Molecular Orbital Theory of the Electronic Structure of Organic Compounds

Transcription

1 Molecular Orbital Theory of the Electronic Structure of Organic Compounds IV. A CNDO/S-CI SCF MO Study on the Lower Electronic States of Large Molecules. Singlet-triplet Transitions of Dioxodiazacycloalkanes Horacio Grinberg* Departamento de Quimica Organica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, 1428 Buenos Aires, Argentina Julio Marañon * and Oscar M. Sorarrain * * Laboratorio de Física Teórica, Departamento de Física, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, 1900 La Plata, Argentina Z. Naturforsch. 37a, (1982); received October 22, 1981 The semiempirical molecular orbital CNDO/S-CI spectral parameterization has been used to elucidate the lower triplet electronic states of a series of dioxodiazacycloalkanes. The l 3 B2(no-T*) and l 3 A 2 (no-t*) triplet spectroscopic states involve intramolecular charge transfer from the oxygen to the carbon atom of the carbonyl group, which is supported by electron density calculations of these excited states. The solvation energy was incorporated in the calculations. Introduction In previous papers some ground state molecular properties of a restricted set of dioxodiazacycloalkanes (Fig. 1) [1] and of the cyclol and bicyclic lactam derived from I (m = n = 3) [2] have been obtained with the CNDO/2 and INDO SCF semiempirical methods. More recently, a detailed analysis of the lower spectroscopic states associated with the singlet-singlet electronic transitions of I using the CNDO/S-CI methodology [3] was performed [4]. 0 This lent credence to the applicability of the CNDO/S-CI parameterization and led us to a consistent account of the electronic spectra of these molecules in terms of the closely related electronic states. The trends in the electronic states were related to trends found in the ground state molecular orbitals (MO's). Thus, the interplay of the experimental and computed transition energies could lead to a better characterization of the excited states if an analysis of the triplet states of I could be performed. To this end a description of the lower singlet-triplet electronic transition of I upon the application of the CNDO/S-CI procedure will be presented. Computational Procedure H Fig. 1. Dioxodiazacycloalkanes (I) computed. m = 2, 3 ; TO = 2, 3, 4, 5. Molecular symmetry group is C 2 v for I (to = 2, n = 2;m = 2,TO= 4) and Cs for I (to = 2,to= 3 ; m = 2, n = 5; TO = 3, n = 2; m = 3,to= 3; m 3, n = 4). * MCIC CONICET, Repüblica Argentina, ** MCIC CIC, Argentina. Provincia de Buenos Aires, Repüblica Reprint requests should be addressed to Dr. Horacio Grinberg, Departamento de Quimica Organica, Facultad de Ciencias Exactas y Naturales, Ciudad Universitaria, 1428 Buenos Aires, Argentina / 82 / $ 01.30/0. Complete computational details were given in [4]. In the present work we follow the same line of reasoning except for the fact that we have choosen the Pariser-Parr interpolation method [5] instead of the Nishimoto-Mataga approximation [6] for the calculation of the two-center electron repulsion integrals. In fact, it has generally been concluded that when these integrals are used with the CNDO/S parameterization, the Nishimoto-Mataga integrals give better estimates of singlet states, while the Pariser approximation is better suited for the calculation of triplet states [7, 8]. - Please order a reprint rather than making your own cop}'. Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschung in Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.v. digitalisiert und unter folgender Lizenz veröffentlicht: Creative Commons Namensnennung-Keine Bearbeitung 3.0 Deutschland Lizenz. This work has been digitalized and published in 2013 by Verlag Zeitschrift für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution-NoDerivs 3.0 Germany License. Zum ist eine Anpassung der Lizenzbedingungen (Entfall der Creative Commons Lizenzbedingung Keine Bearbeitung ) beabsichtigt, um eine Nachnutzung auch im Rahmen zukünftiger wissenschaftlicher Nutzungsformen zu ermöglichen. On it is planned to change the License Conditions (the removal of the Creative Commons License condition no derivative works ). This is to allow reuse in the area of future scientific usage.

2 H. Grinberg et al. Molecular Orbital Theory of the Electronic Structure 233 where D is the dielectric constant [11], Qp is the net charge on the p center evaluated at each excited state from expression (3) (see below) and (pp/qq) (1) AE = 6; et Ju, are the bicentric atomic integrals, was introduced where and ej are the energies of the initial and in each spectroscopic state before and after CI was final orbitals, and Jij is the molecular Coulomb performed. The corresponding correlation of these integral. excited states is given in Figure 2. Solvents used The molecules indicated in Fig. 1 will be identified were chloroform-acetic acid for (2,2), water-dmf as (m, n) where m and n are the numbers of methyl- for (2,3), and acetic acid for the rest of the moleene groups of the carboxylic and amine moieties, cules [12]. respectively. The spectroscopic CNDO/S-CI techsince the atomic charge densities in the different nique developed by Jaffe and co-workers [3] and excited states vary widely, the solvation energy used by us in our initial investigation [4] was ob- shows no regular variation, i.e. an inversion of tained from the QCPE [9]. The program was certain spectroscopic levels was detected when the modified to suit local input-output requirements participation of the solvent was allowed for. Thereand to facilitate changing dimension statements. fore, in order to avoid spurious crossings, the Excited states were generated from the ground correlation as given in Fig. 2 was limited to the state occupied and virtual orbitals through a con- lower lying triplet excited states, and hence the figuration interaction procedure between the 40 subsequent discussion will be restricted to this lowest-energy, singly excited states. The restriction problem. to singly excited states means that essentially the As long as the low-energy n*^-n transitions in IVO-method was used. these compounds cannot be explained on the basis of the amount of interaction of the nonbonding Results and Discussion pairs, we must look elsewhere for an explanation. As long as the Pariser-Parr approximation to the A clue can be obtained from our previously reported two-center Coulomb repulsion integrals is used, the results [4]. However, before discussing the n*-< n eigenvectors will differ from those previously com- triplet transitions in the light of Fig. 2, it should puted using the Mataga method [4], However, the be recalled that every attempt to explain spectrosame ordering and symmetry pattern was observed scopic results on the basis of MO energy diagrams in the present calculations. Therefore, the ordering can be dangerous for two reasons [13]: and energies of selected occupied and virtual orbi(1) Observed spectroscopic states are frequently tals that play a part in the computed electronic mixtures of a number of spin configurations, and transitions as given by the ground state and virtual hence the transitions leading to such states cannot orbitals correlation will be omitted from the present be identified by simple one-electron excitations as discussion. implied in MO diagrams; In discussing the MO's resulting from these (2) MO diagrams do not take into account eleccomputations it is convenient to view each molecule tron repulsions in the excited state. as divided between a carbonyl portion and a donor Even with these limitations, the MO diagrams of portion attached to the carbonyl group, in this case our previous findings [4] did definitely indicate a the lone pairs on nitrogen and oxygen. The MO's that are most important in the com- very important trend. Thus, it was evident that the puted electronic transitions are the lowest unoc- order of the n* <r-n transition energies was detercupied MO's, i.e. >2(71*), a2(n*), and a"(n*), and mined by three important factors: the amount of the highest occupied orbitals of symmetry a\, b\, interaction of the nonbonding electrons, the extent and a', which are labeled a\(no), a'(no) of derealization of these electrons, and the relative energies of the MO's. according to the molecular symmetry group. Neglecting the solvent effect, the lowest energy The solvent effect as computed from the exprestriplet excited states calculated for each structure sion [10] (1 3 A 2, 1 3 B 2, 1 3 A", 2 3 A") is comprised almost Esoi = 1/2(1 - D-1) J4QPQq(pp!qq), (2) entirely of the nn* configuration. Thus, the (2,2) P,Q In the Hartree-Fock scheme the singlet-triplet transition is given by the expression

3 Fig. 2. Singlet-triplet correlation diagram of the lower electronic states of dioxodiazacycloalkanes (I). The symmetry of the spectroscopic states is given by the direct product of the symmetry of the molecular orbitals involved in the transition. See Ref. [4] for the symmetry of the different molecular orbitals. Experimental energies are shown in broken lines. N3 CO if^

4 236 H. Grinberg et al. Molecular Orbital Theory of the Electronic Structure triplet electronic transitions near 3 e V result f r o m the transitions &2-< «1, Q>i<-b\, bz-^-bi w h i c h in turn lie a t higher energies t h a n t h e corresponding transitions in (2,4). T h e same transition e n e r g y pattern is observed f o r 1 3 A ", 2 3 A " e x c i t e d s t a t e s of the lower s y m m e t r y compounds i n v o l v i n g t h e transition. E x c i t a t i o n s a t lower w a v e l e n g t h ( 3 A i a n d 3 A ' s y m m e t r y ) are m a i n l y due t o t h e nf=0-4-n-$, no transitions. These assignments are consistent w i t h the experimental f a c t t h a t the c a r b o n y l bands are blue-shifted b y t h e f o r m a t i o n of a h y d r o g e n b o n d with acetic acid as the proton donor, because these hypsochromic shifts characterize t h e absorption bands due to the n*< n transitions [14]. T h e triplet a n * states are c a l c u l a t e d t o be a t t h e same energy as the singlet a n * states of the same molecular configuration, since e x c h a n g e integrals vanish in the C N D O a p p r o x i m a t i o n. T h e n n * e x c h a n g e integrals w h i c h are included in the C N D O a p p r o x i m a t i o n, are e x p e c t e d t o be much larger t h a n t h e a n * e x c h a n g e integral. T h e 3B2(%OTT*) singlet-triplet splitting is calculated t o be 0.5 e V. Since t h e 3 A 2 ( n o n * ) is similar t o t h e 3 B i (nn*), it p r o b a b l y also has a splitting on t h e order of 0.5 e V. T h u s, t h e incorporation of an* exchange, although i t w o u l d increase t h e e n e r g y of the l 3 B2(wo7r*) a n d 13A2(WOTT*) b y a b o u t 0.5 e V, would not change t h e order of t h e first three triplet states. T h e triplet state energies, w i t h o u t the large positive contribution of the e x c h a n g e integral t o the nn* state, w h i c h is f o u n d for t h e singlet states, follow the same ordering as t h a t of the occupied MO's from which t h e y arise. T h u s, t h e trends in t h e energies of t h e l 3 B 2 ( % o ^ * ), l 3 A 2 (wo7r*), a n d 1 3 A " ( a n * ) are t h e same as those g i v e n for their singlet counterparts. T h e in-plane polarization of the l 3 A i, l 3 B i, 1 3 A ' spectroscopic states i n v o l v e d in the nn* transitions p r o b a b l y result f r o m intensity stealing f r o m transitions t h a t occur a t higher energies of the same molecular configuration induced b y a combination of solvent p e r t u r b a t i o n and non p l a n a r i t y of the amide g r o u p in the e x c i t e d state. T h e triplet spectroscopic state l 3 B i is c o m p u t e d to occur a t a n d e V for the molecules (2,2) and (2,4), respectively, w h e n t h e solvent effect is incorporated (Figure 2). T h e associated electronic transitions bz-t-az, (i2-* b2 are predicted t o be of 235 r e l a t i v e l y high intensity since t h e y are n * < n. T h e same trend is observed for the spectroscopic s t a t e l 3 A i. Here the computed energy (neglecting t h e solvent effect) is also calculated t o be lower for (2,4) t h a n for (2,2). T h e c o m p u t e d triplet transitions b e t w e e n 2 a n d 3 e V for the other molecules (C s s y m m e t r y ) in t h e series result f r o m transitions t o t h e a " states w h i c h arise f r o m transitions f r o m the a' (no) t o t h e a" (n*) orbitals a t energies below 4.5 e V (Figure 2). T h e electronic transitions of A ' symmetry (a" <-a") i n v o l v e lone pairs on nitrogen a n d o x y g e n. T h e stabilization of the a" (n*) is less t h a n t h a t of t h e a" (n). This results in the l3a'(nn*) being calculated in the narrow energy range e V (Fig. 2) along the series, e x c e p t for (2,5) where t h e spectroscopic transition is of the t y p e a' <-a' w i t h a high C I composition (0.9). F r o m F i g. 2 it can be observed t h a t t h e first t w o triplet spectroscopic states are m u c h lowered b y C I, since o n l y % of the composition of these excited states can be a t t r i b u t e d t o one molecular configuration. T h e a\(a) M O remains close in e n e r g y to the 62 (^0) e v e n t h o u g h the CL\ (a) no longer contains a high percentage of nitrogen lone pair character. I n particular, since the l 3 A2(?&o^*) a n d 1 3 B2(^On*) b o t h result f r o m a transition f r o m M O ' s t h a t are p r e d o m i n a n t l y on the o x y g e n t o a M O b2(n*) containing a b o u t 4 5 % C = 0 character [4], transitions t o these states result in charge transfer f r o m the o x y g e n t o the carbon a t o m of t h e c a r b o n y l group. Transitions t o the b2(n*) a n d ci2(n*) (symmetries A i a n d B i ) also i n v o l v e carb o n y l, o x y g e n t o carbon charge transfer, b u t t o a lesser degree since the a,2 (n) a n d &2 (n) eigenvectors possess more nitrogen character [4]. T h e same t r e n d is observed for the 1 3 A " and 2 3 A " e x c i t e d states of t h e lower s y m m e t r y compounds. These excited state properties can be rationalized b y analazing the charge densities of the amide group a t o m s in the various triplet spectroscopic states. I n the v i r t u a l orbital a p p r o x i m a t i o n t o t h e e x c i t e d state w a v e functions the atomic charge densities in t h e e x c i t e d state are g i v e n b y [3] q(a)f = q(a) + 2 2CZ(Clr-Cfr), m=l r= 1 (3) where q(a) is the ground state charge d e n s i t y on a t o m A, C'im is the C I coefficient for t h e contribution t o the i t h state of the mth electronic configura-

5 236 H. Grinberg et al. Molecular Orbital Theory of the Electronic Structure 236 Molecule b (TO, n) State C 0 N H (2, 2) 1 3 A2 2.9 io-i io-i I 3 B2 2.9 io-i io-i AI 3.2 io-i io-i 0.0 l 3 BI 3.2 io-i io-i AI 2.9 io-i io-i (2, 4) l 3 AI 3.4 io-i l 3 BI 3.4 io-i A I 3 B B (2, 3) 1 3 A' A' A" A" A" io-i (3, 2) 1 3 A' io-i 2 3 A' A" " " A" A" (3, 3) 1 3 A' A" io A" " A' A' (2, 5) 1 3 A' A" A" IO" A' IO" A' (3, 4) 1 3 A' A' A" IO" A" A" IO" 2 Table 1. CNDO/S-CI electron density difference for excited triplets (Pariser-Parr integrals) from ground state a. a Carbon and hydrogen atoms are those of the amide group. b See Figure 1. tion which arises from the excitation of an electron from orbital j to virtual orbital k' and Cjr and Ck'r are the coefficients of the rth atomic orbital on atom A in molecular orbitals j and k', respectively. In this work, the sum over m includes the 40 lowest energy electronically excited configurations, and the sum over r is again over the atomic orbitals on atom A. Results of the CNDO/S-CI atomic electron density differences for excited triplets from the ground state of the atoms involved in intramolecular charge transfer are compared in the Table 1. They follow the same general trend as found for the singlet states [4]. Particularly, the decrease of charge on the carbon is accompanied by a corresponding increase in electron density on the hydrogen atom in 1 3 A2, 1 3 B2, 1 3 A", and 2 3 A" excited electronic states. Thus the electronic repulsions which result from the yr-system are relieved by a derealization of electrons from the ring onto the hydrogen atoms through the a framework. The oxygen atom is predicted to be the most basic heteroatom in both the ground and triplet excited states. These results predict the nitrogen to bear a slight negative charge and the oxygen atom, while still the most negatively charged site in the molecule, to be less basic in the l 3 Ai, l 3 Bi, 1 3 A', 2 3 A', 1 3 A", and 2 3 A" excited states than in the ground state. Conclusions This study has demonstrated that CNDO/S calculations provide a reasonable description of

6 236 H. Grinberg et al. Molecular Orbital Theory of the Electronic Structure 237 various triplet state properties of the seven dioxodiazacycloalkanes considered. In particular, this method provided an excellent opportunity to examine the nature of the "lone pair" of electrons in this kind of molecules. It was further found that transitions to the n* states result in intramolecular charge transfer from the oxygen to the carbon atom of the carbonyl group. This probably causes a lengthening of the C=) bond distance due to the excitation of the lone pair electron to the TI*=0 orbital. It is hoped that this study will further serve as a reference for the investigation of ring contrac- tions via cyclol in excited states. Work along these lines is underway and will be published at a later time *. A cknoviledgem ents This work has been made possible by grants in aid from CONICET, Repüblica Argentina, and CIC (Provincia de Buenos Aires), Repüblica Argentina. * Note added in proof: Additional material associated with this work (extensive singlet-triplet correlations, solvation energy for the various spectroscopic triplet states, etc.) is available upon request. [1] J. Maranon, 0. Sorarrain, H. Grinberg, S. Lamdan, and C. H. Gaozza, Z. Naturforsch. 31a, 1677 (1976). [2] J. Maranon, 0. Sorarrain, H. Grinberg, S. Lamdan, C. H. Gaozza, Tetrahedron 34, 53 (1978). [3] J. Del Bene and H. H. Jaffe, J. Chem. Phys. 48, 1807 (1968); ibid. 48, 4050 (1968); ibid. 49, 1221 (1968). [4] H. Grinberg, N. S. Nudelman, J. Maranon, 0. Sorarrain, and C. Gömez, Z. Naturforsch. 36a, 494 (1981). [5] R. Pariser and R. G. Parr, J. Chem. Phys. 21, 767 (1953). [6] K. Nishimoto and N. Mataga, Z. Phys. Chem. Frankfurt 12, 335 (1957). [7] C. A. Masmanidis, H. H. Jaffe, and R. L. Ellis, J. Phys. Chem. 79, 2052 (1975). [8] H. M. Chang and H. H. Jaffe, Chem. Phys. Lett. 23, 146 (1973). [9] Calculations were performed on an IBM 360/50 computer with a QCPE program CNDO/S-CI. The modified closed-shell CNDO/S-CI program (QCPE No. 315) was obtained from the Quantum Chemistry Program Exchange, Indiana University, Bloomington, IN The molecular symmetry groups were automatically generated from the cartesian coordinates. [10] I. Jano, C. R. Acad. Sei. Paris 261, 103 (1965). [11] The dielectric constants used to calculate the solvation energy are intended to be the limiting values at low frequencies, the so-called static values. Solvents used are given in [12]. [12] H. Grinberg, S. Lamdan, and C. H. Gaozza, J. Heterocyclic Chem. 12, 763 (1975). [13] H. H. Jaffe, D. L. Beveridge, and M. Orchin, J. Chem. Ed. 44, 383 (1967). [14] G. J. Brealey and M. Kasha, J. Amer. Chem. Soc. 77, 4462 (1955).