摘要:
1976 621 Theoretical Study of Charge Distribution in the Benzoyl and Alkynoyl Cations By Charles U. Pittman, jun.,' Bong Kim, Quock Y. Ng, and Lowell D. Kispert," Department of Chemistry, The University of Alabama, University, Alabama 35486, U.S.A. SCF-MO calculations.inthe INDO approximation, have been carriedout on benzoyl (1). propynoyl (2), but-2-ynoyl (3). 3-phenylpropynoyl (4), and 3-fluoropropynoyl cations (5). In each case the geometries were optimized to minimum energy. In aggreement with 13C n.m.r. observations, significant charge delocalization into the phenyl rings occurs in (1) and (4). From an examination of the charge distributions, x-bond orders, and x-ando-electron distributionsit is shown that extensive delocalization of charge occurs in (Z), (3).and (5), contrary to previous+interpretations of 13C spectra. Thus all the mesomeric forms RCZC-C-O+ ++ RC-C-C + = 0 RC=C=C=O contribute significantly in (2). (3), and (5) as well as in (4), and the charge is not substantially localized on oxy-gen as previously claimed. PROTONn.m.r. spectroscopy has been widely used to probe the structure and charge distribution in carbo- ~ations.l-~ Recently 13C n.m.r. spectroscopy has been used to probe more directly the charge distributions along the carbon frame~ork.~ Recently, Olah reported the 13C n.m.r. spectra of the benzoyl cation (1)5 and alkynoyl cations (2)-(4).6 It was concluded from an examination of the 13C chemical shifts (shown in the Scheme in p.p.m. from tetramethylsilane) that in (1) and (4) substantial charge delocalization into the phenyl rings occurred. Surprisingly, however, the authors also concluded that the charge was mainly localized on the oxygen atom in (2) and (3) and that vinylic resonance hybrids made no appreciable contribution.6 132.9 141.3 \-I I I\ 110.7 20.9 12C.1 (2) 131.2 lL1.8 However, parallels between 13C chemical shifts and charge distribution must be viewed cautiously because C.U. Pittman, jun., and S. P. McManus, ' Carbonium Ions,' in ' Reactive Intermediates in Organic Chemistry,' ed. S. P. McManus, Academic Press, New York, 1973. G. A. Olah and C. U. Pittman, jun., Adv. Phys. Org. Chem., 1966, 4, 306. G. A. Olah and J. A. Olah, in ' Carbonium Ions,' eds. G. Olah and P.von R. Schleyer, Wiley-Interscience, New York, 1970, VO~.11, pp. 715-782. G. A. Olah and P. 147.Westerman, J. Anzer. Chem.SOC., 1973, 95, 7530. G. A. Olah and P. W. Westerman, J. Amer. Chem. SOC., 1973, 95, 3706. G. A. Olah, R. J. Spear, P. W. Westerman, and J. M. Denis, J. Amer. Chem. Soc.,,1974, 96, 5855. 7 J. B. Stothers, Carbon-13 NMK Spectroscopy,' Academic Press, New York, 1972 (see ch. 4). 8 J. A, Pople, D. L. Beveridge, and P. A. Dobosh, J. Chem. Phys., 1967, 47, 2026.* J. A. Pople and D. L. Beveridge, ' Approximate Molecular Orbital Theory,' McGraw-Hill, Highstown, 1970 13C chemical shifts depend on hybridization, charge, and anisotropic diamagnetic and paramagnetic shielding effects.' Furthermore, the choice of model compounds, with which one compares the chemical shifts of (1)-(4) presented some difficulty. Thus, we decided to perform a series of SCF-MO calculations, in the INDO approxim-ati~n,~.~on cations (1)-(4), as well as the 3-fluoropropyn- oyl cation (5),to probe the charge distribution as a test of the conclusions derived from the 13C n.m.r.data. TO allow a rigorous test, within the INDO framework, geometry optimization calculations were performed in a systematic manner. RESULTS Geometry optimization was carried out as described previously.11-18 All calculations employed the CINDO program, QCPE number 141. The optimized geometries and charge distributions of carbocations (1)-(5) are given in the Figure. All bond lengths and bond angles were systematically varied.The minimum in the energy varied by <0.05 kcal mol-1. Thus the bond lengths have been obtained to 0.001 A. In (1) and (4) the substituent axes bisect the ring angles. The results clearly indicate substantial charge delocaliz- ation occurs in cations (2) and (3) (in contrast to Olah's conclusions based on 13C n.m.r. studies 6) as well as in cations (l), (a), and (5). The following items support this con- clusion : (i) substantial positive charge is calculated at C,, in (2) and (3); (ii) the magnitude of the positive charge calculated at C,, is about the same for (2)-(4); (iii) the magnitude of the charge on oxygen is about the same for (1)-(5). The charge is almost zero. Considering that the lo C. U.Pittman, jun., Q. Y. Ng, and L. D. Kispert, unpub- lished work. l1 L. D. Kispert, C. Engelman, C. Dyas, and C. U. Pittman, jun., J. Amer. Chem. Soc., 1971, 93, 6948. l2 C. U. Pittman, jun., C. Dyas. C. Engelman, and L. D. Kis-pert, J.C.S. Faraday II, 1972, 5979. l3 L. D. Kispert, C. U. Pittman, jun., D. L. Allison, T. B. Patterson, jun., C. W. Gilbert, jun., C. F. Wains, and J. Prather,J.Amer. Chem.Soc., 1972, 94, 5979. l4 C. U. Pittman, jun., T. B. Patterson, jun., andL. D. Kispert,J. Org. Chem.,1973, 38, 471. l5 C. U. Pittman, jun., L. D. Kispert, andT. B. Patterson, jiin., J. Phys. Chem., 1973, 77, 494. l6 G. R. De Mar4 S. Lapaille, L. D. Kispert, and C. U. Pitt-man, jun., J. Mok. Structure, 1973, 17,417. l7 C. U. Pittman, jun., A.Kress, T. B. Patterson, P. Walton, and L. D. Kispert, J.Org. Chern.,1974, 39, 373. l8 C. U. Pittman, jun., A. Kress, and L. D. Kispert, J. Org.Chem., 1974, 39, 378. H\ 1.380 /" 1.207-0 119.50 17\/ \ H*0*035 H+0.015 (1) 1.218 1.365 + 1.200H-C-C-C-0 1.100 '0.139 *0.270 -0.055 '0,629 '0018 (1.823 (0.778) (1.685) (21 HwH1.303 1.298 1.225 1.352 -k 1.205 Optimized geometries, charge distributions, and x-bond orders of cations (1)-(6). Numbers in parentheses are the sum of the p. and p, x-bond orders carbonyl oxygens in the neutral model molecules, propenal and propynal, exhibit charges of -0.228 and -0.277.10@ respectively, in INDO calculations, one can conclude that J.C.S. Perkin I1 the oxygen atoms of cations (1)-(5) have donated sub- stantial x-electron density to the a-carbon.Examining the electron densities in the pzand p, orbitals of oxygen and the a-carbon revealed this trend. The Ca-0 o-bond remained strongly polarized towards oxygen: and (iv) The Ca-0 x-bond orders in (1)-(5) and the CgC, and C,-Cs x-bond orders in (2)-(5) were remarkably consistent (see Figure). They show that substantial x-delocalization from oxygen to C, occurs in (2)-(5). It is obvious that charge is highly delocalized in (2), (3), and (5) and that this delocalization resembles that in (4) [or(l)], contrary to conclusions based on 13C n.m.r. shifts.6 Significant charge delocalization into the phenyl rings occurs in (1) and (4) in agreement with Olah's 13C n.m.r.chemical shift interpretations.6 This is clearly evident from (a) the higher charge densities calculated at the orfho-and para-positions, (b) the obvious bond alternation effect where the optimized phenyl ring geometries of (1) and (4) have a distinct quinonoid character, slid (c) the p, x-bond order alternation in the phenyl rings (see Table). No n.m.r. data is yet available on (5). However, the trend here is the same as in the other cases except that the charge at C,, is quite high due to polarization of the C-F u bond toward fluorine. Conclusions.-Geometry optimized INDO calculations demonstrated that charge delocalization is extensive in alkynoyl cations (2), (3), and (5),in addition to (4). Thus, a Pzx-Bond orders for phenyl rings in cations (1) and (4) Bond (1) (4) cip**-corfAo 0.662 0.692 Corllm--c*fo 0.706 0.693 Cmefa-Cpcira 0.638 0.648 substantial contribution is made by each of the resonance hybrids a-c in these species. There appears to be no unique difference in charge delocalization between (4) and (2), (3).or (5) over the oxygen and a,p, or y-carbons. ' Keten-like ' R-CeC-&o --+R-C=C-C& -R-&=C=O mesomeric forms of (1) [i.e. (la)] make a substantial contri- bution as do the vinylogous (4b) and cumulene (4a) meso- meric forms of (4). Calculations, of course, do not take into account solvent effects. A specific solvent may stabilize mesomeric forms (Zb), (3b), or (5b) by co-ordinating to oxygen. However, if that was actually taking place, so that the charge distribu- tion pattern was seriously changed going from the gas phase to solution, one might expect it would also occur in (4).Thus mesomeric structure (4d) should be stabilized in the same manner, and one would not expect a fundamentally different delocalization pattern for (4) vem'szcs (2) or (3) in .b 1976 solution. Thus, INDO calculations support the expect- ations of simple resonance considerations for structures (1)-(5) and strongly suggest that the n.m.r. spectra of (2) and (3) should be re-examined or reinterpreted. It could be argued that other quantum chemical approaches might be more effective. Thus, preliminary MIND0/2 calculations 19 were performed on (2) and (3). As with INDO, the MIND0 results predicted significant charge distribution onto C, and strong contributions by (2a and c) and (3a and c).Recently, ab initio calculations have been successfully applied to studies of the energies of isodesmic reactions involving charged and neutral species.20-22 Rad~m,~~using both the STO-3G 24 and extended 4-31G 25+ basis sets, successfully predicted the GObond length in the acetyl cation as 1.141 and 1.107 respectively vemus experi-mental values of 1.108 26 and 1.109 A 27 found in X-ray + crystal structures. The optimized GO lengths in (1)-(5), l9 M. J. S. Dewar, ' The Molecular Orbital Theory of OrganicChemistry,' McGraw-Hill, New York, 1969. 2o L. Radom, J. A. Pople, and P. von R. Schleyer., J. Amer. Chem. SOG.,1972, 94, 6935.21 W. J. Hehre, R. Ditchfield, L. Radom, and J. A. Pople,.T. Amer. Chem. SOC.,1970, 92, 4796. 22 L. Radom, W. J. Hehre, and J. A. Pople, J. Chem. SOG.(A),1971, 2299. 23 L. Radom, Austral. J. Chem., 1974, 27, 231. as calculated by INDO, were in the range 1.200-1.211 A. The use of the 4-31G basis set is known to be more reliable in predicting geometries of charged but no+ optimized C=O lengths have yet been calculated using that method for conjugated cations such as (1)-(5). Thus, comparison with INDO geometries for (1)-(5) cannot yet be made. However, conjugated systems such as (1)-(5)+ would be expected, on first principles, to have longer GO lengths than the acetyl cation. We thank the University of Alabama computer centre for donation of free computer time and to the University of Alabama Research Committee to C. U. P. for partial support of this work. [6/640 Received, 4th April, 19761 24 W. J. Hehre, R. F. Stewart, and J. A. Pople, J. Chem. Phys., 1969, 51, 2657. 25 R. Ditchfield, W. J. Hehre, and J. A. Pople, J. Chern. Phys., 1971, !54, 724. 2e F. P. Boer, J. Amer. Chern. SOC.,1968, 90, 670. 27 J. M. LeCarpentier and R. Weiss, Acta Cryst., 1972, B28,1421. 28 W. A. Lathan, W. J. Hehre, and J. A. Pople J. Amer. Chem. SOC.,1971, 93, 808. 29 W. A. Lathan, W. J. Hehre, L. A. Curtis, and J. A. Pople,J. Amer. Chew. SOC.,1971, 93, 6377.
ISSN:1472-779X
DOI:10.1039/P29760000621
出版商:RSC
年代:1976
数据来源: RSC