Generation and Reactions of an Octacyclic Hindered Pyramidalized Alkene

Octacyclo[10.6.1.0.0.0.0.0.0]nonadeca-5,8,14-triene (27), a hindered pyramidalized alkene, has been generated from a diiodide precursor. Contrary to the usual behavior of known pyramidalized alkenes, no Diels−Alder adducts were obtained from the present alkene when it was generated by different standard procedures in the presence of different dienes. However, products derived from the reduction, t-BuLi addition, condensation with the solvent, or dimerization were isolated from these reactions, depending on the conditions used to generate it. No [2 + 2] cross product among this pyramidalized alkene and tricyclo[3.3.1.0]non-3(7)-ene was formed when a mixture of the corresponding precursor diiodides was reacted with sodium amalgam. The analysis of selected geometrical and orbital parameters determined from quantum mechanical calculations indicates that the degree of pyramidalization of this alkene and its higher steric hindrance compared with other polycyclic pyramidalized alkenes may explain its peculiar reactivity.


■ INTRODUCTION
The chemistry of highly pyramidalized alkenes has been the subject of several reviews, 1−7 and new highly reactive intermediates of this kind have been recently described. 8−12 In a pyramidalized alkene, the olefinic carbon atoms are rehybridized by an admixture of additional p-character into the original sp 2 σ-orbitals. This makes the geometry around the olefinic carbon atoms nonplanar. The π bond is now formed from two p-orbitals with some s-character. These orbitals are well aligned (torsion angle = 0), but they are not parallel. Consequently, the distance between them increases, and the neat overlap decreases, affecting the intrinsic reactivity properties of the pyramidalized double bond.
From a structural point of view, the degree of pyramidalization of syn-pyramidalized alkenes belonging to the C 2v point group of symmetry (left structure of Figure 1) can be described by the pyramidalization angle (Φ), which corresponds to the angle between the plane containing one of the olefinic carbon atoms and the two substituents attached to it and the elongation of the CC bond. Its value can be obtained according to the formula given in Figure 1 from the RCC (α) and RCR (β) angles. For alkenes belonging to the C s point group of symmetry (right structure of Figure 1), the flap or hinge angle (Ψ), corresponding to the dihedral angle among the R 2 CCR 2 and R 1 CCR 1 planes, or its supplementary angle (ζ) is usually used. While pyramidalized alkenes are generally related to the geometrical strain of the olefinic carbon atoms, it is worth noting that alkene pyramidalization may also occur due to electronic effects. 13 With regard to the synthetic accessibility, the more pyramidalized alkenes are usually generated by reaction of a vicinal double bridgehead diiodide or dibromide with an organolithium reagent in THF, sodium/potassium alloy, or sodium amalgam in an ether solvent or molten sodium in boiling 1,4-dioxane. Scheme 1 collects several of the most representative reactions of pyramidalized alkenes.
Lukin and Eaton 19 generated 1,2-dehydrocubane (cubene, 15), one of the most highly pyramidalized alkenes ever Recently, we have described 21 the preparation of octacycle 28 as a possible precursor of pyramidalized alkene 27 (Scheme 2) by reaction with fluoride anions, following the procedure described by Lukin and Eaton 19 to generate cubene 15. However, all attempts to generate 27 on reaction with CsF alone or in combination with AgF in the presence of dienes, such as 1,3diphenylisobenzofuran, tetraphenylcyclopentadienone, furan, or anthracene at different temperatures, left the starting compound unchanged. When compound 28 was reacted with dimethyl acetylenedicarboxylate 29, CsF, and AgF in the presence of tris(dibenzylideneacetone)dipalladium(0)·CHCl 3 [Pd(dba) 2 · CHCl 3 ] as catalyst, product 30 was obtained as a result of cocyclotrimerization of 29 and 28, at the CC bond further from the iodine and trimethylsilyl groups. In all the reactions, the obtained products always contained the trimethylsilyl group. The lack of reactivity of the trimethylsilyl group toward fluoride anions is likely due to the steric hindrance experienced by this group.
Reaction of 31 with the less electrophilic, but much more readily available, dienophile, 2-iodoethynyl(phenyl)iodonium triflate, 24 in acetonitrile at reflux for 64 h, followed by treatment with NaI/CuI, did not give the expected diiodide 34. Reaction of 34 with methanol catalyzed by K 2 CO 3 gave in good yield the corresponding diol 35, which was transformed into the corresponding bismesylate 36 on reaction with MsCl in the presence of Et 3 N. Reaction of 36 with NaI in acetone at reflux gave the tetraiodide 37 in good yield. Reaction of 37 with potassium cyclopentadienide in a mixture THF/DMF in the presence of a catalytic amount (5 mol %) of 18-crown-6 gave in high yield the desired octacyclic diiodide 38. As in the preparation of 28, this transformation implies a double domino nucleophilic substitution of neopentyl-type iodides by the cyclopentadienide anion followed by a double intramolecular Diels−Alder reaction, with formation of six new C−C bonds and four new cycles apart from those of the cyclopentadienide. Thus, the octacyclic product 38 is formed from the bicyclic precursor 37 in only one step. Although the structure of the symmetric compound 38 (C s point group) was clearly deduced from the 1 H and 13 C NMR data, it was confirmed by X-ray diffraction analysis (see the ORTEP structure of diiodide 38 in the SI). 25 When a cold solution of diiodide 38 and 1,3-diphenylisobenzofuran 42 in THF was treated with a pentane solution of t-BuLi, the expected Diels−Alder adduct from the reaction of the pyramidalized alkene 27 and diene 42 was not observed in the crude product by 1 H NMR. After column chromatography, the only isolated product was 39 (26% yield), which must derive from the reaction of 27 with t-BuLi followed by protonation during the quenching of the reaction mixture. 17 Similar results were obtained when diene 42 was replaced by anthracene or diene 14 in the above reaction. In both cases, the 1 H NMR Scheme 2. Attempted Generation of 27 from 28 Scheme 3. Preparation of Octacycle 38 from Cyclopentadiene 31 The Journal of Organic Chemistry Article spectrum of the crude product did not show the expected signals for the corresponding Diels−Alder adducts, and after column chromatography, the only isolated product was always 39 (Scheme 4).
To solve this problem, a solution of diiodide 38 and diene 14 in 1,4-dioxane was added to an excess of 0.48% sodium amalgam. The 1 H NMR spectrum of the crude product from this reaction did not show the presence of the expected Diels−Alder adduct from pyramidalized alkene 27 and diene 14. By column chromatography, two hydrocarbon products were isolated, the reduction product 40 (5%) and the [2 + 2 + 2 + 2] dimer 43 (16%). Worthy of note, when a solution of diiodide 38 in 1,4dioxane was reacted with 0.48% sodium amalgam in the absence of diene 14, the formation of dimer 43 was not observed. Compound 40 was the only isolated product (61% yield). The structure of 40 was easily deduced from its NMR data, which show the high symmetry of this compound (C 2v point group of symmetry), and later confirmed by X-ray diffraction analysis (see ORTEP structure of compound 40 in the SI). 26 In the case of dimer 43, the structure was first obtained by X-ray diffraction analysis, and the data showed that the unit cell of 43 contains one molecule of each enantiomer ( Figure 2). 27 Keeping in mind the C 2 symmetry point group of 43 and the fact that all signals of the different protons and 13 C atoms of 43 appear clearly separated, except for both pairs of methylenic protons, we could fully assign its 1 H and 13 C NMR spectra with the aid of the 1 H/ 1 H homocorrelation spectra (COSY and NOESY) and 1 H/ 13 C heterocorrelation spectra (sequence gHSQC for one-bond correlations and gHMBC for long-range correlations). Especially significant to carry out this assignment was the observation of correlations among pairs of protons belonging to a different half of the molecule, such as 13 (30) When diiodide 38 was added to molten sodium in boiling 1,4-dioxane, the standard conditions used by our group to obtain dimers from pyramidalized alkenes, 5,18,20 once again no dimer was observed in the crude reaction product ( 1 H NMR), and after column chromatography, compounds 40 (55% yield) and 41 (25% yield) were the only isolated products. The last one is a formal addition product of pyramidalized alkene 27 to the solvent.
Attempted formation of a bis(triphenylphosphine)Pt(0) complex derived from 27 was carried out by reacting a THF solution of 38 and ethylene-bis(triphenylphosphine)Pt(0) with liquid (0.47%) sodium amalgam, following the procedure described by Borden et al. 15,16 to prepare complex 4 from pyramidalized alkene 2 (Scheme 1). However, a complex mixture of products was obtained, from which the expected complex could not be isolated.
The preceding results might be explained by assuming the formation of pyramidalized alkene 27 on reaction of diiodide 38 with t-BuLi, sodium amalgam, or molten sodium in boiling 1,4-dioxane. Thus, although it does not react with the different studied dienes, it does not experience cross coupling with tricyclo[3.3.1.0 3,7 ]non-3(7)-ene 2, and it does not form a complex with bis(triphenylphosphine)Pt(0); however, it reacts with t-BuLi to give the addition product 39, with 1,4-dioxane to give the reduction product 40 or the addition product 41 and,

Scheme 4. Transformations of Octacycle 38
The Journal of Organic Chemistry Article on only one occasion, dimer 43, a kind of reaction also observed for other pyramidalized alkenes. On the contrary, pyramidalized alkene 6 (Scheme 1), generated by reaction of diiodide 5 with liquid sodium amalgam, gave in good yields Diels−Alder adducts with many dienes, such as 14 (Scheme 1), 1,3-diphenylisobenzofuran 42, furan, or 2,5-dimethylfuran, 28 as well as a cross-coupling product with pyramidalized alkene 10 (see structure on Scheme 1). 29 Moreover, the related pyramidalized alkene 18 from diiodide 17 also gave crosscoupling reaction with pyramidalized alkene 10 (Scheme 1) and a Diels−Alder adduct with 1,3-diphenylisobenzofuran 42. 5 The results herein described can be partly explained on the basis of the lower pyramidalization and greater steric hindrance of alkene 27 compared with the related alkenes 6, 18, or 25. Using the optimized geometries of pyramidalized alkenes 27, 6, and 25 obtained from M06-2X/6-311+G(d) 30,31 calculations ( Figure 4), the pyramidalization angle (ϕ) of alkenes 6 and 25, which contain the substructure of tricyclo[3.3.0.0 3,7 ]oct-1(5)-ene, were calculated to both be equal to 68.3°. Since for symmetry reasons the pyramidalization angle of alkene 27 is not applicable, a comparison will be performed on the basis of their flap (ψ) or supplementary flap angles (ζ). The supplementary flap angles of alkenes 6 and 25 were calculated to be 70.1°and 70.2°, respectively, quite close to their pyramidalization angles. However, in alkene 27, a value of 52.0°calculated for the supplementary flap angle is indicative of a much lower degree of pyramidalization, in concordance with the fact that this alkene contains the substructure of tricyclo[3.3.1.0 3,7 ]non-3(7)ene (2), for which a pyramidalization angle (ϕ) of 53.7°had been calculated with the B3LYP/6-31G(d) basis set. 5 The differences in the degrees of pyramidalization of alkenes 27, 6, and 25 are also reflected in the orbitalic features of the double bond ( Figure 5), as revealed from the analysis of the natural bond orbitals (NBOs) 32 derived at the CISD/ 6-31G(d) 33 level. For standard double bonds (i.e., without geometrical strain), the sp 2 hybridization implies a spatial orientation of the π orbital of 90°. In pyramidalized alkenes, however, geometrical distortion introduces a deviation in the angle formed by the atomic hybrid orbital. The results indicate that the deviation angle for alkenes 25 and 6 amounts to 124°a nd 123°, respectively, while the deviation angle for alkene 27 is 114°. This confirms that the structural stress in alkene 27, which contains the tricyclo[3.3.1.0 3,7 ]non-3(7)-ene subunit, is lower than in the more reactive alkenes 6 or 25, which contain the tricyclo[3.3.0.0 3,7 ]oct-1(5)-ene moiety.
Furthermore, the biradical character of compounds 27, 25, and 6 was examined following the method of Takatsuka et al., 34 which relies on the number of unpaired electrons (N e ; eq 1) determined from the occupancy of the natural orbitals obtained from broken-symmetry calculations at the UHF/6-31G(d) level.
where n i denotes the occupation number.
The results point out that N e increases from 27 (1.37) to 6 (1.51) and to 25 (1.85). This can be interpreted from the larger degree of pyramidalization of the double bond in the two latter compounds (see above). Nevertheless, keeping in mind the similar geometrical features of 25 and 6 (see Figure 4), the larger value of N e determined for alkene 25 compared to compound 6 reveals the contribution played by the double bonds located at the two ends of the molecule (between atoms 5 and 6 as well as 11 and 12; Figure 4). A similar effect can be expected for compound 27 due to the presence of the double bond (between atoms 5 and 6; Figure 4). Indeed, calculations performed for the compound obtained upon saturation of this double bond lead to an estimated N e value of 1.18. Accordingly, it can be concluded that the geometrical differences introduced by the distinct bridges in the skeleton of these alkenes increase effectively the biradical character of the most strained compounds.
Finally, to evaluate the relative steric hindrance of alkenes 6, 25, and 27, the external angles among the planes defined by the atoms C8, C9, C11, and C12, on one hand, and C8, C9, C5, and C6, on the other hand, were calculated (see Figure 4). For alkenes 6 and 25, which contain the same carbocyclic skeleton, angles of 192.3°and 193.0°were calculated. However, for alkene 27, the corresponding value was 170.7°. These values show that the external face of the pyramidalized CC bond in compounds 6 and 25 is much more accessible than in alkene 27.
Overall, the combination of a lower degree of pyramidalization and higher steric hindrance can explain the reactivity observed for the hypothetical pyramidalized alkene 27. From the

The Journal of Organic Chemistry
Article obtained results, it may be assumed that alkene 27 might be generated from diiodide 38 under the different reaction conditions studied: (a) reaction with t-BuLi in THF since product 39 derived from the addition of the t-butyl group to 27 was obtained; (b) reaction with sodium amalgam since the reduction product 40 and dimer 43 were isolated; and (c) reaction with molten sodium in boiling 1,4-dioxane since the reduction product 40 and the product of addition of 27 and the solvent were isolated. The lack of reactivity of 27 toward different dienes, ethylene-bis(triphenylphosphine)Pt(0), or tricyclo[3.3.1.0 3,7 ]non-3(7)-ene 2 (generated simultaneously in situ) can be explained on the basis of the steric hindrance and the possibility of alternative transformations. For instance, in the attempted cross coupling among pyramidalized alkenes 2 and 27, compound 3, the cyclobutane dimer of 2, and the reduction product 40 were the only isolated products. Reasonably, the formation of these products must be faster than the cross coupling of 2 and 27, mainly due to the steric hindrance of 27. However, alternative mechanisms to explain these results can not be ruled out.

Article
The mixture was stirred at rt for 17 h. The solution was cooled to −35°C. Powdered NaI (828 mg, 5.52 mmol) and CuI (1.05 g, 5.51 mmol) were added, and the mixture was stirred at rt for 20 h. The solvent was distilled under reduced pressure, and to eliminate the formed iodobenzene, toluene (10 mL) was added. The solvent and volatiles were distilled off under reduced pressure, repeating this process three more times. The black solid residue (4.03 g) was subjected to automatic column chromatography [  (199 mg, 0.49 mmol) in anhydrous CH 2 Cl 2 (4.5 mL) was prepared in a round-bottomed flask provided with Ar atmosphere and magnetic stirring. Anhydrous Et 3 N (0.27 mL, 1.97 mmol) was added dropwise; the solution was cooled to 0°C (ice−water bath); MsCl (90 μL, 1.18 mmol) was added dropwise; and the reaction mixture was stirred at 0°C for 2 h. Saturated aqueous solution of NaHCO 3 (0.5 mL) was added. The aqueous phase was separated, and the organic one was washed with more saturated aqueous solution of NaHCO 3 (3 × 5 mL). The combined aqueous phases were extracted with CH 2 Cl 2 (3 × 5 mL), and the combined organic phase and extracts were washed with water (7 mL), dried (anhydrous Na 2 SO 4 ), and concentrated in vacuo to give crude dimesylate 36 (298 mg), which was subjected to automatic column chromatography (35−70 μm of silica gel, 12 g, hexane/EtOAc mixtures) to give dimesylate 36 (271 mg, 98% yield) as a yellow oil, on elution with hexane/EtOAc from 65:35 to 10:90. The analytical sample of 36 (203 mg) was obtained as yellow solid by crystallization of the above product from a 1:3 mixture of CH 2 Cl 2 /pentane (2 mL).   1 mL) was placed in a round-bottomed flask provided with Ar atmosphere, magnetic stirring, and reflux condenser. Powdered NaI (550 mg, 3.65 mmol) was added, and the reaction mixture was heated to reflux for 17 h. The solvent was evaporated under reduced pressure to give a yellow residue (770 mg) that was subjected to column chromatography (35−70 μm silica gel, 3 g, hexane) to give tetraiodide 37 (190 mg, 85% yield) as a yellow viscous oil. R f = 0.67 (silica gel, 10 cm, hexane/EtOAc 8:2). 1  In a 10 mL flask, KH (30% in mineral oil, 134 mg, 1.00 mmol) was washed with anhydrous THF (5 × 5 mL) under an Ar atmosphere. To the washed KH, anhydrous THF (5 mL) was added, and the suspension was cooled to 0°C (ice−water bath). Freshly distilled cyclopentadiene (120 μL, 99 mg, 1.5 mmol) was added, and the mixture was stirred at this temperature for 10 min. 18-crown-6 (13 mg, 49 μmol, about 5% with respect to KH) was added, and the mixture was stirred at 0°C for 10 min and at rt for 15 min to give a pink suspension.
Substitution Reaction. In a 25 mL flask provided with magnetic stirring, reflux condenser, and Ar atmosphere, a solution of tetraiodide 37 (187 mg, 0.30 mmol) in anhydrous DMF (2.2 mL) was prepared. The solution was cooled to 0°C (ice−water bath), and then part of the above solution of potassium cyclopentadienide (3.3 mL, 0.66 mmol) was added dropwise. The mixture was stirred at 0°C for 5 min and at rt for 10 min, and then it was heated at 90°C for 17 h. The mixture was allowed to cool to rt; MeOH (0.1 mL) was added; and the mixture was stirred for 10 min. Then, EtOAc (5 mL) and water (5 mL) were added, and the organic phase was separated. The aqueous phase was extracted with EtOAc (3 × 8 mL), and the combined organic phases were washed with saturated aqueous solution of NaHCO 3 (3 × 8 mL), water (2 × 8 mL), and brine (8 mL), dried (anhydrous Na 2 SO 4 ) and concentrated in vacuo to give a brown oily residue (213 mg), which was subjected to column chromatography [35−70 μm silica gel (4 g) pentane/EtOAc mixtures] to give, on elution with pentane, octacycle 38 (128 mg, 85% yield) as white solid. An analytical sample of 38 (97 mg) was obtained as white solid, by crystallization of the above product from CH 2 Cl 2 /MeOH 1:3 (2 mL). R f = 0.62 (silica gel, 10 cm, hexane/EtOAc 9:1); mp 236.8−237.5°C (CH 2 Cl 2 /MeOH). 1  .0 8,19 .0 11,16 .0 13,17 ]nonadeca-5,14-diene 39. A solution of octacycle 38 (90 mg, 0.18 mmol) and diene 42 (58 mg, 0.22 mmol) in anhydrous THF (2.9 mL) was prepared in a two-necked round-bottomed flask provided with Ar atmosphere, magnetic stirring, and lowtemperature thermometer. The solution was cooled to −67°C, and a solution of t-BuLi in pentane (1.7 M, 110 μL, 0.19 mmol) was added dropwise. The color of the solution changed from yellow to dark brown. The mixture was stirred at this temperature for 30 min, and it was allowed to heat to rt for 30 min. MeOH (0.15 mL), water (2 mL), and Et 2 O (3 mL) were successively added, and the organic phase was separated. The aqueous one was extracted with Et 2 O (3 × 4 mL). The combined organic phase and extracts were dried (anhydrous Na 2 SO 4 ) and concentrated in vacuo to give a yellow oil (108 mg) that was subjected to column chromatography (35−70 μm silica gel, 3. X-ray Crystal-Structure Determination of Compound 38. A colorless prism-like specimen of C 19 H 18 I 2 , approximate dimensions 0.214 mm × 0.226 mm × 0.365 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured on a D8 Venture system equipped with a Multilayer monochromator and a Mo microfocus (λ = 0.71073 Å). The frames were integrated with the Bruker SAINT software package 35 using a narrow-frame algorithm. The integration of the data using an orthorhombic unit cell yielded a total of 10601 reflections to a maximum θ angle of 30.53°(0.70 Å resolution), of which 4596 were independent (average redundancy 2.307, completeness = 98.0%, R int = 3.26%, R sig = 6.07%) and 4000 (87.03%) were greater than 2σ(F 2 ). The final cell constants of a = 11.9655(3) Å, b = 15.3718(4) Å, and c = 16.7007(4) Å and volume = 3071.78(13) Å 3 are based upon the refinement of the XYZ-centroids of reflections above 20σ(I). Data were corrected for absorption effects using the multiscan method (SADABS). 35 The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.6280 and 0.7461. The structure was solved and refined using the Bruker SHELXTL Software Package, 36 using the space group Pbca, with Z = 8 for the formula unit, C 19 H 18 I 2 . The final anisotropic full-matrix least-squares refinement on F 2 with 190 variables converged at R1 = 2.53%, for the observed data, and wR2 = 7.11% for all data. The goodness-of-fit was 1.040. The largest peak in the final difference electron density synthesis was 0.700 e Å −3 , and the largest hole was −1.040 e Å −3 with an RMS deviation of 0.167 e Å −3 . On the basis of the final model, the calculated density was 2.163 g cm −3 and F(000), 1904 e. For more details, see Table 1 in the Supporting Information.
X-ray Crystal-Structure Determination of Compound 40. A colorless prism-like specimen of C 19 H 20 , with approximate dimensions of 0.058 mm × 0.103 mm × 0.440 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured on a D8 Venture system equipped with a multilayer monochromator and a Mo microfocus (ë= 0.71073 Å). The frames were integrated with the Bruker SAINT software package, 35 using a narrow-frame algorithm. The integration of the data using a triclinic unit cell yielded a total of 71539 reflections to a maximum θ angle of 30.57°(0.70 Å resolution), of which 7702 were independent (average redundancy 9.288, completeness = 99.2%, R int = 4.49%, R sig = 2.43%) and 6307 (81.89%) were greater than 2σ(F 2 ). The final cell constants of a = 5.9104(4) Å, b = 11.1856(8) Å, c = 19.9679(14) Å, α = 105.046(2)°, β = 96.566(2)°, γ = 90.113(2)°, and volume = 1265.80(15) Å 3 are based upon the refinement of the XYZ-centroids of reflections above 20σ(I). Data were corrected for absorption effects using the multiscan method (SADABS). 35 The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.6714 and 0.7461. The structure was solved and refined using the Bruker SHELXTL Software Package, 36 using the space group P−1, with Z = 4 for the formula unit, C 19 H 20 . The final anisotropic full-matrix least-squares refinement on F 2 with 343 variables converged at R1 = 5.45%, for the observed data, and wR2 = 17.00% for all data. The goodness-of-fit was 1.044. The largest peak in the final difference electron density synthesis was 0.505 e Å −3 , and the largest hole was −0.338 e Å −3 with an RMS deviation of 0.064 e Å −3 . On the basis of the final model, the calculated density was 1.303 g cm −3 and F(000), 536 e. For more details, see Table 1 in the Supporting Information.
X-ray Crystal-Structure Determination of Compound 43. A colorless plate-like specimen of C 38 H 36 , with approximate dimensions 0.062 mm × 0.204 mm × 0.307 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured on a D8 Venture system equipped with a multilayer monochromator and a Mo microfocus (λ = 0.71073 Å). The frames were integrated with the Bruker SAINT software package, 35 using a narrow-frame algorithm. The integration of the data using a triclinic unit cell yielded a total of 64261 reflections to a maximum θ angle of 30.66°(0.70 Å resolution), of which 7215 were independent (average redundancy 8.907, completeness = 99.4%, R int = 4.19%, R sig = 2.51%) and 5824 (80.72%) were greater than 2σ(F 2 ). The final cell constants of a = 7.3827(3) Å, b = 11.4509(5) Å, c = 14.5547(6) Å, α = 77.556(2)°, β = 86.847(2)°, and γ = 77.599(2)°and volume = 1173.44(9) Å 3 are based upon the refinement of the XYZ-centroids of reflections above 20σ(I). Data were corrected for absorption effects using the multiscan method (SADABS). 35 The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.7149 and 0.7461. The structure was solved and refined using the Bruker SHELXTL Software Package, 36 using the space group P−1, with Z = 2 for the formula unit, C 38 H 36 . The final anisotropic full-matrix least-squares refinement on F 2 with 343 variables converged at R1 = 4.45%, for the observed data, and wR2 = 12.41% for all data. The goodness-of-fit was 1.018. The largest peak in the final difference electron density synthesis was 0.414 e Å −3 , and the largest hole was −0.240 e Å −3 with an RMS deviation of 0.058 e Å −3 . On the basis of the final model, the calculated density was 1.394 g cm −3 and F(000), 528 e. For more details, see Table 1 in the Supporting Information.
Computational Methods. Full geometry optimizations were performed with the M06-2X density functional method 30 by using the 6-311+G(d) 31 basis set. The nature of the stationary points was verified by inspection of the vibrational frequencies within the harmonic oscillator-rigid rotor approximation. Molecular electrostatic potential analysis was performed from the optimized geometries. The natural bond orbital analysis (NBO) 32 was carried out at the CISD/6-31G(d) level of theory, 33 in order to evaluate the orbitals of the pyramidalized double bounds. All DFT computations were carried out using the keyword Integral(Grid = Ultrafine) as implemented in Gaussian09, 37 which was used to carry out these calculations. The biradical character of pyramidalized alkenes was examined following the method of Takatsuka et al., 34 which relies on the number of unpaired electrons determined from the occupancy of the natural orbitals obtained from broken-symmetry calculations at the UHF/6-31G(d) level.
Nomenclature. The complex name of these polycyclic compounds have been obtained by using the POLCYC program. 38