Heterocyclic dyes displaying excited-state intramolecular proton-transfer reactions (ESIPT): computational study of the substitution effect on the electronic absorption spectra of 2-(2′-hydroxyphenyl)-1,3-benzoxazole derivatives

Heterocyclic dyes displaying excited-state intramolecular proton-transfer reactions (ESIPT): computational study of the substitution effect on the electronic absorption spectra of 2-(2′-hydroxyphenyl)-1,3-benzoxazole derivatives

Please download to get full document.

View again

of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Sheet Music

Publish on:

Views: 2 | Pages: 5

Extension: PDF | Download: 0

  J. Chem. Soc ., Perkin Trans. 2 , 1999, 1123–1127 1123 Heterocyclic dyes displaying excited-state intramolecular proton-transfer reactions (ESIPT): computational study of thesubstitution e ff  ect on the electronic absorption spectra of 2-(2  -hydroxyphenyl)-1,3-benzoxazole derivatives† Maximiliano Segala, Nei Sebastião Domingues Jr., Paolo Roberto Livotto and Valter Stefani* Instituto de Química, Universidade Federal do Rio Grande do Sul, Av. Bento Gonçalves, 9500,Porto Alegre, RS, Brazil 91501-970 Received (in Cambridge) 10th February 1999, Accepted 14th April 1999 Semi-empirical molecular-orbital methods were used to simulate the electronic absorption spectra of a series of 2-(2  -hydroxyphenyl)-1,3-benzoxazole derivatives, namely AM1 and MNDO-PM3 for geometry optimization andINDO/S-CI and HAM/3 for spectroscopic features. Wavelengths of maximum absorption that agree better withexperimental data were found when INDO/S-CI was applied to PM3-generated inputs. Chemical substitution red-shifted the absorption spectrum of all the model compounds, a feature discussed based on the calculated energy levelsof frontier orbitals and charge redistribution upon electronic excitation. Introduction Chemical compounds displaying excited-state intramolecularproton-transfer reactions (ESIPT) have long been known fortheir uses as UV-stabilizers, 1  lasing dyes, 2  and, more recently,biological probes, 3,4  as well as their plastic scintillation appli-cations 5  and optical nonlinearities. 6  Much has been establishedabout the mechanisms which underlie their photophysicalbehaviour, 7  but the tuning of electronic spectra is still a matterof trial and error, often involving laborious and costly labor-atory preparations. In a previous paper, 8  we used compu-tational tools to investigate the molecular geometry andspectroscopic properties of 2,5-bis(1,3-benzoxazol-2  -yl)hydro-quinone (BBHQ, Fig. 1) and its monomethyl derivative(2,5-bis(1,3-benzoxazol-2  -yl)-4-methoxyphenol, BBMP),chosen as model compounds to study ESIPT, comparing theresults to experimental data. We found that structural and elec-tronic features provided by semi-empirical AM1 calculationsdescribe well the mechanism for ESIPT, yielding conformerassignment consistent with expensive high level ab initio  studies.Spectroscopic calculations using the INDO/S-CI and HAM/3methods were proved to be valuable for the prediction andinterpretation of the visible absorption spectrum of BBHQ andBBMP.2-(2  -Hydroxyphenyl)-1,3-benzoxazole (HBO, 1 , Fig. 2) iswell known to be of importance as the basic skeleton of molecular structures which can undergo ESIPT, includingBBHQ, and many experimental studies on its photophysics areavailable. 9,10  Calculated electronic ground and excited state Fig.1 2,5-Bis(1,3-benzoxazol-2  -yl)hydroquinone dyes (BBHQ,R  = H; BBMP, R  = Me). ONONORHO †Tables containing calculations of optimized geometries, transitionenergies and oscillator strengths for 13 model compounds are availableas supplementary data. For direct electronic access see http://www.rsc.org/suppdata/p2/1999/1123, otherwise available from BLDSC(SUPPL. NO. 57550, pp. 6) or the RSC Library. See Instructions forAuthors available via  the RSC web page (http:www.rsc.org/authors). energies suggest a qualitative four-level diagram (Scheme 1).Bhattacharyya et al  . 