«xy- «1..-. 93?." n. wkm91~7~i .. :NL'R k: . .2 . . we: M£~th .. 21333:: n. 2.. W ‘L... 35.; {IS-I. gun. I". -.~ 233L173": ‘4‘.qu . ......;‘~;-- 223%: Wu“... .. ,, 2.2::- :: 41:212.: . 'z: ' " :21“ "3/3"“ 2. ‘14. W W “,2...“ 3...; L523. . “MW #52:“ :..¢~....-L*-- , h . . , “1:21.... _ u w; x .N b "y ”-A“ [‘3’ utmw W‘fitmfic‘“ :2 .. $33, 375‘ .2: W w, oi‘mgmz‘; ‘7“ "(mi Lns‘JfA wan... I \m m uq. .3. “:25 LL 2.23%“ W"; .325. 2;" w...w:.‘“a..“2” W . ‘ghm IA fl.\‘~2¥-). :12.“ . . . 2. .. 2 - 22%;. 52.1% “‘“m... . .r. .2: Lu. vfixwuu 1.; 2 k» ; ~ ' " egg.- ‘72:} .' 23;.“1‘1 u.?~»}':..,.u“ 'Z'NTTR‘Z‘L‘ 2 - - - LP. .‘1-37‘3‘“ 3L 5.195va - 4423.“ «35?? 35: ~ " "Ha-wit. Hi ’1 W 1"”“‘ “.2122 L We: ‘ ka ‘ ‘2?» .22.“? L25 532.2% Chum; ~:~- “.mv-‘lfi. eke." W. yv'l- "-2- "231:“ 21:31”; ' ‘11 ' ~ .' Y:- a . at .01; ”$11351. wa‘xfx‘i ‘ \I‘ N v] 9- 4'va” -f;-dm:“51h»\ \E-IN\MN-\.\\V_ \Séa. .- wywut‘ Liv" J‘w «4.3:: :2 LLLLLLLLLM}: ' W‘ m . W- 2: 5:?“ "W2“? $7»\ .28; W": 31"" . .~ 23:; 3% - 112% ?é~wd. , u :q in} ‘ ’ “Lynx” Nun.“ ~. .u - ,L ~ u WK.“ “#:2212322, - g - A 1w WWW , "54%? 2 2.2:; 422"}: 22$? ‘ ‘2' W2?“ . L ”“222 “guifi’é . . .. - ~ '--2 . WW 33591:.22'2. 17: » 2' ‘2 ,. ~ L521 W11 x- ‘15.“ flW‘ L" 2 ., ”@225 . 535%.. ‘ 2 mm‘h‘itm?“ 32;: ... ”5&9? “‘2' '2 222222 g; . 2 » 2» 313;} 2 ". I . ~ . 9 .. if: 'v’ ' \- ~ . \. “‘“ 1:2", - “ ~ - "9. . In , - fiqm:“n mu. . - ' , » ‘9‘ . ‘ ‘ - :v::. :an“ w Lam-2129,13}: 2:;- “W ‘25 'Q'k “ . r6 ., :1: *2. 22 » » - ' 72.; ~ 4 - - .22 22222225222732.2552 » V. , . q. - . ~52: :32: A ,- 2. ‘ . - *2 , 2 2 33"” .223“- 2'2. ‘2‘ = ‘ .. 2 A: ' . ' ‘ ‘Al'. x: w: v ‘ ‘ V ' gffl' 2 .1 : I. .4 1‘3"" 2 ‘ . ' 2, ”WW' :2? 22:99.: 7121...; ' w "2‘2‘2 2?": 2 - '.: . ‘ " ' ' ‘ ~ ' n: ‘ ' ‘ 2 ;. LL; 2.“ » L . ”$535: 12:9 2 222 -. _ A .2 _ , w ,. .2 ll 3 . L, - L Vu LL 9. 7 -~ ‘ _ LLL .' L: . . L L L ‘L-Lth:..uh‘ .: a}. _ waif“ ' %,fi? ' ‘ . . i . _ ‘ v" . . '. -' ‘L' 2, ' 11.921. 2&1“). - Liv? ' 7' “a? “-1.1“: A: .1232.» 5%de ‘2 ”2&1? U 2 .~ V" -»’ ' A “ » 2 , . 2 7,. a fLL‘e. ' ' . . Mgr-,5: Eggafik 272,}: L :4»; ~ ". -.' .- — , - -~. ar‘ . c 532$?“ 32% 2}}?3... E: ’ . ’- 2 ‘ ‘ ' u ' High'g'f‘" 1’ -. ‘W m? 25:21 Hug-:3 2511;“. 2: 5‘. 3-35 ,' 2:3 - 2 2452222222' '2 2‘2””? ~ , .:~ ~ -. , : . .3 :2, ' 2 -; 22 32222922: “WW 212 lghh. n.“ w 3“. . \. -. t 722:. 2 :2 : : Iv... 1, . 2“ 2 22 2:22 . 2 4: waw 01$ { mi§iggifil§lw 2.. .Ji," 2 '25:“ . “71: .-v-. «gag? vi; :2," 3727‘? . 2'35 . L'r—Lsycg’ : L u’da 1 . v... 1 If $7.. .15.". 7f 4'; . ' ' f, =.. .'. '., $2: .30 - :2!- 3333. 172:1: -' min? ' xv“ “3:1? .-. a?” 431;}. my? . 2:;er ' 11 2 .5272 ‘:' Lp‘f : .‘f-EW'E: .i - a .1391." 22.; . L -J‘ .r} 'r' 2 n-~.-. - ”is. “£22” '3; . ‘ . 32:" , m f. .3 . ~11“ ' m... is, .' 337:1}:- .457 713‘ ‘ "‘14:; 2. 2:22.2- mmrui... ' MEm: $35.3” .1; . )‘3r' __».4.»".a. If _ ‘tm THESIS . LIBRHI‘: Y ,7 . MIChlgan Stam This is to certify that the thesis entitled SPECTROSCOPIC STUDIES OF SOME MODEL KETOPORPHYRINS presented by Gabriel Boktos has been accepted towards fulfillment of the requirements for M.S. Chemistry degree in I Major professor Date Februar 20 1981 0-7639 . OVERDUE FINES: 25¢ per du per item RETURNING LIBRARY MATERIALS: Place in book return toremo charge from circulation records l SPECTROSCOPIC STUDIES OF SOME MODEL KETOPORPHYRINS By Gabriel Boktos A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1981 wise ,3 } ABSTRACT SPECTROSCOPIC STUDIES OF SOME MODEL KETOPORPHYRINS By Gabriel Boktos Absorption, resonance Raman and infrared spectroscopies were employed to study the carbonylated derivatives of chlorins and isobacteriochlorins, viz. Cu-gemini monoketo— porphyrin, Cu—gemini diketoporphyrins, and Cu—a, y-dioxo- porphodimethene. This investigation focused on the observation of the stretching frequency and on the effect of the position of the carbonyl(s) on the macrocyclic ring on the electronic and vibrational spectra. Resonance Raman spectroscopy also provided information concerning the stretching frequencies of the CaCmCa and C8-CB bonds, the location of the carbonyl and how the perturbation by the carbonyl substituents relates to the overall structure of the molecule. These studies indicate that the positioning of the carbonyls around the pyrrole rings has only negligible effect on the CaCmCa stretching frequency. However, the 08—08 bond is perturbed by the carbonyl substituent(s) as exemplified by the great fluc— tuation of the 08—08 stretching frequency between 1567— 1583 cm_1' 7 a ii- I ' Gazer-me WIJN‘T' fig 11 luff Gabriel Boktos The infrared spectra also provided information about carbonyl stretching. The non-splitting or splitting of the infrared peak was attributed to the existence or lack of a mirror plane through the molecule, which makes the carbonyls equivalent or non—equivalent respectively. The infrared studies thus supported and supplemented the resonance Raman results. Finally, two additional compounds: Cu-transoctaethyl~ chlorin [Cu(OEC)] and Cu—3,13-diformyl,8,18—di—n—pentyl-2, 7,12,17,-tetramethylporphine [Cu(2F-6F) for short] were introduced for purposes of comparison. Raman spectra of both were obtained under Soret excitation. In Cu(OEC) vibrations above 1300 cm_1 are intensified under these conditions, in contrast to intensification of lower frequency modes under Q-band excitation. The characteristic C=O vibrations in Cu(2F—6F) and Cu(2—6) diketo differ by more than 35 cm_l; thus the formyl and ketone groups are easily distinguished in the substituted pcrphyrins. scams-E's" nit «92.. items be: "'H-QQIJB ..':.=.-'!. asfbuis u a "'7’ I . I l':'1 Dedicated to my mother. ii ACKNOWLEDGMENTS I would like to acknowledge the contributions of Professors George E. Leroi and Gerald T. Babcock, under whose guidance and support I completed this undertaking. In addition, the help of Professor Chris Chang, coupled with the preparative work by Richard Young, is gratefully appreciated and acknowledged. I would also like to express my gratitude to fellow students who took the time to converse with me and exchange ideas. So to B. Ward, J. McMahon, R. Blevins and to all my compatriots I express my deep appreciation and a big thanks. Finally, a special thanks to Margy Lynch and Beverly Adams, who ably and cheerfully helped in the preparation of this manuscript. '10 aflbiiflflii‘finéo 6d? eafiolwmzns 2w .--2'.' Ham 1 .-.. ' p' - -.- .5 — . a. - - . . 3.1m: dues-:53? n." net-29¢ r - :.- - . . -"E'."-7 2:1" 2:213 ' ’ TABLE OF CONTENTS List of Tables.......................................... List of Figures......... ..... .. ..... .......... ...... .... Chapter 1 Introduction.............. ............. ..... Chapter 2 Methods..................................... A. Absorption spectra of chlorins and related compounds. ......... ................. 1. Some theoretical aspects of heme absorption. ..... . ........... . ... ...... ..... 2. Results.... ........................... ...... B. Resonance Raman spectroscopy of chlorins and related compounds.. ........ .... 1. Some general theoretical aspects..... ..... .. 2. Resonance Raman results... ................ .. C. Infrared spectroscopy - Results ......... .... D. Comparison of resonance Raman bands of Cu(OEC) upon excitation at 413.1 nm with those of Ozaki, et al. at Aexc = 488.0 nm and other models............ Chapter 3 Conclusions........... ......... .. .......... . 1. Absorption spectra. ..................... .... 2. Resonance Raman spectra...... .............. . 3. Infrared spectra. ....... ... ............. .... Suggestions for further research. ........... .... References ........ . ....... . .......................... ... iv 28 28 35 51 60 67 67 68 7o 71 72 .selfisT ta talJ V nu...Ino.-no.-dsotno:.cbplnnouv|-nInc-I.... iv lugs-AII-COIICIIIII[III-Ill- -' .—.-_ LIST OF TABLES Energies and wavelengths of the chlorin electronic transitions............................. AU values under Soret excitation................... Energies of the carbonyl signals in both IIR. andR-RSIIIIII llllll O IIIIII IIIIIIIIIIIIIIIIC RRS peak positions and intensities of the Cu(OEC) bands under Soret and visible excitation.. ........................... ........... Symmetry correlation between C2V and Duh groups for the in-plane vibrations..... ...... ...... 25 49 59 63 65 'II' 3:13:55 .. ;.'_' 1‘) ...."'II.. ’3' "“5 "1319113 ”ifiuwfneta -! 1a. 1b. 10. \OCD'QOUI 10. 11. 12. 13. 14. 15. 16a. 16b. 17a. LIST OF FIGURES Schematic representation of laser Raman experimental arrangement................... Optical spectra of Ferrocytochrome C in CHZCl2 (conc. 0.5 mM)... .......... ............ Optical spectra of Fe(OEC)Cl. C in H Cl (conc. 0.3 mM) ............. ............... 2 2 Four orbital model (from reference #7)........... Shape and nodal characteristics of the four orbitals (from ref. #7) ....... .......... The 4- orbital model for OPP- THP and ADJ- TH P. ... Optical spectra of Cu(OEC)............ ....... .... Optical spectra of Cu(2F-6F)... ....... . ...... .... Optical spectra of Cu monoketo. ............... ... Optical spectra of Cu(2-3) diketo. ......... ...... Optical spectra of Cu(2—4) diketo................ Optical spectra of Cu(2-5) diketo..... ........ ... Optical spectra of Cu(2—6) diketo .......... ...... Optical spectra of Cu(a-y) dimeso......... ..... .. 22 nelectron conjugation pathway. ........... .... RR spectra RR spectra RR spectra RR spectra RR spectra of Cu(OEC). ............. .. ........ .... of Cu(2F-6F)....... ......... .. ...... .. of Cu monoketo ..... . ............... ... of Cu monoketo.. ...................... of Cu(2—3) diketo... ...... . ...... ..... vi 11 13 15 16 18 19 21 22 24 27 36 38 39 no 42 r"! “39355 ".c- -:L\.i.in.:nasu'2.q:.tt QIIOIOIIIIIII.IlI‘E’—’:Qn‘H'-Ii(ii:-n:3 .- . .f ' 5‘s? 1‘ 17b. 18. 19. 20. 21. 22. 23a. 23b. 23c. 23d. 23e. 23f. 24a. 24b. 25. RR RR RR RR RR spectra spectra spectra Spectra spectra of Cu(2-3). of Cu(Z—A) diketo.. ..... .. of Cu(2-5) diketo.... of Cu(2-6) diketo... of Cu(a—y) dimeso diketo.. Structure of isobacteriochlorin...... IR spectra of Cu (monoketo).. I—R spectra of Cu(2-3).... I-R spectra of Cu(2-4)..... out-una- I-R spectra of Cu(2—5).. ..... . ............. ...... I-R spectra of Cu(2-6)... ................... ..... IR spectra of Cu(2F-6F)..... RR spectra of Cu monoketo..... RR spectra of Cu monoketo..... Chemical structure of metallo— trans- octaethylchlorin (from ref. #61 43 an 125 46 47 50 52 53 54 55 57 58 61 61 62 -F . .---..-I-n‘o ...-1!",‘a r.-"-.'£-.‘- ---I 1:. ?' 5~LII 5L. RRS: IRS: Cumonoketo: Cu(2-3): Cu(2-u): CU(2-5): Cu(2—6): Cu(OEC): Cu(OEP): Cu(2F-6F): Fe(2F-4F): Cu(a-Y): ABBREVIATIONS Resonance Raman Spectroscopy (Spectra) Infrared SpectrOSCOpy Cu-gemini monoketoporphyrin or Cu—gemini porphoketone Cu-2, 3—gemini diketoporphyrin or Cu—2, 3—gemini porphodiketone Cu—2, u—gemini diketoporphyrin Cu—2, 5—gemini diketoporphyrin Cu—2, 6—gemini diketoporphyrin Cu-transoctaethyl chlorin Cu—octaethylporphyrin Cu—3, 13—diformyl,8,18—di—n—pentyl-2,7,12, 17,-tetramethylporphine 2,4-diformyl-iron protoporphyrin IX Cu-a,Y-dioxo-porphodimethene viii (221399213) 2.1051201: mtg-3' 35.118“? sensuous gear! - H.- " .J'J' '. .' ".. -2. :a..' CHAPTER 1 Introduction One of the major challenges of modern biochemistry has been the elucidation of biological function in terms of molecular structure. Numerous spectroscopic methods have been introduced to monitor structural features and detect changes which accompany biological function. Among them, vibrational spectroscopy offers high expectations, since vibrational frequencies available from infrared or Raman spectra are sensitive to bonding and geometric arrangements of localized groups of atoms in a molecule. Among molecules of biological importance, metallo— porphyrins and metallochlorins are two of the best candidates. Metalloporphyrins are essential to the life of plants, animals, fungi and bacteria. Metallochlorins appear in many forms such as chlorophylls, or as gemini ketoporphyrins, which have been recently synthesized in our bioorganic lab by Professor Chris Chang. These molecules are of biological importance because they are related to synthetic analogues of sirohydrochlorin which is the demetallated siroheme prosthetic group of nitrite and sulfite reductase. The synthesis of the gemini diketoporphyrins involves the oxidation of octaethylporphin using hydrogen peroxide. Each diketoporphyrin was then separated and purified by 1 chromatography. Experimental In this particular project the techniques employed were absorption, infrared, and resonance Raman spectroscopies. The optical spectra were taken by using a Cary 17 spectro- photometer in the visible and ultraviolet regions in order to obtain both the Q and Soret bands. The infrared spectra were recorded using a Perkin Elmer 283 B Infrared spectro— photometer, with 0.1 mm (cavity) cells. Resonance Raman spectra of Cu—gemini-ketoporphyrins in CHZClZ were recorded by using two Soret lines (406.? and 413.1 nm) from a Spectra Physics 164—11 Kr ion laser in conjunction with a Spex 1401 double monochromator and associated Ramalog electronics. A prism, external to the laser cavity, was utilized in order to separate the two frequencies. The rotating cell technique 'was also used to avoid the local heating of the samples. (See Figure 1a for schematic representation.) The compounds were obtained from our biorganic lab and they were 80-95% purity. The solutions used were 5-30 pmolar in CHZClZ' Visible excitation (488.0 nm, 514.5 nm) was also employed but fluorescence obscured the weakly-enhanced Raman scattering; therefore Q—band excitation was abandoned. i951“ n2 anoiaen islalc "-F. ' ‘”5 ‘nFHhfl' ET. -fi*t35' .oytqgmfl; 5 q we?!” has .r-mw‘lni .2101th - -o'13:29qa Y1 WEEDS 3:21:22. 52:! was: .2-:aw 5 3'9qu Lash-«p 911‘! -1‘ .; 1ainrfififidfl Rotating Cell Sample ( Laser ‘3 Collection Optics —[|——00m[}l: Pol. Filter Pol. Scrambler | I ! PMTl\ 4/1”?“ \\ l’\' ‘l "\l Double \, \i I], \: Monochromator \/ \| 'I \l l, 3,11 l Figure la. Schematic representation of laser Ramon experimental arrangement. CHAPTER 2 A. ABSORPTION SPECTRA OF CHLORINS AND RELATED COMPOUNDS 1. Some theoretical aspects of heme absorption In general, the electronic absorption spectra of porphyrins and related compounds are dominated by two strong transitions as in Figure 1b [1]. The transition in the 500—700 nm visible region is referred to as the Q-band, whereas the more intense transition in the near uv region at approximately 400 nm is the Soret, or B—band. Between the B and Q transitions there is a satellite band (on the Q-band) which arises from the vibronic mixing of bands B and Q [2,3]. Many times there is a plethora of bands between these two main bands, arising from the perturbation of the centrosymmetric cyclic model due to lowering of symmetry or the existence of peripheral substituents on the ring as in Figure la [4]. The 16—member cyclic polyene mentioned above is a convenient model for the treatment of porphyrin absorption Spectra [4]. Simpson [5] was the first to discuss the porphyrin spectra. His approach was based on a successful free— electron-model, characterized by the assumption that the w—electrons of porphin are free particles on a 16—lattice atom ring, providing a conjugated electron current. In this model, the 18 n—electrons are paired in orbitals of L, m 0 ob- ..moamt. > x» av owl $00 000 mmvo NLOO . 133V 3 ONT ..<.m:u_m.. O._2 OO _ _ — ~ _ — _ _ _ . wOO $00 000 000 N00 v, 331 726:3 _. E 02.8. «323 oa noaqonfionzaoam n .... OInQn A82”. 00 32: 3 02.8. 33.3 e.“ nlomQQ. n 5 01%.» A830 Cu 32: 6 increasing angular momentum with the system filled up to the l = :4 level. When a transition occurs from 1 = :4 to the next highest 1 = :5 level, then it can be allowed or forbidden depending on whether A1 = :1 or A1 = :9, thus giving rise to the B and Q bands respectively. A more detailed and explanatory model, first introduced by Moffitt [4], Platt [6] and later presented by Gouterman [7], considered only the two highest filled cyclic polyene orbitals as shown in Figure 2. This model which was also reaffirmed by a number of Russian spectroscopists [3,8] is the famous "four orbital model". The two lowest unfilled orbitals are the degenerate eg states, whereas the two highest filled orbitals also can be accidentally degenerate, which is a rather general argument pertaining in many cases. The spatial shape and nodal characteristics of these four orbitals are shown in Figure 3 for purposes of comparison. There are basically two doubly—degenerate M0 configura- )2 (a2u21 (eg)1 and (a1u>1 (a2u>2 (eg>1 for the two lowest excited states of the fi -electrons emanating tions [9], (alu from the porphyrin ground state configuration (alu)2 (a2u)2. It has been shown [10] that these two singly—excited con- figurations give equally mixed linear combinations, as a result of configuration interaction. Therefore, the x and y components of the Q and B states can be extracted from the relationships involving the CI parameter a and the appropriate 50/50 mixtures of the excited configurations 0:, Q3 B: and B; as follows: , and Cl (69) 7 a) Y bl (Ozu) — b2(0|u) PORPHYRIN X H C2 (eg) Y D) Y b| (02") "J CHLORIN Figure 2. Four orbital model (from reference # 7) Figure 3. Shape and nodal characteristics of the four orbitals. (from ref. # 7) 9 |Q > cosalQ: >— sinalBg > 0 . 0 >= >.. B > IQ cosule Sinal y [B >= cosulBg >+ sinalQ: > 23> O . O B >+ >, _1 cosal y Sinale (2 ) Parameter a was demonstrated to unmix the configurations, and it differs for each central metal and porphyrin complex. Gouterman [8] pointed out that consideration of the various factors which influence a can give very fruitful results with regards to the properties of porphyrin absorption or emission spectra. The Q0 and B0 components are: 0 ._ Q = /2 [a e + a e ] x -E 2u gX lu gy Q0 = /2 Ia e — a e ] y -E 2u gy lu gX O _ B=/2[ae —ae] x 2u gX lu g 0 B = V2 [a e + a e ] (2.2) y —E 2u gy lu gX 2. Results Metallochlorins [11] differ from metalloporphyrins [12-15] merely by the reduction of one CB — CB bond of a pyrrole ring in the conjugated macrocycle; however, this gives rise to distinctly different absorption spectra. 10 Isochlorins and bacteriochlorins, which have two saturated CB — CB bonds, also exhibit electronic spectra with great differences in comparison with metallochlorins and metal- loporphyrins. Two consequences in the absorption spectra follow when changing from a porphin to a chlorin, with the consequent lifting of the degeneracy of the lowest excited configura- tions (Figures 2 and 4): (1) the spectrum and mainly the Q band shifts to the red, and (2) the Qy band intensifies as in Figures 5, 7, 8, 10, and 11. However, the QX band of chlorins is not strengthened sufficiently to stand out from the vibrational overtones of Qy [16]. Both consequences mentioned above have been quantitatively shown by Platt [17,18] and later by Gouterman [8], who predicted that the "true" measure of intensity is the dipole strength q2 (in square Angstroms) given by: _ a q ‘250062 where A is wavelength of band—peak, AA is the halfwidth and e is molar extinction coefficient of the sample in that particular region. The chlorin Soret band is found at almost the identical energy as that of porphin and its BX and By components are almost degenerate in unsubstituted metal chlorins. Hence, the CI pattern of chlorin as well as its spectrum likely resemble those of tetrabenzaporphin [19—22]. ENERGY 11 OPP " THP ADJ " THP Figure 4. The 4- orbital model for OPP-THP and ADJ-THP 12 The x—allowed states, which are not constrained by symmetry, appear to be lowered by CI more than y-allowed states. Some calculations performed by McHugh et al. [23] indicate that the Soret band of free base chlorins includes three major allowed electronic transitions as in free-base porphin. Another very prominent feature of the chlorin substituted spectra is the "split" Soret. This phenomenon is due to the presence of an electron-withdrawing group such as a carbonyl directly attached to the pyrrole ring - which makes forbidden transitions allowed. These bands arise from transitions between B and N states. The N band itself, as seen in Figure 4, arises from transitions between the "a" orbital and the eg orbital. The "a" orbital which lies below the alu and a orbitals is a descendant of a cluster of closely 2u spaced orbitals such as 2a2u and.2b2U of porphin. The bands that arise from the coupling of B and N are indicative of the new-n-r* state which involves the electron—withdrawing group. They were also observed and identified by Caughey et al. [24] in the spectra of metal 2,4—diacetyldeutero- pcrphyrins. A brief assignment of the visible and uv spectra follows: a) The Cu(OEC) optical spectra, as shown in Figure 5, exhibit an intense band at