UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Asymmetric induction in solid state photochemistry Gudmundsdottir, Anna D. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1993_spring_phd_guomundsdottir_anna.pdf [ 8.11MB ]
Metadata
JSON: 831-1.0059590.json
JSON-LD: 831-1.0059590-ld.json
RDF/XML (Pretty): 831-1.0059590-rdf.xml
RDF/JSON: 831-1.0059590-rdf.json
Turtle: 831-1.0059590-turtle.txt
N-Triples: 831-1.0059590-rdf-ntriples.txt
Original Record: 831-1.0059590-source.json
Full Text
831-1.0059590-fulltext.txt
Citation
831-1.0059590.ris

Full Text

ASYMMETRIC INDUCTION IN SOLID STATEPHOTOCHEMISTRYByAnna Dora GuomundsdottirB.Sc. Chemistry University of Iceland, 1985M.Sc. Chemistry University of British Columbia, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESCHEMISTRYWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAApril 1993© Anna Dora Guomundsdottir, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of  Chef/7706yThe University of British ColumbiaVancouver, CanadaDate  NT ^'73DE-6 (2/88)ABSTRACTThe enantioselectivity of the di-rr-methane photorearrangement in chiral crystalswas studied. Three achiral dibenzobarrelene carboxylic acid derivatives and one achiralamino dibenzobarrelene derivative were synthesized for this work. These compoundswere induced to crystallize as chiral crystals by forming salts with chiral counterions.Photolysis of the chiral crystals resulted in asymmetric induction in the products.The regioselectivity of the photorearrangements of the starting materials prior tosalt formation was studied in solution as well as in the solid state. In addition, thephotochemistry of the methyl and ethyl esters of the carboxylic acids was alsoinvestigated. The photochemistry of some of these dibenzobarrelene derivatives wasfound to be medium-dependent. Possible structure-reactivity correlations for thesecompounds, based on X-ray crystallographic data, are discussed.The chiral salts of the dibenzobarrelene derivatives were photolyzed in solutionand in the solid state. The extent of asymmetric induction in the crystalline phase wasstudied by measuring the enantiomeric excess of the photoproducts. No optical activitywas observed for the solution photoproducts, presumably because the salts haddissociated. The enantioselectivity of the di-rc-methane photorearrangement in the solidstate varied from poor to good, depending on the chiral salt in each instance. The chiralcounterion ensures chiral crystals, but the crystal lattice alone is accountable forasymmetric induction, and different salts crystallized in different crystal packingarrangements leading to varying degrees of asymmetric induction.The absolute steric course of the di-n-methane rearrangement in the solid statewas studied. Chiral salts for which the photorearrangements were both regio- andenantioselective were selected for this study. The reaction pathway was mapped byricomparing the absolute configuration of the starting material and the photoproduct. Afterthe reaction pathways had been determined, the crystal structures of the starting materialswere analyzed in order to identify the crystal forces that control the enantioselectivity ofthe di-n-methane rearrangement. A general explanation of the factors that control theasymmetric induction of the di-m-methane rearrangement in dibenzobarrelene derivativeswas proposed. It was concluded that forming chiral salts of achiral molecules is aconvenient and an effective way to study reactions in chiral crystals.Finally, solid state photochromism was discovered for some of the bridgehead-substituted dibenzobarrelene derivatives. Possible intermediates responsible for thephotochromic phenomena are suggested.iiiTABLE OF CONTENTSAbstract^ iiTable of Contents^ ivList of Tables viiiList of Figures^ xAcknowledgment xivDedication^ xvChapter 1 Introduction^ 11.1 Photochemistry in the Solid State^ 21.2 The Topochemical Principle 31.3 Photoreactions in Organic Salt Crystals^ 101.4 Asymmetric Photoreactions in the Solid State 111.5 The Di-it-methane Rearrangement^ 211.6 Research Objectives^ 27Chapter 2 Results and Discussion 322.1 Preparation of Substrates^ 322.2 Photochemical Studies of the Starting Materials Prior to Salt Formation^ 352.2.1 Photolyses of Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic acid (36)^ 35iv2.2.2 Photolyses of 12-Methy1-9,10-dihydro-9,10-ethenoanthracene-11-carboxylate (37) and Methyl 12-Methy1-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid (41) 432.2.3 Photolyses^of^Dimethyl^9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)^ 472.2.4 Photolyses of Trimethyl 9,10-Dihydro-9,10-ethenoanthracene-9,11,12-tricarboxylate (43) and Dimethyl 9-Ethoxycattony1-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (44) 542.2.5 Photolyses^of^Dimethyl^9-Carboxy-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (38)^ 632.3 Photochemistry of Salts of Starting Materials 682.3.1 Photolyses of Salts of Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-calboxylate-12-carboxylic acid (36)^ 682.3.2 Photolyses^of^Salts^of^12-Methy1-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid (37)^ 812.3.3 Photolyses of Salts of Dimethyl 9-Carboxy-9,10-dihydro-9,10-ethanoanthracene-11,12-dicarboxylate (38)^ 882.3.4 Photolyses of Salts of Dimethyl 9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)^ 952.4 Absolute Steric Course of Solid State Di-a-methane Rearrangements^ 1002.4.1 Absolute Configurations of S-(-)-Proline tert-Butyl Ester Salt 96and Photoproduct 32a^ 1012.4.2 Absolute Configurations of Salts 103, 105a and Photoproduct 52^ 1092.4.3 Absolute Configurations of R-(-)-Camphorsulfonic Acid Salt116a and Photoproduct 60^ 1172.5 Solid State Photochromism of Dibenzobarrelene Derivatives 38, 43 and44^ 1222.6 'y-Ray Irradiation of Salt 134^ 131Chapter 3 Experimental^ 1333.1 General^ 1333.2 Preparation of Starting Materials^ 1363.2.1 Synthesis of Starting Materials 1363.2.2 Salt Formation of Starting Materials^ 1513.2.2.1 Salt^Formation^of Ethyl^9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic acid (36)^ 1513.2.2.2 Salt Formation of 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylic acid (40)^ 1623.2.2.3 Salt Formation of 12-Methy1-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid (37)^ 1633.2.2.4 Salt Formation of 9-Carboxy-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (38)^ 1663.22.5 Salt Formation of Dimethyl 9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)^ 1683.3 Photochemical studies^ 1713.3.1 General Procedures 1713.3.2 Diazomethane Workup^ 1723.3.3 Photolyses of Starting Materials^ 1733.3.4 Photolyses of Salts^ 1883.3.4.1 Photolyses of Salts Formed with Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylicAcid (36)^ 1883.3.4.2 Photolyses of Salts Formed with 12-Methy1-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic Acid (37)^ 193vi3.3.4.3 Photolyses of Salts Formed with 9,10-Dihydro-9,10-ethenoanthracene-11-cathoxylic Acid (40)^ 1963.3.4.4 Photolyses of Salts Formed with Dimethyl 9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)^ 1983.3.4.5 Photolyses of Salts Formed with Dimethyl 9-Carboxy-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate(38)^ 200^3.3.5 Absolute Configuration of Some Photoproducts^ 2023.3.5.1 Absolute^Configuration^of^4b,8b,8c,8d-tetrahydrodibenzo[aXcyclopropa[c,d]pentalene-8b,8c-dicarboxylic acid, 8c-methyl 8b-ethyl ester (32a)^ 2023.3.5.2 Absolute Configuration of Methyl 8b-Methy1-4b,8b,8c,8d-tetrahydrodibenzo[a,ficyclopropa[c,d]-pentalene-8c-carboxylate (52)^ 2043.3.5.3 Absolute Configurations of Dimethyl 8d-Amino-4b,8b,8c,8d-tetrahydrodibenzo[a,ficyclopropa[c,d]-pentalene-8b,8c-dicarboxylate (60)^ 206REFERENCES^ 210viiLIST OF TABLES2-1 Medium Dependent Photochemistry of Ester-Acids 36 and 51^ 372-2 Medium Dependent Photochemistry of Tri-Esters 43 and 44 562-3 Regioselectivity of the Di-n-methane Rearrangement of Acid 38^ 642-4 Photoproduct Mixture Composition for Salts 85 to 89^ 692-5 Photoproduct Mixture Composition for Complexes and Salts 90a to 90cand 92 to 98 ^ 762-5 Photoproduct Mixture Composition for Complexes and Salts 90a to 90cand 92 to 98 (Continued)^ 772-5 Photoproduct Mixture Composition for Complexes and Salts 90a to 90cand 92 to 98 (Continued)^ 782-6 Photoproduct Mixture Composition for Salts 103 to 105^ 822-7 Photoproduct Mixture Composition for Salts 108 and 109 882-8 Enantiomeric Excess in Photoproduct 60 from Salts 116a to 117^ 973-9 Medium Dependent Photochemistry of Ester-Acid 36^ 1753-10 Photoproduct Mixture Composition of Salts 85 to 89 as a Function ofthe Photolysis Medium^ 1893-11 Solid State Photoproduct Mixture Composition for S-(-)-Proline Salt90a as a Function of Temperature^ 1913-12 Solid State Photoproduct Mixture Composition for Photolyses of Salts90b, 90c and 92 to 98^ 1923-13 Solution Photoproduct Mixture Composition for Salts 90a, 90b, 90c, 92and 98 ^  193viii^3-14^Solid State Photoproduct Mixture Composition for Salts 103, 104 and105b ^ 1953-15^Solution Photoproduct Mixture Composition for Salts 103, 104 and105b ^ 1953-16 Enantiomeric Excess in Photoproduct 60 as a Function of PhotolysisMedium and Sulfonic Acid Structure^ 2003-17^Solid State Photoproduct Mixture Composition for Salts 108 and 109,after Diazoethane Work-up^ 202ixLIST OF FIGURES1-1 Photodimerization of trans-Cinnamic Acid (1) in the Crystalline Stateand in Solution^ 41-2 Pictorial Representation of the "Reaction Cavity" Concept^ 51-3 Photodimerization of Pentanone 5^ 71-4 Medium Dependent Photochemistry of Cyclohexenone 7^ 81-5 Photodimerization of Acridizinium Salt 11^ 111-6 Enantiomeric Conformations of 1,1'-Binaphthyl 131-7 Absolute Asymmetric [2+2] Photodimerization of Compounds 14 and15  ^151-8 Asymmetric [2+2] Photodimerization of Compound 19^ 161-9 Absolute Asymmetric Photodimerization of Compound 24 in ChiralCrystals  ^181-10 Absolute Asymmetric Norrish Type II Reactions in Chiral Crystals^ 191-11 Solid State Hydrogen Abstraction for Compound 30^ 201-12 Representation of a Di-it-methane Rearrangement 221-13 Regioselectivity of the Di-it-methane Rearrangement in Esters 31a to31d  ^231-14 Four Different Pathways for the Di-x-methane Rearrangement in Ester34  ^251-15 Crystal Conformation of Compounds 34 and 35^ 261-16 Dibenzobarrelene Derivatives^Selected^for Studying AsymmetricInduction in Chiral Salt Crystals 281-17 The Four Di-it-methane Systems in the Dibenzobarrelene Skeleton^ 30x1-18 Dibenzobarrelene Derivative 40^ 312-19 Preparation of Dibenzobarrelene Derivatives via the Diels-AlderReaction ^ 322-20 Preparations of Acids 37 and 40^ 332-21 Preparation of Ester-Acid 37 342-22 Preparation of Ester-Acid 38^ 342-23 Photolysis of Di-Ester 31a 352-24 Photolysis and Work-up of Ester-Acid 36 and 51^ 362-25 Different Hydrogen Bonded Forms of Ester-Acids 36 and 51^ 382-26 1,2 Aryl Shift of Ester-Acids 36 and 51^ 402-27 Medium Dependent Infrared Spectra of Ester-Acid 36^ 412-28 Inductive Effect of the Alkyl Substituent on the Infrared CarbonylStretching Frequency of Alkyl Benzoates^ 422-29 Photolysis of Compounds 37 and 41 432-30 Structure of Compounds 54 and 55^ 442-31 Comparison^of Radical^Stabilization^Energies^of^9-SubstitutedFluorenyl and Substituted Methyl Radicals 462-32 Photolysis of Monosubstitued Dibenzobarrelenes 56 and 57^ 472-33 Photolysis of Amine 39^ 482-34 Equilibrium between Norcaradiene and Cycloheptatriene^ 492-35 Crystal Structure of Amine 39^ 512-35 Crystal Structure of Amine 39 (continued)^ 522-36 Photolysis of Ketone 65^ 542-37 Photolysis of Tri-Esters 43 and 44^ 552-38 Di-n-methane Rearrangement of Ester 74 572-39 Crystal Structure of Tri-Ester 43^ 592-40 Photorearrangement of Labeled Benzobarrelene via the S i^ 60xi2-41 Photolysis of Compound 77^ 612-42 Singlet State Photorearrangement of Dibenzobarrelene Derivatives 83and 84 ^ 622-43 The Di-it-methane Rearrangement of Acid 38^ 642-44 Crystal Structure of the Ethanol Solvate of Acid 38 662-44 Crystal Structure of the Ethanol Solvate of Acid 38 (Continued)^ 672-45 Photolysis of Salts of Ester-Acid 36^ 702-46 Hydrogen Bonded Complex of Proline and Acid 91 in the Solid State^ 712-47 11-1-NMR Spectra of a Mixture of Photoproducts 32a and 33a^ 732-48 Product Ratio of Photolyses of S-(-)-Proline Salt 90a at DifferentTemperatures^ 742-49 Enantiomeric Excess of Photoproduct 32a at Different Temperatures ^ 752-50 Photolysis of Dibenzobarrelene Derivatives 99 and 100^ 802-51 Photolysis of Salts of Acid 37^ 832-52 11-1-NMR of Photoproduct 52 842-53 Diastereomers 106 and 107^ 852-54 11-I-NMR Spectra of Diastereomers 106 and 107^ 862-56 1H-NMR Spectra of Product 70^ 912-57 Crystal Structure of (S,S)-Ephedrine Salt 108^ 932-58 Photolysis of Compounds 39, 112 and 113 962-59 1H-NMR Spectra of Photoproduct 60^ 982-60 Crystal Structure of Proline tert-Butyl Ester Salt 96^ 1022-60 Crystal Structure of S-(-)-Proline tert-Butyl Ester Salt 96 (continued)^ 1032-61 Transesterification of Di-Ester 32a into Di-Ester 35^ 1042-62 Circular Dichroism Spectra for Compounds 32a and 35 1052-63 Some Torsion Angles of S-(-)-Proline ten-Butyl Ester Salt 96^ 1062-64 Type II Reaction of Macrocyclic Di-ketones^ 107xii2-65 Crystal Structure of Di-Ester 34^ 1082-66 Crystal Structure of (S,S)-(+)-Pseudoephedrine Salt 105a.^ 1092-66 Crystal Structure of (S,S)-(+)-Pseudoephedrine Salt 105a (continued)^ 1102-67 Synthesis of Compound 106^ 1112-68 Crystal Structure of Compound 106 1122-69 Some Torsion Angles in (S,S)-(+)-Pseudoephedrine Salt 105a^ 1132-70 Crystal Structure of S-(+)-Prolinol Salt 103^ 1142-71 Some Torsion Angles in S-(+)-Prolinol Salt 103 1152-72 Photochemistry of Salt 117^ 1162-73 Crystal Structure of R-(-)-Camphorsulfonic Acid Salt 116a^ 1182-74 Circular Dichroism Spectrum of Photoproduct 60^ 1192-75 Crystal Structure of Compound 120^ 1202-76 Absolute^Configuration^of (-)-Enantiomer^of Product^60^andCompound 120 1212-77 Some Torsion Angles in R-(-)-Camphorsulfonic Acid Salt 116a^ 1222-78 Dibenzobarrelene Derivatives 38, 43 and 44 which Display Solid StatePhotochromism^ 1232-79 Solid State UV-Visible Absorption Spectra for Compounds 38 and 43^ 1242-80 ESR Spectra of Irradiated Crystals of Tri-Ester 43^ 1252-81 Bridgehead^Substituted^Dibenzobarrelene^Derivatives^and^theirPhotochromism 1262-82 Radical Ions 126-128^ 1272-83 Styrylpyridinium Tetraphenylborate Complex, 129^ 1282-84 Zwitterion 130^ 1292-85 Biradicals 131 and 132^ 1302-86 Compound 133 1312-87 Photolysis of Piperidine Salt 134^ 132ACKNOWLEDGMENTI would like to express my sincere thanks to my research supervisor, Dr. J.R.Scheffer for his guidance and encouragement throughout the course of my research andin the preparation of this thesis. I appreciate assistance from Dr. S. Rettig and Dr. J.Trotter in preparing the crystallographic work for this research. I addition I wish to thankall my co-workers in solid state chemistry for their friendship and support. The staff ofthe departmental service laboratories deserve acknowledgment for their assistance.Thanks are due to the proof-readers of this thesis, Melvin Yap, GuOrtinHelgadOttir and Jana Mica. Further more, Jana, I am grateful for all your help inpreparing this thesis, your patience with my "English", but most of all for yourfriendship. GuOrtIn and Helgi, I do appreciate your help and friendship as well.Special gratitude to my uncle Siguraur Halldársson and his wife SigrtinMagniisdOttir for their support.Finally, Kristinn, I appreciate all your help in preparing this thesis. I am thankfulfor all your encouragement, support and sense of humor throughout this work.xivDEDICATIONFor Oskar, GuOmundur and SvanhildurFyrir handan fj011in sj6Ma dverganir sj6bf8a tin Mjallhvftme8 sj6 grOuga munnasj6faldar kvartanirn$ elf tila8 *Lira.Er ekki betraa8 lata skera dr sir hjarta8en 0 grata sig lifandi1318andieftir einhverjum kOngssynisem hefur lif 'AftI hendi ser uppfra InfI glerkistusofandisvefni vanans.126rdis Richardsdöttir* . Evintframerall.xvChapter 1 IntroductionCHAPTER 1 INTRODUCTIONThe design of reactions that yield optically active products from achiral startingmaterials is one of the core topics in organic chemistry. Numerous asymmetric reactionshave been reported.' The strategy most of these reactions have in common utilizes thedissymmetric influence of a resolved chiral agent on the reaction pathway to accomplishasymmetric induction, which results in diastereomeric transition states. The dissymmetricinfluences arise from the chiral characteristics of the resolved reagent, catalyst, solvent orhost molecule in most instances. Asymmetric reactions have also been carried out in theabsence of any external chiral material. Such reactions are referred to as absoluteasymmetric syntheses2 and are obvious candidates for an explanation of the origin of thefirst chiral organic material in nature.3Asymmetric induction in fluid phase photochemistry has not been widely studied,even though the hypothesis that circularly polarized light may have led to the generationof optical activity on earth dates back to the 19th century. 4 Direct irradiation withcircularly polarized light has been shown to lead to optically active products, but only inextremely low optical yields. The most common method of generating dissymmetricinfluences in the photochemistry of fluid solutions is the use of chiral reagents.Methodologies such as chiral solvents, sensitizers and complexes have also been studied.Nevertheless these methods are limited owing to low optical yields.In recent years topochemically controlled reactions in organic crystals havebecome of interest for asymmetric synthesis. 5 The literature lists examples ofphototransformations in the solid state that occur with good optical yields. Most of thesereactions make use of resolved chiral substituents that are chemically bonded to thereagents. In solution photochemistry, the chiral center alone is responsible for all1Chapter 1 Introductionasymmetric induction, but for reactions in the solid state, it is not only the chiral centerthat resides in a chiral environment; the entire molecule is influenced by the chiralsurrounding of the crysta1. 5 Moreover, examples of absolute asymmetric transformationsin the solid state have been reported that lead to high optical yields in the photoproducts. 5In these instances, achiral molecules that form chiral crystals have been transformed intochiral products using the chiral environment of the crystals as the sole source of chiralityin the photoreaction.1.1 Photochemistry in the Solid StatePhotoreactions in the solid state are capable of giving products that are verydifferent from those observed in fluid phase owing to the fact that the crystal latticeinfluences the reactivity. Two important factors govern reactions in the solid state: theconformation and the packing arrangement of the reacting molecules. Compared toisotropic phases, in which a flexible organic molecule may adopt many conformationsowing to the fast equilibrium between them, in crystals the molecule will rarely adoptmore than one conformation, and it is most common that organic molecules crystallize inor near their minimum energy conformation. 6 The limitation on motion in crystals willconsequently affect reactions that are sensitive to the conformation of the reactant andthis can lead to increased selectivity. The packing of the reactant into a crystal isimportant, since the anisotropic environment of the crystal lattice can affect the course ofa reaction.Despite an interesting history that dates back to the early 20th century, 7 organicsolid state photochemistry did not become a subject of systematic study until the 1960swhen X-ray crystallography became readily available. With the aid of X-ray2Chapter 1 Introductioncrystallography, it is possible to analyze in detail the molecular conformation as well asthe packing arrangement in the crystals, leading to a greater ability to interpret structure-reactivity relationships.Considerable progress has been made in this field and many solid state reactionshave been reported in recent years. In response, a number of review articles and bookshave been published. 8 A serious limitation of solid state chemistry is the inability topredict the packing arrangement of molecules in a crystal lattice, thus the study of solidstate photochemistry tends to be of the "discover-and-explain" variety. This can beobserved to be a normal procedure in laying out the foundations for a new field andundoubtedly, with a better understanding of crystal packing and topochemistry, solidstate reactions will eventually be rationally planned and exploited.1.2 The Topochemical PrincipleThe first principle of solid state chemical reactivity was proposed by Kohlshutter 9in 1918 and was termed the topochemical postulate. According to Kohlshutter atopochemical reaction in the solid state recognizes the restriction of the rigid three-dimensional environment in the crystal and therefore all reactions take place withminimum atomic and molecular movement.Many years later, Schmidt and his co-workerso carried out pioneering studies onthe photodimerization of trans-cinnamic acid derivatives in the crystalline phase whichallowed them to refine the topochemical postulate and establish it on a valid experimentalbasis. In their investigation they discovered that trans-cinnamic acid (1) crystallized inthree polymorphic forms, a, 13, y, and each form showed characteristic photochemicalbehavior (Figure 1-1).3Chapter 1 IntroductionPI\^HC3C) -C\--=\ COONPh1a-form tw^HccfriccalPh2PI\ ^ by\Pt\^ COON\COCH13-form 3Pt\_—^\Ft\ CCOHCOWhvNO REACTIONrformPt\ ^by___________•„,^Pti‘%.=-__/ C°:*-14Figure 1 - 1^Photodimerization of trans-Cinnamic Acid (1) in the Crystalline State andin Solution\coOH Solution4Chapter 1 IntroductionThe centrosymmetric intermolecular arrangement in the a-type crystals led tocentrosymmetric a-truxilic acid 2, whereas the parallel translation arrangement in the 13-form gave the mirror-symmetric ii-truxilic acid 3. Finally the y-form was found to beunreactive because of the poorly overlapped arrangement of the two double bonds. Insolution, however, trans-cinnamic acid 1 underwent only trans-cis photoisomerizationand no dimerization was observed.On the basis of crystallographic and photochemical studies of trans-cinnamic acid(1) and other derivatives, Schmidt suggested that the packing of the molecules in thecrystal lattice, which determines the distance and orientation of the reactive centers withrespect to one other, is the dominant factor in controlling solid state reactions." Thistopochemical postulate has been viewed as the basic foundation of organic solid statephotochemistry, even though some exceptions exist. 5b, 8i DisallowedFigure 1-2^Pictorial Representation of the "Reaction Cavity" Concept. ReactionCavity before Reaction (solid line) and at the Transition State (brokenline)5Chapter 1 IntroductionAn extension of the topochemical postulate which was introduced by Cohen isreferred to as the reaction cavity. 12 It considers a reacting molecule in a crystal as asubstance lying in a cavity formed by the presence of its adjacent molecules and theshape of the cavity is set by the packing of the crystals (Figure 1-2). Atomic andmolecular movements necessary for a reaction cause pressure on the cavity wall whichmay become disordered, but any distortion in the shape of the cavity would be restrictedby the closely packed environment. Cohen redefined the topochemical postulate to meanthat those reactions which proceed under lattice control do so with minimal distortion ofthe surface of the reaction cavity. In cases where more than one reaction pathway ispossible, the pathway leading to least disruption of the cavity would be favored.Ramamurthy et al. 13 have further extended the concept of the reaction cavity andapplied it to reactions in organized media other than crystals. This allowed them torationalize reactivity in media such as organic inclusion hosts, silica and zeolite surfaces,micelles and liquid crystals.The close fit of the reaction complex in the surrounding crystal results in productmolecules which are a substitutional solute in the parent crystal. As the reactionproceeds, the product concentration increases until the solubility limit is exceeded andthe reaction stops or the product crystallizes out in its own structure. In exceptional cases,where there is a structural similarity between the reactant and the product, the reactionpreserves its crystallinity during the entire conversion of the starting material to theproduct. Such reactions are called topotactic or single crystal-to-single crystaltransformations. 14 Very few examples of topotactic reactions are known. One that hasbeen intensively studied, by Jones et al., 15 is the photodimerization of pentanone 5(Figure 1-3). It was observed that the packing arrangement of the starting material andthe product are such that the reaction requires very little atomic motion. Mostinterestingly, the reaction was monitored directly during the dimerization by X-raycrystallography.6Chapter 1 IntroductionbyCrystal56Figure 1-3^Photodimerization of Pentanone 5Scheffer, Trotter and co-workers80j.m have systematically investigatedunimolecular reactions in the solid state. They have identified a specific intermoleculareffect, steric compression, that hypothetically modifies the solid state reactivity in a waysimilar to the reaction cavity concept. 16 They studied the photochemistry ofcyclohexenone 7 and found that in solution, irradiation afforded quantitative yields ofcage product 8, whereas irradiation of crystals of 7 gave compound 9 as the only product(Figure 1-4). The different reactivities in solution versus the solid state was explained inthe following way: in solution relatively fast intermolecular [2+2] dimerization formsproduct 8 via the high energy conformation 7B. Since cyclohexenone 7 is locked in theminimum energy conformation 7A in the crystal, formation of product 8 is prevented andonly the hydrogen abstraction-initiated product 9 is formed. This was unexpectedbecause close analogues of compound 7 transfer a hydrogen to C(3), presumably to formthe more stable radical center at C(2). 17 Biradical closure then leads to products similar to10. X-ray crystallography revealed that there is close contact between the methyl groupat C(3) and the methyl group of a neighboring molecule located below it in the lattice.7Chapter 1 Introduction4^ conformer 7Bconformer 7AI polymer bycrystal I solution8I C2 - 05C  boning^ 1 C3 - Cs boning10^ aFigure 1-4 Medium Dependent Photochemistry of Cyclohexenone 78Chapter 1 IntroductionConsequently the authors suggested that this contact raises the activation energy of thenormally favored hydrogen transfer to C(3) through steric hindrance caused by thedownward motion of the methyl group which accompanies pyramidalization at thiscenter. There is no such contact involving the methyl group attached to C(2), whichexplains formation of the observed product 9.Further support for this hypothesis came from photolysis of enone 7 in polymerfilms." It was expected that polymer matrices would provide a medium midway betweensolution and the solid state that lacks the specific close contact to the C(3) methyl groupand also slows the conformational isomerization to 7B. Therefore, if steric compressionwas responsible for the unusual reactivity of enone 7 in the solid state, then in polymerfilms it should react normally to give compound 10 upon photolysis. Such was indeedwhat was observed experimentally.Gavezzotti suggested that free volume or space in crystals determines reactivity inthe solid state."' He explained that in the initial stage of a reaction the free volume or theavailable non-occupied space around the reactive centers controls the selectivity of thereaction because of the presence of relatively stationary neighbors. The free volumeconcept is by no means a departure from Cohen's reaction cavity concept, but can beviewed as an extension of it. The advantage of the Gavezzotti free volume concept is thatit is better defined and can be quantitatively calculated. Furthermore it has been usedsuccessfully to explain solid state reactivity.19, 20The effects of a crystalline medium on a photoreaction can be divided into twocategories. A primary effect controls the conformation of the reagent. In solutionreactions can proceed from high energy conformations while in crystals reactions aregenerally limited to one low energy conformation. A secondary effect arises when thecrystal lattice restricts the motions required for a given reaction through anintermolecular steric effect which depends on the crystal packing arrangement. Most ofthe postulates discussed above fall into this latter category.9Chapter 1 Introduction13 Photoreactions in Organic Salt CrystalsSolid state photochemistry has focused mainly on reactions in molecular crystalswhile photoreactions of crystalline organic salts have not received much attention. Themajor difference between molecular crystals and crystals made from salts of the typeRCOO-M+ and RNH3+X- is the force holding the crystal lattice together. Molecularcrystals are held together by relatively weak dipole-dipole forces, van der Waals forcesand in some cases by