11  have discussed its preferred conform-ations and assigned a planar conformer as that responsible forabsorption. According to them, this should be the result of aconjugative trend towards coplanarity of aromatic rings. Infact, thermal activation at room temperature is of the sameorder as the calculated energy di ff  erence between rotamers, sotwo signi fi cant species should coexist, giving rise to di ff  erentabsorptions and emissions. A broad absorption centered at318.7 nm has been assigned to the most stable rotamer (form E 0 in Scheme 1). 12  An intense fl uorescent emission with maximumyield at 486.2 nm follows, with the large Stokes-shift beingexplained in terms of an ultra-fast ESIPT process occurring onE 1  which leads to keto–enol tautomerization. The observed Fig. 2 Benzoxazole structures (see Table 1). NOOHRR'R''  54216'5'3'2'1' Scheme 1 ESIPT mechanism. NOOHNOOHHN ONOOHNOOHE 0 E 1 absorptionESIPTemissionK 0 K 1  1124 J. Chem. Soc ., Perkin Trans. 2 , 1999, 1123–1127 fl uorescence is then the result of a transition from K 1  to theunstable K 0  form, which regenerates E 0  once the proton trans-fer has been reversed. Theoretical studies using ab initio  andsemi-empirical methods have been carried out on this com-pound, with the emphasis on the relative stabilities of the enoland keto tautomers and their conformers in the S 0  and S 1 states. 13,14 Spectroscopically parameterized computational methodshave made possible a di ff  erent approach to photophysicalinvestigations. Codes such as INDO/S-CI and HAM/3 havebeen successfully applied to systematic studies involving abroad range of organic compounds. 15–19  These methods havemainly been used for predicting the maximum wavelengths of absorption and emission using sets of molecular orbital param-eters empirically developed for this purpose. 20  Both methodsalso provide an indication of transition intensity by means of a calculated oscillator strength, 21  and, although less satisfac-torily, have been used to calculate other spectroscopic relatedproperties. 15,16,22 The successful results obtained on BBHQ prompted us toapply semi-empirical methods to examine the e ff  ect of chemicalsubstituents on the fi rst excitation energy (wavelength of max-imum absorption) of the HBO skeleton from a theoreticalstandpoint. In this respect, HBO is smaller and should requireless time-consuming calculations than BBHQ. A set of samplestructures has been conceived, their geometries fully optimizedin their fundamental states according to the semi-empiricalmethods PM3 and AM1, and their visible absorption spectrasimulated with INDO/S-CI and HAM/3. The validity of thesemethods and their combinations is judged on the basis of avail-able experimental data and some conclusions are drawn con-cerning the e ff  ect of the substitution pattern on the wavelengthof maximum absorption for the chosen skeleton. Molecular orbital calculations The model compounds were investigated in the ground state byusing the AM1 23  and MNDO-PM3 24  semi-empirical methods,as implemented on the UniChem 4.1 package. 25  All structureswere fully optimised at the Hartree–Fock level on the basis of our previous experience. 8  The S 0  state of the E 0  conformer wasassumed to be the most populated, hence it was used as theinput for the spectroscopic calculations. Computations wereperformed at the HAM/3 (Hydrogenic Atoms in Molecule, Ver-sion 3) 26  and INDO/S-CI (Intermediate Neglect of Di ff  erentialOverlap/Spectra-Con fi guration Interaction, as implemented onthe Argus package) 27  levels to evaluate S 0  –S 1  transition ener-gies. The 100 con fi gurations of lowest energy were taken intoaccount, namely those obtained by exciting one electron fromone of the 10 highest occupied to one of the 10 lowest un-occupied molecular orbitals, using the half-electron approachin the CI treatment. Calculations were carried out on CrayY-MP2E and T94 supercomputers. Table 1 Structures investigatedStructure 12345678910111213 RHHHHHCOOHCOOHCOOMeCOOHCOOHCOOHCOOEtCOOHR  HNH 2 NCSHNCSNH-tBuHHOHOHNO 2 NO 2 NH 2 OHHR  HHHHNH 2 NO 2 HHNH 2 HNO 2 HNH 2 Molecular geometries Geometries were fully optimised for thirteen molecules (Table1) as their most stable species (E 0  in Scheme 1). Completereports are available as supplemental material.† There is in gen-eral a good agreement among the results yielded by PM3, AM1,and those reported in previous theoretical studies. 28,29  Nonethe-less, PM3 reproduces better the H-bond distances in the site of proton transfer, although it predicts smaller valence angles.Table 2 shows the average values of the geometry datayielded by PM3 and AM1 calculations for the thirteen modelcompounds. Even though AM1 values for N-3   –H bond dis-tances and O–C-2–C-1 bond angles (see Fig. 2) are larger thanthose indicated by PM3, the small discrepancy between datarendered by the di ff  erent methods should be pointed out. Inparticular, PM3 computations predict a planar structure forthe conformer E 0 , in agreement with MNDO. 10,30  Electronicabsorption and jet-cooled 1 H-NMR spectra support thisresult. 30  Concerning structures 9  and 13 , AM1 renders adihedral angle within 1 and 2 degrees between the aromaticrings.The AM1 method has been extensively tested and proved tobe very useful in reproducing several properties among manyorganic compounds. 31  PM3 still lacks con fi rmatory evidence of its adequacy in a broad sense. On the other hand, it has sup-posedly enhanced functions for calculating bond e ff  ects at dis-tance. 24  While the AM1 and PM3 methods both use the samesemi-empirical implementation of the NDDO approximationthey di ff  er in the parametrization procedure. These facts justifythe application of both methods to the same set of molecules,since data for direct comparison thus become available. Of relevance to the present work is the fact that apart from smallchanges in the H-bond site neither of the methods indicatesthat the introduction of substituents on HBO a ff  ects the planarconformation and geometric features of its basic skeleton. Electronic absorption spectra The INDO/S-CI and HAM/3 computed S 1 ← S 0  transition ener-gies and oscillator strengths for the absorption spectra of themodel compounds are reported in Table 3, together with thecorresponding experimental data. The complete set of data isavailable as supplemental material.† Energy values refer tovertical transitions starting from AM1 and PM3 optimizedgeometries for the E 0  conformer with an H-bond involving theN atom, in agreement with the assignment of the most popu-lated species in the ground state. Spectroscopic calculations onthe HAM/3 level were not performed on structures 3  and 4  dueto a lack of parameters for the S atom.Direct comparison of the results yielded by INDO/S-CI andHAM/3 shows a similar trend for the e ff  ect of substituents onthe electronic spectra of the structures under investigation. Agood linear correlation is observed between the calculated Table 2 Average of the measures obtained in both geometric methodsBondLength/ÅAngle/  Dihedral/  N-3   –HO–HO–C-2C-2–C-1C-1–C-2  C-2   –N-3  C-2   –N-3   –HN-3   –H–OH–O–C-2O–C-2–C-1C-2–C-1–C-2  C-1–C-2   –N-3  C-2–C-1–C-2   –N-3  AM12.1660.9701.3661.4101.4521.33390.78139.05110.51126.16122.52130.920.70PM31.8460.9671.3531.4141.4531.33896.26145.20108.98124.35119.37125.780.10  J. Chem. Soc ., Perkin Trans. 