approximately 615 nm, assignable according to Gouterman [7], to the Qy(0—0) transition, whereas the remaining weak bands around 500—600 nm are 13 Sword co 8.8% .8160 .n 83.... ES A cos com 08 cow 14 assignable to the Qx(O-O) and the Qy(0—1) and Qx(O—1) transitions. The Soret band appears at 400 nm with a weak shoulder to the blue, possibly arising from the lowering of symmetry by the saturation of one CB - CB bond in Cu(OEC). The Cu(OEC) absorption spectra recorded here are in agreement with the spectra of Fe(OEC)Cl and [Fe(OEC)Im2]Cl and in general of M(0EC), recorded by Kitagawa previously [25]. b) The absorption spectra of 2F — 6F are shown in Figure 6. Due to the presence of the electron-withdrawing formyl groups on positions 2 and 6, both Soret and Q bands are shifted to the red as in cytochrome oxidase [26] and in 2-4 diformyl iron protoporphyrin IX [27]. The visible bands appear at 615, 580, 560 and 510 nm. The lowest energy band at 615 nm probably corresponds to the Qy(0—0) transition. The other three bands appear to be extremely weak — especially the 580 and 560 nm bands — which indicates cancellation of the dipole moments for the corresponding transitions. The Soret is very sharp but slightly split at 420 nm and 415 nm, which possibly signifies the presence of the two formyl groups. c) The Cu—monoketo spectrum illustrated in Figure 7 clearly exhibits the ”chlorin type" absorption. The two most important features are the very intense Qy(O—0) band at 618 nm and the strong "split type" Soret band at 415 nm. The Qy(0—0) band is also accompanied by two - and not four - not very well resolved humps at higher energies; 570 and 512 nm. 15 .CwinU ..o 28on .6230 .m 2:92.. AECZ 00K Auom nxxm 00a _ _ _ _ u n mIo . «Iuézmxowrnflu .. a . ouo Io nIU U\I \Nxo/N \Nzo/Nxo ”Io 1NIo Io 16 28.2.9: :0 .o 283... .6230 N 2:9... .854 com 000 Omm 00v _ _ 17 The most striking feature, however, if the multiple splitting of the Soret. This effect has been observed previously by Weiss [16], who concluded that the multiple splitting of the Soret band is attributable to the existence of a substituent on the porphyrin ring which lowers the symmetry and makes "forbidden" bands allowed. These bands, which appear in the spectrum at energies above that of the Soret band (see Figure 6), are forbidden in square (D4h) symmetry. d) As illustrated in Figure 8, the Cu(2-3) diketo spectrum is characterized by a very intense Qy(0—0) transition at 688 nm and a dramatically "split Soret". Again the Qy(0—0) transition is accompanied by three higher energy peaks at 630 nm, 580 nm and 540 nm. The 580 nm band appears to be somewhat more intense than either of the other two. The experimental Soret band is rather complex. It shows three obvious components, at 442 nm, 430 nm and 395 nm, with the middle component being most intense. This pattern of splitting is again attributable to the existence of two carbonyls on two adjacent pyrroles. Another striking effect to be observed in the Cu(2-3) diketo spectrum is the dramatic shift of the Qy(0—O) band to 688 nm, and also its increase in intensity relative to the Soret intensity. e) The Cu(2—4) diketo optical spectrum is shown in Figure 9. The most obvious characteristic of the spectrum is the band which lies farthest to the red. This band, 18 00h 8m 6.9.6 3.85 “.o 9.8QO .0230 .m 8:9“. .854 00m 00v _ 19 ....oEoco 3.8: ... 6.8.5 Stu. 5 .o 88on .0230 .m 822“. .85.. 00h com com 00.» — . a‘aUT. ... ...... 20 which has been assigned [9] to the Qy(O-0) transition, is of lower intensity than the band to its blue. Due to this anomaly we will refrain from any discussion of this parti- cular optical spectrum. f) Figure 10 illustrates the spectrum of Cu(2—5) diketo, where the two carbonyls are on Opposite pyrroles. Again, the two most prominent features of the spectrum are the very intense Qy(0—0) band and the quadruply split Soret. The Q transitions occur at 670 nm, 615 nm, 585 nm and 540 nm, whereas the Soret bands occur at 442 nm, and 420 nm, with two satellite bands at 398 nm and 380 nm. In the visible region the band at 585 nm appears to be more intense than either the 615 nm or 540 nm bands. The same effect was previously seen by Seely [28] in the hexahydrotetraphenyl porphin spectra, and by Coyne et al., [29] in bacteriochlorphyll a. Magnetic circular dichroism studies that followed [30] assigned this band to the Qy(O—1) overtone. The Soret band is the most dramatically split of all the observed spectra, which indicates tremendous lifting of the degeneracy where "forbidden" bands become "allowed". g) Similarly, the Cu(2—6) diketo spectrum shown in Figure 11 reveals the same effect, particularly the increased intensity of Qy(0—0) in comparison to the Soret. The visible bands occur at 702 nm (the most highly intense and red shifted Qy band of all the recorded spectra) and as small humps at 666 nm, 648 nm and 520 nm. 21 98.6 6-8 :0 ..o 2.6QO .0230 .o_ 230.... 2.5.. CON. OOm 00m 00.» . _ 22 .29.... 8-8.8 .5 2.89.. _8_ao .: 2:2... .854 00h 000 000 00¢ Omm _ _ _ _ _ 23 The Soret is doubly split into a very intense component at 427 nm - with a shoulder to its blue at 400 nm - and a weaker component at 380 nm. Again, this spectrum needs further investigation by fluorescence [31] and MCD [38]. studies. h) Finally, the Cu (a-Y) dimeso spectrum shown in Figure 12 is atypical of chlorin and typical of porphin spectra, with a rather broad Soret band at 440 nm and a visible transition at 600 nm, with a satellite band at 560 nm. The Soret band is more intense than the low energy Q band and very broad, possibly due to the burial of one weak B component within the other. The most interesting feature, however, is the very slight hump at 400 nm. This hump could signify the existence of an electron—withdrawing group attached to the ring. (The wavelengths and energies of the various absorption bands are listed in Table 1 for purposes of comparison.) Now that the various transitions have been assigned, one may consider the relative positions of the bands in relation to the conjugation pathway of each molecule. Note, first, that when two adjacent rings are reduced, such as in Cu(2—3) and Cu(2—4) diketo, the two x—polarized transitions are nearly equal in energy. The very distinct feature about the optical spectra of both of these compounds is the dramatic shifting of the Qy band towards the red. This effect is clearly illustrated by the increase of the "box" — or of the conjugation pathway — to 22 n instead of .E 511131! 24 63,2616. :0 ..o 26QO 66.30 .m. 239“. BE... CON Com 00m . 00? Con _ . _ _ 25 so mmm.mmfiwm HI HI H..- so www.mfimom an own So www.mmfimw a: man 80 :Nm.mowmm EC om: So :m:.:mwmm so one. mammm as man so saw. mmmmm H- an own H- as we: as on: also wmfl.:m:mm Hiso omm..ommfi also mmm..mo:~ H.so :m:.:mmmw an em: as mom sq mo: as we: H.8 mom.ommmfi H.so m.m.w.mm. Hiso wmfi.:s:mm H.so afim.mfimmfi a: omm a: cam as ow: as can Huse mmo.mmsmfi H.so 5.0.:mom. so ewe. mammfi H.so mum..smm. H-50 owm.oaoem ea mam s: mwm as men as 0mm 5: ma: H-30 mfio.mfiom. H-50 mefi.oeme. H-50 som.eeee. H-EO efio.mswm. H-eQ msfi.mmwsfi as see as was as coo as one as can HiEo :Ho.m:N:H HiEo mmm.mmm:a :80 :wm. :mmda HiEo :ww.:mm:H HiEo mor.woooa s: mom as one as mmo sq mwo as com AQINVSO Amimvs A:2NV5 Animvso oCoE so .mQOHPHmsmpP camopeomao sfisoaso exp mo mamewHw>m3 use moHMposm i H mqm3 (o. ) I (2 3) kn ———z.—— p ,0 pa kn o - 3c where 10’ v0 are respectively the intensity and frequency of incident, exciting radiation, and v0 i v is the kn 29 frequency of the scattered radiation, which we will abbre— viate vL; where the (+) sign refers to anti-Stokes scattering and the (—) sign to Stokes scattering. Finally, (God) is the po Cartesian coordinate component of the polarizability tensor, and which relates the initial to the final eigen- states [40] according to: P a a a E x xx xy xz x P = a on a E y yx yy yz y Pz azx azy O‘zz Ez (2'4). where Pi and Ei are the polarization i.e., induced dipole, and electric field respectively. This scattering tensor takes the following form according to Kramers and Heisenberg—Dirac, being derived by a second order perturbation (dispersion) formula: > ‘ 5’ + ———————° 0 DO kn hc m Vkm — v0 + 1Tm vnm + V0 + lrm where “o’ua are the electric dipole moment operators; i.e., up = —E e(Sl)p and (Sl)p is the 9th component of the lth electron's position vector. Fm is the natural half-width of the state im>. The summation runs over all intermediate states \m>, neglecting states ik> and ln>. In Equation (2.4) above, both terms in the brackets must be considered in the case where v0 is far from resonance, i.e., when Vk — V0 >> 0, whereas the energy In 3o denominator of one vibronic manifold decreases, thus dominating the summation over states m, as resonance is approached. One of the major challenges that many scientists [41,42,43] successfully met was the elucidation of the vibronic nature of the states and operators in the scattering tensor. For this purpose the adiabatic Born-Oppenheimer (ABO) approximation will be used [44], in which the vibronic states are formed as products of pure vibrational states i(R€)>, with pure electronic states lk(Rg.q)>- R and q E are the vibrational normal coordinates and the electronic coordinates respectively. Writing: lk> = lkli(R€)>. (2.5) then i( )= T—l‘l (U) (2.6) R: a s a where Ug is a set of normal coordinates and wgare the harmonic oscillator eigenstates. By substitution of the ABO approximation into Equation (2.4). (a ) for the resonance case becomes: ackgmi . (0. ) . : — Z *2 _ . 2. pa kg+k1 c m w hvkg,ki hvo+1rmw ( 7) where ’5 'h‘. l- a ‘-'-"" .-._.-.;.--.- --" - '3 .:-;'.--1~.1;i' ‘3‘” 31 9mm = ’k(U€,q)upm(Ug,q)da (2.8) and similarly: fin) = Ikw,enigma},.qldq. (2.9) Equations (2.8) and (2.9) both represent the electronic transition dipoles between the two electronic states \k > and |m> respectively. Now expanding 9(Ug) or + pkmomk + Dim0$k okmomk (2.10) where pkm = (igégil)U=O , 0°: p(U ) E WV 777’ ' -l I “ '1. r . I Tl-H'. E If h I " ~" im- '..':..'.) ancusupfi if [V ------ --. ”'4‘: ‘. '_J. ___"__ ’ 5*“ 113-1. sin-”Hr: em. 3 . - ~ - -;-'.:.I:er.a-n ’:'-'_ 32 The first term in the square brackets of Equation (2.10) above corresponds to the so called Albrecht A term [41] and the second and third terms (neglecting the fourth term as too small relative to them) correspond to the Albrecht B term. As resonance is reached by simply allowing the energy hv to approach hv the denominator of expression (2.7) 0 becomes extremely small, which makes apo very large. Hence, sw,kg’ when hv — hvo becomes of the order of a vibrational sw,kg energy level difference, hm, then neglecting the dependence of the denominators on the excited state vibrational quantum numbers is no longer valid or appropriate. Basically two types of expansions can be derived from Equation (2.l0) (the expansion of the polarizability tensor), the Franck-Condon 0th order [45] and the Herzberg- Teller lst order [46] perturbation terms. The former term involves no vibronic coupling and is the A—term mentioned previously, whereas the latter term involves vibronic coupling of states oém, oék, is called the B—term, and takes the form: ' :2 z o 8H _ -1 pkm €p+mpkp (Em Ep> (2.11) where H is the Hamiltonian for the total electronic energy of the molecule, |p> corresponds to other than lk> and lm> states, and §%3 is the vibronic coupling operator [47]. E . .,- . .... .. .- . -- -. . . - . -.:.-.- "r? ..vfl L 33 The mixing of states by this operator depends upon the energy separation between these states and upon the relative coupling strength of< p|%%—| m>. The form that a takes now is: hv _ J; _ - . 0 “pa kgfiki ‘ cz sw,kg h"o + lrmw pkmomk x (2.12) Now substituting the expression for pkm into the above expression we obtain: ( ) l 2 Z pkmomk o, .= "' . p0 kg,k1 c m w hvkg,mw - hvo + 1me 3H -l 2 Z 2 3U5‘ m p + . m p+m w hvm,p + Omkopk where the first term is the A—term and the second and third the B—term again. Far from resonance, the F—C vibrational integrals could take the form: —h\)o + irmw 0L = 6g,ic (2'13) OIH Z w hvkg,mq The above term signifies the application of closure to the F—C vibrational integrals far from resonance, where the A—term contributes only to Rayleigh scattering; closure 34 vanishes as the resonance denominator is affected by the vibronic energy separations. The F-C overlap integrals for Raman scattering become non—zero with the removal of orthogonality between a ground state vibrational level and an excited state level with a different vibrational quantum number [38]. Hence, the orthonormality of the vibrational levels is preserved and the F—C integrals give rise to Rayleigh scattering even at resonance, if and only if, along some particular coordinate, the potential curve for nuclear vibration in the excited electronic state can be obtained from that for the ground state; this is accomplished by a translation along the energy axis by a purely electronic excitation energy. Therefore, the removal of orthogonality and the occurrence of Raman-scattering may be due to: a) the shift in equilib- rium position upon excitation and b) a difference in vibrational frequency between the ground and excited states. In reference to preceding Equation (2.12) large scattering intensities are affected by a strong intrinsic transition moment, Omk’ where the modes enhanced in resonance are those most responsible for a "forbidden" intensity contribution in an allowed electronic band. This is true in the case where the H-T term dominates the scattering mechanism, which implies that totally symmetric modes are enhanced when two allowed electronic bands of the same symmetry vibronically couple. 35 2. Resonance Rgmgn results The resonance Raman spectra of Cu(OEC), Cu(2F-6F), Cumonoketo, Cu(2-3), Cu(2-4), Cu(2-5), Cu(2-6), and Cu(a-Y) in CHZCl2 upon excitation at 406.7 nm are shown in Figures 14 through 21. The Raman lines between 1500 and 1750 cm—1 appear to be most enhanced, with the exception of Cu(a-Y) where the lower frequency lines between 600 and 800 cm"1 1500 cm_1 are most enhanced. and 1300— In previous studies the high frequency bands above 1400 cm‘1 were found to contain significant CaCm, CB'CB and CaCB contributions [48,49]. The band at 1580—1590 cm-1 was found to be mostly due to the CB—CB stretching [50,51]. Also, the band between 1620-1650 cm—1 was found to contain a great deal of CaCmCa contribution. In the present study we will focus our attention primarily on the carbonyl stretching frequency itself. We will also investigate the effect of the position of the carbonyl around the pyrrole rings, and its implications on the resonance Raman spectra. Assignment of the various other bands at this point cannot be made without normal coordinate analysis (NCA) [52,53]; thus it will be omitted. The Cu(OEC) Raman spectrum exhibit no bands above 1650 cm_l, where the carbonyl stretch is expected. This is due to the absence of a C=0 group on the ring, as illustrated in Figure 14. .36 l I00 99”— 2 |200 OVZI — f0 0) g I 1 I300 X exc. = 4l3.l nm 212:- oeeli— g 22:72 — — - “2 27792 - EBSI — 8 869i — - Q 2279: — I700 Figure l4. RR spectra of Cu (OEC). 37 The Cu(2F-6F) spectra, however, exhibit a low intensity band at 1671 cm—1, as shown in Figure 15. This band is attributed to the H—C=0 stretching frequency itself. Lutz [54] observed a similar line at 1664 cm-l in the type b chlorophyll spectra which he also assigned to the C=O stretching of the formyl group. A similar line was observed by Gradyushko, et al. [55], and by Fischer, et al. [56] for the type b chlorophylls. Kitagawa [27] recently reported a similar line around 1650-1670 cm-1 in the RRS of monoformyl and diformyl iron protoporphyrin IX. This Raman line, as noted earlier, is assignable to the formyl C=0 group stretching mode and corresponds to the characteristic line of reduced cytochrome oxidase at 1670 cm—1 observed by Salmeen, et al. [26]. The stretching frequency of the C=O bond depends on the delocalization of the formyl n electrons to the porphyrin ring which affects the bond order of the C=O bond. Thus the higher the delocalization of the formyl n electrons to the porphyrin the less the bond order and hence the lower the stretching frequency. This effect is observed on going from Cu(2F-6F) to (2F-4F) iron proto— porphyrin IX, to chlorophyll b. The Cumonoketo spectra, as illustrated in Figure 16a, b exhibit a broad band at 1710 cm_l. A similar band was also observed by Lutz [54] in chlorophyll b spectra between 1695—1705 cm“1 which he assigned to the ketone C=O stretching motion. We assign the band at 1710 cm—1 to the same motion. -'i an? 58’ Eil- - Le: E _ 7.3-- ._| 1‘12. .4. (if; in“ 3.1-! 551318 I111m*_» I. = '-.;.’+ ‘I-ms 5:361 .75 32:11 all-fl" . ‘ I“ 11 38 212!— 992| — 1782! — 9le- SHI— |400 I300 I200 l |00 l600 I500 I700 = 4l3.l nm. X exc. Figure l5. RR spectra of Cu(2F-6F). 39 Cuz" Monoketochlorin OSQI — b691— 0‘l79l - l 2300 2200 (”00 I500 I400 I700 |600 |800 Az7(cm") Figure l60. RR spectra of Cu monoketo. Xexc. = 406.7 nm. 4O OQQI— 069l- O€9l - IOOO l|00 |200 300 xexc. = 423.2. |600 I500 I400 I700 Figure l6b. RR spectra of Cu monoketo. 41 The Cu(2-3) spectra are shown in Figure 17a, b. A prominent feature is the band at 1705 cm-1 which is again assigned to the carbonyl stretching motion. This band is broad and appears symmetric which indicates that the two carbonyls on positions 2 and 3 are symmetrically located; i.e. there is a mirror plane in the molecule as shown in Figure 22. On the contrary, the Cu(2-4) spectra as shown in Figure 18 exhibit a highly unsymmetrical band at 1718 cm-1, which corresponds to the C=O stretching frequency. The fact that this band is unsymmetrical suggests that the two carbonyls are non—equivalent and that there is no mirror plane through the molecule as seen in Figure 22. Similarly, the Cu(2—6) spectra illustrated in Figure 20 show a very broad unsymmetrical band at 1700—1708 cm-1 which again indicates the non-equivalenoe of the two carbonyls and the lack of a mirror plane in the molecule. However, the Cu(2—5) spectra, as shown in Figure 19 exhibit a low intensity rather symmetrical broad band at 1707 cm'1 assignable to the carbonyl stretching motion. This indicates again that the two carbonyls are equivalent, due to the presence of a mirror plane as shown in Figure 22. Finally, the Cu(a—Y) spectra show absolutely no l (see Figure 21) for reasons scattering above 1650 cm- that are at the present unknown to us. We also checked the region where —0H stretching occurs (3000-4000 cm‘l), in case enolization of C=O via hydrogen bonding occurred 1.2 I500 I400 I300 l200 989] —' l |600 9179| ‘- l I700 SOLI '- l800 Figure l7a. RR spectra of Cu (2'3) diketo. Xexc. = 406.7 nm. 43 .E: 36 u 9.2 89 08. ooh com. i _ _ _ _ 3-3 :0 S 88QO mm at 8:9... OON_ 009 003 _ . \\ 1+4 Cu" 2.4 diketo O isochlorin O Q 009l- 4 839l- L I500 I600 I8CX) I700 A27(cm") Figure l8. RR spectra of Cu (2'4) diketo. 45 OOm_ ATEUVRQ Oom— Qoxs 3-8 5 .o 2.8% mm .m. 23: 00: 1.60! — 8Q“ —- crecuofl 8N. |8|I — BIZI — O 29% m .m :5 892I - C D 00.3 ZLEI — 1.6EI — 22M — 817M— t'9‘bl — ISM — VISI— 6179'- 009 _ EBQI - 669i - 00b 00m. d — — H79I — 901.! — _. (UZQBI— 46 QLQI — I 200 I700 I600 I500 I400 I300 I800 Figure 20. RR spectra of Cu (2'6) diketo. Xexc. = 406.7 nm. .oEvzu 0356 1-3 :0 .o 2....QO mm ..N 959“. ATEUVRQ .83 — q — q - d - a - 1706'- 1.96 — soon— 680l- 47 Oti— VOL- _ 860: — .33. 992!- 06H- 022L— bOEL- 99H— VH3- £92?- 862%- OLbP‘ ZZbP— o 9 Ed 29% x638 ESP- #8 during preparation of the sample, and again no signal was detected. Upon excitation at 413.1 nm the corresponding Raman lines for Cumonoketo and Cu(2-3) appeared at 1710 and 1710 1 cm- respectively. The spectra for the remaining chlorins with Xexc = 413.1 nm were not recorded in the present study. Let us now examine how the position of the exciting line effects the relationship of the absorption spectra to the observed Raman vibrational frequencies (see Table 2). The Cumonoketo spectra exhibit a uniform enhancement pronounced in both the high and low frequency modes. The Cu(2—3) spectrum exhibits enhancement of the modes around 1 1300 cm_ and mainly towards the higher frequency region. The Cu(Z—u) and Cu(2—6) spectra Show enhancement of the 1 (middle region) in resonance with modes around 1100 cm- the Soret (A max. = 426 nm). The most prominent feature of the latter spectrum is the high intensity of the bands around 1500-1600 cm-1 and the uniformity of the remaining high and low frequency modes. Finally, in the Cu(2-5) = 406.7 nm with A = 442 nm the A spectrum under e max xc high frequency modes around 1960 cm~l should be enhanced in resonance with the Soret. The spectrum though between 1100-1700 cm_1 appears to be uniform. r: 1' vi bns nth-gs?- _. rig-W . ... iris-7gb. an} «aw ' I- . . . eniucifio 3ntrléme1;§d:'wa& suioaqs 9d? .iio?fifi3qili - 11 L:-';-:. . ;*¢u ;. L“ :J“A d? -'-- l 35-1 .Uoppooon Coon Po: o>d£ whpommm* .ANME «v Moog Eseflxms map onMchflm vcw pohom pHHQm m mopwoflvcfl oEflpm oza++ .onEoonsso mocmcommsmsm %o o>HPmOflwQH mfi msaw> o>HpmmoQ ose+ III: III: III: Hqu mmman bq « * * also HHHH b< HIEo mmmln+b< H..Eo mmm u b4 49 a: fin; u 8.3 HIEo mum n b< H180 doman b< also :HHHH ba 7:5 $3" 3 also i3“ 3 Tea an??? Tao m3“ 3 TE NEIJE 3.35 3.85 3-35 8-35 0:055 2585 E: m.©o: n oXo< .QOflPMPHoxo Pwhom sous: mozam> D< I N mgm A, modes Big C2 OXIS A29 . . anti - symmetrIc B } to C2 axis ’ 8? modes 20 TABLE 5. from Ref. # 6|. Symmetry correlation between C2,, and 04h groups for the in-piane vibrations. .I .1. 921-: 79* ... (s30) 03 .. . - -A-uouao ._ ”mu—- 66 correspond to those of Ni(0EP) at 1602 cm-1 and 1576 cm'1 which were assigned to the CB—CB stretching mode. Lutz, et al. [54] studied the resonance Raman spectra of chlorophyll and its isotopically substituted derivatives. The Raman lines of chlorophyll a at 1115, 1147, 1290, 1350 l and 1380 cm- recorded by Lutz are consistent with the lines of Cu(OEC) which fall at 1130, 1156, 1293, 1362 and 1371 cm'l. The small differences in frequency may be attributable to differences in the peripheral substituents, the central atom, and/or the number of reduced rings. Thus it becomes obvious that under preresonance enhancement for Xe = 413.1 nm both the low and high XC frequency modes are enhanced and the spectrum appears more or less uniform. . art-:1: 27515-1.- sf L .0335. 5 F iltquIOIHG 15 ..fljs nl-B‘E IL. 1:! 'i' .----s': "so 08C]:- '- ‘3‘ .- ‘.-.-, anti {3"- _ .1 (TE: -_ 1. '- .1 CHAPTER 3 Conclusions The results obtained from the application of a variety of spectroscopic methods to chlorins, isochlorins and iso- bacteriochlorins have been described in the preceding chapter. These methods provide insight into the structure of the molecules and offer help to others who study similar compounds. This chapter serves as a synopsis of the involvement of the carbonyl moiety in the compounds investigated, in an effort to provide a consistent picture of each molecule. Three major aspects are summarized: 1. Absorption spectra 2. Resonance Raman spectra 3. Infrared spectra 1. Absorption spectra The most prominent feature of the optical spectra of all the carbonylated derivatives of chlorin is the shifting to the red of both the Soret and Q bands (as in cytochrome oxidase [26]) due to the presence of the electron—withdrawing ketone group. This effect is most notably observed in the Cu(2-6), Cu(2—3) and Cu(2—4) diketo spectra, where the visible band can be clearly distinguished at 702, 688, and 67 68 688 nm respectively; in Cu(OEC) the Q band maximum lies at 615 nm. This tremendous shift towards lower energy indicates lifting of the degeneracy of the eg states. The effect of carbonylation on the Soret band is rather hard to distinguish because the latter is multiply split. The Cumonoketo spectra project a different and less dramatic picture than the compounds mentioned above. The Qy(O—O) band appears minimally red shifted, at 618 nm. This is an indication that the carbonyl mono-substitution causes little apparent perturbation on the ring with respect to the energy of the Qy(O—O) transition. The Soret also appears to be split for both Cumonoketo and Cu(OEC), but with a higher degree of splitting for the former, which indicates the presence of an electron withdrawing group. This less dramatic red shifting can be attributed to the smaller conjugation pathway (involving 20 n bonds) for Cumonoketo in comparison to the disubstituted ketoporphyrins. In contrast, carbonyl disubstitution causes more extensive effects (see Figure 22) which are translated as a skewing in the structure of the molecule. This effect is more obvious for Cu(2-4) and Cu(2-6) diketos as was revealed by RRS and IRS. 2. Resonance Raman spectra The presence or absence of the ketone vibrational band(s) was interpreted in terms of the involvement or non-involvement of the C=0 in the conjugation pathway of the ring. ‘16 ”$613? Viefitgai-et . :15; ughi‘i'eib o: 0112:? ”serum-fl fire-:- 31108 iii? no ..- ‘ IF» - p I l - h . . . - - .- ‘ , ' . _' 1- I" . ' 5 _ I' .3 ' u ..Jl-".‘- 69 The most prominent feature of the diketo Raman spectra is the broadness of the C=0 band, which indicates the presence of both carbonyls. The Cu(2—4) and Cu(2—6) diketo spectra gave the broadest C=0 signal. Recently, Tsubaki, et al. [27] showed that when both positions 2 and 4 are occupied by formyl groups, donation of electrons from the individual C=O groups becomes less than in the case of monosubstitution, probably due to mutual repulsion of the electrons. The C=0 groups may possibly have a similar stretching frequency, giving rise to one broad band. Indeed, a single C=O stretching frequency was observed, at higher wavenumber (1668 cm_l) than either of the absorptions for mono—substitution at position 2 or 4 (1648 and 1660 cm"1 respectively). This effect is clearly shown in the spectrum of Cu(2F-6F) (1661 cm-l), and also in the spectrum of Cu(2-4) diketo, where the carbonyl signal is at 1720 cm-1 which is higher than the signal for Cumonoketo which is at 1710 cm'l. The carbonyl signal for Cu(2-4) and Cu(2—6) diketos also appeared to be split in the Raman spectra, which indicates non—equivalence of the two carbonyls. This non— equivalence of the two carbonyls for each molecule can be translated into the ring as a higher degree of perturbation of these structures in the form of skewing or bending, and the lack of a mirror plane through these molecules. (Note, however, that the splitting for Cu(2-6) is much less dramatic due to the C2 axis which makes the two carbonyls _-_-3 ‘18s a? -__.'-;= WW? mtg 0:0 mm we I“ . "" -:.-:,-5 a bra: '.i FsflCiIZ".-:'E'0L- £1330 rem. 59:13 mm {‘5}; .iflsdua'r actresses --.':. .-.---: .- .- _i -.-_-..-;I Lyric}? are hot I' if — if- .. . .:_-. :suirlwbmt- '. FFT’SYJ-m 7O "interchangeable".) The two marker bands for CB—CB and CaCmCa stretching are worthy of special mention. The CB-CB stretching frequency is normally found between 1580-1590 cm"1 [50,51]. In our Raman spectra, this band fluctuates greatly (16 cm_l) which indicates a varying pattern of perturbation from one molecule to the other. The methine bridge, CaCmCa stretching frequency appears to be approximately the same for each molecule (1640—1646 cm_l). This could indicate a slight or negligible perturbation of the methine bridge by the carbonyl mono- or disubstitution at the 08—08 bond. The frequency difference is largest between Cumonoketo and Cu(2-3) diketo in comparison to the remaining diketos. When all the above postulations are considered one may see that the structure-sensitive region (above 1500 cm‘l) primarily is perturbed under carbonyl mono— or disubstitution. ThlS region includes CaCme 08—08 and CaCB stretching vibrations. 3. Infrared spectra Infrared spectroscopy was employed in order to specifically detect the carbonyl band(s) and to supplement resonance Raman spectroscopy. In this regard IRS proved to be very fruitful. The bands for Cumonoketo, Cu(2-3), Cu(2—5) and Cu(2F-6F) were broad but not split, whereas the bands for Cu(Z-Q) and Cu(2—6) were doubly split. The splitting signified the non-equivalence of the carbonyls #4- 35. 9; as..."50£bfl1 -!'§1.f"”. ff ...; -: aid-35:10:! "-'-' 71 and also the structural differences between the molecules. Suggestions for Further Research Although all the vibrational spectra recorded during this project are of good quality, additional work is needed for further justification of our results. More Soret excitation lines should be utilized and a resonance Raman excitation profile should be obtained in order to accurately determine which Raman lines are resonance enhanced. Isotopic (15N, 17O) substitutions should be made in order to clearly distinguish the various vibrational stretching frequencies of the various bonds such as C=O, Ca-N, CB—C etc. 8’ Different metals should also be used such as Sn, which will tell us which bands are metal sensitive. (A detailed and complete normal coordinate analysis should be also performed for specific assignment of each spectroscopic line to a normal mode.) Finally, the Cu(2-4) diketo and Cu(a—y) dimeso must be reinvestigated due to the anomalies observed in their spectra, which we attributed to impure samples. “an mentally I69 I Em: p noun ’2!» m flint“ v.3. his} .1931.” - 1.": ofmfioxl - 1.1!: my :ci :5: . .u-e LI ST OF REFERENCES 10. 11. 12. 13. 14. 15. LIST OF REFERENCES Gouterman, M., in "The Porphyrins", Vol. III, (Dolphin, D. ed), pp. 1-165 (1978), Academic Press. Lutz, M., Spectroscopy Letters, 2, 133-145 (1974). Gurinovitch, G.P., Sevchenko, A.N., Solovyov, K.N., "Spectroscopy of Chlorophyll and Related Compounds", Izdatel‘stvo Nauka; Tekhnika: Minsk, 1968 (English translation; AEC—TR-7199). Moffitt, W., J. Chem. Phys., 22, 320, (1954). Simpson, W.T., J. Chem. Phys., 11, 1218 (1949). Platt, J.R., "Radiation Biology", A. Hollander, ed. Vol. III, Chapter 2, McGraw Hill, New York, 1956. Gouterman, M., J. Chem. Phys., 39, 1139 (1959). Ksenofontova, N.M., Solovyov, K.N., Tsenter, M.Y., Bobovich, Y.S., Shkirman, S.F., and Kachura, T.F., Zh. Prikl. Spektr., 11, 918 (1972). Gouterman, M., Journal of Molecular Spectroscopy, 6, 138 (1961). Shelnutt, J.A., "A simple interpretation of Raman excitation spectra of metalloporphyrins", to be published. Ozaki, Y., Kitagawa, T., and Ogoshi, H., Inorganic Chemu l_8. 1772 (1979)- Spiro, T.G., Accts. Chem. Res., 339 (1974). Spiro, T.G., and Strekas, T.G., Proc. Nat. Acad. Sci. USA, 69, 2622 (1972). Verma, A.L., and Bernstein, H.J., J. Chem. Phys., 61, 2560 (1974). Solovyov, K.N., Ksenofontova, N.M., Shkirman, S.F., and Kachura, T.F., Spec. Lett, 6, 455 (1973). 72 ";.---. . . . .tmnstuol- 6' .. .' ui; - 1 . . f;. i .qq .!39 .3 fl” . .rful .S . . F _. I 16. 17. 18. 19. 20. 21' 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33- 34. 73 Weiss, Jr., Charles, J. Mol. Spec., 44, 37 (1972). Platt, J.R., J. Chem. Phys., 22, 263 (1951). Platt, J.R., Klevens, Modern Physics, 26, 182 (1944). Weiss, C., Kobayalshi, H., and Gouterman, M. J. Mol. Spec., 11, 108 (1963). McHugh, A., Gouterman, M., and Weiss, C., Theor. Chim. Acta, 24, 346 (1972). Goedheer, J.C., and Siero, P.J., Photochem. Photobiol., 6. 509 (1967)- Knop p, J. V. , and Stichtenoth, H., Z. Naturforsch., 22A, 639 (1972) Sundbom, M., Acta. Chem. Scand., 22, 1317 (1968). Caughey, W.S., Deal, R.M., Weiss, C., and Gouterman, M., J. Mol. Spec., 26, 451 (1965). Weiss, C., and Gouterman, M., J. Chem. Phys., 42, 1838 (1965). Babcock, G.T., and Salmeen, I., Biochemistry, 22, 2493 (1979). _ Tsubaki, M., Nagai, K., and Kitagawa, T., Biochemistry, 12» 379 (1980)- Seely, G.T., and Calvin, M., J. Chem. Phys., 23, 1068 (1955). Coyne from Reference #16. Houssier, C., and Sauer, K., Biochim. Biophys. Acta, 112. 492 (1969)- Houssier, C., and Sauer, K., J. Amer. Chem. Soc., 92, 779 (1970)- Briat, B., Schooley, D.A., Recordo, R., Bunnenberg, B., and Djerassi, C., J. Amer. Chem. Soc., §9, 6170 (1967 . Longuet—Higgins, H.C., Rector, C.W., and Platt, J., J. Chem. Phys., 1g, 1174 (1950). r0, T. G. and Strekas, T. C., J. Amer. Chem. Soc., fi_, 338 (1974) ““ 35- 36. 37. 380 39- 40. 1+1. 42. 43. [+4. 45. 46. 47. 48. 2+9. 50. 51. 52. 74 Spaulding, L.P., Chang, C.C., Nai—Teng, Y., and Felton, R.H., J. Amer. Chem. Soc., 91, 2517 (1975). Shelnutt, J. A., Rousseau, D. L., Friedman, J. M., Simon, S. B., Proc. Natl. Acad. Sci., Z6, 4409 (1979). 0ndrias, Mark R., Ph.D. Thesis, Michigan State University, 1980. Robin, J.H., Clark, and Stewart, B., Structure and Bondin , 36 (1979). Tang, J., and Albrecht, A. C., in "Raman Spectroscopy", Vol. 2 (Szymanski, H. A., ed), pp. 33— —69, (1970), Plenum Press. Guillory, W.A., in "Introduction to Molecular Structure and Spectroscopy", (Allyn and Bacon, Boston, 1977 Albrecht, A.C., J. Chem. Ph sics, 33, 156 (1960). Nafie, L.A., Pezolet, M., and Peticolas, W.L., Chemical Physics Letters, 26, 4043 (1976). Shelnutt, J.A., O'Shea, D.C., Yu, N.T., Cheng, L.D. and Felton, R.H., J. Chem. Ph 8., 64, 1156 (1976). Born, M., and Huang, K., in "Dynamical Theory of Crystal Lattices", (Oxford, 1956). Condon, E.W., Phys. Rev., 92, 858 (1928). Powell, J.L., and Crasemann, B., in "Quantum Mechanics", (Addison-Wesley, Massachusetts, 1961). Levine, I.N., in "Quantum Chemistry", (Allyn and Bacon, Boston, 1970). Kitagawa, T., Ozaki, Y., Kyogoku, Y., Adv. Biophys., ll. 153 (1978)- Abe, M., Kitagawa, T., Kyogoku, Y., J. Chem. Phys., £2, 4526 (1978)- Kitagawa, T., Abe, M., Ogoshi, H., J. Chem. Phys., 62. 4516 (1978) Felton, R.H., Yu, N.T., O'Shea, D.C., and Shelnutt, J.A., J. Am. Chem. Soc., 96, 3675 (1974). Bright, E., Wilson, Jr., J.C., Decius, J.C., and Cross, P.C., in ”Molecular Vibrations", (Dover Publications, New York, 1955). 53- 54. 55- 56. 57- 58. 59. 60. 61. 62- 75 Mackintosh, D.F., Michaelian, K.H., Can. J. S ec., 3%, 1 (1979). Lutz, M., J. Ram. Spec., 2, 497 (1974). Gradyushko, A. T., Mashenkov, V. A., and Solovyev, K. N., Bio h sics, 14, 869 (1969 ). Fischer, M. S., Templeton, D. H., Zalkin, A.,a Calvin, M., J. Am. Chem. Soc., 94, 3613 (1972)d Mason, S.F., J. Chem. Soc., 976 (1958). Ogoshi, H., Saito, Y., and Nakamoto, K., J. Chem. Phys., 52, 4194 (1972). Thomas, D.W., and Martell, A.E., J. Am. Chem. Soc., Zé, 1338 (1956)- Thomas, D. W. , and Martell, A. E., J. Am. Chem. Soc., 81. 5111 (1959) Ogoshi, H., Masai, N., Yoshida, Z., Takemot and Nakamoto, K., Bull Chem. Soc., Japan 44, 49 (1971). Burger, H., Burczyk, K., and Fuhrhop, J.H., Tetrahedron, 27, 3257 (1971). I i . "‘2" _ . '- '_ _. _; . ...-'5 IHIVI‘Q " '.I. I II. 9 :."-. 3.1.45.1!" . 33-153 (“fie-5“ I. 0 .u - If .. IS" I. 4.; . ' .53"! IV: