Article pubs.acs.org/IC
Structural and Energetic Properties of Closed Shell XFn (X = Cl, Br, and I; n = 1−7) and XOnFm (X = Cl, Br, and I; n = 1−3; m = 0−6) Molecules and Ions Leading to Stability Predictions for Yet Unknown Compounds K. Sahan Thanthiriwatte, Monica Vasiliu, and David A. Dixon* Department of Chemistry, The University of Alabama, Shelby Hall, Box 870336, Tuscaloosa, Alabama 35487-0336, United States
Karl O. Christe Loker Research Institute and Department of Chemistry, University of Southern California, Los Angeles, California 90089-1661, United States S Supporting Information *
ABSTRACT: Atomization energies at 0 K and heats of formation at 0 and 298 K were predicted for the closed shell compounds XF, XF2−, XF2+, XF3, XF4−, XF4+, XF5, XF6−, XF6+ (X = Cl and Br) and XO+, XOF, XOF2−, XOF2+, XOF3, XOF4−, XOF4+, XOF5, XOF6−, XO2+, XO2F, XO2F2−, XO2F2+, XO2F3, XO2F4−, XO3+, XO3F, XO3F2− (X = Cl, Br, and I) using a composite electronic structure approach based on coupled cluster CCSD(T) calculations extrapolated to the complete basis set limit with additional corrections. The calculated heats of formation are in good agreement with the available experimental data. The calculated heats of formation were used to predict fluoride affinities, fluorine cation affinities, and F2 binding energies. On the basis of our results, BrOF5 and BrO2F3 are predicted to be stable against spontaneous loss of F2 and should be able to be synthesized, whereas BrF7, ClF7, BrOF6−, and ClOF6− are unstable by a very wide margin. The stability of ClOF5 is a borderline case. Although its F2 loss is predicted to be exothermic by 4.4 kcal/mol, it may have a sufficiently large barrier toward decomposition and be preparable. This situation would resemble ClO2F3 which was successfully synthesized in spite of being unstable toward F2 loss by 3.3 kcal/mol. On the other hand, the ClOF4+ and BrOF4+ cations are less likely to be preparable with F2 loss exothermicities of −17.5 and −9.3 kcal/mol, respectively. On the basis of the F− affinities of ClOF (45.4 kcal/mol), BrOF (58.7 kcal/mol), and BrO2F3 (65.7 kcal/mol) and their predicted stabilities against loss of F2, the ClOF2−, BrOF2−, and BrO2F4− anions are excellent targets for synthesis. Our previous failure to prepare the ClO2F4− anion can be rationalized by the predicted high exothermicity of −17.4 kcal/mol for the loss of F2.
■
INTRODUCTION Halogen fluorides and oxofluorides have been studied extensively because they have a wide range of coordination numbers up to 8 and of formal oxidation states up to +VII.1−4 They are strong oxidizing agents and are amphoteric as they can act as both fluoride ion donors and acceptors. Therefore, halogen fluorides offer a unique opportunity to study a wide range of coordination numbers and oxidation states and represent an ideal case for the study of the relationship of structure and bonding in hypervalent molecules. To this end, a better understanding of their structures and thermochemical properties, including heats of formation and fluoride affinities, is crucial. There have been a number of experimental studies of the structures,5−32 of XFn (X = Cl and Br; n = 1−6) and XOnFm (X = Cl, Br, and I; n = 1−3, m = 0−6) compounds. The structures of ClF, ClF3, and ClF5 have been determined in the gas-phase using microwave spectroscopy and electron diffraction.6,9,15,16 The structures of ClF2+,33 ClF4+,17 ClF6+,20 ClOF2+,30 and ClF4−34 have been determined by X-ray diffraction. The © 2012 American Chemical Society
molecular structures of ClOF3 and ClO3F have been determined by gas electron diffraction13,24 and those of ClOF,35 ClO2F,36 and ClO3F37 from their rotational spectra. The structures of BrF, BrF3, and BrF5 in the gas-phase have been determined by microwave spectroscopy and electron diffraction.9,8,11,12 The structures of BrF2+,38 BrF4+,17 BrF6+,20 BrF4−,10 BrOF2+,30 and BrF6−39 have been determined by X-ray diffraction. The crystal structures of BrOF3 and BrOF4− were obtained from [NO2]+[BrF4]−·2BrOF3 and [NO]+[BrOF4−], respectively.22 The gas-phase structure of BrO3F was determined using electron diffraction.13 The structure of BrO3F2− was determined from an X-ray crystallography study of [NO]2[BrO3F2][F].23 Solid-state structures of IOF3 and IO2F were determined by X-ray crystallography25,26 and the gas-phase structure of IOF5 by a combined microwave-electron diffraction study.40 The structures of IOF4−, IOF6−, and IO2F2− have been determined by X-ray crystallography Received: July 3, 2012 Published: September 25, 2012 10966
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Figure 1. Calculated CCSD(T)/aVTZ geometries of the chlorine fluoride molecules and ions (bond lengths in Ǻ and bond angles in degrees).
of CsIOF4,27 N(CH3)4+IOF6−,41 and N(CH3)4+IO2F2−,28 respectively. Although a crystal structure has been reported for IO2F3,42 the structure is that of an oxygen-bridged dimer. Because these compounds are difficult to synthesize and handle, they have often been characterized only by vibrational spectroscopy.43−64 For example, the structure of ClO+ was determined using spectroscopic techniques65−67 in combination with high-level computational methods.68 The ClO2+ and BrO2+ structures have also been determined by X-ray diffraction69−72 and spectroscopic observations,50,73−79 and have been studied theoretically.21,80 There are a number of summaries of the heats of formation of some of these compounds from experiment.81−84 A number of theoretical studies17−23,85−100 for these compounds are also available. The levels of theory used for these studies include density functional theory (DFT), the Gaussian-3 (G3) and Gaussian-3X (G3X) methods, and CCSD(T) (coupled-cluster singles and doubles substitutions with perturbatively connected triples) calculations extrapolated to the complete basis set (CBS) limit. The direct prediction of reliable heats of formation101 requires a high-level treatment of electron correlation using methods such as CCSD(T).102 This approach has emerged as the most accurate computationally affordable method that can be applied to small to moderate sized systems and is considered the “gold standard” for chemical accuracy,103,104 though it is unfortunately limited by its O(N7) complexity in its standard form where N is the number of basis functions. In collaboration with Washington State University, we have been developing a composite approach101,105 for the prediction of the thermodynamic properties of molecules based on molecular orbital theory using coupled cluster methods at the CCSD(T) or higher levels with correlation-consistent basis sets.106 These calculations have been extended to heavier elements using the
new effective core potential/correlation consistent basis sets developed by Peterson and co-workers.107 The resulting total energies are extrapolated to the CBS limit to minimize the basis set truncation error. Smaller corrections for the core−valence (CV) correlation, molecular scalar relativistic (SR) corrections, and atomic spin−orbit corrections (SO) are also included in the total atomization energy (TAE) calculations. A correction for zero-point vibrational energies (ZPEs) is included to obtain zero-point inclusive atomization energies (ΣD0). Given ΣD0 and the heats of formation of the elements, the heat of formation of a given compound can then be calculated. Our composite approach (with no empirical parameters for the electronic energies) assumes that the effects of the smaller corrections are additive to the extrapolated CBS “valence” electronic energies. In general, this composite CCSD(T) approach is capable of achieving near chemical accuracy (i.e., ± 1 kcal/mol with respect to experiment) in thermochemical calculations for chemical systems composed of first and second row elements, as documented for nearly 300 compounds in Feller’s Computational Results Database.101,108 We have used such a computational approach to calculate heats of formation from total atomization energies for the closed shell compounds XF, XF2−, XF2+, XF3, XF4−, XF4+, XF5, XF6−, XF6+, XO+, XOF, XOF2−, XOF2+, XOF3, XOF4−, XOF4+, XOF5, XOF6−, XO2+, XO2F, XO2F2−, XO2F2+, XO2F3, XO2F4−, XO3+, XO3F, XO3F2− (X = Cl, Br, and I). A number of the values for the iodine compounds have been reported previously including IF7 and IF8−.88 These calculations enable the prediction of F+ detachment energies (a measure of the oxidizer strength), fluoride affinities (a measure of Lewis acidity) with the fluoride affinity (FA) defined as the negative of the enthalpy of the reaction A + F− → AF−, and F2 binding energies. 10967
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Figure 2. Calculated CCSD(T)/aVTZ geometries of the chlorine oxofluoride molecules and ions (bond lengths in Ǻ and bond angles in degrees).
■
space with 25 electrons to be handled explicitly. All CCSD(T) valence computations were performed with the core electrons frozen except for the compounds with Br and I. The calculations for the Br and I compounds had all of the electrons correlated and were performed with the aug-cc-pwCVnZ/aug-cc-pwCVnZ-PP basis sets. Additional single point “valence” energy calculations for compounds with Br were carried out with the aug-cc-pVnZ/aug-cc-pVnZ-PP basis set. The atomic energies were calculated at the R/UCCSD(T) level starting with a restricted open shell Hartree−Fock and an unrestricted CCSD(T).114 The converged energies were extrapolated to the CBS limit using two schemes. Feller et al. have discussed the different extrapolation schemes and evaluated them.115 The aug-cc-pVnZ energies were extrapolated using a mixed exponential/Gaussian function of the form
COMPUTATIONAL PROCEDURE
Equilibrium geometries for all of the halogen and oxo-halogen compounds were optimized at the CCSD(T) level with the augmented correlation-consistent polarized valence double-ξ and triple-ξ basis sets.109 Single point energies were calculated with the quadruple-ξ and quintuple-ξ basis sets at the triple-ξ geometries. The augmented correlation-consistent polarized weighted core−valence basis sets with Stuttgart small-core relativistic effective-core potentials (RECP) (aug-cc-pwCVnZ-PP) were used for I,110 following our prior work on iodine complexes.88 Augmented correlation-consistent polarized valence basis sets with effective core potentials (aug-cc-pVnZ-PP) and aug-cc-pwCVnZ-PP (CV calculations) were used for Br,111 augmented correlation-consistence polarized valence basis sets with tight-d functions (aug-cc-pV(n+d)Z) were used for Cl,112,113 and augmented correlation-consistent polarized valence basis sets (aug-ccpVnZ)109 were used for F and O. For Br, the RECP subsumes the (1s2, 2s2, 2p6) orbital space into the 10-electron core set, leaving the (3s2, 3p6, 4s2, 3d10, 4s2, 4p5) space with 25 electron to be handled explicitly, with the (4s2, 4p5) electrons active in valence correlation space. The RECP for I subsumes the (1s2, 2s2, 2p6, 3s2, 3p6, 3d10) orbital space into the 28-electron core set, leaving the (4s2, 4p5, 5s2, 4d10, and 5p5)
E(n) = ECBS + A exp[− (n − 1)] + B exp[− (n − 1)2 ]
(1)
as first proposed by Peterson et al.116 with n = 2(D), 3(T), and 4(Q). In the second approach, the CBS limit was obtained by using a twopoint extrapolation scheme,117
E(lmax ) = ECBS + B /lmax 3
(2)
with lmax = Q, and 5. 10968
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Table 1. Calculated Geometry Parameters for Selected Compounds Compared to Experiment molecule ClF (C∞v) ClF3 (C2v)
ClF2+ (C2v) ClF4− ClF4+
(D4h) (C2v)
ClF5 (C4v)
ClF6+ (Oh) BrF (C∞v) BrF2+ (C2v) BrF3 (C2v)
BrF4+ (C2v)
BrF4− (D4h) BrF5 (C4v)
BrF6− (Oh) BrF6+ (Oh) ClOF (Cs)
ClOF2+(Cs)
ClO2F (Cs)
ClO2+ (C2v) ClOF3 (Cs)
parametera r(Cl−F) r(Cl−Fax) r(Cl−Feq) ∠Fax−Cl−Feq r(Cl−F) ∠F−Cl−F r(Cl−F) r(Cl−Fax) r(Cl−Feq) ∠Feq−Cl−Feq ∠Fax−Cl−Fax r(Cl−Fax) r(Cl−Feq) ∠Fax−Cl−Feq r(Cl−F) r(Br−F) r(Br−F) ∠F−Br−F r(Br−Fax) r(Br−Feq) ∠Fax−Br−Feq r(Br−Fax) r(Br−Feq) ∠Feq−Br−Feq ∠Fax−Br−Fax r(Br−F) r(Br−Fax) r(Br−Feq) ∠Fax−Br−Feq r(Br−F) r(Br−F) r(Cl−F)c r(ClO)c ∠O−Cl−Fc r(Cl−F) r(ClO) ∠O−Cl−F ∠F−Cl−F r(Cl−F) r(ClO) ∠O−Cl−O ∠O−Cl−F r(ClO) ∠O−Cl−O r(Cl−Fax) r(Cl−Feq) r(ClO) ∠O−Cl−Feq ∠O−Cl−Fax ∠Fax−Cl−Feq
calculatedb 1.6396 1.6052 1.7054 87.1 1.552 101.7 1.806 1.602 1.540 105.6 173.9 1.6045 1.6660 85.6 1.5603 1.7713 1.6910 98.7 1.7295 1.8167 85.9 1.7214 1.6704 104.4 168.8 1.8994 1.7071 1.7713 84.5 1.8707 1.6767 1.6961 1.4950 110.3 1.5665 1.3921 108.9 93.9 1.6984 1.4299 114.9 101.6 1.4227 121.4 1.7210 1.6159 1.4137 109.2 96.2 86.0
experiment
molecule
9,63
1.628313 1.598151.58416 1.698, 1.703 87.515 1.566(3)33 96.4(3) 1.771−1.814(1)34 1.618(2)17 1.530(2) 103.08(12) 173.92(13) 1.5716 1.669 86.0 1.550(4)20 1.7589879,63 1.69(2)38 93.5(2.1) 1.7218 1.810 86.2 1.728(3)7,19 1.664(3) 97.5(2) 168.9(2) 1.890,10 1.890(4)39 1.680,11 1.699(6)12 1.780, 1.768(1) 84.5, 85.05(43) 1.854(1)39 1.666(11)20 1.6926(43)35 1.4878(46) 110.56(8) 1.522(2), 1.543(2)30 1.455(2) 105.5 98.7 1.690(7)36 1.419(9) 115.0(3) 101.6(9) 1.385(5),21 1.410(2)78 117.8(4),21 118(9)78 1.713(3)24 1.603(4) 1.405(3) 108.9(0.9) 94.7(2.0) 87.9(1.2)
ClO3F (Cs)
BrOF2+(Cs)
BrOF3 (Cs)
BrOF4− (C4v)
BrO2+ (C2v) BrO3F (Cs)
BrO3F2− (D3h) IOF3 (Cs)
IOF4− (C4v)
IOF5 (C4v)
IOF6− (C5v)
IO2F (C2v)
IO2F2− (C2v)
IO2F4− (D4h)
parametera
calculatedb
experiment
∠Fax−Cl−Fax r(ClO)d r(Cl−F)d ∠O−Cl−Fd ∠O−Cl−Od r(Br−F) r(BrO) ∠O−Br−F ∠F−Br−F r(Br−Fax) r(Br−Feq) r(BrO) ∠O−Br−Feq ∠Fax−Br−Fax r(Br−F) r(BrO) ∠O−Br−Feq r(BrO) ∠O−Br−O r(Br−O) r(Br−F) ∠O−Br−F ∠O−Br−O r(BrO) r(Br−F) r(IO) r(I−Fax) r(I−Feq) ∠O−I−Feq ∠O−I−Fax r(I−F) r(IO) ∠O−I−F r(I−Fax) r(I−Feq) r(IO) ∠O−I−Feq r(IO) r(I−Fax) r(I−Feq) ∠O−I−Feq r(I−F) r(IO) ∠O−I−F r(IO) r(I−F) ∠O−I−O ∠O−I−F r(IO) r(I−Feq)
166.8 1.4091 1.6205 102.1 115.7 1.7069 1.5615 106.5 92.6 1.8108 1.7419 1.5784 107.2 167.4 1.8899 1.5857 94.2 1.6137 115.4 1.5913 1.7503 101.8 115.9 1.6150 1.8553 1.7543 1.9266 1.8701 104.7 94.3 1.9938 1.7615 91.1 1.8454 1.8499 1.7539 97.2 1.8114 1.90137 1.9654 95.9 1.9196 1.7794 99.9 1.7927 2.0238 108.3 92.9 1.7775 1.9137
170.5(4.1) 1.404(2),13 1.4014(6)37 1.619(4),13 1.6195(9)37 100.8(8),13 101.99(4)37 116.6(5),13 115.80(3)37 1.736(4), 1.733(4) 1.549(5) 102.8(3) 89.6(2) 1.820, 1.83922 1.725 1.569 103.3 169.8 1.846(2), 1.912(2)22 1.575(3) 93.4, 92.1 1.595(2),21 1.613578 111.9(1),21 117.578 1.582(1)14 1.708(3) 103.3(3) 114.9(3) 1.602(6)23 1.861(16) 1.71(4)25 1.91(3) 1.84(3) 99.5(1.8) 91.2(1.4) 1.98/1.95(1)27 1.72(1) 89.9/88.1(4) 1.863(4)40 1.817(2) 1.715(4) 98.0(3) 1.75−1.7741 1.82 1.88 94−96 1.903(5)26 1.773(6) 94.6(2) 1.774(2)28a 2.0025(2) 101.98(12) 91.01(7) 1.771(1)e 1.872(1)
a
Bond lengths are given in angstroms (Å) and bond angles are given in degrees (deg). bGeometries optimized using CCSD(T) with the aug-ccpVTZ/aug-cc-pVTZ-PP basis sets. cExperimental rz value from microwave spectroscopy. dSecond experimental is rz value from microwave spectroscopy. eHaiges, R.; Christe, K., unpublished.
The following additional additive corrections to the TAE were used: zero-point vibrational energies (ΔEZPE), core−valence effects (ΔECV) for the Cl and Br compounds for the valence only calculations, a correction for scalar relativistic effects (ΔESR), and spin− orbit corrections (ΔESO). Zero-point vibrational energies were computed using second-order Møller−Plesset perturbation theory,118
MP2/aug-cc-pVTZ. Core−valence (CV) corrections were calculated for all compounds containing Cl at the CCSD(T)/aug-cc-pwCVTZ119,120 and at the CCSD(T)/aug-cc-pwCVTZ/aug-cc-pwCVTZ-PP121 level for compounds containing Br. Scalar relativistic effects were evaluated by using expectation values for the two dominant terms in the Breit− Pauli Hamiltonian, the so-called mass-velocity and one-electron 10969
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Table 2. Observed and Calculated Frequencies (cm−1) at the MP2/aug-cc-PVTZ/aug-cc-pVTZ Level molecular symmetry
mode symmetry
ClF ClF2+
C∞v C2h
σg a1
ClF2−
D∞h
molecule
ClF3
C2v
b1 σg+ σu+ πu a1
b1
ClF4+
C2v
b2 a1
a2 b1 b2 ClF4−
ClF5
D4h
C4v
a1g a2u b1g b2g b2u eu a1
b1 b2 e
ClF6+
ClF6−
BrF BrF2−
BrF3
Oh
Oh
C∞v D∞h
C2v
a1g eg t1u t2g t2u a1g eg t1u t2g t2u σg σg+ σu+ πu a1
b1 b2
frequency 795.1 (785.5)a 885.8 364.4 891.6 483.8 459.0 285.5 782.7 547.1 322.9 739.1 437.7 322.5 837.0 628.1 516.6 184.2 488.5 867.8 546.2 879.2 373.3 507.6 425.8 418.6 262.5 182.5 635.0 225.2 751.2 556.4 491.0 378.0 278.7 514.1 767.1 504.7 299.7 729.9 658.9 934.6 598.0 518.8 348.0 526.8 377.8 746.4 307.1 259.2 191.4 672.4 (664.5)a 456.1 426.4 224.7 690.9 557.8 239.7 625.7 356.8 250.7
experiment
molecule
78363 810,4380644 406, 384 818, 821 47645 470
BrF4+
molecular symmetry
mode symmetry
C2v
a1
a2 b1
760,47 76146 538, 535 328, 332 702, 703 442, 434 328, 364 800,48 80249 571, 574 510, 515 237, 235 475, 475 795, 803 537, 534 829, 822 385, 386 50552 425 417 288 inactive 590
b2 BrF4−
D4h
a1g a2u b1g b2g b2u eu
BrF5
C4v
a1
b1 b2 e
BrF6+
709,53 72254 541, 539 486, 493 375, 375
BrF6−
480, 487 732, 725 482, 484 302, 299 679,50 68851 630, 631 890, 890 582, 590 513, 517 353, 353 52555 384
Oh
Oh
a1g eg t1u t2g t2u a1g eg t1u
ClOF
Cs
t2g t2u a′
ClOF2+
Cs
a′
a″ ClOF3
Cs
a′
289 67063 46045 450 236 672,47 67546 547, 552 235, 242 597, 614 347, 350 252, 242
a″
ClOF4−
C4v
a1
b1
10970
frequency 762.1 638.7 383.8 151.4 403.7 748.6 432.9 764.3 276.8 520.8 322.6 233.5 446.6 173.6 516.5 181.5 699.2 587.5 373.5 314.9 555.2 233.9 658.6 420.2 240.0 692.8 682.0 793.2 442.4 408.7 272.1 564.5 442.9 575.9 211.3 235.0 166.7 1174.5 (1039.9)a 585.6 (602.9)a 324.7 (307.4)a 1443.3 (1371.6)a 804.7 (801.0)a 523.7 (499.9)a 380.5 (369.8)a 756.7 (780.6)a 393.7 (369.1)a 1315.2 (1233.3)a 696.8 (681.1)a 490.6 (494.5)a 480.4 (484.8)a 317.2 (324.1)a 225.1 (227.7)a 696.7 (681.2)a 511.3 (505.5)a 419.1 (422.0)a 1281.7 466.0 368.6 364.2 192.8
experiment 72348 606 385 216 704 419 736 369 52356 317 246 449 i.a 542 183 68247 587 369 312 535 644 415 237 65820 668 775 427,433 405 565,55 56157 450, 445
240,241 103858 597 310 1334, 132329 734 512 405 694 383 122259 694 482 489 319 224 684 501 414 120360 456 339 356
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Table 2. continued molecule
molecular symmetry
mode symmetry b2 e
ClO2F2+
C2v
a1
a2 b1 b2 ClO2F2−
C2v
a1
a2 b1 b2 ClO3F
C3v
a1
e
ClO2+
C2v
a1
BrOF3
Cs
b2 a′
a″
BrOF4− a
C4v
a1
frequency 257.0 640.2 444.6 201.8 1317.3 791.2 545.2 387.5 383.9 1579.6 534.0 848.7 527.9 1149.1 564.6 363.8 187.2 338.0 1277.0 241.5 561.1 370.0 1118.3 (1067.9)a 705.2 (715.7)a 553.3 (549.1)a 1376.8 (1324.6)a 587.4 (583.3)a 403.7 (402.4)a 1065.7 481.4 1335.1 1102.3 (1009.5)a 628.8 (636.5)a 531.6 (538.9)a 367.1 (357.2)a 237.7 (243.7)a 185.8 (187.6)a 617.7 (611.7)a 414.4 (401.6)a 344.0 (345.5)a 1044.8 492.2
experiment
molecule
278 600 414 194 124129 756 514 390 390 1479 530 830 514 119159 510 363 198 337 1225
molecular symmetry
mode symmetry b1 b2 e
BrO2+
C2v
a1
IOF4−
C4v
b2 a1
b1 b2 e
IOF5 378 106361 717 549 1314 573 414 106523 519 1308 99559 625 531 345 236 201 601 394 330 930,59 93060 500, 499
C4v
a1
b1 b2 e
IO2F4−
D4h
a1g a2u b1g b2g b2u eg eu
frequency 285.5 417.6 173.0 228.4 529.3 385.8 170.9 984.4 331.3 1009.8 958.4 518.7 256.9 471.2 167.2 199.4 513.0 335.1 132.9 985.7 690.7 645.7 356.4 646.5 292.3 235.2 717.5 368.4 346.6 201.3 863.4 560.3 945.3 354.6 544.7 242.4 240.9 374.9 608.1 380.4 166.0
experiment 302, 299 417, 413 205, 178 235, 236 505, 505 395, 395 179, 178 87823 372 943 88760 537 279 480 219 482 374 140 92762 681 640 329 647 307 712 375 341 208 82464 569 885 349 555 255 380 590
Values in parentheses calculated at the CCSD(T)/aug-cc-pVTZ level. ΔESO(Br) = −3.50 kcal/mol, and ΔESO(I) = −7.24 kcal/mol from Moore.129 The calculated ΣD0 values are obtained from the following expression
Darwin (MVD)122 corrections from configuration interaction singles and doubles (CISD)123 calculations. The CISD(MVD)/aug-cc-pVTZ expectation values are generally in good agreement with spin-free oneelectron Douglass−Kroll−Hess computations, for most calculations. Since we employed a RECP for Br and I atoms, there is a possibility of “double counting” the relativistic effects when computing MVD corrections to the energy which already includes most of the relativistic effects. Because the MVD operators mainly sample the core region where the pseudo-orbitals are small, it could be assumed that any double counting is very small.124 Molecular spin orbit corrections (second-order as the molecules are all closed shell singlets) for the I containing molecules were carried out using density functional theory (DFT)125 with the B3LYP exchange correlation functional126,127 and the aug-cc-pVTZ-PP-SO basis set for I128 and the aug-cc-pVTZ for F and O. The atomic spin−orbit corrections are ΔESO(O) = −0.22 kcal/mol, ΔESO(F) = −0.39 kcal/mol, ΔESO(Cl) = −0.84 kcal/mol,
ΣD0 = ΔEelec(CBS) − ΔEZPE + ΔECV + ΔESR + ΔESO
(3)
for the Cl compounds and with the valence only CBS extrapolation for the Br compounds. Equation 4 was used to calculate ΣD0 for the compounds with Br and I when the core−valence corrections are included in the CBS extrapolation, ΣD0 = ΔEelec(CBS) − ΔEZPE + ΔESR + ΔESO
(4)
By combining the ΣD0 with the known heats of formation ΔH f at 0 K for elements, ΔH0f (O) = 58.99 ± 0.02 kcal/mol, ΔH0f (F) = 18.47 ± 0.07 kcal/mol, ΔH0f (Cl) = 28.59 ± 0.01 kcal/mol, ΔH0f (Br) = 28.18 ± 0.01 kcal/mol, ΔH0f (I) = 25.61 ± 0.01 kcal/mol, gas-phase ΔH0f values can be derived for compounds in the current 81
10971
0
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Table 3. Components for CCSD(T) Atomization Energies for XFn (X = Cl and Br; n = 1−6) in kcal/mola ΣD0 (0 K)i
CBS b
reaction
DTQ
ClF → Cl + F ClF2+ + e → Cl + 2F ClF2− → Cl + 2F + e ClF3 → Cl + 3F ClF4+ + e → Cl + 4F ClF4− → Cl + 4F + e ClF5 → Cl + 5F ClF6+ + e → Cl + 6F ClF6− → Cl + 6F + e BrF → Br + F BrF2+ + e → Br + 2F BrF2− → Br + 2F + e BrF3 → Br + 3F BrF4+ + e → Br + 4F BrF4− → Br + 4F + e BrF5 → Br + 5F BrF6+ + e → Br + 6F BrF6− → Br + 6F + e
62.54 −168.45 188.95 127.97 −111.57 265.15 186.01 −46.17 311.66 63.40 −148.25 199.79 151.65 −68.25 301.81 239.99 −1.48 383.05
c
Q5
62.67 −167.98 188.76 128.25 −110.67 264.82 186.39 −45.05 310.95 63.36 −148.43 199.37 151.17 −68.84 300.74 238.94 −2.33 380.86
DTQCVd
ΔEZPEe
ΔECVf
ΔESR
g
ΔESOh
DTQ
Q5
DTQCV
63.23 −149.96 199.41 150.04 −72.34 300.44 236.20 −9.26 380.90
−1.14 −3.06 −2.16 −4.51 −7.61 −5.03 −8.74 −13.22 −8.28 −0.96 −2.57 −1.90 −3.89 −6.52 −4.42 −7.72 −11.16 −7.17
0.12 0.61 0.18 0.26 −0.22 1.02 0.27 −0.12 0.58 −0.03 −1.04 0.02 −0.83 −2.69 −0.25 −2.23 −5.59 0.02
−0.25 −0.42 −0.26 −0.72 −1.58 −0.24 −1.80 −4.03 0.03 −0.04 0.22 −0.24 0.05 0.60 −0.28 0.29 1.06 −0.28
−1.23 −1.62 −1.62 −2.01 −2.40 −2.40 −2.79 −3.18 −3.18 −3.89 −4.28 −4.28 −4.67 −5.06 −5.06 −5.45 −5.84 −5.84
60.17 −172.47 184.90 121.27 −122.48 258.18 173.33 −65.61 300.10 58.48 −155.92 193.39 142.30 −81.92 291.80 224.88 −23.00 369.78
60.05 −172.94 185.09 121.00 −123.38 258.51 172.95 −66.73 300.81 58.44 −156.09 192.98 141.83 −82.52 290.73 223.83 −23.86 367.59
58.35 −156.56 192.99 141.53 −83.33 290.68 223.32 −25.20 367.61
a
The atomic asymptotes were calculated with R/UCCSD(T). bExtrapolated by using eq 1 with aVDZ, aVTZ, and aVQZ (frozen core). Extrapolated by using eq 2 with aVQZ and aV5Z (frozen core). dExtrapolated by using eq 1 with awCVDZ, awCVTZ, and awCVQZ without frozen core approximation. eThe zero point energies for the polyatomics were taken as 0.5 of the sum of the CCSD(T) with aug-cc-pVTZ or aug-cc-pVTZPP harmonic frequencies. fCore−valence correction calculated as the difference in energy between the valence electron correlation calculation and that with the appropriate core electrons included at the CCSD(T) with aug-cc-pwCVTZ or aug-cc-pwCVTZ-PP. gThe scalar relativistic correction is based on a CISD(FC)/VTZ MVD calculation and is expressed relative to the CISD result without the MVD correction, i.e., including the existing relativistic effects resulting from the use of a relativistic effective core potential. hCorrection due to the incorrect treatment of the atomic asymptotes as an average of spin multiplets. Values are based on C. Moore’s Tables. Cf. ref 129. iThe theoretical value of ΔD0 (0 K) was computed with the CBS estimates. c
study. We obtain enthalpies of formation at 298 K by following the procedures outlined by Curtiss et al.86 Heats of formation at 298 K were obtained by combining the atomic thermal corrections (1.04 kcal/mol (O), 1.05 kcal/mol (F), 1.10 kcal/mol (Cl), 2.93 kcal/mol (Br), and 1.58 kcal/mol (I)) with the molecular thermal corrections in the appropriate statistical mechanical expressions.130 All calculations were performed with the MOLPRO2010131 package of ab initio programs. Molecular spin orbit correction computations were carried out with NWCHEM132 program. The DFT transition state calculations for ClF7 and BrF7 were done with the Gaussian09 program system.133
CCSD(T)/ aug-cc-pV(6+d)Z, and the latter value is in excellent agreement with experiment. The computed geometry parameters for ClF3 are in good agreement with the experimental values, with the predicted bond lengths being within 0.007 Å and the bond angle within 0.4°. The structure of the ClF4+ cation from the X-ray crystal structure17 of ClF4+SbF6− is a pseudo trigonal bipyramid. The four fluorine atoms occupy the two axial and two of the equatorial positions with a sterically active lone valence electron pair occupying the third equatorial position. The calculated geometry of the free ion is in very good agreement with the experimental structure. For ClF5, the predicted Cl−Fax bond length is 0.033 Å longer than experiment and the Cl−Feq bond length is 0.004 Å shorter than experiment;6 the calculated bond angle is 0.4° smaller than experiment. The calculated geometry of ClO2+ agrees with the experimental values, with the bond length and bond angle only differing by 0.01 Å and 2.0°, respectively.21,78 For ClF6+ and BrF6+, the experimental21 and predicted bond lengths agree within 0.01 Å. The calculated geometry of ClOF2+ agrees with the X-ray crystallographic determination with the bond length and angle differing by 0.05 Å and 3.0°, respectively.30 For ClOF, ClO2F, and ClO3F, the calculated and experimental35−37 bond lengths and angles are in good agreement with the experimental microwave structures within 0.01 Å for the bond distances and better than 1° for the bond angles. The calculated geometry parameters for BrF and BrF3 are in good agreement with experiment within 0.01 Å and 0.3°.8,9,63 The calculated geometry of BrF4+ is in good agreement with the experimental crystal structure values reported by Christe et al.,7,19 except for the Feq−Br−Feq bond angle which is
■
RESULTS AND DISCUSSION Geometries. The calculated geometries are summarized in Figures 1 and 2 for compounds where the center atom is Cl. The basic molecular shapes are the same for the XFn (X = Cl, Br, and I; n = 1−8) and XOnFm (X = Cl, Br, and I; n = 1−3, m = 1−6) compounds with respect to their central atom, so only the structures of the chlorine fluorides and oxofluorides are illustrated. The only difference between the chlorine and bromine couple and the iodine is the steric activity of the lone valence electron pair on the central atom in the XF6− anions. In ClF6− and BrF6− it is sterically inactive, and the anions are octahedral, whereas in IF6− it is sterically active and the anion is distorted from Oh symmetry.88 Detailed calculated geometry parameters for all compounds are given in the Supporting Information. A comparison of the calculated geometries with those experimentally observed is given in Table 1. ClF has a 1Σ+ ground state with an experimental bond distance of 1.628313 Å.9,63 The calculated bond length improves systematically from 1.6396 Å to 1.6284 from CCSD(T)/aug-cc-pV(T+d)Z to 10972
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Table 4. Components for CCSD(T) Atomization Energies for XOnFm (X = Cl and Br; n = 1−3, m = 1− 6) in kcal/mola ΣD0 (0 K)i
CBS b
reaction
DTQ
ClOF → Cl + O + F ClOF2+ + e → Cl + O + 2F ClOF2− → Cl + O + 2F + e ClOF3 → Cl + O + 3F ClOF4+ + e → Cl + O + 4F ClOF4− → Cl + O + 4F + e ClOF5 → Cl + O + 5F ClOF6− → Cl + O + 6F + e ClO2F → Cl + 2O + F ClO2F2+ + e → Cl + 2O + 2F ClO2F2− → Cl + 2O + 2F + e ClO2F3 → Cl + 2O + 3F ClO2F4− → Cl + 2O + 4F + e ClO3F → Cl + 3O + F ClO3F2− → Cl + 3O + 3F + e BrOF → Br + O + F BrOF2+ + e → Br + O + 2F BrOF2− → Br + O + 2F + e BrOF3 → Br + O + 3F BrOF4+ + e → Br + O + 4F BrOF4− → Br + O + 4F + e BrOF5 → Br + O + 5F BrOF6− → Br + O + 6F + e BrO2F → Br + 2O + F BrO2F2+ + e → Br + 2O + 2F BrO2F2− → Br + 2O + 2F + e BrO2F3 → Br + 2O + 3F BrO2F4− → Br + 2O + 4F + e BrO3F → Br + 3O + F BrO3F2− → Br + 3O + 3F + e
103.78 −98.72 228.74 182.45 −73.03 316.49 222.08 294.11 179.05 −51.50 302.46 218.97 328.14 235.71 344.50 103.44 −88.69 241.39 205.13 −54.71 355.41 261.14 366.76 168.77 −74.20 309.51 224.79 369.86 201.40 343.38
c
Q5
DTQCVd
ΔEZPEe
ΔECVf
ΔESR
g
ΔESOh
DTQ
Q5
0.31 0.22 0.41 0.44 0.46 0.60 0.26 0.09 0.61 0.59 0.72 0.53 0.47 0.88 0.85 −0.46 −1.99 −0.35 −1.75 −4.56 −1.10 −4.57 −4.49 −1.58 −3.76 −1.38 −3.82 −3.70 −3.31 −3.43
−0.66 −1.20 −0.72 −1.61 −3.27 −1.37 −4.14 −4.18 −1.64 −2.84 −1.74 −3.57 −3.78 −3.29 −3.59 0.13 0.63 −0.14 0.37 1.05 −0.03 0.74 0.26 0.52 1.10 0.17 0.80 0.38 0.95 0.57
−1.45 −1.84 −1.84 −2.23 −2.62 −2.62 −3.01 −3.40 −1.67 −2.06 −2.06 −2.45 −2.84 −1.89 −2.28 −4.11 −4.50 −4.50 −4.89 −5.28 −5.28 −5.67 −6.06 −4.33 −4.72 −4.72 −5.11 −5.50 −4.55 −4.94
99.00 −107.69 222.62 171.68 −89.51 305.23 202.67 274.47 169.97 −65.69 292.16 202.42 310.50 221.24 328.58 96.40 −99.62 232.92 192.53 −72.72 342.11 240.79 345.05 158.16 −89.27 297.23 207.44 350.79 186.64 326.49
99.83 −106.23 223.18 172.85 −87.77 305.78 204.05
102.54 −91.72 240.36 202.21 −61.06 352.82 254.62 359.93 166.4 −79.33 306.86 219.38 364.32 196.95 338.48
−2.98 −6.15 −3.96 −7.37 −11.05 −7.87 −12.52 −12.15 −6.38 −9.89 −7.22 −11.05 −11.49 −10.17 −10.89 −2.60 −5.07 −3.48 −6.33 −9.23 −6.88 −10.85 −11.43 −5.21 −7.69 −6.11 −9.21 −10.24 −7.85 −9.11
104.61 −97.26 229.29 183.61 −71.29 317.04 223.45 181.06 48.51 304.07 221.14 329.71 238.70 347.01 103.77 −88.59 241.37 204.93 −55.06 354.57 260.67 169.43 −73.83 309.26 224.88 369.65 202.30 343.86
171.98 −63.11 293.77 204.59 312.07 224.23 331.09 96.72 −99.52 232.90 192.33 −73.07 341.28 240.32 158.82 −88.90 297.47 207.53 350.58 187.54 326.96
DTQCV
95.96 −99.84 232.24 191.35 −73.25 340.63 238.84 342.70 157.38 −89.98 296.20 205.85 348.95 185.49 325.01
a
The atomic asymptotes were calculated with R/UCCSD(T). bExtrapolated by using eq 1 with aVDZ, aVTZ, and aVQZ (frozen core). Extrapolated by using eq 2 with aVQZ and aV5Z (frozen core). dExtrapolated by using eq 1 with awCVDZ, awCVTZ, and awCVQZ without frozen core approximation. eThe zero point energies for the polyatomics were taken as 0.5 of the sum of the CCSD(T) with aug-cc-pVTZ or aug-cc-pVTZPP harmonic frequencies. fCore−valence correction calculated as the difference in energy between the valence electron correlation calculation and that with the appropriate core electrons included at the CCSD(T) with aug-cc-pwCVTZ or aug-cc-pwCVTZ-PP. gThe scalar relativistic correction is based on a CISD(FC)/aug-cc-pVTZ or aug-cc-pVTZ-PP MVD calculation and is expressed relative to the CISD result without the MVD correction, i.e., including the existing relativistic effects resulting from the use of a relativistic effective core potential. hCorrection due to the incorrect treatment of the atomic asymptotes as an average of spin multiplets. Values are based on C. Moore’s Tables. Cf. ref 129. iThe theoretical value of ΔD0 (0 K) was computed with the CBS estimates. c
compressed in the solid state by fluorine bridges to the anions. For BrF4−, the calculated bond length is 0.009 Å longer than that from the X-ray crystal structure10 of KBrF4. For BrF5, the calculated triple-ζ geometry is in good agreement with the X-ray crystal structure11 and the electron diffraction and microwave gas-phase structure.12 The calculated structure for BrO3F is in good agreement with the electron diffraction structure14 within 0.01 Å for the bond lengths and 1° for the bond angles. The calculated geometry of BrOF2+ agrees with the X-ray crystallography determination with the bond length and angle differing by 0.02 Å and 3.0°, respectively.30 The calculated geometry of BrO2+ agrees with experiment with the bond length and angle differing by 0.01 Å and 2.0°, respectively.21,78 For IOF3, the calculated structure is in reasonable agreement with the experimental one, considering the size of the experimental error bars.25 The calculated bond lengths of IO2F and IO2F2− are in excellent agreement with the experimental values.26,28 The calculated structure for IOF4− is in fair agreement
with the average experimental values27 for the I−F bond distance and the ∠O−I−F bond angle, considering the large experimental error bars. However, the calculated I−O bond distance is somewhat surprisingly 0.04 Å longer than experiment. We only report the structure for trans-IO2F4− as the cis-isomer is ∼10 kcal/mol higher in energy. Vibrational Frequencies. The harmonic vibrational frequencies for the molecules with experimental values calculated at the MP2/AVTZ level are given in Table 2, where they are compared with experiment. The harmonic vibrational frequencies for all of the XF n and XO n F m compounds are listed in the Supporting Information. The overall agreement between the computed and the experimental frequencies is very good, especially considering that the calculated values are harmonic values and the experimental values include an anharmonic component. Although the calculated values for ClF2+, ClF4+, ClF6+, and BrF4+ show considerable deviations from experiment, the experimental data are strongly affected by fluorine bridging in the solid state.48−51 10973
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Table 5. Components for CCSD(T) Atomization Energies for IOnFm (n = 1−3, m = 1− 6) in kcal/mola ΔESO
CBS b
reaction
DTQ
IOF → I + O + F IOF2+ + e → I + O + 2F IOF2− → I + O + 2F + e IOF3 → I + O + 3F IOF4+ + e → I + O + 4F IOF4− → I + O + 4F + e IOF5 → I + O + 5F IOF6− → I + O + 6F + e IO2F → I + 2O + F IO2F2+ + e → I + 2O + 2F IO2F2− → I + 2O + 2F + e IO2F3 → I + 2O + 3F IO2F4− → I + 2O + 4F + e IO3F → I + 3O + F IO3F2− → I + 3O + 3F + e
120.65 −36.78 270.73 267.05 37.50 428.82 369.30 511.85 199.89 −20.15 357.73 294.81 478.33 240.05 412.63
Q5
ΔEZPEd
ΔESR
e
atomic
120.79 −35.63 270.80
−2.46 −4.37 −3.34 −5.92 −8.45 −6.48 −10.18 −11.36 −4.75 −6.87 −5.73 −8.00 −9.73 −7.00 −8.42
−0.48 −0.39 −0.60 −0.69 −0.72 −1.04 −1.06 −1.55 −0.43 −0.44 −0.78 −0.76 −1.20 −0.52 −0.92
−7.85 −8.24 −8.24 −8.63 −9.02 −9.02 −9.41 −9.80 −8.07 −8.46 −8.46 −8.85 −9.24 −8.29 −8.68
c
200.22
f
ΣD0 (0 K)g molecular
DTQ
Q5
1.40 1.21 1.01 1.41 2.50 1.38 2.81 2.53 1.20 1.58 1.32 2.09 2.62 1.65 2.08
111.26 170.1 259.56 253.22 136.7 413.66 351.46 491.66 187.83 214.9 344.08 279.30 460.78 225.89 396.69
110.00 258.61
186.96
a
The atomic asymptotes were calculated with R/UCCSD(T). bExtrapolated by using eq 1 with awCVDZ, awCVTZ, and awCVQZ without frozen core approximation. cExtrapolated by using eq 2 with awCVQZ, and awCV5Z without frozen core approximation. dThe zero point energies for the polyatomics were taken as 0.5 of the sum of the CCSD(T) with aug-cc-pVTZ or aug-cc-pVTZ-PP harmonic frequencies. eThe scalar relativistic correction is based on a CISD(FC)/ aug-cc-pVTZ or aug-cc-pVTZ-PP MVD calculation and is expressed relative to the CISD result without the MVD correction, i.e., including the existing relativistic effects resulting from the use of a relativistic effective core potential. fCorrection due to the incorrect treatment of the atomic asymptotes as an average of spin multiplets. Values are based on C. Moore’s Tables. Cf. ref 129. gThe theoretical value of ΔD0 (0 K) was computed with the CBS estimates.
Table 6. Components for CCSD(T) Atomization Energies for XOn+ and XOn− (X = Cl, Br, I; n = 1−3) in kcal/mola ΔESO
CBS b
Reaction
DTQ
ClO+ + e → Cl + O ClO2+ + e → Cl + 2O ClO3+ + e → Cl + 3O ClO− → Cl + O + e ClO2− → Cl + 2O + e ClO3− → Cl + 3O + e BrO+ + e → Br + O BrO2+ + e → Br + 2O BrO3+ + e → Br + 3O BrO− → Br + O + e BrO2− → Br + 2O + e BrO3− → Br + 3O + e IO+ + e → I + O IO2+ + e → I + 2O IO3+ + e → I + 3O IO− → I + O + e IO2− → I + 2O + e IO3− → I + 3O + e
−211.97 −113.13 −84.91 117.64 175.17 262.29 −204.23 (−205.38) −126.42 (−128.35) −119.87 (−123.50) 114.02 (113.97) 168.99 (168.19) 249.42 (247.10) −184.97 −104.65 −93.17 117.14 182.29 278.75
Q5
c
−211.30 −111.17 −82.17 118.10 176.43 264.66 −203.81 −125.57 −118.71 114.35 169.77 250.55 −184.80 −104.17 −92.42 117.46 182.90 279.53
ΔEZPE
d
−1.47 −4.12 −7.26 −0.96 −2.91 −6.43 −1.50 −3.20 −6.20 −0.93 −2.61 −5.37 −1.41 −3.00 −5.64 −0.97 −2.61 −5.08
e
ΔESRf
ΔEcv
atomic
−0.13 0.33 0.47 0.17 0.44 0.75 −0.74 −1.23 −2.61 0.07 −0.46 −1.63
0.00 −0.79 −2.00 −0.50 −0.96 −1.89 0.20 0.64 1.04 −0.09 0.03 0.39 −0.05 −0.15 −0.17 −0.16 −0.34 −0.54
−1.06 −1.28 −1.50 −1.06 −1.28 −1.50 −3.72 −3.94 −4.16 −3.72 −3.94 −4.16 −7.46 −7.68 −7.90 −7.46 −7.68 −7.90
g
ΣD0 (0 K)h
molecular
DTQ
Q5
1.05 1.07 1.48 1.68 1.98 1.32
−214.63 −118.99 −95.19 115.29 170.45 253.22 −209.99 −134.15 −131.80 109.35 162.01 238.64 −192.84 −114.41 −105.41 110.24 173.64 266.55
−213.96 −117.03 −92.46 115.74 171.71 255.59 −209.57 −133.29 −130.64 109.68 162.79 239.78 −193.72 −115.00 −106.13 108.87 172.27 266.00
a
The atomic asymptotes were calculated with R/UCCSD(T). bExtrapolated by using eq 1 with aV(D+d)Z, aV(T+d)Z, and aV(Q+d)Z for Cl and aVDZ, aVTZ, and aVQZ for O in Cl compounds (frozen core); with aVDZ-PP, aVTZ-PP, and aVQZ-PP for Br and aVDZ, aVTZ, and aVQZ for O in Br compounds (frozen core); with awCVDZ-PP, awCVTZ-PP, and awCVQZ-PP for Br and awCVDZ, awCVTZ, and awCVQZ for O in Br compounds without frozen core approximation given in parentheses; with awCVDZ-PP, awCVTZ-PP, and awCVQZ-PP for I and awCVDZ, awCVTZ, and awCVQZ for O in I compounds without frozen core approximation. cExtrapolated by using eq 2 with appropriate QZ and 5Z type basis as stated in footnote b. dThe zero point energies for the polyatomics were taken as 0.5 of the sum of the CCSD(T) with aug-cc-pVTZ or augcc-pVTZ-PP harmonic frequencies. eCore−valence correction calculated as the difference in energy between the valence electron correlation calculation and that with the appropriate core electrons included at the CCSD(T) with aug-cc-pwCVTZ or aug-cc-pwCVTZ-PP. fThe scalar relativistic correction is based on a CISD(FC)/ aug-cc-pVTZ or aug-cc-pVTZ-PP MVD calculation and is expressed relative to the CISD result without the MVD correction, i.e., including the existing relativistic effects resulting from the use of a relativistic effective core potential. gCorrection due to the incorrect treatment of the atomic asymptotes as an average of spin multiplets. Values are based on C. Moore’s Tables. Cf. ref 129. hThe theoretical value of ΔD0 (0 K) was computed with the CBS estimates.
to 100 cm−1 lower than the calculated ones, appears to be quite general. To determine if this is due to the computational
The fact that in many oxofluorides the observed XO stretching frequencies at MP2/AVTZ level tend to be about 50 10974
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Table 7. Calculated Heats of Formation (kcal/mol) at 298 K molecule
DTQa
Q5
Δ(DTQ-Q5)
ClF ClF2+ ClF2− ClF3 ClF4+ ClF4− ClF5 ClF6+ ClF6− ClF7 BrF BrF2+ BrF2− BrF3 BrF4+ BrF4− BrF5 BrF6+ BrF6− BrF7 IFg IF2+g IF2−g IF3g IF4+g IF4−g IF5g IF6+g IF6−g IF7g ClOF ClOF2+ ClOF2− ClOF3 ClOF4+ ClOF4− ClOF5 ClOF6− ClO2F ClO2F2+ ClO2F2− ClO2F3 ClO2F4− ClO3F ClO3F2− ClO+ ClO2+ ClO3+ ClO− ClO2− ClO3− BrOF BrOF2+ BrOF2− BrOF3 BrOF4+ BrOF4− BrOF5 BrOF6− BrO2F BrO2F2+
−13.0 237.9 −119.8 −38.0 224.1 −157.0 −54.1 202.8 −163.0 20.9 −13.7 (−13.5) 218.2 (−218.8) −130.2 (−129.8) −61.3 (−60.5) 180.7 (182.1) −192.2 (−191.1) −107.9 (−106.4) 157.5 (159.6) −233.7 (−231.5) −62.8(−60.3) −19.7(−19.7) 176.3(177.2) −143.6(−143.1) −104.3(−102.9) 89.6(91.7) −239.6(−238.5) −205.0(−202.6) 22.5(27.8) −326.1(−323.3) −234.6(−229.0) 6.6 231.0 −98.7 −30.3 248.5 −145.3 −25.8 −78.8 −6.2 247.2 −110.0 −2.9 −92.6 0.7 −88.4 302.2 264.9 299.4 −27.7 −24.4 −49.0 7.0 (7.5) 221.0(221.2) −111.1 (−110.5) −53.1 (−51.9) 229.9(230.5) −184.1 (−182.6) −65.7 (−63.8) −151.5 (−149.2) 3.7 (4.5) 269.0(269.8)
−13.1 237.5 −119.6 −38.3 223.2 −156.6 −54.5 201.7 −162.2 20.7 −13.6 218.4 −129.8 −60.8 181.3 −191.2 −106.9 158.3 −231.5 −61.2 −19.4 176.7 −143.3 −103.7 90.6 −238.8 −203.7
0.1 0.4 −0.2 0.3 0.9 −0.4 0.4 1.1 −0.7 0.2 0.1 −0.2 −0.4 −0.5 −0.6 −1.0 −1.0 −0.8 −2.2 −1.6 −0.3 −0.4 −0.3 0.6 −1.0 −0.8 −1.3
5.8 229.5 −99.3 −31.4 246.7 −145.8 −27.2 −79.5 −8.2 244.5 −111.6 −5.1 −94.2 −2.3 −90.9 301.5 263.0 296.7 −28.1 −25.6 −51.4 6.7 220.9 −111.1 −52.9 230.3 −183.3 −65.2
0.8 1.5 0.6 1.2 2.2 0.5 1.4 0.7 2.0 2.7 1.6 2.2 1.6 3.0 2.5 0.7 1.9 2.7 0.4 1.2 2.4 0.3 0.9 0.0 −0.2 −0.4 −0.8 −0.5
3.0 268.7
0.7 0.3
experiment −13.3 ± 0.1,b −12.0 ± 0.1,c −14.3 ± 2.9e, f
−39.3 ± 1.2,b −38.0 ± 0.7c
−56.9 ± 1.7,b −59.9 ± 3.9,d −57.2 ± 4.2,e −56 ± 4.5,f −59.2 ± 4.2e, f
−14.0 ± 0.4c
−61.1 ± 0.7c
−102.5 ± 0.5c
−22.6 ± 0.9c
−200.8 ± 0.4c
−229.7 ± 0.6c
−35.4 ± 1.5e, f
−7.8 ± 2.5e, f
−5.12 ± 0.68,c,h −5.3 ± 4,c,i −5.6 ± 1.2e, f
10975
dx.doi.org/10.1021/ic301438b | Inorg. Chem. 2012, 51, 10966−10982
Inorganic Chemistry
Article
Table 7. continued molecule −
BrO2F2 BrO2F3 BrO2F4− BrO3F BrO3F2− BrO+ BrO2+ BrO3+ BrO− BrO2− BrO3− IOF IOF2+ IOF2− IOF3 IOF4+ IOF4− IOF5 IOF6− IO2F IO2F2+ IO2F2− IO2F3 IO2F4− IO3F IO3F2− IO+ IO2+ IO3+ IO− IO2− IO3−
DTQa
Q5
Δ(DTQ-Q5)
−117.1 (−116.0) −9.7 (−8.1) −134.9 (−133.1) 33.5 (34.7) −88.2 (−86.8) 295.3 (295.7) 278.0 (278.7) 334.1 (335.1) −24.0 (−23.9) −18.1 (−17.8) −36.4 (−35.8) −9.0 168.9 −139.0 −114.8 134.6 −256.7 −177.2 −299.3 −27.1 213.2 −165.0 −82.4 −245.8 −6.6 −159.4 276.9 257.1 306.7 −26.1 −30.9 −66.3
−117.3 −9.8 −134.7 32.6 −88.7 294.9 277.1 332.9 −24.3 −18.9 −37.6 −7.8
0.2 0.1 −0.2 0.9 0.5 0.4 0.9 1.2 0.3 0.8 1.2 −1.3
−26.2
−0.9
277.8 257.7 307.4 −24.7 −29.6 −65.7
−0.9 −0.6 −0.7 −1.4 −1.4 −0.5
experiment
a
Calculations from DTQCV extrapolated results are given in parentheses for compounds with Br. bReference 82. cReference 81. dReference 83. Reference 84a. fReference 84b gValues obtained from reference 88a. hReference 135. iReference 136.
e
the oxidation state of the central atom of the molecule, with the corrections increasing with increasing oxidation state. The ΔESR corrections for Cl in the formal +VII oxidation state are in the range of −3.3 to −4.2 kcal/mol. For the CCSD calculations, an estimate of the potential for significant multireference character in the wave function can be obtained from the T1 diagnostic.134 The values for the T1 diagnostics are small (