2 , 1999, 1123–1127 1125 Table 3 Transition 0–0 and oscillator strengthExperimentalAM1/INDOPM3/INDOAM1/HAMPM3/HAM 12345678910111213 E  /eV3.75 a 3.37 b 3.56 b 3.68 b  —  — 3.21 c 3.23 c  —  —  — 3.19 d   — log ε 4.173.494.054.07 —  — 4.003.91 —  —  — 3.84 —  E  /eV3.793.663.693.723.573.573.523.533.353.683.303.543.36  f  0.710.540.580.660.791.030.690.690.720.680.510.680.88 E  /eV3.833.683.713.773.683.613.513.523.553.783.343.523.54  f  0.670.490.530.610.760.960.530.520.730.640.480.520.82 E  /eV3.362.86 —  — 2.902.772.792.782.052.721.862.802.35  f  0.480.27 —  — 0.820.400.260.250.600.450.210.260.84 E  /eV3.382.89 —  — 2.992.632.772.772.252.861.822.782.54  f  0.360.27 —  — 0.730.370.280.290.580.480.190.290.78 a Ref. 12. b Ref. 4. c Ref. 38. d  Ref. 39. wavelengths of maximum absorption for HBO and its deriv-atives and experimental values (Fig. 3). Correlation coe ffi cientsfor the linear regressions are: for INDO/S-CI, 94 and 96% forcalculations using AM1 and PM3-optimised inputs, respect-ively; for HAM/3, 95 and 98% for data based on the samemethods, in the same sequence. This indicates that inputgeometries generated with PM3 better reproduce experimentalexcitation energies under both spectroscopic methods. Thesame behaviour is presented for oscillator strength.Concerning HAM/3, relative values ( i.e.  compared to HBO)of the transition energies are quite good compared to experi-mental fi ndings and showed the same trend as those generatedby INDO/S-CI. Absolute energies, however, are low comparedto both experimental and INDO/S-CI yielded data. The bestapproximations to experimental data were always given by thecombination of PM3 and INDO/S-CI. The excitation energiescalculated at the HAM/3 level for the fi rst π * ← π  transitions areunderestimated by circa  0.5 eV. Underestimation of theseenergies is a characteristic not only of HAM/3, but also of CNDO. 32  This observation can be explained in terms of theFranck–Condon principle. It is known that electronic excitationinduces geometry changes on the absorbant species—here, E 0 .According to our calculations, the C-2–O and C-1–C-2   bondlengths are specially a ff  ected, suggesting a slight localization of double bonds. 8  Vertical excitation energies are then greater thanthe 0–0 transition energy.The electronic state associated with each transition can bequantitatively evaluated through the oscillator strength, whichre fl ects the typical relationship between molar absorption co- Fig. 3 Maximum absorption against fi rst excitation energies (eV):   AM1/INDO,   AM1/HAM,  ; PM3/INDO and   PM3/HAM.(The numbers in the fi gure refer to the compounds in Table 1). e ffi cients and allowed, partially allowed, or forbidden transi-tions. 21  According to our calculations, the electronic transitionsunder consideration are of the π * ← π  type. It is known that thepresence of H-bonds in solution destabilises the n, π * states andstabilises the π , π * ones, leading in some cases to energy inver-sion. Our calculations show such an inversion for HBO and itsderivatives, which should be due to intramolecular interactionbetween the H-atom and the free electron pair of N-3  . There-fore, π * ← n transitions are shifted to higher frequencies. Similarsystems have shown the same behaviour. 33  The wavelength of maximum absorption is always predicted to srcinate from asinglet–singlet transition. Even though HAM/3 yields nearlydegenerate triplet states, a low yield of intersystem crossing(ISC) is expected due to reduced spin–orbit coupling. 1 a ,34 Calculated data show that all S 1 ← S 0  transitions are closelyrelated to the HOMO–LUMO orbitals, and both INDO/S-CIand HAM/3 indicate charge-transfer from aromatic rings to theoxazole ring (Fig. 4). This is also supported by the variation of dipole momentum observed as a consequence of these transi-tions. The e ff  ect of substituents on color depends largely on theelectron-donor/electron-acceptor power of the substituents.There is a red-shift in the fi rst absorption band of all derivativescompared to that of HBO, probably caused by destabilizationof the HOMO orbital due to the greater electronic density.Table 3 shows small changes in the theoretical descriptionof singlet electronic absorption spectra on going from HBO toits derivatives; more speci fi cally, there is a slight variation of energy and oscillator strength for π * ← π  transitions. As a gen-eral trend, all substituents lead to a red-shift, but replacementof H at the R, R  , or R   position (see Fig. 2) does result indistinct bathochromic shifts for the wavelengths of maximumabsorption. Again the e ff  ect of chemical substitution dependson the electron-donor/acceptor power of the substituents. Thered-shifts calculated for HBO derivatives as compared to theparent compound can be interpreted in terms of the energylevels of the frontier orbitals as an indication of HOMOdestabilization.According to our calculations, the wavelength of maxi-mum absorption undergoes greater shifts when an electron-withdrawing group is placed at position R   ( 6   versus   10 ), andthis e ff  ect increases when position R   is occupied by anelectron-donor moiety ( 7   versus   10 ). It has been reported that Fig. 4  π -Electronic density changes in HBO and derivatives usingINDO/PM3 calculations for the electronic S 0  –S 1  transitions (greycircles  = increase; open circles  = decrease). NOOH  1126 J. Chem. Soc ., Perkin Trans. 2 , 1999, 1123–1127modi fi cation of 2-(2  -hydroxyphenyl)benzimidazole (HPB) 35 by the introduction of an electron-withdrawing substituentat the position corresponding to R   considerably shifts itswavelength of maximum absorption to the red end of thevisible spectrum. 2 d  Electron-withdrawing groups such as ethoxycarbonyl andnitro placed at R and R   ( 11 ) seem to be more e ff  ective inpromoting the red-shift, although at the expense of a reduc-tion in oscillator strength. The electron-donor groups amino,isothiocyanate, and N-tert -butylthioureide ( 2 , 3  and 4 ) placedat position R   have little e ff  ect on the absorption maximum of HBO. The existence of such an e ff  ect should be due to aninternal electronic resonance.The good agreement between experimental and calculateddata (Fig. 3) proves the usefulness of the employed theoreticalmethods in predicting the optical properties of the class of compounds under investigation. One could expect that theenergy e ff  ect of a double substitution should be roughly the sumof the e ff  ects of the two single substitutions, however there is asynergistic e ff  ect, probably due to the two aromatic rings beingconnected to each other, not acting as an isolated system. Alsoto a lesser extent it is due to the intramolecular H-bonding inter-action which stabilises the system. 36,37  Unsynthesized structure 11 , for instance, is expected to red-shift the simulated absorp-tion maximum of HBO (3.83 eV according to PM3-INDO) by0.49 eV (to 3.34 eV according to the same methods). The e ff  ectcan be attributed to electronic delocalization promoted by theelectron-donor moiety amino at position R   added to the e ff  ectof the nitro group at R  . This assumption is supported by theabsorption values found for molecules 2  and 6 . Electronic densities The results calculated for all the studied compounds can beexplained on the basis of electron donation and withdrawal.INDO/S-CI simulations also provide some insight into the elec-tronic rearrangement which follows a 1 S 1 ← 1 S 0  transition. Thisis illustrated for HBO in Fig. 4. The π -electron density on the6-membered ring of the benzoxazole moiety decreases. In theirturn, O-2, C-2  , and N-3   gain electronic density upon molecu-lar excitation, while C-1, C-2, and O-1   have the oppositee ff  ect—characterizing a typical donor–acceptor chromogen.Proton transfer is expected to be strongly favored in the fi rstexcited state, since data indicate an increase of the π -electrondensity on N-3   and a reduction on O-1  . Indeed, di ff  erences inacidity and basicity are expected upon excitation. 36  Derivatives 9  and 13  showed the largest positive electron load variations onN-3  , but without a corresponding reduction of electronic dens-ity over O-2. As judged by the spectroscopic results observedfor the whole set of simulated structures, this suggests a specialrole for N-3   in the red-shift of absorption maxima. The acid– base reasoning also explains the greatest red-shift calculated forstructures presenting an electron-donor at R   and an electron-acceptor at R  . Conclusion Semi-empirical methods, namely MNDO-PM3 and AM1 (forgeometry optimization) and INDO/S-CI and HAM/3 (for spec-troscopic analysis), were applied to simulate the structural andelectronic features of a series of HBO derivatives. The calcu-lated quantities were compared among themselves and withexperimental data. From the results obtained, one notices that:(i) spectroscopic calculations depend on the input geometry,PM3 being superior to AM1 for structural simulation; (ii)compared to HAM/3, INDO/S-CI gives values for the energiesof maximum absorption which better reproduce experimentaldata and (iii) the e ff  ect of substituents on the calculated elec-tronic absorption spectra of the structures under investigationis well rationalized with its basis in electron-donor and electron-acceptor characteristics and electronic loading promoted byoptical excitation.Although uncertainties arise from the neglect of solvente ff  ects and many approximations have been made in both spec-troscopic methods, the results obtained are in good agreementwith experimental data. Our fi ndings support all the availabledata concerning HBO and its derivatives and can possibly beextended to a wider range of compounds displaying ESIPT.It seems to be clear that suitable substitution can open a wide fi eld of research for HBO derivatives and related compounds,and that INDO/S-CI will allow new approaches on the theor-etical building of new materials for a broad range of appli-cations. This highlights the complementarity between theoryand experiment for the solution of complex problems. Withtheoretical guidance, one can select a few candidates foractual synthesis, saving the time and money which would beused in doubtful synthetic work. At the moment, our group isinvestigating synthetic routes to derivatives 5  and 9 , as well assimulating the photophysical behaviour of a series of benzimidazoles. Acknowledgements The authors are grateful for fi nancial support and scholar-ships from The Conselho Nacional para o DesenvolvimentoCientí fi co e Tecnológico (CNPq) and The Fundação deAmparo a Pesquisa do Estado do Rio Grande do Sul(FAPERGS). Computational facilities were made available byThe Centro Nacional de Supercomputação da UniversidadeFederal do Rio Grande do Sul (CESUP-UFRGS). We wouldalso like to thank Paulo Fernando Bruno Gonçalves andCristiano Krug for their important contributions. References 1( a ) J. E. A. Ottersted, J. Chem. Phys. , 1973, 58 , 5716; ( b ) W. Klöp ff  er, Adv. Photochem. , 1977, 10 , 311; ( c ) D. B. O’Connor, G. W. Scott,D. R. Coulter, A. Gupta, S. P. Webb, S. W. Yeh and J. H. Clark, Chem. Phys. Lett. , 1985, 121 , 417.2( a ) A. U. Acuña, A. Costela and J. M. Muñoz, J. Phys. Chem. , 1986, 90 , 2807; ( b ) A. Costela, F. Amat, J. Catalán, A. Douhal, J. M.Figuera, J. M. Muñoz and A. U. Acuña, Opt. Commun. , 1987, 64 , 457; ( c ) A. U. Acuña, F. Amat-Guerri, A. Costela, A. Douhal,J. M. Figuera, F. Florido and R. Sastre, Chem. Phys. Lett. , 1991, 187 , 98; ( d  ) M. L. Ferrer, A. U. Acuña, F. Amat-Guerri, A. Costela,J. M. Figuera, F. Florido and R. Sastre, Appl. Opt. , 1994, 33 , 2266;( e ) R. Sastre and A. Costela, Adv. Mater. , 1995, 7 , 198.3A. Sytnik and M. Kasha, Proc. Natl. Acad. Sci. USA , 1994, 91 ,8627.4M. G. Holler, MSc Thesis, Universidade Federal do Rio Grande doSul, 1997.5A. Pla-Dalmau, J. Org. Chem. , 1995, 60 , 5468.6X. Shang, G. Tang, G. Zhang, Y. Liu, W. Chen, B. Yang andX. Zhang, J. Opt. Soc. Am. B  , 1998, 15 , 854.7M. A. Rios and M. C. Rios, J. Phys. Chem. A , 1998, 102 , 1560.8N. S. Domingues Jr., C. Krug, P. R. Livotto and V. Stefani, J. Chem.Soc. , Perkin Trans. 2 , 1997, 1861.9G. J. Woolfe, M. Melzig, S. Schneider and F. Dorr, Chem. Phys. ,1983, 77 , 213.10Th. Arthen-Engeland, T. Bultmann and N. P. Ernsting, Chem. Phys. ,1992, 163 , 43.11K. Bhattacharyya, D. Nath, D. Majumdar, A. Ghosh, N. Sarkarand K. Das, J. Phys. Chem. , 1994, 98 , 9126.12A. Mordzinski and A. Grabowska, Chem. Phys. Lett. , 1982, 90 , 122.13K. C. Hass, W. F. Schneider, C. M. Estevez and R. D. Bach, Chem.Phys. Lett. , 1996, 263 , 414.14I. A. Z. Al-Ansari, J. Lumin. , 1997, 71 , 83.15A. E. Obukhov, Laser Phys. , 1996, 6 , 890.16A. L. Marzinzik, P. Radenacher and M. Zander, J. Mol. Struct. ( THEOCHEM  ), 1996, 375 , 117.17K. Endo, Y. Kaneda, M. Ainda and D. P. Chong, J. Phys. Chem.Solids , 1995, 56 , 1131.18J. T. Francis and A. P. Hitchcock, J. Phys. Chem. , 1994, 98 , 3650.19A. Grabowska, K. Kownacki and L. Kaczmarek, J. Lumin. , 1994, 60  –  61 , 886.  J. Chem. Soc ., Perkin Trans. 2 , 1999, 1123–1127 1127 20E. Linddhonlm and L. Asbrink, Molecular Orbitals and theirEnergies, Studied by the Semi-empirical HAM Method  , Springer-Verlag, New York, 1985, p. 142.21P. W. Atkins, Molecular Quantum Mechanics , 2nd edn., OxfordUniversity Press, New York, 1983.22Y. Takahata, J. Mol. Struct.  ( THEOCHEM  ), 1995, 335 , 229.23M. Dewar, E. Zoebisch, E. Healy and J. Stewart, J. Am. Chem. Soc. ,1985, 107 , 3902.24J. Stewart, J. Comput. Chem. , 1989, 10 , 209.25W. Thiel, Program MNDO94, version 4.1, 1994.26L. Asbrink, E. Lindholm and C. Fridh, Chem. Phys. Lett. , 1977, 52 ,63.27J. Ridley and M. C. Zerner, Theor. Chim. Acta , 1973, 32 , 111; 1976, 42 , 223.28M. A. Rios and M. C. Rios, J. Phys. Chem. , 1995, 99 , 12456.29S. Nagaoka, A. Itoh, K. Mukai and U. Nagashima, J. Phys. Chem. ,1993, 97 , 11385.30N. Ernsting, Th. Arthen-Engeland, W. Thiel and M. Rodriguez, J. Chem. Phys. , 1992, 97 , 3914.31J. J. P. Stewart, J. Comput. Aided Mol. Des. , 1990, 4 , 1.32Y. Takahata, J. Mol. Struct.  ( THEOCHEM  ), 1993, 283 , 289.33J. Catalán, J. Palomar and J. L. G. Paz, J. Phys. Chem. A , 1997, 101 ,7914.34L. Lavtchieva, V. Enchev and Z. Smedarchina, J. Phys. Chem. , 1993, 97 , 306.35A. Douhal, F. Amat-Guerri, M. P. Lillo and A. U. Acuña, J. Photochem. Photobiol. A , 1994, 78 , 127.36I. M. Brinn, C. E. M. Carvalho, F. Heiser and J. A. Miehe, J. Phys.Chem. , 1991, 95 , 6540.37J. J. Novoa and M.-H. Whangbo, J. Am. Chem. Soc. , 1991, 113 ,9017.38V. Stefani, A. A. Souto, A. U. Acuña and F. Amat-Guerri, DyesPigm. , 1992, 20 , 97.39D. S. Correa, DSc Thesis, Universidade Federal do Rio Grande doSul, 1999. Paper 9 / 01158G 
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks