengths and angles in I5+, describe the bonding in terms of two -center and three-center s bonds and account for the structure in terms of the VSEPR model. 26. Until the work of K.O. Christe (Inorg. Chem. 25, 3721 (1986)), F2 could be prepared only electrochemically. Give chemical equations for Christe's preparation and summarize the reasoning behind it. 27. Predict the line-patterns in the 19F NMR spectrum of the following halogen fluorides. Assume that the axial position(s) is(are) deshielded more, and that there is no chemical exchange: a) ClF3 b) BrF5 c) IF7 28. Provide an MO interpretation of the bonding in ClF3. Treat the F atoms as simple spheres (pseudo s orbitals), but consider all the valence s and p orbitals of the central atom, as usual. 29. The Se 2I42+ cation in [Se 2I4][Sb2F10]2 has the unusual shape shown below. Like many halogen compounds, it is extremely electron rich. This means that there are many electrons per nucleus. One bonding model for SeI2+ involves the p bonds indicated at the right. Use these p orbitals as a basis set to construct an MO description of the dimer in Se 2I42+. It is not necessary to provide symmetry labels for the dimer orbitals, but provide an energy-level diagram and orbital topologies. Is your model consistent with the very weak bonding between the two halves of the dimer indicated by the observed bond distances? (Reference: T. Klapotke and J. Passmore, Acc. Chem. Res. (1989) 22, 234-240).
30. Provide an MO description of the bonding in XeF4. Use the example done in class for BrF5 as a guide. 31. The structure of XeF6 has been the subject of continuing discussion and debate among chemists. It is not an easy molecule to study! Describe the idealized VSEPR and actual gas phase structures of XeF6. What structure does XeF6 adopt in solution (NMR evidence) and in the solid state? Reference: K. Seppelt, Acc. Chem. Res. (1979) 12, 211-216.
Problem Set #7
32. Interpret the NMR spectra of the C5F5N-Xe-F+ cation. (a) shows the 19F signal for the Xe-F fluorine atom only, while (b) shows the 129Xe spectrum. C5F5N is a pentafluoropyridine molecule, attached to Xe via the N atom. The nitrogen is natural abundance.
33. Interpret the 125Xe NMR spectrum of OXe(OTeF5)4 shown below. The structure resembles that of OXeF4, with the OTeF5 groups replacing the fluorine ligands. OTeF5 has the structure indicated below. Each of the larger lines in spectrum is split into quintets by 4 Hz. The separation between the quintets is 55 Hz, while the separation between the centres of the three clusters is 640 Hz.
34. Is the poorly resolved 125Xe NMR spectrum shown below consistent with the proposed structure Of C6F5X+?
Volume 25 Number 21 October 8, 1986
0 Copyright 1986 by the American Chemical Society
Chemical Synthesis of Elemental Fluorine
The chemical synthesis] of elemental fluorine has been pursued for at least 173 years2 by many notable chemists, including Davy,2 F r e m ~M~ i ~ s a nand Ruff.5 All their attempts have failed, and , o ,~ the only known practical synthesis of F2 is Moissan's electrochemical process, which was discovered exactly 100 years ago.6 Although in principle the thermal decomposition of any fluoride is bound to yield fluorine, the required reaction temperatures and conditions are so extreme that rapid reaction of the evolved fluorine with the hot reactor walls preempts the isolation of significant amounts of fluorine. Thus, even in the well-publicized case of K3PbF7,7*s only trace amounts of fluorine were i ~ o l a t e d . ~ , ~ These failures, combined with the fact that fluorine is the most electronegative element and generally exhibits the highest single bond energies in its combinations with other elements,1° have led to the widely belief that it is impossible to generate fluorine by purely chemical means. The purpose of this communication is to report the first purely chemical synthesis of elemental fluorine in significant yield and concentration. This synthesis is based on the fact that thermodynamically unstable high-oxidation-state transition-metal
In the context of this communication, the term "chemical synthesis of elemental fluorine" implies the generation of F, by purely chemical means and excludes either techniques such as electrolysis, photolysis, discharge, etc. or the use of elemental fluorine for the synthesis of any of the starting materials. The regeneration of fluorine from materials prepared from fluorine obviously is just a method for chemically storing but not for chemically generating fluorine. Davy, H. Phil. Trans. R. Soc. London 1813, 103, 263. Fremy, M. E. Ann. Chim. Phys. 1856, 47, 44. Moissan, H. C. R. Hebd. Seances Acad. Sci. 1886, 102, 1543; 1886, 103,202,256,850 1884,99,655,874; 1885,100,272, 1348; 1885,101, 1490; 1886,102,763, 1245; 1886,103, 1257; 1889,109,862,637; Ann. Chim. Phys. 1887, 12, 472; 1891, 24, 224; Bull. So?. Chim. Fr. 1891, 5 , 880. Ruff, 0.Z. Angew. Chem. 1907.20, 1217; Z . Anorg. Chem. 1916,98, 27. Moissan, H. C. R. Hebd. Seances Acad. Sci. 1886, 102, 1543. Brauner, B. Z. Anorg. Chem. 1884, 7, 1 . Clark, G. L. J . A m . Chem. SOC.1919, 41, 1477. Argo, W. L.; Mathers, F. C.; Humiston, B.; Anderson, C. 0. Trans. Am. Electrochem. SOC. 1919, 35, 335. See for example: Cotton, F. A,; Wilkinson, G. Advanced Inorganic Chemistry; Interscience: New York, 1972; p 1 1 3 . Schmitz-Dumont, 0.;Opgenhoff, P. Z. Anorg. Allg. Chem. 1952, 268,
fluorides can be stabilized by anion formation. Thus, unstable NiF,, CuF,, or MnF, can be stabilized in the form of their corresponding MF62-anions. Furthermore, it is well-known that a weaker Lewis acid, such as MF,, can be displaced from its salts by a stronger Lewis acid, such as SbF,. K2MF6 2SbF5 2KSbF6 [MF,] (1)
If the liberated MF4 is thermodynamically unstable, it will spontaneously decompose to a lower fluoride, such as MF, or MF,, with simultaneous evolution of elemental fluorine. [MFd
Since a reversal of (2) is thermodynamically not favored, fluorine can be generated even at relatively high pressures. Consequently, the chemical generation of elemental fluorine might be accomplished by a very simple displacement reaction, provided a suitable complex fluoro anion is selected which can be prepared without the use of elemental fluorine and is derived from a thermodynamically unstable parent molecule. The salt selected for this study was K2MnF6. It has been known16 since 1899 and is best prepared from aqueous H F s01ution.l~ 2KMn04 + 2KF
+ lOHF + 3 H 2 0 2
+ 3 0 2 (3)
The literature yield of 30% was increased to 73% and can Probably be improved further by refining the washing procedure (use of acetone instead of HF)." The other starting material, SbF5, can be prepared]9 in high yield from SbCI, and HF. SbCI, + 5 H F SbFs 5HC1 (4)
Since both starting materials, K2MnF6and SbF5,can be readily prepared without the use of F2 from H F solutions, the reaction K,MnF6
+ MnF3 + ll2F2
Ryss, I. G. In The Chemistry of Fluorine and Its Inorganic Compounds; USAEC Translation 3927; AEC: Oak Ridge, T N , 1956; p 3 1 . Cady, G.H. In Fluorine Chemistry; Simons, J. H., Ed.; Academic: New York, 1950; pp 293-294. O'Donnell, T. A. In Comprehensiue Inorganic Chemistry; Bailar, J. C., Emeleus, H. J., Nyholm, R., Trotman-Dickenson, A . F., Eds.; Pergamon: Oxford, U.K., 1973; pp 1010-1013. Naumann, D. In Fluor und Fluorverbindungen; Spezielle Anorganische Chemie in Einzeldarstellungen, Vol. 2; Schneider, A,, Ed.; Steinkopff Darmstadt, W. Germany, 1980; p 3.
represents a truly chemical synthesis of elemental fluorine. The displacement reaction between K2MnF6 and SbF5 was carried out in a passivated Teflon-stainless-steel reactor at 150 " C for 1 h. The gas, volatile at -196 O C , was measured by PVT and shown by its reaction with mercury and its characteristic odor to be fluorine. The yield of fluorine based on (5) was found to be reproducible and in excess of 40% but most likely can be improved upon significantly by refinement of the experimental conditions. Fluorine pressures of more than 1 atm were generated in this manner. In summary, the purely chemical generation of elemental fluorine can be achieved in high yield and concentration by a very simple displacement reaction between starting materials that can
(16) Weinland, R. F.; Lauenstein, 0.Z. Anorg. Allg. Chem. 1899, 20, 40. ( 1 7 ) Bode, H.; Jenssen, H.; Bandte, F. Angew. Chem. 1953, 65, 304. (18) Chaudhuri, M. K.; Das, J. C.; Dasgupta, H . S. J . Inorz. Nucl. Chem. _ . 1981, 43, 85. (19) Ruff, 0. Ber. Dtsch. Chem. Ges. 1906, 39, 4310.
0 1986 American Chemical Society
Inorg. Chem. 19186, 25, 3122-3124
oxidation potential for the formation of ('+TMP)Fe'"O. The two l e oxidation potentials that they report are + I .01 and + I .40 V. They attributed our results to the presence of chloride ion impurity and to the absence of N a 2 C 0 3and water. We show here that chloride ion is not present in our system; we provide additional data in support of our assignment of potentials for l e oxidation of iron(1V) porphyrins to the corresponding iron(1V) porphyrin K cation radicals; and we show that the potential (+1.01 V)6 assigned by Groves and Gilbert for the l e oxidation of (TMP)Fe'I'OH to (TMP)Fe"O is in actuality due to two l e oxidations. The following observations establish the absence of all chloride ion. In our experiments pure (TMP)Fe"'OH was used and the solvent and electrolyte system was devoid of chloride ion. Reactions were carried out at -71 OC where CH2CI2solvent does not undergo oxidation. In Figure 1 there is shown repetitive visible spectral scans of the first two sequential l e oxidations of (TMP)Fe"'OH (conditions, positions of isosbestic points, peak heights, etc. provided in the caption). That the spectral changes of parts A and B of Figure 1 are associated with l e oxidations follows from their generation by controlled-potential coulometry. From the isosbestic points there is seen to be no competitive change of ligand nor accumulation of intermediate. Inspection of Figure 1A reveals the absence of the spectral characteristics (absorbance at 380 and 510 nm) of (TMP)Fe"'CI. Indeed, spectroelectrochemistry at -71 "C with (TMP)Fe"'CI shows that l e oxidation at 1.18 V is accompanied by a decrease in the Soret absorbance at 420 nm and an increase in absorbance at 398 nm with an isosbestic point at 528 nm. Additional evidence for the absence of chloride ion in our experiments is shown by the observation that the presence of trace concentrations (10-5-10-4 M) of [ ( n C4H9)4N+] [CI-] in a solution of (TMP)Fe"'OH results in a CV where the first oxidation is no longer reversible. Such is not the case with the CV's we have r e p ~ r t e d . ~ . ~ known' that the It is reduction of iron(II1) porphyrin to iron(I1) porphyrin is strongly influenced by the axial ligand. Employing (TMP)Fe"'OH, we find that the le-reduction potential is at -1.05 V while the potential for l e reduction of (TMP)Fe"'CI occurs at -0.75 V (dry CH2CI,, 25 "C). The samples of (TMP)Fe"'OH employed in the electrochemical and spectroelectrochemical studies showed no evidence of a peak potential at -0.75 V. In our hands the electrochemical oxidations of (TMP)Fe"'OH have been found to be both chemically and electrochemically reversible. Groves and Gilbert (working in a solvent composed of CH2CI2 wet with water and saturated with Na,C03) reported that the oxidation of (TMP)Fe"O to ('+TMP)Fe'"O is irreversible and occurs at a higher potential (+1.40 V ) than the potential ( + I . I 4 V) for the (TMP)Fe"'CI to ('+TMP)Fe"'(CI), oxidation. A CV similar to theirs, with the exception of the absence of the irreversible peak at +1.40 V, is obtained for (TMP)Fe"'OH in CH2C12 that has been wet by being passed through air-equilibrated alumina (Figure 2A). By simple visual observation of the CV, it might appear as though the first oxidation wave (1.01 V) represents a single l e process, as they assumed. Controlled-potential coulometry (at 1.06 V) at -71 "C showed that there are two electrons ( n = 2.1 f 0.2) associated with this wave. For the low-temperature controlled-potential coulometry, the system was first calibrated by using (TMP)Fe"'CI. The coulometric oxidation was monitored by change of current measurement with time and also by running a CV when n is equal to 1. Therefore, the first oxidation wave must represent two le oxidations that are so close in potential that they cannot be distinguished by CV.' A possible explanation for
(7) (a) Scheidt, W. R.; Reed, C. A. Chem. Rev. 1981, 81, 543-555. (b) Jones, S. E.; Srivatsa, G. S.; Sawyer, D. T.; Traylor, T. '3.; Mincey, T. C. Inorg. Chem. 1983, 22, 3903-3910. (8) There is no reason that two l e oxidations so close in potential should resemble a 2e oxidation with CV. Bard and Faulkner (Bard, A. J.; Faulkner, L. R. Electrochemical Methods; Wiley: New York. 1980; pp 233-235) point out that when the potential difference between two l e processes is less than 100 mV the individual waves are merged into a broad wave. Also, they show that if the difference is 35.6 m V (theoretical), which occurs when there is no interaction between the redox groups on the substrate, then the observed wave has all the characteristics of a l e transfer.
be prepared in high yields from H F solutions and have been known for 80 years or longer. As in the cases of noble gas20or N F 4*] chemistry, the successful chemical synthesis of elemental fluorine demonstrates that one should never cease to critically challenge accepted dogmas. Acknowledgment. The author is grateful to R . D. Wilson for his assistance with some of the experiments, to Drs. C. J. Schack, W. W. Wilson, and L. R. Grant for help, and to the U S . Army Research Office and Office of Naval Research for financial support.
(20) Bartlett, N. Proc. Chem. SOC., London 1962, 218. (21) Christe, K. 0.;Guertin, J. P.; Pavlath, A. E. Inorg. Nucl. Chem. Left. 1966, 2, 83.
Rocketdyne A Division of Rockwell International Canoga Park, California 91303
Received August 20, 1986
Karl 0. Christe
Electrochemical Generation of Iron(1V)-Oxo Porphyrins and Iron(1V)-Oxo Porphyrin a Cation Radicals
Sir: With weakly basic ligands, such as chloride or perchlorate, the first electrochemical l e oxidation of an iron(II1)porphyrin is porphyrin-centered, resulting in the formation of an iron(II1)~, porphyrin a cation radical.] We were the first to s h ~ wthat ~ when the ligands are the strongly basic oxy anions, HO- and CH30-, the first electrochemical l e oxidation is iron-centered, providing an iron(1V) porphyrin. In our investigations (dry CH2CI2solvent), potentials were determined by cyclic voltammetry, coulometry was determined by controlled-potential oxidation at the potentials corresponding to the CV peak positions, and the nature of the products was established by low-temperature spectroelectrochemistry and comparison of the spectra to those of known porphyrin species. In addition, the identification of electrochemically generated iron( IV) porphyrin species was verified by chemical conversion to known species at the same oxidation level. To obviate p-oxo dimer formation all investigations employed (meso-tetrakis( 2,6-disubstituted pheny1)porphinato)iron(111) hydroxide and methoxide salts. In this manner, we showed that (tetrakis(2,4,6-trimethylphenyl)porphinato)iron(III) hydroxide ((TMP)Fe"'OH) on 1e oxidation (+ 1.O 1 V)4 provides an iron(1V) porphyrin ((TMP)FeiVO),S and the second l e oxidation (+1.13 V) gives an iron(1V) porphyrin *-cation radical (('+TMP)FelVO). In a recent communication in this journal, Groves and Gilbert reexamined the electrochemistry of (TMP)Fe"'OH in wet CH,CI, saturated with Na2C03.6 Their results substantiated our original discovery that the first and third oxidation peaks observed with (TMP)Fe"'OH are for the formation of (TMP)Fe"O and (2+TMP)FeIVO,respectively, but disputed the value of our second
( 1 ) (a) Phillipi, M. S.; Goff, H. M . J . Am. Chem. SOC.1982, 104, 6026-6034. (b) Phillippi, M. A.; Shimomura, E. T.; Goff, H. M. Inorg. Chem. 1981, 20, 1322-1325. (c) Cans, P.; Buisson, G.; Duee, E.; Marchon, J.-C.; Erler, B. S.;Scholz, W. F.; Reed, C. A. J . Am. Chem. SOC. 1986, 108, 1223-1234. (2) Lee, W. A.; Calderwood, T. S.; Bruice, T. C. Proc. Natl. Acad. Sci. U . S . A . 1985, 82, 4301-4305. (3) Calderwood, T. S.;Lee, W. A.; Bruice, T. C. J . Am. Chem. SOC. 1985,
107, 8272-8273. (4) All potentials given in this paper are vs. a saturated caromel electrode (SCE). The potentials given by Groves and Gilbert are corrected to S C E by using the conversion factor in their paper. (5) The nature of the oxo ligand is not known. Though the iron(1V)-oxo porphyrin is written as (TMP)Fe"O, the oxo ligand may be -OH or perhaps a second oxo ligand may be present since it is virtually impossible to free a polar organic solvent of all traces of water. In our previous publications (ref 2 and 3), we used the notation (TMP)Fe"OH. (6) Groves, J. T.; Gilbert, J. A. Inorg. Chem. 1986, 25, 123-125.
0 1986 American Chemical Society
Acc. Chem. Res. 1989,22, 234-240
Sulfur and Selenium Iodine Compounds: From Nonexistence to Significance
T.KLAPOTKE J. PASSMORE* and
Department of Chemistry, University of New Brunswick, Fredericton, New Brunswick, Canada E3B 6E2 Received May 6, 1988 (Revised Manuscript Received March 21, 1989)
It has been our goal to prepare quantitatively in one-step reactions simple compounds that are novel in terms of their stereochemistry and bonding, are first examples of new classes of compounds, and open up new areas of chemistry. Such achievements are often discoveries rather than planned syntheses, and the sulfur and selenium iodine cations (Table I) described here were prepared as a result of an unsuccessful search for S13AsF6.' Binary sulfur iodides are unstable under ambient condition^,^-^ and selenium iodides5 are unknown. Before our work, there were no examples of stable species at room temperature containing covalent S-I ar Se-I except Seh2-.6 We have now prepared, usually quantitatively, a number of salts of the sulfur and selenium iodine (and bromine) cations shown in Table I, all of which contain covalent S-I and Se-I bonds. In addition, these cations provide examples of stable derivatives of S7,Se6,thermodynamically stable npr-npr bonds ( n I3), and r*-+ bonds. Many of the cations maximize intercationic halogen-chalcogen contacts and thus have cluster-like characteristics, e.g., the cube-like SQI,~+(Figure 10) and the distorted right(Figure 4). The bonding triangular prismatic S214,+ encountered in these cations has been helpful in understanding the puzzling geometries of more complex : + and a related species, e.g., S , Sei+, S4N4, S202-,fuller account of which is given in ref '7. Instability of Neutral Sulfur and Selenium Binary Iodides Solid S212has been characterized at -90 0C,2and SI, at 9 K,34 but no structural evidence has been presented for the corresponding binary selenium iodide^.^ The instability of the S-I and Se-I bonds can be attributed to their very low ionic resonance stabilization energies as the electronegativity of iodine is about the same as H that of sulfur and selenium. Thus A (eq 1) and AH (eq 2) are -18.0 and -1.3 kJ mol-l, respectively.8 They 2 -s-I(g) -s-s- (g) + Iz(g) (1) 2 -Se-I(g) -Se-Se- ( g ) + 12(g) (2) are even more unstable in the solid state due (in part)
Thomas Klapotke was born in Gijttingen in 1961. He studied chemistry at t h e Technical University of Berlin under the supervision of Prof. Kopf and obtained his Ph.D. in 1986. In 1987 he won a Feodor-Lynen-Scholarship of the Alexander von Humboldt Foundation and spent a year as a visiting scholar with Prof. Jack Passmore at the University of New Brunswick in Canada. Since 1988 h e has been at the Technical University of Berlin working on fluoro organometallic chemistry. Jack Passmore was born in Barnstaple, Devon, England, and received his B.Sc. and D.Sc. from the University of Bristol, England, and his Ph.D. degree (with Dr. Neil Bartlett) from the University of British Columbia in Vancouver. H did postdoctoral work during the year 1968-1969 at MacMaster Universie ty, HamiRon, Ontario, Canada (with Dr. Ronald J. Giliespie). He then joined s the faculty at the University of New Brunswick in 1969, where h e i presently Professor of Chemistry.
Table I CharacterizedoBinary Sulfur and Selenium Halogen Cations F c1 Br I
Br2S+SSBr [(S71)21]3+ SeF3+ SeC13+ SeBr3+ Se13+ Se7+SeSeC1 SezBr6+ Se212+ Br2Se+SeSeBr 12Se+SeSeSe+Izb (Se I+ See$n
Structure of cations determined by X-ray crystallography. Identified in solution by 77Se NMR.
to the large sublimation energy of solid I, (62.3 kJ mol-l). For example, CF3SI is detected as a gas, but readily disproportionates in the solid state above -100 "C according to eq 3.9a CH3SI behaves similarly and is also only stable in the solid state at very low temperatures, decomposing to CH3SSCH3and 2CF,SI(s)
The structure of Ph3CSI, which is stable10cin the solid state at -78 "C and in solution in the dark, has been determined.lobEvidence for RCOSI (R = aryl) has been presented,loabut the material has not been structurally characterized. No neutral compound containing a room
(1) Passmore, J.; Taylor, P.; Whidden, T. K.; White, P. S. J. Chem. SOC., Chem. Commun. 1976,689. (2) (a) Dasent, W. E. Non-Existent Compounds;Marcel Dekker: New York, 1965; p 162. (b) Daneky, J. P. In Sulfur in Organic and Inorganic Chemistry; Marcel Dekker: New York, 1971; Vol. 1, p 327. (c) Peach, M. E. Int. J. Sulfur Chem. 1973,8(1), 151. (3) (a) Vahl, G.; Minkwitz, R. 2.Anorg. Allg. Chem. 1978,443,217and references therein. (b) Manzel, K.; Minkwitz, R. 2.Anorg. Allg. Chem. 1978,441, 165. (4) Feuerhahn, M.; Vahl, G. Inorg. Nucl. Chem. Lett. 1980, 16, 5. (5) Behrendt, U.; Gerwarth, U. W.; J e e r , S.; Kreuzbichzer, 1.; Seppelt, K. In Gmelin Handbook of Inorganic Chemistry, 8th ed.; Springer-Verlag: Berlin, Heidelberg, New York, Tokyo, 1984; Supplement Vol. B2, No. 10. (6)Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; Pergamon Press: Oxford, New York, Toronto, Sydney, Paris, Frankfurt, 1986 and references therein. (7) Burford, N.; Passmore, J.; Sanders, J. C. P. From Atoms to Polymers: Isoelectronic Analogies. In Molecular Structure and Energetics; Leibman, J. F., Greensberg, A., Eds.; VCH Vol. 8, in press and references therein. (8) (a) Johnson, J. P.; Murchie, M. P.; Passmore, J.; Tajik, M.; White, P. S.; Wong, C.-M. Can. J. Chem. 1987, 65, 2744. (b) Steudel, R. In Chemistry of the Non-Metals; Walter de Gruyter: Berlin, New York, 1977. (9) (a) Minkwitz, R.; Lekies, R. Z. Anorg. Allg. Chem. 1987,544, 192. (b) Shum, L. G. S.; Benson, S. W. Int. J.Chem. Kinet. 1983,15(5), 433. Benson, S. W. Chem. Reo. 1978, 78, 23. (10) (a) Kato, S.; Hattori, E.; Mizutta, M.; Ishida, M. Angew. Chem., Int. Ed. Engl. 1982,21,150.(b)Minkwitz, R.; Preut, H.; Sawatzki, J. 2. Naturforsch. 1988,438,399. (c) Guaraldi, G.; Ciuffariu, E. J. Org. Chem. 1970,35,2006. (d) Maloof, M.; Smith, S.; Soodak, M. Mech. React. Sulfur Compd. 1969, 4, 61. Frankel-Conrat, H. J. Biol. Chem. 1955,217, 373.
0001-4a42/~9/0122-0234$01.50/0 0 1989 American Chemical Society
Sulfur and Selenium Iodine Compounds
Acc. Chem. Res., Vol. 22, No. 7, 1989 235
Figure 2. S71+cation in S71SbFs.
Figure 1. Se13+cation in SeIaMF6(M = As,Sb). Weak contacts are indicated as here and in other figures.
S71MF6(M = As, Sb), vastly different from those calculated for SI3MF6. Subsequently S71MF6were prepared quantitatively, in liquid SO2 or AsF, solution16 according to eq 6-8.
4s '/ 8 8
temperature stable covalent S-I bond has so far been characterized: although the human thyroid is thought to contain such a compound.lW It is likely part of a protein, isolated from other S-I bonds, and thus kinetically stabilized. The first synthesis and structure of a stable neutral iodo selenide RSeI (R = 2,4,6tB~3C6H2) recently been reported;'l is it likely kihas netically stabilized by the bulky substituent (cf. kinetically stable RP=PR and R2Si=SiR2; R bulky groups).l2
+ 312 + 10SbF5
6S7ISbF6 + (SbFs)&bF,
= ca. 19)
2s71AS.F~+ [ ( X - 14)/8]s8 (8)
Attempts were made to prepare S81+according to eq 9 and 10, but S71AsF6was formed in both reactions; I2 and s (for eq 9), and KAsF, and s (for eq lo), were 8 a also quantitatively produced. Presumably S81+is I3ASF6 + nS8 Ss(ASF6)z + KI
Sulfur and Selenium Iodine Cations Preparation and Characterization of Se13MF6 (M = As, Sb). Although neutral binary selenium iodides are unstable, salts of Se162-6 have been known for some time. More recently we prepared SeI3MF6 according to eq 4 and 5 as well as various other routes.13
S7IASF6 + 12 + ( n - 7/8)s8 (9) (10)
S7IAsF6 + KASF6 + '/es8
+ 312+ 3AsF5 SOZ(1) 6Se + 91z + 10SbF5
6Se13SbF6+ (SbF3)3.SbF5 ( 5 )
The heat of reaction 4 was estimated to be -100 kJ The Se-I bonds in Se13+ are probably only slightly stronger than those in Se12. The crystal lattice energy stabilizes Se13AsF6substantially.sa Presumably SeI,2- salts are also stabilized by their crystal lattice energies. The X-ray crystal structures of SeI3MF6 confirmed the identity of the pyramidal Se13+cationsa (Figure 1). The average selenium-iodine bond distance is 2.510 A and is similar to the sum of the covalent radii of Se and I (2.493 A). SBr3MF614 (N.B. SBr4 is not known) and Te13MF68a*15 also been prepared. have Preparation of S71MF6, (S71)4S4 ( A S F ~ ) and ~, (S71)21(SbF6)3.2AsF3. The successful preparations of salts of Se13+led naturally to the attempted synthesis of the analogous SI3+compounds.' Initial reactions were carried out with an excess of sulfur relative to the stoichiometric amounts indicated in eq 4 and 5. The chemical analyses of these products corresponded to
(11) du Mont, W.-W.; Kubiniok, S.; Peters, K.; von Schnering, H.-G. Angew. Chem., Int. Ed. Engl. 1987,26, 780. (12) (a) Cowley, A. H. Ace. Chem. Res. 1984, 17, 386. (b) West, R. Angew. Chem., Int. Ed. Engl. 1987,26, 1201. (c) Yoshifuji, M.; Shima, I.; Inamoto, N.; Hirotsu, K.; Higuchi, T. J. Am. Chem. SOC. 1981,103, 4581. (13) Passmore, J.; Taylor, P. J.Chem. SOC., Dalton Trans. 1976,804. (14) (a) Passmore, J.; Richardson, E. K.; Taylor, P. Inorg. Chem. 1978, 17,1681. (b)Murchie, M. P.; Passmore, J. Inorg. Synth. 1986,24,76. (c) Brooks, W. V. F.; Passmore, J.; Richardson, E. K. Can. J. Chem. 1979, 57, 3230. (d) Passmore, J.; Richardson, E. K.; Whidden, T. K.; White, P. S. Can. J. Chem. 1980,58, 851. (15) Passmore, J.; Sutherland, G.; White, P. S. Can. J. Chem. 1981, 59, 2876.
formed initially but it disproportionates to the more thermodynamically stable (estimated at ca. 26 kJ mol-') S71MF6and sulfur.16 An important factor in the stability of S71+, relative to S81+, the lower ionization is 8 energy of SI (836.4 kJ mol-') relative to that of s (872.4 kJ m0l-I).l7 The ionization energy of S5 (830 kJ mol-') is less than that of s (868 kJ mol-') or S4 (1000 kJ 6 and consistently the radical cation S5+has been detected in solution but not S4+or S6+.l8 Other S7+ derivatives to have been characterized are S71+,1J6 S7Br+,19(S71)213+,20 S7+-S6-S7+(S192+).21 This and suggests that the odd unipositively charged rings SI+ and S5+(with or without substituents) are most stable than the even-membered rings S8+or s6+.This is in contrast to the situation for neutral rings, where evenmembered rings are the most stable, with the stability 8 sequence s > s6 > S7 >>> S5 (not isolated).22 We have been unable to synthesize S81+(and S8Br+)1v16y19 (and S6Br+),19,20 our attempts to make and or S61+ (and S5Br+)led to the isolation of (S71)4S4(A~F6)6,20,23 (S71) (SbF6) 2I 3.2AsF3,20 (S7Br)4S4(A~F6) These and 6. l9 compounds were prepared quantitatively according to eq 11 and 12 from sulfur, iodine, and the corresponding pentafluoride.
(16) Passmore, J.; Sutherland, G.; Taylor, P.; Whidden, T. K.; White, P. S. Inorg. Chem. 1981,20, 3839. (17) Wagman, D. D.; Evans, W. H.; Parker, V. B.; S c h u " , R. H.; Halow, I.; Bailey, S. M.; Chruney, K. L.; Nuttall, R. L. J.Phys. Chem. Ref. Data 1982, Suppl 2, 11. (18) Low, H. S.; Beaudet, R. A. J. Am. Chem. SOC. 1976, 98, 3849. (19) Passmore, J.; Sutherland, G.; Whidden, T. K.; White, P. S.; Wong, C.-M. Can. J. Chem. 1985,63,1209. (20) (a) Passmore, J.; Sutherland, G.; White, P. S. J. Chem. SOC., Chem. Commun. 1979,901. (b)Passmore, J.; Sutherland, G.; White, P. S. Inorg. Chem. 1982,21, 2717. (21) Burns, R. C.; Gillespie, R. J.; Sawyer, J. F. Inorg. Chem. 1980,19,
1423. (22) (a) Steudel, R.; Mausle, H.-J. 2.Anorg. A&. Chem. 1981,478,156 and 177. (b)Steudel, R.; Strauss, R.; Koch, L. Angew. Chem., Int. Ed. Engl. 1985, 24, 59. (23) Passmore, J.; Sutherland, G.; White, P. S. J. Chem. SOC.,Chem. Commun. 1980, 330.
236 Acc. Chem. Res., Vol. 22, No. 7, 1989
+ 212 + 9AsF5
+ 312 + lOSbF5
2( (S7I)2I) (SbF6)3*2AsF,+ (SbF3)3SbF5 (12)
Klapotke and Passmore
(S71)4S4(ASF&rj + 3AsF3
We were surprised to find that Sz+ had been formed in reaction 11, as a large excess of AsF6 in SOz or AsF, was oxidizes sulfur only to Sg2+. In fact S42+ only pre8 pared by heating s and liquid SbF6 at 120 "C for several days!24-26 It was postulated that iodine facilitates the oxidation to S4'+ and subsequently s was 8 by quantitatively oxidized to S4(A~F6)2 AsF5 in liquid SOz in the presence of trace halogen (Iz,Br2, C12)within minutes of the reaction mixture warming up to room temperat~re.~~-~~ S t r U C t U r e S O f S7IMF6, (S71)4S4(ASF6)6, and (s7I)zI(SbF6)3.2AsF3 The structures of all four salts were determined by X-ray crystallography. The S71+ cations in both S71MF61p16 salts and in (S71)4S4(AsF6)620 are essentially identical; S71+(Figure 2) consists of a seven-membered homoatomic sulfur ring in a slightly distorted chain configuration similar to those in y- and 6-S7,27 S70,28 Slg2+,21 with an exocyclic iodine. and but The geometries of S70 and S71+ similar, with similar are bond-length a l t e r n a t i ~ n s . These ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ alternations may be viewed as arising from the alternations present in S7and the presence of a positively charged tricoordinate sulfur atom (connected to the iodine).'J6 The extent of the lengthening and shortening is greatest near the source of the perturbation: S(l)-S(7) is very long (2.389 (4) A, bond order of 0.37)31and S(7)-S(6) very short (1.900 (5) A, bond order of 1.76) (Figure 2). In valence-bond terms, the structure can be viewed as consisting of structure A and a number of other resonance structures that delocalize the charge into the ring, the most important of which is B.
Figure 3. (S71)23+cation in (S71)zI(SbF6)3.2AsF,.
Figure 4. The S2I4'+cation in S414(AsF6)z.
Figure 5. The Szand 21z dimers joined via x*-x* interactions to give SzI,2+.
s-s / s-s
The S+-I distances (2.30-2.37 A)16 in all S71+salts (including (S71)4S4(AsF6)6) correspond to a bond order of 1 (sum of covalent radii: 2.37 A) and are all shorter than the s-1 bond length (2.406 (4) A) in (C6H6)3CSI.10" These are the only examples of structural determinations of covalent S-I bonds. (CH )2SISbF6,(CH3)&SbC16,32a (CH3)(CF3)SIMF632 and fi have recently been
(24) Murchie, M. P. Ph.D. Thesis,University of New Brunswick, 1986. (25) Ruff, 0.;Graf, H.; Heller, W.; Knock, Eer. Dtsch. Chem. Ges. 1906, 39, 4310. (26) Gillespie, R. J.; Passmore, J. Adu. Znorg. Chem. Rudiochem. 1976, 17, 49 and references therein. (27) (a) Steudel, R.; Reinhardt, R.; Schuster, F. Angew. Chem., Int. Ed. Engl. 1977,16, 715. (b) Steudel, R.; Steidel, J.; Pickardt, J.; Schuster, F. 2. Nuturforsch. 1980,35B, 1378. (28) Steudel, R.; Reinhardt, R. Angew. Chem., Znt. Ed. Engl. 1977,16, (29) Steudel, R. Angew. Chem., Znt. Ed. Engl. 1975,14, 655. Steudel, R.; Schuster, F. J.Mol. Struct. 1978,49, 143. (30) Gillespie, R. J. Chem. SOC. Rev. 1979, 8, 315. (31) Campana, C. F.; Lo, F. Y.-K.;Dahl, L. F. Inorg. Chem. 1979,18, 3060. (32) (a) Minkwitz, R.; Preutzel, H. 2. Anorg. A&. Chem. 1987,548, 97. (b) Minkwitz, R.; Werner, A. 2. Nuturforsch. 1988,43E, 403.
reported to be stable up to -20 O or -35 OC, respecC tively, and were characterized by Raman and NMR spectroscopy. (S71)4S4(AsF6)6 contains discrete S71+ Sz+ cations and and AsF, anions.m The+ cation has a square-planar S : geometry similar to those in S4(AsF6)z.0.6S0~3 and cation (S7Br)4S4(&F6)2.19 structure of the (S7&I3+ The consists of two equivalent S71+ units that have geometries similar to those observed in S71MF616and (S71)4S4(~F6)620*233), joined by a bridging iodine (Figure atom. The structure is approximately described in and terms of two resonance structures S7IZ2+ S71+,and thus the bridging sulfur-iodine bond (2.674 (7) A) has a formal bond order of 0.5. The I(l)-S(6) intercationic distance (3.394 (3) A) in S71+itself and the corre(3.381 sponding I(l)-S(3) interaction in (s7I)J3+ (9) A) are significantly less than the corresponding sum of the van der Waals radii of 4.0 A. In addition, the bridging iodine atom also has a very weak contact with each of the S71+ units (1(2)-S(6), 3.777 (8) A). Thus both S71+ and (s71)z13+ cluster-like characteristics. have Preparation and Structure of S214(AsF6)2. an In attempt to prepare other salts of novel sulfur-iodine cations (e.g., s21(AsF6),cf. SzFAsF6), we reacted s4(AsF,), with an excess of iodine. One product was characterized as SZI4(AsF6), was subsequently and synthesized quantitatively in liquid sulfur dioxide according to eq 13.33 Systematic attempts to prepare f/4S8 + 212 + 3AsF5 SzI,(AsFrj)z + AsF3 (13)
S13AsF6were unsuccessful. All reactions lead to S&(AsF,),(s) and Iz(s) although S13AsF6formation was
(33) Passmore, J.; Sutherland, G.; Whidden, T. K.; White, P. S. J. Chem. SOC., Chem. Commun. 1980, 289.
Sulfur and Selenium Iodine Compounds
Acc. Chem. Res., Vol. 22, No. 7, 1989 237
.. .- . .. . - . .... .
0 - I
Figure 6. The Se2142+ cation in Se21,(Sb2Fll)2.
Figure 8. Two Se12+radical cations joined via weak n*-a* interactions to give Se2142+.
SezIz+cation has an eclipsed structure (Figure 6) simwith two SeI, units 'oined by a ilar to that of S204z-42 weak selenium-selenium bond (2.841 (2) ) and very weak iodine-iodine interactions (3.756 (2), 3.661 (2) A) (Figures 6 and 8)41(i.e., the structure of Sez142+ only n Bonding , n; Nonbonding l'ly Antibondlng superficially resembles that of Sz142+). Occupancy 2 e 2c le The Sez142+ cation may be regarded as two SeIz+ Figure 7. The Se12+radical cation n MOs derived from pz selradical cations, joined, in part, by overlap of the odd enium and iodine AOs. electron in each of the P* SeIz+ molecular orbitals (Figure 7) resulting in some bonding between all six shown to be thermodynamically feasible; the disproatoms and a formal selenium-iodine bond order of portionation to solid Sz14(AsF6)z Iz(s) was shown and 1.25i4l (Figure 8). Consistently, the selenium-iodine to be even more thermodynamically f a v ~ r a b l e . ~ ~ - ~ ~ bond distances in Sez142+ (2.436 (2)-2.450 (2) A) are The structure of Sz14(AsF6)z consists of discrete SzI?+ significantly shorter than those in Sei,' (2.510 (2)-2.513 and AsF6- ions with weak anion-cation contact^.^^^^^ and (2) A).& Thus Sz142+ Sez12+are different from one The SzI2+ cation has a unique distorted right triangular another, and from their isoelectronic counterparts PZI4= prismatic structure (C, symmetry) (Figure 4)33and, and A s ~ Iwhich have classical all Q eclipsed geome~ , ~ ~ unexpectedly, does not have the structure as the isotries. However, they both are cluster-like, and both electronic Pz14.34 contain npr-npa (n > 2) and P*-P* bonds. That the The S-S (1.828 (1) A,cf. Sa,Sz(X3Cg) 1.8894 A3,) and structure of Sez12+is different from Sz12+ supports our 1-1 bond lengths (2.597 (2) A,cf. Iz(g),2.662 As and 12+, contention that the geometry and bonding in the sulfur 2.557 (4) A37)in Sz142+ indicate bond orders of 2.33 and cation are dependent on the equality of the IEs of Sz 1.33, respectively. The S-S bond distance in this cation and Iz. is the shortest reported. The S-I distances of 2.858 (6) Preparation O f ( S e 6 I ) , ( A S F 6 ) , and S e & ( A S F 6 ) p A and 3.195 (6) A are comparable to sulfur-iodine Several allotopes of sulfur and their derivatives have distances in sulfur-iodine charge-transfer complexeszc been isolated and characterized, including S, (n = 6-13, (2.629 and are longer than that in [(HzN)zCS]zI+ 18,20, and SnO (n = 6-10),28 and SlzOzin which has a formal S-I bond order of 0.5, but they are S1202-2 SbC15,46 and in various cations, for example, less than the sum of the van der Waals radii (4.00 A).39 S71+,16 S7Br+,19 (s71)z13+,20 and (s8)zAg+.47 Selenium The Sz142+ cation may be regarded as consisting of forms48aonly the unstable rings Seg,48bSe6,48c and S2@+and 218%' units, weakly bonded together via two Se7,48d3e349 addition to polymeric gray selenium. Dein mutually perpendicular sets of P*-P* orbitals (Figure rivatives of selenium rings had not been reported prior 5) by electrons in P* orbitals. Thus, P bonding in the to our work. (S71)4(S4)(A~F6)620~23 greatest has the cation is maximized. The equidistributions of charge thermal stability, of the salts containing sulfur-iodine over all three dimer units (S2.66+,2I,O."+) and the recations, and therefore we attempted to prepare (Se7sulting bonding situation may arise from the near I)4Se4(AsF6)6.50 Selenium and iodine were reacted with equality of the ionization energies of Sz (9.36 eV) and AsF, in liquid AsF3 as indicated in eq 11 but by using Iz (9.3995 eV).40 Therefore, Sz142+ an example, par is selenium instead of sulfur. However, the reaction excellence, of a thermodynamically stable species that proceeded according to eq 15, and subsequently Se61contains a npr-npr (n I 3) bond.7 Preparation and Structure of S e z 1 4 ( S b z F 1 1 ) 2 . It was postulated that the structure and bonding in Sz12+ (41) Nandana, W. A. S.; Passmore, J.; White, P. S.; Wong, C.-M. J. were a consequence of the near equality of the ionizaChem. SOC., Chem. Commun. 1982, 1098. tion energies (IEs) of Sz and Iz. To test this hypothesis, (42) Dunitz, J. D. Acta Crystallogr. 1956,9, 579. Kiers, C. Th.; Vos, A. Acta Crystallogr. 1978, B34, 1499. we attempted the synthesis of Sez12+(IE of Sez = 8.33 (43) Baudler, M.; Stassen, H.-J. 2.Anorg. Allg. Chem. 1966,343,244. eV),40 and subsequently Se214(Sb2F11)2 prepared was (44) Steudel, R. Top. Curr. Chem. 1982,102,149. Steidel, J.; Steudel, according to eq 14 and its structure determined.4l The R. J. Chem. Soc., Chem. Commun. 1982, 1312.
(34) Leung, Y. C.; Waser, J. J . Phys. Chem. 1956, 60, 539. (35) Fink, E. H.; Kruse, H.; Rameay, D. A. J . Mol. Spectrosc. 1986, 119. 337. (36) Karle, I. L. J. Chem. Phys. 1955, 23, 1739. (37) Davies, C. G.; Gillespie, R. J.; Ireland, P. R.; Sowa, J. M. Can. J. Chem. 1974,52, 2048. (38) Lin, G. H.-Y.;Hope, H. Acta Crystallogr. 1972, B28, 643. (39) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry,4th ed.; Interscience: 1982. (40) Rosenstock, H. M.; Droxl, K.; Steiner, B. W.; Herron, J. T. J. Phys. Chem. R e f . Data 1977, Suppl. 1 , 6 .
(45) Steudel, R. Comments Znorg. Chem. 1982,1, 313. (46) Steudel, R.; Steidel, J.; Pickardt, J. Angew. Chem., Znt. Ed. Engl. 1980, 19, 325. (47) Roesky, H. W.; Thomas, M.; Schimkowiak, J.; Jones, P. G.; PinChem. Commun. 1982, 895. kert, W.; Sheldrick, G. M. J. Chem. SOC., (48) (a) Cherin, P.; Unger, P. Inorg. Chem. 1967,6,1589. (b)Cherin, P.; Unger, P. Acta Crystallogr. 1972, B28, 313. Marsh, R. E.; Pauling, L.; McCullough, J. D. Acta Crystallogr. 1953,6, 71. Foss, 0.;Janickis, V. J. Chem. SOC., Dalton Trans. 1980, 624. (c) Miyamoto, Y. Jpn. J. Appl. Phys. 1980, 19, 1813. (d) Steudel, R.; Strauss, E. M. 2.Naturforsch. 1981,36B, 146. (e) Steudel, R.; Papavassiliou, M.; S t r a w , E.-M.; Laitinen, R. Angew. Chem., Znt. Ed. Engl. 1986, 25, 99. (49) Steudel, R.; Papavassiliou, M.; Krampe, W. Polyhedron 1987, 7(7), 581. (50) Nandana, W. A. S.; Passmore, J.; White, P. S. J. Chem. Soc., Chem. Commun. 1983, 526.
238 Acc. Chem. Res., Vol. 22, No. 7, 1989
Klapotke and Passmore
to that in [Phzses2'],53 which has an Se6ring with a boat conformation. The iodine substituents are in the 1,4 axial positions. The tricoordinate selenium atoms in the Se6IZ2+ cation are positively charged, but there is delocalization of charge into the ring, resulting in bond alternation51 (2.482 (2) A, 2.227 (2) A) and the formation of 4pv4pn bonds. Each of the two iodine atoms makes two intraionic contacts with both the dicoordinate, but partially charged, selenium atoms within the ring and the contacts (3.719 (2) A and 3.709 (2) A) that are substantially shorter than the sum of the van der Waals radii of Se and I (4.15 A). Thus the Se6IZ2+ a defhas inite distorted cube cluster-like geometry, which it probably retains in solution (see below). Identification of Se412+ Se612+ 77Se and by NMR Spectroscopy in Solution. The characterization of the sulfur-iodine cations is seriously hindered by the lack of a suitable spectroscopic technique, but 77Se NMR can be used in the selenium system. We therefore systematically searched for selenium-iodine cations in SOz solution, by natural abundance 77SeNMR. As part of this investigation, we followed the reaction of Se42fwith varying amounts of I2 and found that the equation proceeded according to eq
Figure 9. View of the polymeric cations (SesI+), in (SesI),.n( AsF,), representing weak intercationic selenium-selenium contacts (3.591 (3) A).
.=I Figure 10. Se81,2+ cation in SesIz(AsFs)z.2SOz.
AsF6 was prepared quantitatively according to eq 16 in liquid SOz solution.50 Large crystals of Se61AsF6can 32Se + 21z + 9AsF5 ~ S ~ G I A S FSe8(AsF6):,+ 3AsF3 (15) +G 12Se + I2 + 3AsF5
Se4(AsF6)z 212 --* Se414(ASF&
be prepared by thermally cycling the reaction mixture. The crystals appeared ruby red in transmitted light and have a golden appearance in reflected light. The successful synthesis of the polymeric (Se61),n(Ad?,) suggested that Se612(AsF6)2 might be preparable. Subsequently, selenium and iodine were reacted with AsF5 in liquid SOz according to eq 17.51 6Se + Iz i- 3AsF,
SeGIz(AsF6)z-2S0z AsF3 + (17) The 77SeNMR of this solution showed the presence of about 11 different selenium cations (see below), and many attempts to produce crystals at room temperature 2se4142++ 2Se13++ Se6IZ2+ (19) were unsuccessful. However, when the solution was cooled to -70 "C for 1 h and left at room temperature, The 77SeNMR spectra of solutions of Se612(AsF6)2 at then 80% of the selenium crystallized out as highly various temperatures show that Se6IZ2+ itself is in crystalline Se61z(AsF6)z~2S0z.51~52 according to eq 20.62 Structure O f (Se61)n.n(ASF,) and S ~ ~ I Z ( A S F ~ ) Zequilibrium with SeJ+ and Se4142+ -~so2.The structure of (Se61),.n(AsF6)consists of AsFG2Se612+* SeS2+ Se41d2+ + (20) anions and polymeric strands of [Se61+],cations with It is also in equilibrium with several other species, which some cation-cation and cation-anion interactions50 are presently under investigation. Se412+and Se612+ (Figure 9). The cation contains hexaselenium rings in undergo Se+-I and Se-Se bond redistribution reactions, a chair conformation similar to that of cyclohexaand since the various combinations do not differ greatly selenium.48c The rings are joined to two neighboring in enthalpy, the formation of the large number of hexaselenium rings by two weak (2.736 (3) A) exocyclic species is probably entropy driven. It is likely that the 1,4 diaxial Se-I bonds (Figure 9) of bond order ca. 0.5; sulfur-iodine cations also give complex equilibrium each tricoordinate Se atom carries a charge of 0.5. The mixtures in solution. [Se6I+In cation was the first example of a derivative of Chloro a n d Bromo Cations of S u l f u r and Selea selenium ring; and it is also polymeric, unlike the nium. Al binary chalcogen-chlorine cations of the type l known sulfur-iodine cations. XC13+ (X = S, Se, Te) have been prepared, and the The discrete centrosymmetric Se6IZ2+ cation contains hexaselenium rings of chair conformation (Figure 10) (53) Faggiani, R.; Gillespie, R. J.; Kolis, J. W. J . Chem. Soc., Chem. similar to those in Se648b [Se61+],,wbut in contrast and Commun. 1987, 592.
(51) Passmore, J.; White, P. S.; Wong, C.-M. J . Chem. SOC.,Chem. Commun. 1985, 1178. (52) Wong, C.-M. Ph.D. Thesis, University of New Brunswick, 1988. (54) Burns, R. C.; Chan, W. L.; Gillespie, R. J.; Luk, W. C.; Sawyer, J. F.; Slim, D. R. Inorg. Chem. 1980, 19, 1432. (55) Carnell, M. M.; Grein, F.; Murchie, M. P.; Passmore, J.; Wong, C.-M. J . Chem. Soc., Chem. Commun. 1986, 225.
The major features of the spectrum were two peaks of equal intensity and satellite peaks showing TeJ7Se couplings consistent with an AXX'A' spectrum. These data are consistent with an IzSe+SeSeSe+12 formulation for the cation. In addition to the two major peaks attributable to Se412+, there were three other less intense peaks present, one attributable to Se13+and the other two due to Se612+,whose integrated areas were in a ratio of 1:2. The latter two peaks have satellites due to 77Se-77Secouplings consistent with an AX2X2/Af spectrum and therefore attributable to a symmetric Se6IzZ+.Se4142+ shown to be in equilibrium with was %I3+ and Se612+according to eq 19, and AHo and ASo have been estimated to be 20 kJ mol-' and 60 J K-' mol-1.55
Sulfur and Selenium Iodine Compounds
Ace. Chem. Res., Vol. 22, No. 7, 1989 239
6 s (Or / 8 g
-iI P 7-
6Se) -k 3AsF5 Br2XXXBr(AsF6)+ AsF3 (23)
O = s e
Figure 11. Se9C1+cation in Se9C1SbCls.
less than 1 (Se(l)-Se(2), 2.554 (6) A) to ca. 1.5 (Se(2)-Se(3), 2.211 (6) A). This implies substantial 4pn4pa bonding between Se(2) and Se(3) and, in valencebond terms, suggests that the bonding may be represented by valence-bond structures C and D.
Figure 12. Se2Br6+ cation in Se2Br5AsFs.
The cluster-like geometry of this cation maximizes intracationic contacts, charge delocalization,the number of Se+-Br bonds (cf. MeSeSe+(Me)SeMe)61, Se-Se bond alternation, and P bonding.
@ = 8r
Figure 13. SeSBr3+ cation in Se3Br3AsF6.
structures of the SC13+& and SeC13+57 cations are similar to that of Se13+(Figure 1). Gillespie et al. prepared and characterized by X-ray crystallography the Se7+Se2Clcation in Se9C1SbC&.58This compound contains the first example of a structurally characterized seven-membered selenium ring. The chair conformation of the Se7 ring with the Se2Cl in the endo position is shown in Figure 11. The geometry of the Se7+SeSeC1 is similar to that of Br2Se+SeSeBr(see below) with the Se7+replacing the Br2Se+in Br2Se+SeSeBr. The bromine cations SBr3+ and SeBr3+l k p d are also of interest as these are the simplest binary bromine cations which contain the X(IV)+-Br (X = S, Se) bond. The X-ray Crystal StI'UctureSOf SBr3MF6 and SeBr3MF6 confirm the identity of the pyramidal cationssa (Figure 1). S7BrMF6 and ( S ~ B ~ ) ~ S ~ ( AwereS prepared SF )~ quantitatively by routes similar to those of the iodine counterpart^^^ (see above). The structure of S7Br+in (S7Br)4S4(ASF6)6is very similar to that of s71+ (Figure 2). The first example of an X2Ha15+species (X = chalcogen) is the cation Se2Br5+,which was prepared quantitatively according to eq 21 and 22.59 The Se2Br5+ 4Se + 5Br2 + 3AsF5 2Se2Br5AsF6 AsF, + (21)
cation (C2,, symmetry) contains two trans SeBr2units, linked by a bridging bromine atom at an inversion [X center (Figure 12). Recently Br2X+XXBr(AsF6-) = S, Se] have been prepared quantitatively according to eq@ ,3 '2 containing the Br2X+XXBrcation (Figure 13). The Se-Se bond orders vary from substantially
(56) Edwards, A. J. J. Chem. SOC., Dalton Trans. 1978, 1723. (57) Stork-Blaisee, B. A.; Romers, C. Acta Crystallogr. 1971, B27,386. (58) Faggiani, R.; Gillespie, R. J.; Kolis, J. W.; Malhotra, K. C. J. Chem. Soc., Chem. Commun. 1987,591. (59) Murchie, M. P.; Passmore, J.; White, P. S. Can. J . Chem. 1987,
Conclusions A New Class of Compounds Discovered: The Chalcogen Iodine (and Bromine) Cations. As a result of our unsuccessful attempts to prepare S13(MF6), a large number of novel, stable sulfur and selenium iodine and bromine cations have been prepared quantitatively and their structures have been determined (Table I). Thus, whereas stable neutral sulfur iodides and selenium iodides either are not known (selenium) or can only be prepared at low temperatures (sulfur), sulfur-iodine and selenium-iodine cations have been shown to be unexpectedly numerous. These simple cations have novel structures and bonding arrangements, and it may be argued that the goals outlined at the beginning of this Account have been achieved. The Stable >X+-I (X = S, Se) Bond. The crystal lattice component is likely not the only factor responsible for the stability of the X+-I bonds in the salts described in this Account, as (CH3)2SISbF6, (CH3)2SISbC16, and (CH3)(CF3)SIMF632 stable only at low are temperatures. The S+-I (Br) and Se+-I (Br) bond distances in the more complex chalcogen halide cations are all slightly shorter than the sums of the corresponding covalent radii and the corresponding bond lengths in neutral S(I1)-I (Br) and Se(I1)-I (Br) containing compounds.lOcJ1They are also shorter than the observed or predicted8aX-Hal bond lengths in XHa13+ (X = S, Se; Hal = I, Br).8a The S+-I (Br) and Se+-I (Br) bonds in the chalogen halide cations are therefore presumably stronger than those in corresponding neutral compounds, or simple MHa13+salts and their simple derivatives. In addition, the more complex chalcogen halide cations, as a whole, are probably stabilized by charge delocalization, bond alternation, and halogen-chalcogen intercationic contacts, as a consequence of the presence of X+-I (Br) bonds (much less extensive, or not possible, in Se13MF6,(CH3)2SISbF6, (CHJ2SISbC16, and (CH3)(CF3)SIMF632). Neutral sulfur and selenium bromides and chlorides are stable, consistently; XhHal n+ cations containing an Se-C1 bond (in Se7+SeSeC1)& X-Br bond (in Br2X+XXBr, = S, and X Se)60are observed whereas iodine-containing analogues are not.
(61) Laitinen, R.; Steudel, R.; Weiss, R. J . Chem. Soc., Dalton Trans. 1986, 1095.
(60)Passmore, J.; Tajik, M.; White, P. S. J . Chem. SOC., Chem. Commun. 1988,175.
240 Ace. Chem. Res., Vol. 22, No. 7, 1989
Table I1 M-M Bond Distances and Bond Orders M-M shortest bond order
bond distance, 1.92 (2) 1.828 (1) 1.906 (5) 1.897 (10) 2.223 (5) 2.211 (6) 2.841 (2) 2.292 (4) 2.227 (2)
Klapotke and Passmore
S2142+ S71+ (s71)z13+
Se9Cl+ Se3Br3+ Se21t+ SeJ+ Se6IZ2+
(ref 31) 1.7 2.3 1.8 1.8 1.5 1.6 0.2 1.2 1.5
0 .Se e . I
Figure 14. The structure of SeBzt compared with that of Sez142+.
Examples of Thermodynamically Stable n pan pa (n L 3) Bonds. Charge delocalization and bond alternation lead to the presence of particularly short S e s e and S-S bonds in the homopolyatomic sulfur and selenium halogen cations (Table 11)except for Se2142+. Thus these cations can be regarded as containing examples of thermodynamically stable 3pa-3pa bonds is and 4 p t - 4 ~ ~ bonds. S2142+ particularly remarkable in that it maximizes a bonding and contains 3u and 3 n p r n p t bonds (n 1 3) (Figure 5). The bond order of the S2unit corresponds to 2.33, the highest observed for any isolated non second row element containing compound. It is thermodynamically stable with respect to an all u bonded isomer and also with respect to addition of 12(s)(eq 24).24 Se2142+ also thermodynamis S ~ I ~ ( A S F J ~+ S ) ( s ) ~SI~ASFG(S) ( I~ (24) ically stable with respect to an all u bonded isomer. It consists of two Se12+units, joined by a weak t*-a* interaction (Figures 6 4 , and the dimer contains one 4pa-5pt bond delocalized over the four Se-I bonds. A similar situation is found for many homopolyatomic chalcogen and halogen cations (e.g., X42+, = S, Se, X Te; Hal2+,Hal = Br, I, 142+)7 which also contain thermodynamically stable npa-npa (n I3) bonds. This is in contrast with the neutral group 14 and 15 compounds containing npn-npt (n I3) bonds which are kinetically, but not thermodynamically, stable.I2 This is in part because there is charge localization on adjacent positively charged atoms in the alternative a-bonded (X isomer [12X+X+12] = S, Se). The S+-S+ u bond dissociation energy will be significantly less than that in a normal sulfur-sulfur bond, and in addition, there will be an energy loss accompanying charge localization. Thus the energetics of the u versus a bonds in the cations are very different from those in neutral molecules.62 Presence of T*--?T* Bonds in S2142+ and Se2142+and Its Implications for Bonding in Related Species. The dimers in S2142+ joined by two weak naturally are perpendicular sets of a*-t* bonds (Figure 5). This situation is similar to that found in Id2+ in (NO), and dimers in the solid state, and in a variety of other
Figure 15. T h e HOMO-1 of S2'.
sulfur-containing dimer^.^,^^ Similarly, the two Se12+ units are joined via a six-center two-electron t*-t* bond (Figure 8), similar to the bonding in S2042-.42 The long Se-Se bonds in Se2142+ (2.841 A) and Se82+ (2.83 A)7 are similar in length. In addition, both Se21,2+ and, in Set+,the two tricoordinate formally positively charged selenium atoms and the four selenium atoms to which they are joined have the same eclipsed geometries (Figure 14). Thus the six selenium atoms in S+ e : are also joined by a six-center two-electron a*-a* bond. The geometry of sg2+ is also similar, and the HOMO-1 has been shownMto have a*-a* characteristics (Figure 15). Therefore, the presence of a*-a* bonds of both the four-center two-electron type and the six-center two-electron type are likely to be found in a variety of compounds of the electron-rich elements (e.g., in S4N4, which may be viewed as containing a six-center twoelectron bond about each of the two S-S interactions). Where they have been measured, the strengths of the t*-t* bonds are weak (less than 40 k J In contrast, a high bond energy is associated with t bonding within the monomer. Thus the a*-a* bond formation follows that of the t-bonded fragments.
This Account would not have been possible without the fine, dedicated experimental work of former graduate students and postdoctoral fellows (whose names are given in the references), but especially Dr. Peter Taylor, whose outstanding work opened u p this field, and Dr. Peter White, for the determination of many X - r a y structures, which have often been complicated by MF6disorder problems. W e thank Dr. Neil Burford for many enjoyable discussions and ideas on bonding in related cations (see ref 7), U N B and N S E R C (Canada) for financial support, and the Humboldt Foundaton for a L y n e n Fellowship (T.K.). W e also thank Simon Parsons for his help i n improuing the manuscript.
(63) Banister, A. J.; Clarke, H. Y.; Rayment, I.; Shearer, H. M. M. Inorg. Nucl. Chem. Lett. 1974,10,647. Awere, E. G.; Burford, N.; Mailer, C.; Passmore, J.; Schriver, M. J.; White, P. S.; Banister, A. J.; Oberhammer, H.; Sutcliffe, L. H. J. Chem. Soc., Chem. Commun. 1987, 66. (64) Burford, N., private communication.
(62) Schmidt, M. W.; Truong, Phi. N.; Gordon, M. S. J. Am. Chem. SOC.1987, 109, 5217.
Vol. 12, 1979
Chemistry of Electronegative Elements
Recent Developments in the Chemistry of Some Electronegative Elements
Arzorganisch-Chemisches Institut der Uniuersitat, 0-6900 Heidelberg, West Germany Received August 18, 1978
In 1962 a frontier of chemistry was opened by the discovery of noble gas compounds.l Since then this field has experienced vast development, and many reviews cover the first ten hectic years of worka2 Soon it became obvious that noble gas chemistry had to be seen in close relationship to halogen chemistry and to the chemistry of fluorides and oxyfluorides of certain other, heavier nonmetal elements such as selenium, tellurium, and antimony, especially in their higher oxidation states. Although the total number of original papers dealing with the chemistry and structure of nonmetal fluorides and related species exceeds several thousand, there is still no review available that covers the complete literature. But the structural principles of those compounds are now well-known, especially if the coordination number six is not exceededa3 Xenon hexafluoride is one of the molecules with a formally higher coordination number, if one counts the nonbonding electron pair. Thus, its structure has caused much confusion; the most recent theory, based on physical measurements, is presented here. A more chemical way of solving such a problem would be to prepare other hexavalent xenon compounds like XeL6 and to investigate their structures. Out of many choices of suitable ligands L, only -OTeF5 and perhaps -OSeF5 have been found to be appropriate. Fluorine stabilizes high oxidation states (PtF6,XeF6, IF7,ReF7) for several reasons: its small size and strong electron withdrawal produce a strong, partly ionic bond. This effect can be achieved by the ligand -OTeF5 as well. And fluorine is not likely to be eliminated as F2,since the dissociation energy of Fz is very small. On the other hand, the ligand dimers F5TeOOTeF5,F5SeOOSeF5,FS0200SOzF, and others are readily formed, as the peroxide bonds are quite strong. Thus it was to be expected that no ligand would be better than fluorine, but research in this direction has produced much new information.
symmetry is smaller than predicted by the VSEPR model.* The structure is best described in terms of a mobile electron pair that moves over the faces and edges of the octahedron and thus distorts it in a dynamic manner.g From its vapor pressure it can be concluded that XeF6 is associated in the condensed phase. The association in nonpolar solvents like F5SOSF5,n-C5F12, CF2Cl2,and S02C1F has been verified by 19Fand 129Xe nuclear magnetic resonance. The latter NMR method was first used only recently, but is now the best analytical probe for any kind of xenon chemistry in solution. 129Xe a natural abundance of 26.4% (xenon has enriched up to 60% is commercially available), a nuclear spin of 1/2, and a gyromagnetic factor very close to that of 13C. The 129Xe NMR spectra of many xenon compounds have now been observed.'@14 The results of XeF6 in solution'' a t temperatures below -100 " C have caused much skepticism; however, repetition of the measurements has not allowed any conclusions other than those derived from the first NMR spectra.12J3 These spectra (Figure 1) have to be interpreted in terms of a tetramerization of XeF6 in which both the xenon and the fluorine atoms are magnetically equivalent. Each xenon atom couples to 24 equivalent fluorine atoms and each fluorine atom to 4 equivalent xenon atoms. Thus, the fluorine atoms in the cluster Xe4F24are bound in a nonrigid manner and are equilibrated by a scrambling mechanism that is best named a cogwheel mechanism (Figure 2). Attempts to freeze out the fluorine migration have been made but could not be confirmed.lZJ4 Undoubtedly this cluster has the largest known number of nonrigid bonded ligands, namely 24. The low-temperature configuration of Xe4F24 somewhat is like the high-temperature form of Rh4(CO)12.15J6 In
(1) N. Bartlett, Proc. Chem. Soc. London, 218 (1962). (2) See, e.g., N. Bartlett and F. 0. Sladky, "The Chemistry of Krypton, Xenon and Radon", in "Comprehensive Inorganic Chemistry", Vol. 1, Pergamon Press, Oxford-New York, 1976 p 213. (3) K. Seppelt, Angew. Chem., Int. Ed. Engl., 18, 181 (1979). (4) B. Weinstock, E. E. Weaver, and C. P. Knop, Inorg. Chem., 5,2189 (1966). (5) One can estimate that a xenon octafluoride would be thermodynamically unstable toward loss of fluorine atoms. This would indicate kinetic instability, except at cryogenic temperatures. (6) SF,, SeF,, TeF,, MoF,, WF6, TcF,, ReF,, RuF,, OsF,, RhF6, IrF6, PtF6, UF,, NpFcj, PUF,. (7) W. E. Falconer, M. J. Vasile, and F. S. Stevie, J . Chem. Phys., 66, 5335 (1977). (8) J. Gillespie and R. S. Nyholm, Q. Reu. Chem. SOC., 11,339 (1957). (9) K. S. Pitzer and C. S. Bernstein, J . Chem. Phys., 63, 3849 (1975); U. Nielsen, R. Haensel, and W. H. E. Schwarz, ibid., 61, 3581 (1974); S. Y. Wang and C. C. Lohr, Jr., ibid., 60, 3901 (1974); 61, 4110 (1974). (10) K. Seppelt and H. H. Rupp, Z. Anorg. Allg. Chem., 409, 331,338 (1974). (11)H. H. Rupp and K. Seppelt, Angew. Chem.,Int. Ed. Engl., 13,613 (1974). (12) K. Seppelt and N. Bartlett, 2. Anorg. Allg. Chem., 436,122 (1977). (13) G. J. Schrobilgen, J. H. Holloway, and P. Granger, J . Magn. Resonance, in press. (14) G. J. Schrobilgen, J. H. Holloway, P. Granger, and C. Brevard, Inorg. Chem., 17, 980 (1978).
The Xenon Hexafluoride Structure Problem It is now well established that XeF6 is the highest fluoride of ~ e n o n . ~ , ~ In contrast to XeF2 and XeF4, its structure has prompted 15 years of discussion, but is now almost clear. The fundamental problem must be separated into several parts, as the structure differs in all three physical states. This behavior is different from that of other known hexafluorides, 15 altogether, which are plainly octahedral.6 In the gaseous state XeF6 is monomeric7 but not octahedral, though the deviation from octahedral
Konrad Seppelt is Associate Professor of Chemistry at the University of Heidelberg. He was born in Leipzig, Germany, in 1944. Following undergraduate studies at the University of Hamburg, h e went on to graduate work at Heidelberg, where he received the Ph.D. degree in 1970. His research interests are In the synthetic chemistry of small molecules and in the main group chemistry of the chalcogens, halogens, and noble gases.
0001-4842/ 79/01 12-0211$01.00/ 0 0 1979 American Chemical Society
Accounts of Chemical Research
Figure 1. (Above) 19FNMR spectrum of XeF, in CF3C1at -130 "C (natural isotopic abundance, lZ9Xe= 26.4%). (Middle) '$F NMR spectrum of XeF, in CF3C1at -130 O with an enriched C sample of 12$Xe(62.5%). (Bottom) lZ9XeNMR spectrum in CF2Cl2at -120 "C. The interpretation of these spectra is shown on the right-hand side. Under the assumption of a tetramerization to Xe4F, and the equivalency of four xenon and 24 fluorine atoms, five different species are in solution: '29Xe4FZ4, lZ9Xe3XeFZ4, '29(e2XeZFu,lBXeXe3FZ,and Xe4F,. As only has a nuclear spin of l/z, the first species shows in the '$F NMR a 1:4:6:4:1 quintet (Qt), the second a 1:3:3:1 quartet (Q), the third a 1:2:1 triplet (TI, the fourth a 1:l doublet (D), and the fifth a singlet (S). Many of those lines overlap within the line width, and the intensity of the resulting nine lines depends upon the '*$Xe concentration. In the case of 62.5% the line intensities are (calculated) 3.8:18.8:49.8:84.5:100:85.5:49.818.3:3.8 and (observed) 3.1:18.1:48.883.2:10083.450.1:17.8:2.9. The lZ9Xe spectrum shows a binominal function of 24th order: (calculated down from the center line) 100:92:72.5:48.3:27.2:12.8:4.9:1.6 ...; (observed) 100:93.3:68.8:45,8:29.1:14.5:4.1 The fit between experimental ... and calculated lines is striking. As cannot be shown here, no other kind of multiplicity can explain these spectra.
the dynamic rhodium carbonyl there are no problems concerning the binding forces of such a cluster, as there are bonds between the metal atoms, in addition to mobile carbonyl bridges. In Xe4F24, however, the mobile fluorine bridges alone are responsible for the existence of the cluster. While any xenon-xenon interaction is highly speculative,ls one can, just by counting, find enough orbitals and electrons to form four three-center two-electron bonds, each on one plane. A comparable bonding model is that of B4C14. Certainly the general chemistry of xenon does not call for electron-deficient bonds, though the charge of the xenon in XeF6 is found indeed to be some 1.5 positive units.ls Attempts have failed to measure the Xe-Xe distances in the cluster in solution, mainly because of experimental difficulties. I t is interesting to compare the structure of the cluster in solution with the solid-state structure of XeF@ The cubic modification is the only one (out of four) that has been fully ana1y~ed.l~ unit cell contains 24 The tetrahedral (XeF,+F-), and 8 octahedral (XeF5+F-)6 units. The tetrahedras are comparable to the dynamic cluster in the dissolved state. However, the Xe-Xe distances are quite large (4.2 A) here and do not favor
(15) F. A. Cotton, C. Kruczcynski, and B. C. Shapiro, J . Am. Chem. SOC., 6191 (1972). 94, (16) J. Evans, B. F. G. Johnson, J. Lewis, J. R. Norton, and F. A. Cotton, J . Chem. SOC., Chem. Commun., 807 (1973). (17) The very new Xe2+ion is the first one to have an unquestionable xenon-xenon bond: L. Stein, R. Norris, A. J. Downs, and A. R. Minihan, J. Chem. SOC., Chem. Commun., 502 (1978). (18) T. X. Carroll, R. W. Shaw, T. D. Thomas, C. Kindle, and N. Bartlett, J . Am. Chem. SOC., 1989 (1974) 96,
Figure 2. The cluster Xe4FZ4 two of its configurations. In the in first configuration each xenon holds six fluorine atoms, and in the second four F atoms bind the Xe atoms by Xe-F-Xe bridges. The latter picture resembles very much the (XeFS+F-)4 tetramer in the solid state.lg All fluorine atoms take part in this dynamic process, which is best explained as a cogwheel mechanism. (Reprinted with permission from ref 10, 11, and 13. Copyright 1974, Johann Ambrosius Barth.)
the speculative bonding model in solution. On the other hand, in the crystal the fluorine atoms are in fixed positions. For a simple, binary compound the structural behavior of XeFs is without parallel. The species chemically most similar in behavior is SeF4,because it has (including one electron pair) an odd number of ligands, is highly associated in the liquid phase, and easily forms SeF3+and SeF,- ions, just as XeF, forms XeF6+,XeFa2-,and XeF7-. However, even a t -140 " C SeF, is monomeric in solution, and a freezing-out effect occurs in the fluorine NMR spectrum.20 Xe(OSeF5)2,Xe(OTeF5)2, Xe(OTeF5)r, O=Xe(OTeF5)4, Xe(OTeF& There has been hope that ligands other than fluorine could form single covalent bonds to xenon. Indeed, the
(19) R. D. Burbank and G. R. Jones, J . Am. Chem. SOC., 96,43 (1974). (20) K. Seppelt, 2. Anorg. Allg. Chem., 416, 12 (1975).
Vol. 12, 1979
Chemistry of Electronegative Elements
Figure 3. Crystal structure of Xe(OSeF&. This structure determination is complicated by a threefold disorder along the Se-Xe-Se axis.
first ligand of this type was fluorosulfate.21 However, the compounds FXeOSOzF and Xe(OS02F)2decompose a t room temperature or slightly above, and the fluorosulfate of XeF6 turned out t o be ionic: XeF5+S03F-.zz The situation changed when the substances Xe(OTeF5), and Xe(OSeF5)zwere made.z3t24 These do not decompose below 100 "C, and their covalent character is well established by their lz9XeNMR spectralo and their crystal structuresz5(Figure 3). Once it was shown that the ligands -OSeF5 and -OTeF5 have the capacity of stabilizing high oxidation states, almost like fluorine, a large amount of their chemistry was studied (see below). The process of getting higher valent xenon linked to these ligands turned out to be very difficult and was successful only in the case of the -OTeF5 ligand. Only recently Xe(OTeFJ4, Xe(OTeF,),, and O=Xe(OTeFj)4 as stable, oxygen single-bonded xenon(1V) and xenon(V1) compounds have been made for the first time.26 XeF6 + 2B(OTeF5)3 2BF3 + Xe(OTeF5)6,decomp -20
2 3 2 > C l 5s:
. ~ ______.. . .
Figure 4. lZ9Xe NMR spectrum of O=Xe(OTeF,),, solution in C2F4ClZ, 55.33 MHz. The chemical shift of fi = -5121 against atomic xenon is typical for six-valent xenon. The splitting into quintets ( J x ~= 4 Hz), 13 lines out of a multiplet of 17 (JXe-F -~ = 55 Hz), and the satellites (Jx,125Te = 1281 Hz) give an almost complete structure description. (Reprinted with permission from ref 26. Copyright 1978, Verlag Chemie GMRH.)
XeF4 + 4 / 3 B(OTeF5)3 4 / 3 BF, Xe(OTeF5)4,mp 72 "C (dec) XeOF4 + 4/3B(OTeF5)3 4/3BF3+ O=Xe(OTeF5)4, mp 53 "C The lz9XeNMR of O=Xe(OTeF5)4 is shown in Figure 4. I t gives a full structural description of the molecule by the coupling to axial and equatorial fluorine atoms and to the lZ5Teisotopes. The most interesting species, Xe(OTeF5)s,is a dark red, almost violet crystalline solid. The deep color indicates that the molecule is monomeric, like the yellow XeF6. The color has to do with the nonbonding electron pair on xenon, as Te(OTeF5)6lacks this electron pair and is colorless.z7 Due to the sensitivity of Xe(OTeF5)6 to light and elevated temperatures, structural investigations are difficult but are in progress. The ligands -OSeF5 and -OTeF6 were not found to form stable compounds with krypton. KrF2 could not
(21) N. Bartlett, M. Wechsberg, F. 0. Sladky, P. A. Bullinger, G. R. Jones, and R. D. Burbank, Chem. Commun., 703 (1969). (22) D. D. DesMarteau and M. Eisenberg, Inorg. Chem., 11,2641 (1972). (23) F. Sladky, Angew. Chem. Int. Ed. Engl., 8,523 (1969); F. Sladky, Monatsh. Chem., 101, 1559 (1970). (24) K. Seppelt, Angew. Chem.,Int. Ed. Engl., 11,723 (1972);K. Seppelt and D. Nothe, Inorg. Chem., 12, 2727 (1973). (25) L. K. Templeton, D. H. Templeton, N. Bartlett, and K. Seppelt, Inorg. Chem., 15, 2718 (1976). (26) D. Lentz and K. Seppelt, Angew. Chem., Int. E d . Engl., 17,356 (1978); D. Lentz and K. Seppelt, ibid., 18, 66 (1929). (27) D. Lentz, H. Pritzkow, and K. Seppelt, Angew. Chem., Int. Ed. Engl., 16, 729 (1977); Inorg. Chem., 17, 1926 (1978).
be converted into Kr(OTeF5)2. Only the decomposition products Kr and F5TeOOTeF5 (F5SeOOSeF5) were observed, indicating the intermediate formation of an oxygen-bonded krypton compound, as these peroxides are the typical decomposition products of all xenon compounds as well. Thus KrFz and its ionic derivatives KrF+ and Kr2F3+remain the only krypton compounds known.z8 On the other hand, xenon is known to bind even to nitrogen in FXeN(S02F)2.z9 XeC12 can be made only a t cryogenic temperature^.^^ Despite many attempts, a xenon-carbon bond has not yet been achieved.
Group Electronegativity of the -OSeF5 and -OTeF5 Groups The reason why the -OSeF5 and -OTeF5 groups form the most stable xenon compounds after the simple fluorides has not been fully answered. One may define group electronegativities for these ligands. Certainly electronegativity has no clear meaning, as there are several definitions. The concept becomes even more confusing if electronegativity is extended to whole groups. Even so, a qualitative argument is possible with the valence shell electron pair repulsion model.* This predicts that in a trigonal bipyramid the two axial positions are occupied by the more electronegative atom or ligand, whereas in the case of a square pyramid of the IF, type, the equatorial positions are occupied by the more electronegative ligands. There are numerous examples for this substitutional behavior in case of the trigonal bipyramid. A substitution with -OSeF5 or -OTeF5 on a trigonal bipyramid has not yet been achieved. However, we have replaced fluorine atoms of IF5 with those ligands. All other groups like -OCH3 or -CF3 on IF, indeed prefer the axial p o ~ i t i o n , as ~ - ~ ~ ~
(28) H. Selig and R. D. Peacock, J . Am. Chem. Soc., 86, 3895 (1964); D. E. McKee, C. J. Adams, A. Zalkin, and N. Bartlett, J. Chem. Soc., Chem. Commun., 26 (1973); R. J. Gillespie and G. J. Schrobilgen, ibid., 90 (1974); B. Frlec and J. H. Holloway, Inorg. Chem., 15, 1263 (1976). (29) R. D. Le Blond and D. D. DesMarteau, J. Chem. Soc., Chem. Commun., 555 (1974). (30) C. Y. Nelson and G. C. Pimentel, Inorg. Chem., 6, 1758 (1967); 96, W. F. Howard and L. Andrews, J . Am. Chem. SOC., 7864 (1974). (31) G. Oates and J. M. Winfield, Inorg. Nucl. Chem. Lett., 8, 1093 (1972); J. Chem. Soc., Dalton Trans., 119 (1974). (32) G. Oates, J. M. Winfeld, and 0.R. Chambers, J. Chem. SOC., Dalton Trans., 1380 (1974). (33) D. Naumann, M. Schmeisser, and C. Deneken, J . Inorg. Nucl. Chem., H. H. Hyman Mem. Vol., 13 (1976).
a x i a l region
S e p p e 1t
Accounts of Chemical Research
F ~ < F F
'.;=: O A O
Figure 5. 19FNMR spectra of F,I(OTeF5)5_,and F,I(OSeF5)5_,,schematically with the tellurium and selenium bonded fluorine atoms omitted. The assignment is easily made by the appearance of the axial-bonded fluorine atom in a downfield area. The ligands -OTeF, and -OSeF,, as indicated by 0, do not go into the axial position, where normally the less electronegative ligands have to go. (Reprinted with permission from ref 35. Copyright 1978, Verlag Chemie GMBH.)
predicted. The surprise is that in the compounds F,I(OTeF5)5-, and F,I(OSeF5)5-, the substitutional behavior is inverted. These species were made according to the equations: IC& + 3C10TeF5 I(OTeF&
+ Xe(OTeF5I2 I(OTeF5)5+ Xe3j I(OTeF5), + IF5 F,I(OTeF5)5-,35 IF5 + X POF2-OSeF5 F,I(OSeF5)5-,35
- + 3Cl2
The conformational analysis of the mixed substituted molecules was done by the I9F NMR of the iodinebonded fluorine atoms, as shown in Figure 5 . Fluorine keeps the axial position, meaning that in the sense of the VSEPR model the ligands -OTeF5 and -OSeF5 are more electronegative than fluorine. Kinetic factors cannot explain this abnormal substitutional behavior, as prolonged reaction time and heating do not change the results. Nor can it be a steric effect: there is no reason why F41-OTeF5(eq) should be sterically favored over F4FOTeF5(ax), X-ray structures of the closely and related compounds trans-FzTe(OTeF5)t6 and T e ( 0 TeF6)627 show that there is no important steric hindrance between the quite bulky ligands (Figure 6). The high formal electronegativities of those groups are certainly a result of the inductive effect of five fluorine atoms, possibly enhanced by some (pd), backbonding between oxygen and selenium or tellurium. The latter effect is supported by the structural investigation of F5SOSF5,F5SeOSeF5,and F5TeOTeF5.
(34) K. Seppelt, Chem. Ber., 106, 1920 (1973). (35) D. Lentz and K. Seppelt, Angew. Chem., Int. Ed. Engl., 17, 355 (36) H. Pritzkow and K. Seppelt, Angew. Chem., Int. Ed. Engl., 1 5 , 771 (1976); Inorg. Chem., 16, 2685 (1977).
\ - -
Figure 6. Molecular structure of Te5O4Fzz.Most tellurium(V1) oxyfluorides have the general formula TenOn-1F4n+2. Here the species is best formulated as trans-F,Te(OTeF,),. The environment around the tellurium atoms is always octahedral. (Reprinted from ref 36).
Here indeed a sterically unfavorable eclipsed configuration is maintained, probably by some (pd), double bonding.37 Once the high electronegativity of these groups was established, it became clear that the chemistry of those ligands could become as extensive as that of fluorine (Table Only helium,
(37) H. Oberhammer and K. Seppelt, Angew. Chem., Int. Ed. Engl.,
17, 69 (1978); Inorg. Chem., 17, 1435 (1978).
(38) A. Engelbrecht and F. Sladky, Monatsh. Chem., 96 159 (1965). (39) A. Engelbrecht, W. Loreck, and W. Nehoda, Z. Anorg. Allg. Chem., 360, 89 (1968). (40) F. Sladky, H. Kropshofer, and 0. Leitzke, J . Chem. SOC., Chem. Commun., 134 (1973). (41) F. Sladky and H. Kropshofer, J . Chem. SOC.,Chem. Commun.,
fino (1973). . ,
_ _ . / _ .
(42) E. Mayer and F. Sladky, Inorg. Chem., 14, 589 (1975). (43) K. Seppelt, Chem. Ber., 108, 1823 (1975); 109, 1046 (1976); 110, 1470 (1977).
Vol. 12, 1979
Chemistry of Electronegative Elements
Figure 7. I9F NMR spectrum of HOSFSat -60 "C in CH2C1, solution. This is a typical AB4 pattern. The two strong groups belong to the B4 part: the other lines represent the A part. The line S (singlet) represents the exact chemical shift of the A part. (Reprinted with permission from ref 46. Copyright 1977, Johann Ambrosius Barth.)
neon, argon, and krypton are definitely not able to be bound to these groups. The difference from fluorine is seen in the case of krypton, and in some cases the extremely high oxidation states, as in IF7 and PtF6, could not be established with those groups so far.
HOSF5, HOSeF5, and HOTeF5 The main sources of the -OSeF5 and -OTeF5 groups are the acids. Certainly the most simple derivatives are the most interesting ones, especially in terms of their synthetic value. HOTeF5 is readily made by reaction of telluric acid or its salts with fluorosulfuric a ~ i d , ~ ~ , ~ ~ , ~
Te(OH)6 + 5FS03H 3Se02Fz+ 4HF ClOSF5 + HCl
HZSeO4+ 2HOSeF5 Cla
whereas the selenium compound is made from SeOzF2.45,46 sulfuric acid HOSFShas been prepared very The recently by the method of combining partially positive chlorine (in ClOSFJ with partially negative chlorine (in HC1).46r47 Because HOSF6 decomposes above -60 " C , the chemistry of the -OSF5 group is quite limited, in contrast to that of -OSeF, and -OTeF5. All the materials are strong acids for which several physical properties have been r n e a ~ u r e d . ~ Of, special value ~ ~,~~ is 19F NMR spectroscopy, as the square-pyramidal arrangement of the fluorine atoms gives rise to a typical AB4 pattern. This is true for all derivatives as well (Figure 7).49~50 The chemistry of HOSFE,(the only six-coordinated sulfuric acid) is limited to its decomposition reaction. HOSF5
The driving force of the reaction is the coordination
G. Mitra and G. H. Cady, J . Am. Chem. Soc., 81, 2646 (1959). K. Seppelt, Angew. Chen. Int. Ed. Engl., 11, 630 (1972). K. Seppelt, 2. Anorg. Allg. Chem., 428, 35 (1977). K. Seppelt, Angew. Chem., Int. Ed. Engl., 15, 44 (1976). W. Porcham and A. Engelbrecht, Monatsh. Chen., 102,333 (1971). P. Bladon, D. H. Brown, K. D. Crosbie, and D. W. A. Sharp, Spectrochim. Acta, Part A , 22, 2221 (1970). (50) K. Seppelt, 2. Anorg. Allg. Chem., 399, 65 (1973).
(44) (45) (46) (47) (48) (49)
Accounts of Chemical Research
Figure 8. Molecular structure of Se202F8, determined by as electron diffraction. The structure of Te202FB almost identical. is The Se-Se or Te-'re distances are surprisingly short. (Reprinted with permission from ref 59. Copyright 1974, Verlag Chemie GMBH.)
decrease on sulfur by formation of a sulfur-oxygen double bond. This reaction is very similar to the decomposition of trifluoromethanol, which was made very re~ently.~~,~~ CFBOCl
-120 o c
C12 + HOCF3
In general, the SF5group closely resembles the CF3 group. SF5NHz CF3NH2
HOSeF5, however, is stable to 290 "C and then decomposes in a quite different way: HOSeF5
+ 1/202SeF4 +
The oxidation state +6 for selenium is less favored than in the two neighboring elements S and Te (see below). This effect is certainly a result of the incomplete shielding of the nuclear charge by the first filled d shell (transition-metal contraction) and can be seen as related to the late discovery of the p e r b r ~ m a t e and A s C ~ s~~ about 150 years after corresponding compounds of their neighbors in the periodic system. Thus HOSeF5 is a highly oxidative, aggressive fluorinating agent and is preferably handled in materials like Kel-F and Teflon FEP. Only in the absolute absence of H F does it fail to attack stainless steel. (HOTeF5 even exceeds the stability of HOSeF5 and is by far a weaker oxidizer.) The anhydrofluoride of HOSeF5, SeOF4, was very difficult to synthesize. Absolutely pure NatOSeF5decomposes in vacuo a t 200 OC.j7
K. Seppelt, Angew. Chem., Int. Ed. Engl., 16, 322 (1977). G. Kloter and K. Seppelt, J . Am. Chem. Soc., 101, 347 (1979). A. F. Clifford and L. C. Duncan, Inorg. Chem., 5, 692 (1966). G. Kloter, W. Lutz, K. Seppelt, and W. Sundermeyer, Angem Chem., I n t . E d . Engl., 16, 707 (1977).
(51) (52) (53) (54)
NatOSeF5NaF + SeOF4 SeOF4 probably has the same structure as SOF4trigonal bipyramidal with the double-bonded oxygen in the plane. The unusual coordination number 5 can be considered the driving force for its dimerization to Se202F8, molecule with a four-membered ring ~ n i t ~ ~ p a (Figure 8). The analogous compound of tellurium, Te202F8, was prepared by pyrolysis of Li'OTeFL or B(OTeF,), and studied by electron diffraction as well.57-59The fourmembered-ring unit seems to be a general building principle for the heavier nonmetal elements. Ions like Te208(OH)26- 1208(OH2-,as well as As2OzF82-,all or exhibit a four-membered-ring unit.60-62The recently determined crystal structure of I02F3shows it to be a dimer, 1204F6, with a four-membered ring.63 Such a structure maintains the high coordination number 6 and avoids double-bonded oxygen. The monomer TeOF4 has not yet been detected. Here the necessity of the coordination number 6 is strict. Generally, double bonds on Te(V1) are not known, as arithmetically they would lead to lower coordination numbers. It is unnecessary to use orbital theory to explain this effect. In IOF5 a double bond is possible while the CN of 6 is retained. All other tellurium(V1) oxide fluorides have octahedrally coordinated telluriumz7 (Figure 6). (The more extensively discussed problems of double bonds on silicon can be explained similarly: any silicon double-bonded species would have CN 3, and the latter seems to be impossible in stable compounds under normal conditions. If the favored CN 4 is retained, as in POFBor S02F2, double-bonded species are present.) Conclusion My intention has been to show that in the classical field of the electronegative elements there has been some exciting progress, despite the fact that this field has largely been neglected in recent years in favor of transition-metal chemistry. I regret that many other significant discoveries in this field, such as the development of the halogen fluorides and oxyfluorides, could ~ , not be mentioned in the limited space of this Account. ~~
I acknowledge t h e Deutsche Forschungsgemeinschaft and t h e Fonds der Chemischen Industrie for grants i n aid of research. T h e Deutsche Forschungsgemeinschaft is thanked f o r scholarships for several graduate students.
(55) E. H. Appelmann, J . Am. Chem. Soc., 90, 1900 (1968). (56) AsC1, is made from AsC1, and Clz at -100 "C by UV irradiation. It decomposes at -30 "C. The oxydechloride AsOC1, is more stable. K. Seppelt, Angew. Chem., Int. Ed. Engl., 15, 3 i 7 (1976); 15, 766 (1976); 2. Anorg. Allg. Chem., 434, 5 (1977); 439, 5 (1978). (57) K. Seppelt, Angem. Chem., Int. Ed. Engl., 13,91 (1974); Z. Anorg. Allg. Chem., 406, 287 (1974). (58) H. Oberhammer and K. Seppelt, Inorg. Chem., in press. (59) K. Seppelt, Angew. Chem., Int. Ed. Engl., 13, 92 (1974). (60) 0. Lindquist, Acta Chem. Scand., 23, 3062 (1969). (61) H. Siebert and H. Wegener, Angew. Chem. Int. Ed. Engl., 4,523 (1965). (62) W. Haase, Chem. Ber., 106: 734 (1973); 107, 1009 (1974). (63) L. E. Smart, J . Chem. SOC., Chem. Commun., 519 (19771, and