comparatively strong hydrogen bonds as well. As chemical changesin the solid state usually lead to softening and melting of the sample as the reactionproceeds, solid state reactions are stopped before any melting of the crystal is observed inorder to obtain topochemical control. The lattice in salt crystals is held together by strongCoulombic attractive forces which are ionic in nature. This usually results in crystalswith high melting points, increasing the chances of observing single crystal-to-singlecrystal reactions or at least allowing the solid state reactions to be carried out to higherconversions without loss of topochemical selectivity. Jones and co-workers 21 observed atopotactic photoreaction in organic salt crystals when they studied the [4+4]photodimerizations of various acridizinium salts (Figure 1-5). Interestingly the majorityof these reactions were found to be of the single crystal-to-single crystal variety. X-raystructure analysis of the starting materials and dimers showed that dimerization requiresconsiderable movement of the molecules in the crystal lattice. The authors suggested thatconcomitant movement of the anion minimizes the potential energy of the formed dimercrystals which allows topotactic reaction to be observed.10Chapter 1 Introduction byX= a, Br,'X -• (F120)n^ (X1 2 . (H20),11 12Figure 1-5^Photodimerization of Acridizinium Salt 11The mutual solubility of most pairs of organic molecules is limited in the solidstate,8  but the two component nature of organic salt crystals allows introduction of achemically useful second constituent into the molecular crystal lattice. The normallypassive counterion can thus be chosen to act as a sensitizer, quencher or optically activetemplate for asymmetric synthesis. These properties of salts have not been widelyexploited and one of the few example comes from Yap, 22 who utilized 3'-aminoacetophenone as a solid state photosensitizer in an ammonium carboxylate type salt.1.4 Asymmetric Photoreactions in the Solid StateA brief discussion of crystal chirality is necessary for a better understanding ofsolid state asymmetric synthesis. Molecules that are not superimposable on their mirrorimages are chiral and the molecular chirality derives from the dissymmetric, three-dimensional arrangement of the component atoms.23 There are 230 possible ways to packmolecules into a crystal lattice and each corresponds to a different so-called spacegroup. 24 Space groups can be chiral or achiral depending on the presence or absence of11Chapter 1 Introductionsymmetry elements that convert one enantiomorph into the other. Of the 230 possiblespace groups 65 are chiral. Crystal chirality arises from the dissymmetric spatialarrangement of molecules in the crystal lattice. Consequently, all resolved chiralmolecules must crystallize in chiral space groups. Racemic compounds will eithercrystallize in racemic crystals that contain equal amounts of each enantiomer, or they willspontaneously resolve into chiral crystals of each enantiomer. It is far from common thatracemic compounds resolve spontaneously upon crystallization,3b however, thisphenomenon allowed Pasteur25 in 1848 to perform the first optical resolution utilizingracemic sodium ammonium tartrate. Achiral molecules can also resolve spontaneously toyield chiral crystals. One of the most well studied examples of an achiral molecule whichresolves into chiral crystals comes from Pincock et al. 26 They studied crystallizations of1,1'-binaphthyl (13, Figure 1-6). Rapid equilibrium between the two enantiomericconformations in the molten phase makes 1,1'-binaphthyl maintain its overall symmetry.Heating crystals of racemic 1,1'-binaphthyl leads to a metastable liquid, whichcrystallizes to form high melting crystals. When the melt crystallizes, random nucleationleads to the growth of chiral crystals containing mainly one of the enantiomericconformations. The authors demonstrated that heating the low melting crystals withoutmelting them led to formation of the high melting optically active crystals, that is theresolution occurs in the solid state as well.12Chapter 1 Introductionmelt__I.. S-(+)-13^R-(-)-13Figure 1-6^Enantiomeric Conformations of 1,1'-BinaphthylWhen chiral crystals are utilized to perform asymmetric synthesis, it is necessaryto transform the chirality of the crystal into the molecular chirality of the product througha stereospecific solid state reaction. Only a few asymmetric syntheses in the crystallinephase have been reported in the literature due to the strict requirements for suchreactions.5 Absolute asymmetric reactions in the solid state are usually the result ofaccidental discoveries of an achiral compound which crystallizes in a chiral space groupand undergoes a solid state reaction to give optically active products. Asymmetricsyntheses can be studied systematically in the solid state using optically pure reagents toproduce chiral crystals. An advantage of performing asymmetric transformations ofchiral substances in the solid state over solutions is that the chiral center not only ensureschiral crystals but also exerts a second asymmetric influence in the chiral lattice of thecrystal. In other words, it is not only the chiral center that resides in a chiral environment;the entire molecule is influenced by the chiral surrounding of the crystals.It is possible to design an asymmetric reaction because structure-reactivitycorrelations make it possible to understand product formation and selectivity in the solidstate. Planning an asymmetric reaction in the crystalline phase requires the molecules topack into a crystal lattice in a manner which produces the necessary topochemical13Chapter 1 Introductioncharacteristics to lead to the desired product. Crystal packing is hard to predict since theintermolecular forces that control the packing arrangement in crystals are not wellunderstood. Attempts to steer molecules into certain pre-determined arrangements duringcrystallization (termed "crystal engineering") have been made.' 27 Crystal engineeringutilizes empirical packing generalizations, but this approach is not very advanced andconsequently, obtaining specific crystal structures remains mainly heuristic.The possibility of utilizing the chirality of crystals to achieve absolute asymmetricinduction was first considered almost a century ago,5b but a better understanding of andexperience with topochemically allowed reactions in the crystalline phase were neededbefore any success was achieved. The first example of an asymmetric photoreaction thatmade use of chiral crystals came from Schmidt and his co-workers,28 who extended theirstudy on [2+2] photodimerization in the crystalline phase and performed asymmetricphotodimerizations as well. Their objective was to form crystals that contained ahomogeneous distribution of two vinyl compounds. When this type of crystal isirradiated, it is possible to form homodimers as well as non-symmetrical heterodimers;crystals that are chiral may have the potential to exert enough face discrimination in thereaction so that an enantiomeric excess may be observed in the resulting heterodimers.Schmidt and his co-workers29 tested this concept on diarylbutadienes 14 and 15(Figure 1-7). These compounds crystallize in the chiral space group P212121, but yieldachiral head to head dimers 16 and 17 when irradiated. However when a dilute (15%)solid solution of 15 in 14 was prepared and the longer wavelength-absorbing thiophenecompound 15 selectively photolyzed, heterodimers 18 was obtained in 70% enantiomericexcess.14Chapter 1 Introduction1718bhv 15[Th = Ar =Figure 1-7^Absolute Asymmetric [2+2] Photodimerization of Compounds 14 and 15Analysis of the crystal structures 30 of the mixed crystals led Schmidt and co-workers to conclude that the ground state orientations of the reagents did not explain theenantiomeric selectivity. They suggested that compound 15 must deform in the excitedstate to favor formation of one of the enantiomers over the other. This hypothesis wasfurther supported with theoretical calculations 31Schmidt and his co-workers28 recognized another approach for asymmetric [2+2]photodimerization which utilizes the packing arrangement of benzene-1,4-diacrylatederivatives. Hasegawa32 had studied this type of compound earlier and discovered that ittends to crystallize in stacks in such a way that the two non-identical double bondsoverlap.15hvChapter 1 IntroductionP1 Crystals^ P21 CrystalsFigure 1-8^Asymmetric [2+2] Photodimerization of Compound 1916Chapter 1 IntroductionSuch an arrangement in the solid state leads to the generation of non-centric chiralphotodimers. By inserting a chiral handle into this kind of molecule it is possible that theasymmetric induction of a solid state reaction in the chiral crystals yields enantiomericexcess in the photoproducts. Lahav and Addadi 33 tested this concept on benzene-1,4-diacrylate, 19, which contains a resolved sec-butyl group to induce formation of chiralcrystals (Figure 1-8). Compound 19 crystallized in two dimorphic modifications, P1 andP2 1 . When crystals of the P1 form were photolyzed, quantitative diastereomeric yields ofdimer 20 were obtained. In contrast, irradiation of the P21 form gave quantitativediastereomeric yields of dimer 21. Transesterification of dimers 20 and 21 with methanolgave optically pure enantiomers 22 and 23 respectively. The authors argued thatformation of both enantiomers of the same product in quantitative optical yields usingonly one handedness of chiral handle shows that the chiral crystal environment alone isresponsible for the asymmetric induction. The chiral handle only serves to ensure theformation of chiral crystals.Hasegawa34 discovered an absolute asymmetric transformation in the solid statewhen he studied the [2+2] dimerization of compound 24 (Figure 1-9). Dimer 25 wasformed in 100% enantiomeric excess when 24 was irradiated in the solid state. Thephotochemical behavior of compound 24, as well as the asymmetric induction, could bereadily interpreted by X-ray structure analysis. Molecule 24 crystallizes in the chiralspace group P2 1 2 1 21 in a cisoid form. It is apparent from the X-ray structure that dimersof the same chirality are formed by reaction of either pairs of double bonds.17Chapter 1 Introductionby 24^25Figure 1-9^Absolute Asymmetric Photodimerization of Compound 24 in ChiralCrystalsAsymmetric [2+2] photodimerizations have been studied systematically in thesolid state and efforts have been made to design chiral crystals that have certainintramolecular as well as intermolecular features to gain selective reactions. This hasresulted in a number of [2+2] photodimerizations in the solid state which afford productsin good optical yields. Asymmetric unimolecular photoreactions have also beeninvestigated but not as intensively. Two examples of absolute asymmetric Norrish type IIreactions in chiral crystals have been reported. Scheffer, Trotter and co-workers 35discovered that the adamantyl ketone 26 crystallizes in the chiral space group P2 1 2 1 2 1(Figure 1-10). Solid state photolysis yielded cyclobutanol 27 in 80% enantiomeric excessvia the Norrish type II reaction. The other example of an absolute asymmetric Norrishtype II reaction in the solid state comes from Toda et al. 36 They found that compound 28,which crystallizes in space group P21212 1 , yields 13-lactam 29 in high optical yields whenirradiated (Figure 1-10).18Chapter 1 Introductionhv1■ t i...OH26^ 27hvOH _^Ph ^=7/I^N\leo^0 Pr 28^ 29Figure 1-10 Absolute Asymmetric Norrish Type II Reactions in Chiral CrystalsAn excellent example of how reaction pathways can be probed in chiral crystalscomes from McBride and Feng." The achiral compound 30a crystallizes in chiralcrystals and photolysis in the solid state yields a pair of 10-bromodecyl radicals (Figure1-11). These radicals abstract an a-hydrogen to form secondary and primary pairs ofradicals. ESR studies suggest that only one equatorial hydrogen is abstracted to form thea-pair. Further ESR studies of compound 30b with deuterium in the a-positionsdemonstrated that when the a-hydrogens are replaced with deuterium, the two primaryradicals abstract the 8-hydrogen to generate a pair of secondary radicals. This result wasinterpreted as being due to an isotope effect that changes the preferred site of abstractionfrom a to 8. The optically active form of the di-deuterated 30c was synthesized in orderto differentiate between the reactivity of the two a hydrogens in 30a (Figure 1-11). Theperoxide (30c) crystallizes in the chiral space group P4 321 2 or the enantiomorphous space19OBr(cH2)961) 0/a-pair02 Br(CH2)6-CH-HCHACD2) 0/2 Br(CH2)9CR2CR2(CH2)9Br^+^Br(CH2)9CH2(CH2)6BrBr(CH2)6O030cFigure 1-11 Solid State Hydrogen Abstraction for Compound 30Chapter 1 Introductiongroup P4 1212, with almost equal probability since crystallization does not differentiatebetween deuterium and protium.0Br(CH2)9CFt2 0/hvCR2g>12)9Br^---. 2 CO2 + g3r(CH2)9CR230: (a) R = H; (b) R = D8-pair20Chapter 1 IntroductionWhen the molecules crystallize in the space group P4 32 12, all protiums of the chiralcenter reside in the equatorial position and are in close proximity to the primary radical.Photolyses of these crystals and ESR studies confirmed that the a-pair of radicals isformed, presumably as the a-hydrogens are in a position favorable for abstraction. Whencompound 30c crystallizes in the space group P41212, the hydrogen and deuterium atomsexchange positions. In this enantiomorph, the protium is in the axial position which is notin close contact with the primary radical and is therefore not expected to be abstracted.On the other hand the deuterium in the equatorial position is not abstracted due to theisotope effect, which changes the preferred site of abstraction and, as the authorspredicted, formation of the 8—pair is observed.These examples illustrate that asymmetric phototransfonnations in the solid stateare practical. The potential for asymmetric photosynthesis in the solid state is intimatelyconnected to progress in the general field of solid state organic photochemistry. Further,it is noteworthy that asymmetric oxidation 38 has been achieved from molecules thatcrystallize in non-chiral space groups, provided that the crystals contain a polar axis. Noasymmetric photochemical reaction has yet been reported from such crystals.1.5 The Di-x-methane RearrangementThe reaction under investigation in this thesis is the so-called di-n-methanerearrangement, one of the most thoroughly studied photochemical reactions. 39 As itsname implies, the di-ic-methane rearrangement is the rearrangement of a system of twon-bonds connected via a saturated carbon atom. The commonly accepted mechanism isshown in Figure 1-12. This mechanism was first proposed by Zimmerman 49 in 1967 andhas been successful in predicting the results for a large number of reactions, although the21Chapter 1 Introductiondiscrete existence of the initial 1,4-biradical remains questionable. Both aliphatic andaromatic n-bonds are capable of participating in this reaction. The chemoselectivity andregioselectivity of the di-n-methane rearrangement have been studied intensively.'"Several processes are known to be capable of competing with the di-it-methanerearrangement. These include photocycloadditions, cis-trans isomerizations andsigmatropic shifts. The regioselectivity that arises from unsymmetrically substituteddienes depends on the system under investigation, but generally the regioselectivity canbe interpreted as favoring the most stable biradical intermediate in the Zimmermanmechanism.hv r".1,4 - diene^ vinylcyclopropaneFigure 1-12 Representation of a Di-rc-methane RearrangementRecently Scheffer et al. 8:0 , In, 42 have carried out studies on the solid state di-n-methane rearrangement. This research has concentrated mainly on dibenzobarrelene andits various derivatives, which rearrange to form the corresponding semibullvalenederivatives. When the Zimmerman mechanism for the di-n-methane rearrangement isapplied to dibenzobarrelene bearing different substituents on the vinyl bond, tworegioisomeric products are possible. Scheffer et al:" discovered that the regioselectivityis affected by the reaction medium when they investigated the photobehavior of esters31a, 31b, 31c and 31d (Figure 1-13). Solution photolyses afforded photoproducts 32 and33. Regioisomer 33 in which the smaller ester group occupies the more sterically22lwtw32 3331Chapter 1 Introductioncongested position, was slightly favored. In contrast, the solid state irradiations weremuch more selective for each compound but without any general trend. Twoexplanations were considered to interpret the regioselectivity in the solid state.Compound E 32:33 RatioSolution^Solid State31a COOCH2CH3 47:53 55:4531b COOCH(CH3)2 45:55 7:9331c COOC(CH3)3 40:60 85:1531d COOCH(CH3)CH2CH3 40:60 1:99Figure 1-13 Regioselectivity of the Di-n-methane Rearrangement in Esters 31a to 31dFirst, attempts were made to explain the regioselectivity by comparing which ofthe two ester groups was better capable of stabilizing the initial radical on the vinylbridge. The ester group which has a more coplanar arrangement to the vinyl bond in thecrystals is more conjugated and is better capable of stabilizing the radical by resonance.23Chapter 1 IntroductionTherefore initial vinyl bridging should occur at the less conjugated ester group. X-raystructure analysis of the starting materials revealed that this theory does not explain theobserved regioselectivity.The second explanation suggested that steric effects between the reactingmolecule and its neighbors controlled the regioselectivity. The authors hypothesized thatthe ester group attached to the bridging vinyl atom must move considerably during thebenzo-vinyl bridging process and is therefore most likely to experience unfavorable stericinteraction with the crystal lattice. Based on the crystal structure data, non-bondedcontacts between the moving ester group and the crystal surroundings were estimated bya computer simulation of the benzo-vinyl bridging process and used to obtain potentialpacking energy calculations, which showed the increase in potential energy of the latticewas much smaller for the pathways observed experimentally in all cases.One of the most interesting results in these studies done by Scheffer et al 35.43 wasthe mapping of the absolute steric course for a reaction in the solid state. They discoveredthat the achiral dibenzobarrelene derivative 34 crystallizes in two dimorphic forms, Pbcaand P2 12 12 1 , where the latter is a chiral space group. Irradiation of 34 in solution and inthe Pbca dimorph yielded photoproduct 35 as a racemic mixture. In comparison,irradiation of the P212121 crystals gave quantitative optical yields of compound 35. Insolution, the di-lc-methane rearrangement of compound 34 is four fold degenerate. Figure1-14 depicts the four possible pathways, where pathways I and Mead to one enantiomerand pathways III and IV to the other. There must be complete discrimination betweenpaths (I + II) and (III + IV) in the solid state because quantitative enantiomeric excesswas obtained in the photoproducts. In order to differentiate between these pathways, theabsolute configurations of the starting material and its photoproduct were determined byusing the Bijovet method." The absolute configurations were found to be (11M, 12P)and (4bS, 8bS, 8cS, 8dS) for the starting material and the photoproduct, respectively.24E E(4bS, 8cS, 8bS, 8dS)1 1E E(4bR, 8bR, 8cR, 8dR)Chapter 1 Introduction(4bR, 8bR, 8cR, 8dR)I(4bS, 8bS, 8cS, 8dS)IIII IIIV E = 000cH(a-y2Figure 1-14 Four Different Pathways for the Di-It-methane Rearrangement in Ester 3425upmore hinderedless hindered12MCrystal Conformation11 ConformationalisomerzationCrystal Conformation35Chapter 1 IntroductionThe designation 11M, 12P for ester 34 focuses on the site of dissymmetry around theester groups in the molecule and uses the conformational chirality formalism forassigning absolute configuration. 45 The absolute configuration is specified around thesingle bonds, C 11-C13 and C12-C17, by determining the smallest torsion angle betweenthe groups of highest priority attached to each end of these single bonds. M stands for anegative torsion angle and P for a positive one. Comparison of the absolute configurationof the photoproduct and the starting material indicates that pathways I and or II arefollowed in the rearrangement.The shaded circles indicatecarbonyl oxygen atoms.34Figure 1-15 Crystal Conformation of Compounds 34 and 35Furthermore the authors suggested that pathway II is favored over pathway I bycomparing the X-ray conformation of photoproduct 35 with the X-ray conformation of26Chapter 1 Introductionthe starting material. Assuming a least motion process, pathway II produces photoproduct35 directly in its final crystal conformation whereas topochemical reaction via pathway Iwould lead to the unobserved conformer of product 35 (Figure 1-15). Further support forthis interpretation came from inspection of the local lattice environment of the startingmaterial, which shows that the free space surrounding the ester group at C12 is greaterthan the free space around C 1 1 .1.6 Research ObjectivesThe central theme of this thesis is photochemistry in the solid state. The mainemphasis will be on reactions that utilize crystal chirality for asymmetric induction. Asdiscussed previously, achiral molecules can, though rarely, crystallize in chiral spacegroups and undergo chemical processes that transfer the chirality of the crystals tomolecular chirality of the products. Introducing a resolved chiral handle onto the reactantensures chiral crystals, since chiral compounds must crystallize in chiral space groups. Achiral handle can be introduced conveniently onto a substrate by salt formation with anoptically active counterpart. When such a salt undergoes a solid state reaction at least onenew chiral center must be formed in the achiral part of the salt so that the enantiomericexcess in the photoproducts can be used to measure the extent of the asymmetricinduction by the crystalline medium.27R1Chapter 1 IntroductionR2^R336: COOH COOEt H37: COOH Me H38: COOMe COOMe COOH39: COOMe COOMe NH2Figure 1-16 Dibenzobarrelene Derivatives Selected for Studying AsymmetricInduction in Chiral Salt CrystalsDibenzobarrelene derivatives show the characteristics needed for studyingasymmetric induction in chiral salt crystals. First of all dibenzobarrelene derivativesbearing functional groups such as carboxylic acids or amines which are necessary to formsalts with chiral counterions can easily be synthesized. Four unsymmetrically substituteddibenzobarrelene derivatives were selected for this study: three carboxylic acids, 36, 37and 38, and one amino compound, 39 (Figure 1-16). Secondly, as mentioned earlier,photolysis of unsymmetrically substituted dibenzobarrelene derivatives in solution and inthe solid state yields chiral dibenzosemibullvalenes, where four new chiral centers aregenerated. Such dibenzobarrelene derivatives are excellent for studying both the regio-and the enantioselectivity of the di-it-methane rearrangement because thedibenzobarrelene skeleton has four different 1,4-diene systems (Figure 1-17). Each ofthese four systems is associated with a pathway that leads to a different product. Asshown in Figure 1-17 pathways I and Mead to regioisomer A while pathways III and IVgive regioisomer B. Furthermore pathways I and II lead to different enantiomers ofisomer A. Similarly, pathways III and IV yield the opposite enantiomers of isomer B. Itshould be mentioned that enantioselectivity is only expected under conditions where aresolved dissymmetric influence affects the reaction.28Chapter 1 IntroductionThe approach taken in this work to investigate unsymmetrically substituteddibenzobarrelene derivatives allows the regioselectivity of the di-n-methane rearrange-ment to be studied. The first part of this thesis will deal mainly with the regioselectivityexhibited by the starting materials prior to salt formation. The photochemistry of theesters of the carboxylic acids will also be investigated. This allows the regioselectivity ofthe acids to be compared with the regioselectivity of their corresponding esters givingbetter insight into the effect of hydrogen bonding on the regioselectivity. Any observeddifferences between solution and solid state reactivities will be investigated further by X-ray structure analysis of the reactants in an attempt to establish structure-reactivitycorrelations.The second part of this thesis will investigate photoreactions of salts of thedibenzobarrelene starting materials formed with optically active counterparts. The extentof asymmetric induction in the chiral crystalline phase will be studied by measuring theenantiomeric excess of the photoproducts. The chiral crystals make it possible to studythe effects of the crystal lattice on the four different pathways followed by the di-n-methane rearrangement of the dibenzobarrelene derivatives (Figure 1-17).Regioselectivity verifies which two pathways, (I+II) or (II+IV), are favored.Enantioselectivity can then indicate which of the two pathways are followed for a givenregioisomer, but to distinguish between these pathways the absolute configurations of thephotoproduct and the reactant must be known. The main goal of studying asymmetricinduction is to design a salt that produces only one regioisomer in high optical yield. Thiswould allow mapping of the reaction pathway and consequently X-ray structure analysisof the reactant has the potential to identify the crystal forces controlling regio- andenantioselectivity. Studying optically active salts of different dibenzobarrelenederivatives makes it possible to compare these crystal forces in different systems andsearch for a general explanation for the stereoselective pathways the di-it-methanerearrangement follows in these dibenzobarrelene systems29PalhvatylV________•.Chapter 1 IntroductionPathway_I—_____■APathway II APathway IIIBBFigure 1-17 The Four Di-m-methane Systems in the Dibenzobarrelene Skeleton30Chapter 2 Results and DiscussionA property of organic salt crystals is the strong lattice forces which bind them andlead to high melting points. This reduces the likelihood of crystal melting during reactionand may increase the probability of observing topotactic reaction. The final part of thisthesis will concentrate on efforts made to observe single crystal-to-single crystalreactions for dibenzobarrelene acid 40 (Figure 1-18).40Figure 1 -18 Dibenzobarrelene Derivative 4031Chapter 2 Results and DiscussionCHAPTER 2 RESULTS AND DISCUSSION2.1 Preparation of SubstratesDibenzobarrelene ester derivatives 39 to 46 were prepared by addition of anacetylenic ester derivative to the corresponding anthracene derivative according to themethod of Diels and Alders (Figure 2-19).R2 Compound R 1 R2 R339 NH2 COOCH3 COOCH341 H CH3 COOCH342 H H COOCH2CH343 COOCH3 COOCH3 COOCH344 COOCH2CH3 COOCH3 COOCH345 H COOCH3 COOCH346 CHO COOCH3 COOCH3Figure 2-19 Preparation of Dibenzobarrelene Derivatives via the Diels-Alder Reaction32Chapter 2 Results and DiscussionAll of the acetylenic ester derivatives and anthracene itself are commercially available. 9-Anthraldehyde is also commercially available, but the other 9-substituted anthraceneswere synthesized as follows: 9-amino anthracene was obtained by reduction of 9-nitroanthracene with stannous chloride.47 The methy148 and the ethyl49 esters of 9-anthracene carboxylic acid were made via the 9-anthracene acid chloride obtained byreaction of the 9-anthracene carboxylic acid with thionyl chloride.12-Methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid (37) wasobtained by hydrolysis of methyl 12-methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylate (41) with aqueous Na0H. 50 Similarly, hydrolysis of the Diels-Alder adduct42 with aqueous NaOH yielded 9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid(40, Figure 2-20).501.NaOH2.H+41: Ft, = CH3, R2 = CH342: RI = H, R2 = CH2CH337: RI = CH340: R i = HFigure 2-20 Preparations of Acids 37 and 40The sequence followed to prepare ester acid 36 is illustrated in Figure 2-21. Theinitial step is the Diels-Alder addition of dimethyl acetylenedicarboxylate to anthraceneto yield ester 45. Hydrolysis of ester 45 with aqueous NaOH gives the di-acid 47.50 Theacid anhydride 485° was formed by refluxing the di-acid 47 with oxalyl chloride. Finallyaddition of ethanol to anhydride 48 afforded the desired ethyl 9,10-dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic acid (36).33Chapter 2 Results and Discussion1.NaOH----■2. Hs'Oxalyl chloride 45^ 47EtOH-_,48^36Figure 2-21 Preparation of Ester-Acid 37Dimethyl^9-carboxy-9,10-dihydro-9,10-ethenoandwacene-11,12-dicarboxylateacid (38) was obtained in the following manner: aldehyde 46 was formed by Diels-Alderaddition of dimethyl acetylenedicarboxylate to 9-anthraldehyde. 51 Oxidation of aldehyde46 with NaC1O2 according to the method of Lindgren et a1.52 yielded ester-acid 38.NaC10246^ 38Figure 2-22 Preparation of Ester-Acid 3834Chapter 2 Results and DiscussionAll compounds prepared were fully characterized by spectroscopic and analyticalmethods and X-ray structure analyses in some cases. The spectra of previously reportedcompounds were identical to those reported in the literature. The syntheses andcharacterizations are described in greater detail in the Experimental Section (Chapter 3).2.2 Photochemical Studies of the Starting Materials Prior to Salt Formation2.2.1 Photolyses of Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic acid (36)As mentioned earlier, Scheffer et al. 42 studied the photochemistry of the di-ester31a, which is the methyl ester of ester-acid 36. Photolysis of ester 31a yieldedphotoproducts 32a and 33a, with no selectivity in solution or in the solid state (Figure 2-23). According to Zimmerman's mechanism, 53 initial vinyl benzo bridging is the product-determining step. The initial radical on the vinyl bond is formed next to the vinylsubstituent that can better stabilize it. The lack of preference for either product wasexplained in terms of both ester groups having the same radical stabilizing ability.32a^ 33a31aFigure 2-23 Photolysis of Di-Ester 31a35Chapter 2 Results and DiscussionThe authors characterized photoproducts 32a and 33a from 11-1-NMR analyses.One of the isomers has a methyl singlet at 6 = 3.87 ppm and a methylene quartet at 8 =4.18 ppm. The other has the methyl singlet at 8 = 3.73 ppm and the methylene quartet at8 = 4.35 ppm. The 8b benzylic position is more deshielded than the 8c position, thereforethe more deshielded methyl singlet was assigned to isomer 32a which has the methylester in the 8b position. The same holds true for the methylene signals of the ethyl estergroups. The more deshielded signal is assigned to regioisomer 33a where the ethyl esteris in the 8b position. These chemical shift differences seem to be characteristic ofregioisomers of these kinds and their use in assigning structure is also supported bycorrelations of 11.1-NMR spectral interpretation to X-ray crystal structures done byScheffer et al 4249a:R = CHCH349b:R = CH(CH3)236: R = CH2CH351: R = CH(CH3)2 50a:R = CH2CH350b:R CH(CH3)232a:R = CH2CH332b:R CH(CH3)233a:R = CH2CH333b:R = CH(CH3)2Figure 2-24 Photolysis and Work-up of Ester-Acid 36 and 5136Chapter 2 Results and DiscussionThe photochemistry of ester-acid 36 was studied in solution and in the solid state.The results are listed in Table 2-1. Photoproducts 49a and 50a (Figure 2-24) were notisolated but were converted into products 32a and 33a by treating the reaction mixturewith excess diazomethane. This allowed each photoproduct to be isolated by columnchromatography. Product ratios were determined by GC analysis of the reaction mixture.Table 2-1 illustrates that the photochemical behavior of ester-acid 36 is differentfrom that of di-ester 31a. The regioselectivity of the di-it-methane rearrangement ofester-acid 36 is strongly influenced by the photolysis medium. In the solid state isomer33a was the only product obtained, whereas solvents such as acetone and acetonitrilegive mixtures of photoproducts 32a and 33a. In contrast, photolyses in benzene andaqueous NaHCO3 solutions give quantitative yields of 32a.Table 2-1^Medium Dependent Photochemistry of Ester-Acids 36 and51Solvent Ester-acid 36 Ester-acid 51 54Concentration Ratio 32a:33aa Concentration Ratio 32b:33baAqueous NaHCO 3 0.01b 100:0 0.01b 90:10MeCN 0.01b 57:43 0.01b 50:50Benzene 0.05 100:0 0.06 60:40n 0.001 100:0 0.01 72:28ti 0.0005 100:0 0.001 83:17Crystals 0:100 5:95Acetone 0.01 60:40a The estimated error in the GC analysis is ±5%. b The product ratio is unaffected when theconcentration is changed from 0.1 to 0.001M. c Conversion was kept below 10%.37Chapter 2 Results and DiscussionThese photochemical results for 36 are similar to those observed when Garcia-Garibay et al .54 studied the photochemical behavior of ester-acid 51 The regioselectivityfor ester-acid 51 in different media is listed in Table 2-1 as well. The main differencebetween the reactivities of 36 and 51 is in benzene solution. The regioselectivity forester-acid 51 in benzene is concentration dependent, whereas no such effect was observedfor 36.Garcia-Garibay et al.54 suggested that the regioselectivity is controlled bydifferent hydrogen bonded forms of ester-acid 51 in different media. They identified andstudied three hydrogen bonded forms, 51A, 51B, 51C of ester-acid 51. 54> Species 51A isa monomer which is hydrogen bonded to a solvent molecule; 51B is an intramolecularhydrogen bonded form of ester-acid 51; and 51C is an intermolecular hydrogen bondeddimer (Figure 2-25). The authors showed that structure 51C is predominant inconcentrated solutions in non-polar solvents and in the solid state, whereas structure 51Ais dominant in polar solvents and 51B in non-polar solvents at low concentrations.Solvent/in/OA^ B36: RI C342CH351: Ft, 3.1H(CH)Figure 2-25 Different Hydrogen Bonded Forms of Ester-Acids 36 and 5138Chapter 2 Results and DiscussionThe predominance of form 51A in polar solvents such as acetone and acetonitrileexplains the lack of regioselectivity in these media. Neither product 32b or 33b isfavored since the solvent bonded acid group has a similar radical stabilizing ability as theester group. In aqueous NaHCO3 solution, however, ester-acid 51 exists as an anion andGarcia-Garibay et al . 54 hypothesized that a radical center adjacent to an ester substituentis better stabilized than one next to a carboxylate anion, since only product 32b isobserved.X-ray structural analysis of ester-acid 51 showed that 51C is the only speciespresent in the crystal. The authors rationalized that the dimeric hydrogen bonding in thecarboxylic acid group keeps it tightly "frozen" in the crystal lattice and hinders themotions necessary to form the initial vinyl benzo bridging next to the acid group. On theother hand, formation of photoproduct 33b leaves the hydrogen bond relativelyundisturbed and is thus favored.Garcia-Garibay et al.54 showed that structure 51B dominates in benzene solutionsat low concentrations where product 32b is favored. The authors proposed an excitedstate proton transfer interaction for species 51B in which a proton is transferred from oneoxygen to the other to form species 51D (Figure 2-26). Cristol et al. 55 have shown thatdibenzobarrelene with carbocation in the 11 position represents the condition necessaryfor a positive charge initiated 1,2-aryl shift which would favor regioisomer 32b. Withincreased concentration, species 51C becomes more ubiquitous and hence formation ofproduct 33b. The authors concluded that the product ratio seemed to be determined bythe relative amounts of 51B and 51C at a given concentration.The explanation used to interpret the regioselectivity of the di-it-methanephotorearrangement of ester-acid 51 can also be applied to ester-acid 36. Infraredspectroscopy can be utilized to explore the different hydrogen bonded forms of ester-acid36 in different reaction media.39Chapter 2 Results and DiscussionB^D4^36: R = CH2CH3^ 1no product observed^50: R = CH(CH3)2 productFigure 2-26 1,2 Aryl Shift of Ester-Acids 36 and 51The X-ray crystal structure of ester-acid 36 has not been obtained, but the solidstate infrared spectrum suggests a dimeric carboxylic acid structure. The 0-H stretchingabsorption was obtained as a very broad and relatively structureless band in the region3500-2400 cm-1 which is characteristic for carboxylic acid dimers. 56 The carbonyl bondsare observed at 1734 and 1678 cm -1 . The latter can be assigned to the acid carbonylwhich is shifted to low frequency owing to the intermolecular hydrogen bonding (Figure2-27). This is in good agreement with the solid state infrared spectrum for ester-acid 51(3400-2200, 1724, 1680 cm-1).In acetonitrile, only one broad absorption band at 1714 cm-1 is observed for theacid and the ester carbonyls. The 0-H absorption in the region 3600-3500 cm-1 indicatesRC(0)0-H•••solvent interaction (Figure 2-27). This is very similar to the infraredspectrum of ester-acid 51 in acetonitrile (3632, 3544, 1727 cm- 1). Hence a solventbonded structure analogous to 51A can be proposed for ester-acid 36 in polar solvent.40Chapter 2 Results and DiscussionSpectrum recorded in C6H6Spectrum recorded in CH3CNSpectrum recorded in KBr3200^1600^4001/cmFigure 2-27 Medium Dependent Infrared Spectra of Ester-Acid 3641R C=0 (cm- 1 )Me 1727Et 1720CH(CH2)3 1708 0 4.7Chapter 2 Results and DiscussionThe infrared spectra of ester-acid 36 in C 6H6 did not change when theconcentration was increased from 0.005 M to 0.05 M. These spectra show an 0-H bandat low frequency (2800-2600 cm- 1 ) which corresponds to intramolecular 0-H bonding.The carboxylic acid band is observed at 1731 cm-1. A relatively weak carbonyl band isobserved at 1673 cm- 1 which is appropriate for an intramolecular hydrogen bonded ester.This infrared spectrum is very similar to the low concentration spectrum of ester-acid 51in benzene (2780-2749, 1730, 1660 cm-1).The carbonyl stretch of the ester group in ester-acid 36 is observed at higherfrequencies in all spectra compared to ester-acid 51. This difference can be attributed tothe isopropyl substituent being capable of decreasing the stretching frequency of thecarbonyl ester group relative to an ethyl substituent. Such a substituent effect is normallyattributed to the inductive stabilization of the polar resonance form of the carboxylategroup. This is illustrated very well in the frequency of the carbonyl resonance in theinfrared spectra of a variety of alkyl benzoates 57 (Figure 2-28).Figure 2-28 Inductive Effect of the Alkyl Substituent on the Infrared CarbonylStretching Frequency of Alkyl BenzoatesThe infrared spectrum of ester-acid 36 in benzene suggests that the intramolecularhydrogen bonded structure 36B is predominant at all concentrations in benzene solution.This is consistent with the results of the photochemical reaction of ester-acid 36 inbenzene since only product 32a is formed. For unknown reasons the equilibrium between42Chapter 2 Results and Discussion36C and 36B is not shifted towards 36C with increased concentration as is observed forester acid 51, at least over the concentration range tested.2.2.2 Photolyses of 12-Methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylate(37) and Methyl 12-Methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid(41)Crystals of ester 41 exist in two different crystalline morphologies. Crystalsgrown from ethanol solutions melt at 137°C with immediate resolidification to formcrystals which melt at 146°C. 50 Infrared spectra of the high and low melting crystals aredifferent, which further supports that the idea these crystals are dimorphs. hv1 37: R = H41: R = CH o •'52: R = CI%53: R = HNot observedFigure 2-29 Photolysis of Compounds 37 and 4143Chapter 2 Results and DiscussionThe photochemical behavior of both dimorphs is similar, and the only photoproductformed is compound 52. Direct and acetone-sensitized photolyses of ester 41 in solutionsyielded only the photoproduct 52.The di-it-methane rearrangement of ester 41 can give two isomericphotoproducts, and determining which one of the two isomers is formed (Figure 2.29)can be achieved by analyzing the 11-1-NMR spectra of compound 52 in a similar way asfor products 32a and 33a. The 11-1-NMR spectrum of the observed product can becompared to the spectra of the known analogs 54 and 55 (Figure 2-30).58 Compound 54has the ester group in the more deshielded 8b position while compound 55 has the estergroup in the 8c position. The methyl singlet for ester 54 is at 8 = 3.77 ppm while themethyl singlet for ester 55 is at 8 = 3.72 ppm. Photoproduct 52 has the methyl estersignal at 8 = 3.72 ppm in the 11.1-NMR which suggests that the ester is in the 8c position.The structure of photoproduct 52 was confirmed with an X-ray diffraction analysis of acrystalline derivative of photoproduct 52. The X-ray crystal structure of photoproduct 52was obtained after a chiral handle was introduced into the molecule. This is discussed indetail in Section 2.4.2.54^55Figure 2-30 Structure of Compounds 54 and 55The photochemistry of acid 37 was studied in solution as well as in the solid state.The reaction mixtures were treated with excess diazomethane and analyzed by GC.Compound 52 was the only product formed under the various conditions.44Chapter 2 Results and DiscussionThe photochemistry of compounds 37 and 41 is similar in that the rearrangementis not affected by the reaction medium and only one di-n-methane photoproduct with themethyl substituent in the 8b position is observed. The regioselectivity of the di-n-methane rearrangement of dibenzobarrelenes with substituents at the bridgeheadpositions has been investigated thoroughly, but little is known concerning the effect ofsubstituents located on the vinyl bond. 59 Zimmerman53 has suggested that the radicaltermini of the cyclopropyldicarbinyl biradical become electron rich during the di-n-methane rearrangement. Consequently, the regioselectivity of the di-n-methanerearrangement is controlled by the polar nature as well as the radical-stabilizing ability ofthe substituents. In order to explain the regioselectivity for compounds 37 and 41 theradical stabilization ability and the polar effects of the substituents must be compared.Bordwell et a1.60 have developed a semiempirical method for estimating relativeradical stabilization energies. They compared the radical stabilizing ability of variousgroups in the 9-substituted fluorenyl radical with the substituted methyl radical (Figure2-31). In the methyl radical a carboxylic acid group is a slightly better stabilizer than amethyl group. In contrast, a methyl group stabilizes the 9-fluorenyl radical slightly betterthan an ester group. It was suggested that intramolecular steric inhibition of resonancebetween the ester group and the fluorenyl moiety decreased the radical stabilizing affectof the ester group compared to the methyl group. However, the authors came to theconclusion that methyl, acid and ester groups have very similar radical stabilizing effects.Clearly the relative radical stabilizing effects of the substituents in compounds 37 and 41will not explain the regioselectivity.45Chapter 2 Results and DiscussionR—CH2RR 9-Substituted Fluorenyl R Substituted Methyl RadicalCH3 4.5 kcal/mol CH3 3.3 kcal/molCOOCH3 3.9 kcal/mol COOH 5.7 kcal/molFigure 2-31 Comparison of Radical Stabilization Energies of 9-Substituted Fluorenyland Substituted Methyl RadicalsEven though a methyl group has a similar radical stabilizing effect as an estergroup, it is electron donating whereas the ester group is electron withdrawing. 61 If, asZimmerman suggests, 53 the radical center is electron rich, we may suggest thatphotoproduct 52 is formed exclusively owing to the preference of the radical to beformed next to the electron withdrawing ester group rather than the electron donatingmethyl substituent. Formation of photoproduct 53 (Figure 2.29) can be explained in asimilar way.It is interesting to apply the hypothesis used above to explain the regioselectivityof the di-E-methane photorearrangement of the monosubstituted dibenzobarrelenes 56and 57 (Figure 2-32). Ciganek58 demonstrated that dibenzobarrelene 56 gave onlyproduct 55 when irradiated. The initial radical is formed next to the ester group which isboth a better radical stabilizer and a better electron withdrawing group than a hydrogen.Cristol et al.62 studied the photochemistry of compound 57 and found that regioisomers58 and 59 were formed in the ratio 6:4. The radical stabilization ability of the methylgroup favors formation of product 58 whereas the electron donating effect favorsformation of product 59 and hence a mixture of both products is formed.46Chapter 2 Results and DiscussionI 56: R = COOCH357: R = CH3 Jr55: R = COOCH3^ so58: R=0-13Figure 2-32 Photolysis of Monosubstitued Dibenzobarrelenes 56 and 572.23 Photolyses of Dimethyl 9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)Earlier studies by Paddick et al. 63 on the photochemistry of amine 39 showed thatit rearranges to the keto-diester 62 in acetone. This result was interpreted as being due tothe di-n-methane rearrangement to the unstable amino-diester 61 followed by ringopening and hydrolysis under the photolysis conditions.The solution photochemistry of amine 39 was repeated in acetone. This affordedamino-diester 61 as the major product, which was sufficiently stable to be isolated,although it did undergo conversion to keto-diester 62 upon attempted silica gel columnchromatography. A significant amount of the alternative di-m-methane regioisomer 6047Chapter 2 Results and Discussionwas also isolated from the photolysis, although no mention of this compound was madein the original report by Paddick et al.63 Unlike 61, compound 60 could be purifiedwithout difficulty by column chromatography. GC analysis showed that compounds 60and 61 were formed in the ratio 3:7. Photolyses of crystals of amine 39 yieldedcompounds 60 and 61 in the ratio 14:86. No rearrangement product 62 could be detected.hv4.--Path Aby--.Path B63 64I^4so^ 61Rearrangement^Iand Hydrolysis62Figure 2-33 Photolysis of Amine 39The structures of compounds 60 and 61 are based on their spectra. The 11-I-NMRspectra are particularly informative. In the spectrum assigned to product 60, a one proton48Chapter 2 Results and Discussionsinglet at 8 = 4.40 ppm is attributed to the cyclopropyl methine. In contrast, the spectrumof 61 contains a one proton singlet at 8 = 5.45 ppm, which can be assigned to thehydrogen at the doubly benzylic position. These chemical shift differences appear to becharacteristic of regioisomers of this type, and their use in assigning structure restsultimately on correlating 111-NMR spectral interpretations to X-ray crystal structures. 42The spectroscopic data of 62 matched those reported by Richards et al."Paddick et al. 63 and Iwamura et al. 65 investigated the regioselectivity of the di-it-methane rearrangement of bridgehead-substituted dibenzobarrelene derivatives with estergroups on the vinyl bond. These authors suggested that the regioselectivity is controlledby the stabilizing effect of the bridgehead substituents on intermediates 63 and 64(Figure 2-33), in a similar way as the substituents R affect the equilibrium betweennorcaradiene and cycloheptatriene 66 (Figure 2-34). Electron-accepting groups strengthenthe opposite bond in the cyclopropyl ring and the equilibrium is shifted towardsnorcaradiene, whereas electron-donating groups weaken the same bond and favor thecycloheptatriene.Norcaradiene^ CycloheptatrieneFigure 2-34 Equilibrium between Norcaradiene and CycloheptatrieneIwamura65 did find experimentally that electron-donating and electronegativebridgehead substituents such as OCH3 favor path A whereas electron-accepting groupssuch as Ph favor path B (Figure 2-33). Opposing electron-accepting and electronegativeeffects of the same group can operate simultaneously to determine which regioisomer is49Chapter 2 Results and Discussionformed. For example the electron accepting nature of the NO2 group favors path B but itshigh electronegativity destabilizes biradical 64. The latter effect dominates and only pathA is observed.Solution photolyses of amine 39 favored product 61 or path B, in spite of theelectronegative and electron donating effect of the amine group. Paddick et al. 63explained the observed regioselectivity by proposing that there is intramolecularhydrogen bonding between the amino group and the nearest ester group. The authorssuggested that this intramolecular hydrogen bond diminishes the stabilizing effect of theester group on the adjacent radical. In other words, the intramolecular hydrogen bond issupposed to make intermediate 63 less stable. Furthermore they suggested that thedestabilizing effect of the intramolecular hydrogen bonding in intermediate 63 dominatesthe stabilizing effect of the electron donating amine group, and as a result pathway B isfavored (Figure 2-33).The regioselectivity observed for amine 39 in the solid state is similar to that insolution, product 61 is the major one. The X-ray crystal molecular structure"' of amine39 and the crystal packing diagram are shown in Figure 2-35. There are two independentmolecules in the asymmetric unit, 39A and 39B. The crystal structure reveals that there isintramolecular hydrogen bonding between the NH2 group and the nearest carbonyloxygen (02 in molecule 39A and 04 in molecule 39B). There is one intermolecularhydrogen bond between the amino group in 39B and the other carbonyl oxygen (04) inmolecule 39A and another one between the NH2 group in 39A and the carbonyl oxygen02 in 39B. These intermolecular hydrogen bonds make an infinite chain betweenmolecules 39A and 39B in the crystal lattice. Furthermore the carbonyl oxygen, 02, in39A is intermolecularly hydrogen bonded to the NH group in an another molecule of39A. The hydrogen bonds destabilize the radical stabilizing effect of both ester groups inmolecules 39A and 39B and also hinder their movement. However, the hydrogenbonding affects the formation of photoproducts 60 and 61 similarly.50Chapter 2 Results and DiscussionMolecule 39AMolecule 39BFigure 2-35 Crystal Structure of Amine 3951Chapter 2 Results and DiscussionFigure 2-35 Crystal Structure of Amine 39 (continued)As discussed earlier in this thesis, the steric effects of the crystal lattice can be akey factor in controlling the regioselectivity of the di-x-methane rearrangement. Schefferet al 42 suggested that the ester group attached to the bridging vinyl atom must moveconsiderably during the benzo-vinyl bridging process and is therefore most likely toexperience unfavorable steric interactions. Both ester groups in molecules 39A and 39Bare tightly packed which affects the formation of both the di-n-methane rearrangementproducts similarly.The crystal lattice enforces varying degrees of conjugation between the differentester groups and the double bonds between them. Initial benzo-vinyl bridging at the vinyl52Chapter 2 Results and Discussioncarbon atom that is less conjugated to the attached ester group is favored since bonding atthis side leads to the more highly resonance-stabilized biradical intermediate. Analyzingthe intramolecular features of amine 39 illustrates that the torsion angles between thecarbonyl next to the bridgehead substituent and the vinyl bond are -101° in 39A and -87°in 39B. This illustrates that these carbonyls (02 in 39A and 04 in 39B) are largely un-conjugated. In contrast, the carbonyls farther away from the bridgehead substituent are inconjugation; their torsion angles (04 in 39A and 02 in 39B) with the vinyl bonds are158° for 39A and 164° for 39B. Formation of 61 is therefore favored, which is consistentwith the observed regioselectivity.Examples exist in the literature where the degree of conjugation of the vinylbonds with a radical stabilizing group is thought to affect the di-x-methanerearrangement. Such an example comes from Schaffner and co-workers" who studiedthe photochemistry of ketone 65. The infrared spectrum of ketone 65 in glassy solventmatrices showed two carbonyl bands at 1633 and 1644 cm- 1 . These bands were assignedto two rotamers 65A and 65B, one with the carbonyl in conjugation with the vinyl bond,65A (1633 cm-1), and the other with the benzoyl group out of conjugation, 65B (1644cm-1 , Figure 2-36). Irradiation led to disappearance of the band assigned to theconjugated benzoyl group at 1633 cm- 1 and formation of a carbonyl band whichcorresponds to the photoproduct 66. The carbonyl band assigned to unconjugated rotamer65B at 1644 cm- 1 was unaffected. The author suggested that rotamer 65B did not reactdue to lack of radical stabilization by the benzoyl group.53Chapter 2 Results and DiscussionPh 4- - - -- - - ■IR: 1633 cm-1; conjugated C=065AIR: 1644 cm-1 ; unconjugated C=0658Ph66Figure 2-36 Photolysis of Ketone 652.2.4 Photolyses of Trimethyl 9,10-Dihydro-9,10-ethenoanthracene-9,11,12-tricarboxylate (43) and Dimethyl 9-Ethoxycarbony1-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (44)Acetone-sensitized photolyses of tri-ester 43 gave product 67 as the majorproduct and small amounts of the other di-x-methane rearrangement product 68.Similarly, acetone solution photolyses of tri-ester 44 yielded the major product 69 andthe minor product 70 (Table 2-2, Figure 2-37).5443: R . CH 344: R . CH2CH3Chapter 2 Results and DiscussionODOR67: R = CH369: R . CH 2CH368: R = CH,70: R = CH 2CH 371: R = CH,^ 7372: R = CH2CH 3E . COOCH3Figure 2-37 Photolysis of Tri-Esters 43 and 44Irradiation of tri-ester 43 in chloroform (direct irradiation) gave an additionalcyclooctatetraene product 71 as well as products 67 and 68. Direct photolyses of tri-ester44 in chloroform gave two new compounds, 72 and 73, in addition to products 69 and 70(Table 2-2, Figure 2-37).Crystals of tri-ester 44 exist in two dimorphic forms which have similar meltingpoints, 149°C, but their infrared spectra demonstrate that they are different. Irradiation ofboth crystal forms yielded significant amounts of all of the four photoproducts formed inchloroform solution. Photolyses of crystals of tri-ester 43 also yielded significant55Chapter 2 Results and Discussionamounts of all the chloroform solution products (Table 2-2, Figure 2-37). The crystals oftri-esters 43 and 44 are photochromic; irradiation results in colors which fade in the dark.The photochromism of these compounds is discussed in Section 2.5.3.Table 2-2^Medium Dependent Photochemistry of Tri-Esters 43 and 44Medium Tri-ester 43a Tri-ester 44a(67+68):71b 67:68c (69+70):(72+73)b 69:70c 72:73dAcetone 100:0 91:9 100:0 93:7Chloroform 66:34 90:10 44:54 80:20 58:42Crystals (prisms) 74:26 49:51 61:39 48:52 22:77Crystals (needles) 68:32 38:62 33:66a The estimated error in the GC analysis is ±5%. b Ratio between S 1 and T1 products. c Regioselectivityfor the di-n-methane reactions. d Ratio for the S1 products.The structures of the di-it-methane products were determined from theirspectroscopic data, and as before, the 11-1-NMR spectra were most useful.42 The methineprotons on the cyclopropyl ring appear at 8 = 4.5 ppm in 67 and 69. The more deshieldedprotons in the 4b-position in products 68 and 70 give rise to signals at 8 = 5.1 ppm. Thestructural assignment of cyclooctatetraene product 71 is based on its spectroscopic data.For cyclooctatetraenes 72 and 73 the spectroscopic data are strikingly similar, so it isimpossible to assign the structures based on spectroscopic data alone. Crystal structureanalyses of these compounds allowed their structures to be assigned unambiguously. 67, 69Di-n-methane rearrangements of dibenzobarrelene derivatives are thought to takeplace from the triplet excited state, presumably due to rapid intersystem crossing of theinitially formed singlet excited state.39 The literature cites examples of dibenzobarrelenederivatives which react from the S 1 excited state to form dibenzocyclooctatetraene. 39 The56Chapter 2 Results and Discussionsensitization studies of tri-esters 43 and 44 demonstrate that the dibenzosemibullvaleneproducts form from the triplet excited state, whereas the dibenzocyclooctatetraenes comefrom the singlet excited state.In solution the di-n-methane rearrangement of tri-esters 43 and 44 favoredformation of the 4b-substituted products, 67 and 69 respectively. As described earlier,electron-accepting groups on the bridgehead favor formation of the 8b-substitutedsemibullvalene whereas electronegative and electron-donating groups stabilize thebiradical leading to formation of the 4b-substituted semibullvalene. The electronegativeeffect of the ester group must overwhelm the electron-accepting effect resulting information of the 4b-substituted products.The regioselectivity observed for tri-esters 43 and 44 in solution is similar to thatobserved by Ciganek58 for ester 74 (Figure 2-38). Irradiation of ester 74 yieldedphotoproducts 75 and 76 in the ratio 66:33, respectively. Paquette et al." suggested thatthe photobehavior of ester 74 can be interpreted in terms of reluctance to position theelectronegative ester group at the cyclopropane in the biradical intermediate that leads toproduct 76 (Figure 2-38).57Chapter 2 Results and Discussionby hv 7475^76Figure 2-38 Di-m-methane Rearrangement of Ester 74The di-n-methane rearrangement of tri-esters 43 and 44 in the solid state is notregioselective; similar amounts of both the 4b- and the 8b-substituted semibullvalenes areformed (Table 2-2). This differs from the regioselectivity in solution where the 4b-substituted products are favored. The reactivity must therefore be modified to somedegree by the crystal lattice.An X-ray structure67 was obtained of the tri-ester 43 (Figure 2-39) and analysesof the packing arrangement indicate that there is some steric crowding around both vinylester groups, which presumably affects the formation of both di-n-methane products insimilar ways. When the molecular conformation is analyzed, however, it shows that theester group next to the bridgehead substituent is out of conjugation with the vinyl bond,whereas the other ester group is conjugated. The torsion angle between the carbonyl estergroup (02) next to the bridgehead substituent and the vinyl bond is 102°, whereas thetorsion angle for the other carbonyl ester group (06) and the vinyl band is 160°. Asmentioned earlier, the initial biradical is better stabilized next to the ester group that is in58Chapter 2 Results and Discussionconjugation, and thus based on this assumption, formation of photoproduct 68 is favored.We suggest that since the electronic stabilizing effect of the bridgehead substituent andthe radical stabilization capability of the different ester groups on the vinyl bond work inopposing directions in the solid state and neither one is dominant, no regioselectivity isobserved.Figure 2-39 Crystal Structure of Tri-Ester 43It is a general trend for dibenzobarrelenes with ester groups on the vinyl bond andone bridgehead substituent, that the ester group on the vinyl bond next to the bridgehead59Chapter 2 Results and Discussionsubstituent is out of conjugation?' Presumably this occurs to relieve steric interactionsbetween the ester group and the bridgehead substituent. It can be suggested that the samefactors control the solid state regioselectivity for the di-x-methane rearrangement of tri-ester 44 as tri-ester 43. There are only slight differences between the regioselectivities ofthe di-n-methane rearrangement of the two dimorphs of tri-ester 43, which suggests theyhave similar packing arrangements.Before discussing the singlet state photoreactivity of tri-esters 43 and 44, a briefreview of the literature is necessary. Photolysis of dibenzobarrelene via its singlet excitedstate yields dibenzocyclooctatetraene as the major product." The commonly acceptedmechanism involves [2+2] cycloaddition which is followed by thermal reorganization ofthe resulting cage compound (Figure 2-40). Deuterium labeling experiments were carriedout by Zimmerman to support this mechanism." Monobenzobarrelenes with deuteriumon the bridgeheads gave cyclooctatetraene with C2 symmetry (Figure 2-40).Figure 2-40 Photorearrangement of Labeled Benzobarrelene via the S iRecently, Scheffer et al. 8m, 74 and George et al." have demonstrated that tetra- andtri-substituted dibenzobarrelenes form abnormal dibenzocyclooctatetraenes possessing C2symmetry rather than C. symmetry. Scheffer et al.74 suggested a mechanism for theformation of the abnormal cyclooctatetraene after studying the photochemistry ofdibenzobarrelene derivative 77 (Figure 2-41). Acetone-sensitized irradiation of 77yielded di-rc-methane product 78, whereas direct photolysis give 78 andcyclooctatetraene 79. Solid state photolyses resulted in formation of compound 80 as the60 hv via S 1Chapter 2 Results and Discussionmajor product with small amounts of products 78 and 79. The authors suggested thatcompounds 79 and 80 are products of the singlet excited state and that the initial step isthe so called tri-n-methane76 interaction of both aromatic rings with the vinyl bond toform biradical 81. Migration of both the carbomethoxy groups in biradical 81 leads toproduct 80 whereas Grob fragmentation results in cyclooctatetraene 79.77: R COOCH 382: R = COPhhv via T 1$17879Figure 2-41 Photolysis of Compound 77There are only a few instances of abnormal cyclooctatetraene formation in theliterature.Sm Photolysis of dibenzobarrelene derivative 8275 yields photoproducts61Chapter 2 Results and Discussionanalogous to compounds 79 and 80 (Figure 2-41). Dibenzobarrelene derivatives 83 and848m,74 also yielded abnormal cyclooctatetraenes (Figure 2-42). It is very difficult todistinguish whether a cyclooctatetraene is normal or abnormal based on spectroscopicdata alone, and it is possible that some abnormal cyclooctatetraenes have been incorrectlyassigned as normal cyclooctatetraenes in the literature. Some corrections have beenreportedy4, 75,77, 78hv via S 1_____..83: R=CH 384: R = PhFigure 2-42 Singlet State Photorearrangement of Dibenzobarrelene Derivatives 83 and84Tri-ester 44 is the first compound reported to give both normal and abnormalcyclooctatetraenes. The cyclooctatetraene produced from tri-ester 43, however, can beformed through the [2+2] derived mechanism, the Grob fragmentation or a combinationof both, because these mechanisms lead to the same product. This leads to the question ofwhat factors control the competition between formation of the normal and abnormalcyclooctatetraenes. Substitution at position 9, 10, 11 and 12 on the dibenzobarrelenenucleus is a significant factor, since these are the only compounds that have yieldedcyclooctatetraenes with abnormal symmetry. Scheffer et al . 8m. 74 suggested that thereluctance of dibenzobarrelene derivatives 77, 83 and 84 to undergo [2+2]photocycloaddition to produce normal cyclooctatetraenes may be attributed to steric62Chapter 2 Results and Discussionfactors resulting from the presence of the bridgehead substituents. In other words, thereare more unfavorable intramolecular steric interactions developed in the [2+2] pathwaythan in the tri-rr-methane process. The bridgehead substituent(s) also give additionalstability to the intermediate 1,4-biradical (e.g., 81), which may be the driving force forthe alternative rearrangement observed.Interestingly, the solution photolyses of tri-ester 44 give similar amounts of bothcyclooctatetraenes 72 and 73, whereas the abnormal cyclooctatetraene 73 is slightlyfavored in the solid state for both the dimorphs of tri-ester 44 (Table 2-2, page 56).Further studies of this type should provide the possibility of studying stericeffects as well as the electron withdrawing and donating effects of different bridgeheadsubstituents on the formation of cyclooctatetraenes. This would give valuableinformation about the reaction mechanism for forming abnormal and normalcyclooctatetraenes.2.2.5 Photolyses of Dimethyl 9-Carboxy-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (38)Acid 38 forms different crystals depending on the solvent used for crystallization.Crystals grown from ethanol and acetonitrile contain one equivalent of the solvent in thecrystal lattice, whereas crystallization from ethyl acetate or sec-butanol solution yieldscrystals without solvent molecules. All three crystal forms were found to bephotochromic when irradiated in the solid state. The photochromism is discussed inSection 2.5. After photolysis the reaction mixture was treated with diazomethane, whichtransformed the di-it-methane products to methyl esters 67 and 68. Unreacted acid 38was transformed into tri-ester 43 which underwent further reaction with diazomethane toform a pyrazoline derivative. This simplified the chromatographic separation of the63Chapter 2 Results and Discussionstarting material from products 67 and 68 (Figure 2-43). The disadvantage of this methodis that the dibenzocyclooctatetraene derivatives also undergo pyrazoline formation withdiazomethane and cannot be separated from the derivatized starting material. As the di-rt-methane rearrangement is the main focus of this thesis, it was decided to study only thetriplet reactivity of acid 38, even though it undergoes some singlet state reactivity aswell. The regioselectivity of the di-rr-methane rearrangement of acid 38 in the solid stateand in acetone is listed in Table 2-3.1. hv______►2. CH2N238^ 67^ 68Figure 2-43 The Di-lc-methane Rearrangement of Acid 38The regioselectivity of the di-ic-methane rearrangement of acid 38 in acetonesolution is the same as for tri-esters 43 and 44, presumably because the ester and acidgroups have similar electronegativities.Table 2-3^Regioselectivity of the Di-rr-methane Rearrangement of Acid 38Medium Regioselectivity of the di-x-methane 67:68Acetone 100:0Crystals without solvent molecules No reactionCrystals containing ethanol 40:60Crystals containing acetonitrile 10:9064Chapter 2 Results and DiscussionIrradiations of crystals containing ethanol formed products 67:68 in the ratio40:60, which is similar to that observed for solid state reactions of tri-esters 43 and 44.The crystals grown from acetonitrile gave much better regioselectivity for the di-n-methane rearrangement, specifically compounds 67 and 68 in the ratio 1:9 respectively.Crystals of acid 38 without any solvent molecules were found to be photostable. Thecrystal structure has been obtained only of crystals that contain ethanol in the crystallattice.79The crystal structure of acid 38 confirms that crystals grown from an ethanolsolution contain one equivalent of solvent (Figure 2-44). The methyl group in the ethanolwas disordered with two equally occupied positions. The ethanol serves as a bridgebetween two molecules of 38 via hydrogen bonding, forming an infinite chain. Thehydrogen bonds are between the ester carbonyl 02 and the OH group of the ethanol andbetween the oxygen of the ethanol and the OH group of the carboxylic acid.Analyzing the intermolecular features of the X-ray structure reveals that bothester groups are sterically crowded which affects the formation of products 67 and 68similarly. The intermolecular hydrogen bonding of carbonyl 02 does not favor formationof either product 67 or 68. The hydrogen bonding of carbonyl oxygen 02 decreases itsability to stabilize an adjacent radical but this does not favor formation of product 68because formation of product 68 also requires movement of this carbonyl oxygen (02)which would interrupt the hydrogen bonding.Inspection of the intramolecular features of acid 38 shows that, as before, thecarbonyl group next to the bridgehead substituent is out of conjugation, whereas the otherester group is conjugated to the vinyl bond and therefore better capable of stabilizing anadjacent radical. This hypothesis would favor formation of product 68.The low regioselectivity of the di-m-methane photorearrangement of the ethanol-containing crystals of acid 38 can be attributed to the fact that the intramolecular65Chapter 2 Results and Discussionarrangement of acid 38 and the electronic effect of the carboxylic acid on the bridgeheadwork in opposing directions and neither factor dominates. This is similar to what wasobserved for tri-esters 43 and 44 in the solid state.Figure 2-44 Crystal Structure of the Ethanol Solvate of Acid 3866Chapter 2 Results and DiscussionFigure 2-44 Crystal Structure of the Ethanol Solvate of Acid 38 (Continued)X-ray structure analyses of the two other crystal forms of acid 38 are notavailable, and the solid state infrared spectra are too complicated to interpret thestructural arrangement of acid 38 in these crystals The crystals without solventmolecules became dark green when they were photolyzed and the color faded uponstoring in the dark. This confirms that photons were absorbed but no products were67Chapter 2 Results and Discussionformed. It can be suggested that in these crystals the carboxylic acid groups form dimersresulting in very crowded crystal packing and therefore no reaction is observed. Thereare examples in the literature of other benzobarrelene carboxylic acid derivatives that arenot photoreactive in the solid state although they are reactive in solutions.22Any speculation on the high regioselectivity of the acetonitrile-containing crystalsof acid 38 must await an X-ray crystal structure analysis.2.3 Photochemistry of Salts of Starting Materials23.1 Photolyses of Salts of Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic acid (36)Initially, salts of ester-acid 36 were formed with simple non-chiral bases to gaininsight into their characterization and photochemical behavior. The non-chiral saltsformed with ester-acid 36 are listed in Table 2-4.68Chapter 2 Results and DiscussionTable 2-4^Photoproduct Mixture Composition for Salts 85 to 89Counterion Salt Crystal forms Photoproduct ratio 32a:33aCH3OH Solid StateNa+ 85 Powder 100:0Lb 100:0"Ca+2 86 Powder 100:0" 100:0"Viv7N,IH87 Needles 100:0" 100:0"88 Plates 100:0b 100:0b4. )NIHC-.NHI89 Plates 100:0" 100:013a Photolyses carried out in water. b Compound 33a was not detectable within the limits of this method.The estimated error is ±5%.Salts 85 to 89 were formed by mixing equimolar quantities of acid and base inethanol or diethyl ether and filtering the resulting precipitate. The amine salts wererecrystallized from acetonitrile. These salts were shown to be simple 1:1 complexes (1:2for salt 86) by infrared and 1H-NMR spectroscopy, mass spectrometry and elementalanalysis. The infrared spectra of the salts were most informative. The carbonyl stretch ofthe acid group of ester-acid 36 at 1678 cm -1 was replaced by two bands for thecarboxylate anion COi ; a strong one in the 1650 - 1550 cm-1 region and a weaker onenear 1400 cm- 1 . For the salts formed with amine counterions, the OH band of thecarboxylic acid in the region 3500-2400 cm -1 is replaced with multiple combination69Chapter 2 Results and Discussionbands for NII; in the 3200-2200 cm-1 region. The 111-NMR spectra and elementalanalyses confirmed the ratio of the acid to the base.1 .tv2. H+3.CH2N2Salts of ester-acid 36^ 32a^ 33aFigure 2-45 Photolysis of Salts of Ester-Acid 36Salts 85 to 89 were photolyzed in solution and in the solid state. The reactionmixtures were acidified, treated with excess diazomethane to produce the correspondingmethyl/ethyl diesters, and analyzed by GC. In all instances the major product was 32a(Figure 2-45). The photoreactions follow the same mechanism as those discussedpreviously for ester-acid 36 in aqueous NaHCO3 solutions. The initial benzo-vinylbridging is favored at the carboxylate anion-bearing vinyl carbon, a result that can beattributed to preferential radical formation at the ester-bearing vinyl carbon.In this project, the main aim was to find salts of acid 36 in which the di-n-methane rearrangement is very regio- and enantioselective. These initial studies showedthat the salts of ester-acid 36 with passive counterions underwent the di-n-methanerearrangement with good regioselectivity, both in solution and in the solid state. Itremained to fmd optically active bases which form salts with ester-acid 36. These saltsmust be crystalline so that the chirality of the crystals can be transferred via the di-n-methane rearrangement into the photoproducts. The degree of asymmetric induction willthen depend on the molecular arrangement in the crystals. A decision was made to utilizenatural chiral amines and their simple derivatives as a chiral resource. No specialpreferences were made in selecting amines for these studies other than they had to be70Chapter 2 Results and Discussionreadily available. It did turn out that it was easier to form salts with secondary aminesthan with primary or tertiary amines. It can be speculated that secondary amines give thebest results since they are stronger bases than primary amines but not as stericallyhindered as tertiary amines.The initial studieso of chiral salts of acid 36 were done with the S-(-)-proline salt90a. This salt formed as a white powder, and attempted recrystallizations resulted indeposition of the parent acid. The spectroscopic data demonstrate that a 1:1 complex wasformed between the S-(-)-proline and ester-acid 36. Generally, simple carboxylic acidsare weaker acids than amino acids and therefore ester-acid 36 is not expected to becapable of protonating the proline.81 Due to poor crystal quality, it was not possible toobtain information about the structural arrangement of salt 90a by X-ray analysis.However in our laboratory, 82 X-ray structure analysis of a 1:1 complex of proline withacid 91 shows that proline is in its zwitterionic form and acid 91 is in its neutral form(Figure 2-46). The complex is held together by a hydrogen bond between the OH groupin acid 91 and the COi anion in proline. It can therefore be suggested that proline "salt"90a is a complex which is held together by hydrogen bonding between the neutral formof ester-acid 36 and the zwitterionic form of proline.0H HAcid 91^ RolmFigure 2-46 Hydrogen Bonded Complex of Proline and Acid 91 in the Solid State71Chapter 2 Results and DiscussionThe S-(-)-proline complex 90a was photolyzed in the solid state at roomtemperature. The reaction mixture was acidified, treated with excess diazomethane, andsubjected to silica gel column chromatography to isolate the photoproducts. Theregioisomeric and enantiomeric compositions of the products were determined by 400MHz 1H-NMR spectroscopy. For the enantiomeric excess determination use was made ofthe chiral shift reagent (+)-Eu(hfc)3. The signals monitored were the methyl singlets at 8= 3.9 ppm for 32a and 8 = 3.7 ppm for 33a. These studies revealed that products 32a and33a were formed in the ratio 83:17. Product 32a was formed in 43% enantiomeric excesswhereas product 33a was formed without enantioselectivity. The optical rotation of theproduct mixture was measured at the sodium D line and found to be positive.Determining enantiomeric excess with chiral shift reagents requires largechemical shift nonequivalence of the enantiomeric signals. When baseline separation ofthe enantiomeric signals under investigation is obtained, integrating the signals gives adirect measure of the enantiomeric composition. 83 Caution is required for enantiomericvalues above 90% where the error in the measurement has been reported to be of theorder of 10%. 83, 84 Figure 2-47 illustrates that the enantiomeric methyl ester signals ofboth compounds 32a and 33a are well resolved in CDC1 3 solutions.727.5 6.0 4.0'H-NMR Spectra of Photoproducts 32a and 33ain CDC13 after Addition of (+)-Eu(hfc)3 .Chapter 2 Results and Discussionppm'H-NMR Spectra of Photoproducts 32a and 33ain CDC13 .7.0^6.0^4.0^2.0ppmFigure 2-47 'H-NMR Spectra of a Mixture of Photoproducts 32a and 33a7310080604020Chapter 2 Results and DiscussionAfter demonstrating that photolyses of the complex of acid 36 with a chiral basedo indeed lead to optical activity in the product, the next step was to try to optimize theoptical yields. First of all, the dissymmetric influences of the proline moiety on thereaction pathways of S-(-)-proline salt 90a result in diastereomeric transition states forthe different enantiomers of products 32a and 33a. Lowering the temperature shouldtherefore increase the formation of the enantiomer which has lower activation energy.Secondly, although molecular motion is restricted in the solid state it is not completelyeliminated and molecules in a crystal do undergo thermal motions 8a There are examplesin the literature where the entire molecule undergoes a rotation in the crystal lattice.85Upon lowering the reaction temperature some of these movements are frozen out and thiscan increase the selectivity of solid state reactions.20^0^-25^-40Thmperature (oC)Figure 2-48 Product Ratio of Photolyses of S-(-)-Proline Salt 90a at DifferentTemperaturesFigures 2-48 and 2-49 illustrate that the di-x-methane rearrangement of S-(-)-proline complex 90a becomes more regio- and enantioselective at low temperatures. Lowtemperature photolyses have the disadvantage of taking a considerably longer time. As acompromise, no reaction was carried out at temperatures below -40°C. The74Chapter 2 Results and Discussionregioselectivity improved from 83:17 at room temperature to 94:6 at -40°C for products32a and 33a respectively. The enantiomeric excess increased from 43% to 76% forproduct 32a when the temperature was lowered to -40°C. No enantioselectivity wasobserved for product 33a in any instance. ■ (-)-Enantiomer 32a▪ (+)-Emantiomer 32a▪ Enantiomeric excess20^0^-25^-40Ibmperature (oC)Figure 2-49 Enantiomeric Excess of Photoproduct 32a at Different TemperaturesThe R-(+)-proline complex of ester-acid 36 (90b) was prepared as well as the-(±)-proline complex, 90c. Irradiations of the R-(+)-proline complex 90b at -40°C yieldedsimilar product ratios as the S-(-)-proline complex 90a, but the enantioselectivity wasreversed. In other words, product 32a was formed in 80% enantiomeric excess but theoptical rotation was negative. The fact that the sign of rotation of photoproduct 32a canbe reversed by using the optical antipode of the chiral induction agent indicates that thesystem is well behaved. Ester-acid 36 must have enantiomeric arrangements in the solidsof the R-(+) and S-(-)-proline complexes. Solid state photolysis of the racemic (±)-proline complex 90c at -25°C gave optically inactive product 32a and 33a in the ratio84:16.75Chapter 2 Results and DiscussionThe regioselectivity of the solid state di-it-methane rearrangement of prolinecomplexes 90a to 90c is the same as for the non-chiral salts 85 to 89. It can be suggestedthat there is intermolecular hydrogen bonding between the OH group in ester-acid 36 andthe proline moiety, similar to that observed for the proline complex of acid 91 (Figure 2-46). The carboxylic acid in ester-acid 36 would then have considerable anionic characterwhich would explain the observed regioselectivity.Photolyses of proline complexes 90a to 90c in ethanol yielded products 32a and33a in a 1:1 ratio without optical activity. This can be interpreted as being a result of thecomplex between acid 36 and proline dissociating in solution. Similar photochemicalbehavior was observed for the proline complexes 90a to 90c and ester-acid 36 in acetonesolutions.Salts 92 to 101 were prepared and characterized as described for the non-chiralsalts of acid 36 (85 to 89). These salts were photolyzed in solutions at room temperatureand in the solid state at -40°C. The solid state photolyses were stopped before 20%conversion to minimize melting of the crystals. The regioisomeric and enantiomericcompositions of the photoproducts were studied in the same way as for the photolysis ofS-(-)-proline complex 90a. The results are listed in Table 2-5.Table 2-5 Photoproduct Mixture Composition for Complexes and Salts 90a to 90c and92 to 98Optically ActiveAmineSalt orComplexMedium Product 32a Product 33aYield (%)a eekb Yield (%)a eekb()muleOHAS-(-)-Proline90a Solid StateEthanol9647(+)-76nil453nilnil76Chapter 2 Results and DiscussionTable 2-5 (Continued) Photoproduct Mixture Composition for Complexes and Salts90a to 90c and 92 to 9890b Solid StateEthanol9650(-)-80nil450nilnileiillOHHR-(+)-ProlineQw 0OHli90c Solid State 84 nil 16 nil(±)-Proline Ethanol 50 nil 50 nil&N)....1 0OCH3 92 Solid State 93 (+)-58 7 nilHS-(-)-Proline Methyl Ester Benzene 95 nil 5 nil(--- sCH2OH93 Solid State 100 (-)-37 - -H S-(+)-2-Prolinol0NH2 94 Solid State 100 (-)-24 - -S±)-ProlineamideQ■CH2OCH3H95 Solid State 100 (-)-16 - -S4)-2-(Methoxymethy1)-pyrrolidine77Chapter 2 Results and DiscussionTable 2-5 (Continued) Photoproduct Mixture Composition for Complexes and Salts90a to 90c and 92 to 98(+)-e96 Solid State 100 95 - -0HS-(-)-Proline tert-butyl ester Acetone 50 nil 50 nilSC"' CI-13\ HN\ 97 Solid State 87 (-)-87 13 73cH.3S,S-(+)-Pseudoephedrine.98 Solid State 65 (+)-14 35 nil1401^•o(-)-Strychnine Acetone 60 nil 40 nila The estimated error is ±5%. b The sign of the rotation of the predominant enantiomer is shown inparentheses. C Solid state photolysis carried out at -20°C.Several interesting conclusions can be drawn from the results in Table 2-5.Solution photolysis of the chiral salts yielded racemic products. The product ratiodepended on the solvent and corresponded to the regioselectivity of the photoreaction ofester-acid 36 in the same solvent. This indicates that the salts of acid 36 dissociate insolution and explains why no enantiomeric excess is observed in the photoproducts.In contrast, solid state photolysis of the chiral salts favored formation ofphotoproduct 32a in all instances; only minor amounts of product 33a were formed fromsome of the salts. The enantioselectivity of the di-x-methane rearrangement in the solid78Chapter 2 Results and Discussionstate varied from poor to good. Photoproduct 32a was formed in a range of 14% to morethan 95% enantiomeric excess, whereas no optical activity was observed for isomer 33aexcept in the case of S,S-(+)-pseudoephedrine salt 97, where it was formed in 73%enantiomeric excess. The degree of asymmetric induction depends on the nature of thechiral amine which modifies the crystal packing of ester-acid 36. The enantioselectivityof the solid state photoreaction was not affected by the conversion when it was keptbelow 20%.It is interesting to compare the regio- and enantioselectivity for the di-n-methanerearrangement of chiral salts and complexes 90a to 90c and 92 to 98 with the result of astudy by Scheffer et al. 86 on dibenzobarrelene derivatives 99 and 100 which containchiral substituents. These compounds were photolyzed in solution and in the solid stateand the regio- and diastereoselectivities of the di-n-methane rearrangements weredetermined. Different chiral handles yielded different regio- and diastereoselectivities,both in the solid state and in solutions. The diastereomeric selectivity was not alwaysincreased in the solid state. The authors suggested that the molecular and environmentaleffects of the chiral substituents either reinforce or oppose one another in the solid state.Depending on the relative magnitude of the opposing effects, it could lead to eitherreduced or strongly reversed selectivity in the solid state compared to the liquid state.79hv___♦a-i3ccc^r 2CCOCH3Chapter 2 Results and Discussion99^ 101a100^101b^ 102 R1 = R2 .^l'12CCompound Medium Product 101 Product 102Yields (%) dea Yield (%) dea99 Solid State 100 80:20 - -Solution 100 40:60 - -101 Solid State 80 61:39 20 50:50Solution 90 66:33 10 75:25a Diastereomeric excessFigure 2-50 Photolysis of Dibenzobarrelene Derivatives 99 and 10080Chapter 2 Results and DiscussionIn general, photolyzing chiral salt crystals of ester-acid 36 is an effective way toobtain photoproducts in high optical yields. The conversion of the solid state photolyseswas limited to 20% in order to minimize melting of the crystals. Within this range, theenantiomeric excess in the photoproducts was not affected by the degree of conversion.The different chiral amines used for salt formation did not affect the regioselectivity ofthe di-n-methane rearrangement but controlled the enantioselectivity. This is a moreeffective method than introducing chiral substituents into ester-acid 36, which requirestedious synthetic routes and complicated separations of isomers.23.2 Photolyses of Salts of 12-Methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid (37)Chiral salts 103 to 105 were prepared from acid 37 by mixing equimolarquantities of acid and base in diethyl ether and filtering the resulting precipitates. Thesalts were shown to be simple 1:1 complexes by infrared and 1H-NMR spectroscopy,mass spectrometry and elemental analysis. These salts were photolyzed in acetonesolutions at room temperature and in the solid state at -40°C. The reaction mixtures wereacidified, treated with excess diazomethane and subjected to silica gel columnchromatography, which resulted in the isolation of photoproduct 52. The enantiomericcomposition of the photoproduct was determined from 400 MHz 11I-NMR spectrautilizing the chiral shift reagent (+)-Eu(hfc)3. The signal monitored was the methylsinglet at 8 = 1.90 ppm. The optical rotation of photoproduct 52 was also measured at thesodium D line. The results are listed in Table 2-6.81Chapter 2 Results and DiscussionTable 2-6^Photoproduct Mixture Composition for Salts 103 to 105Chiral Amine Salt Enantiomeric excess^in Product 5210Acetone  Solid State103 nil (+)-38CH2OHN1HS-(+)-Prolinol104 nil (+)-26(^....0tsil 0HS-(-)-Proline tert-butyl ester1^3105a nil (+)-95%411 NcH3 \H(S,S)-(+)-PseudoephedrineOH1. /CH3105b nil z (-)-95%N\CH3^H(R,R)-(-)-Pseudoephedrinea The estimated accuracy in these values is ±10%. b The sign of rotation of the predominant enantiomeris shown in parentheses.82Chapter 2 Results and Discussion1.1v2. H+3.CH2N 2Salts of acid 37^ 52Figure 2-51 Photolysis of Salts of Acid 37Addition of the chiral shift reagent to a 11.1-NMR sample of photoproduct 52 inC6D6 led to splitting of the methyl singlet at 8 = 1.90 ppm, but baseline separation ofthese signals could not be obtained (Figure 2-52). Nevertheless this method allowed theenantiomeric excess of the photoproduct 52 to be determined in instances where theenantiomeric excess is low, for example in the cases of irradiation of salts 103 and 104(Table 2-6). When this method was used to determine the enantiomeric excess ofphotoproduct 52 formed by the solid state photolysis of pseudoephedrine salts 105, onlyone enantiomer was observed within the limits of this method. As described earlier, whenenantiomeric excess is determined by use of chiral shift reagents, the accuracy of themethod decreases with increasing enantiomeric excess.83Chapter 2 Results and Discussion11-1-NMR Spectra of Photoproduct 521in C6D6 after addition of (+)-Eu(hfc)37^6^5^4^3^21PPm1111-NMR Spectraof Photoproduct 52 in C6D67^6^5^4^3^2^1^0WinFigure 2-52 111-NMR of Photoproduct 521^084H^,C1-130H3C^PhChapter 2 Results and Discussion106^ 107Figure 2-53 Diastereomers 106 and 107A second method was used to determine the enantiomeric excess of photoproduct52 from the solid state photolysis of pseudoephedrine salts 105a and 105b. Photoproduct52 was isolated as before, hydrolyzed with K2CO3 and refluxed with S-(-)-a-methylbenzylamine to yield compound 106. The same procedure was applied tophotoproduct 52 which was isolated from the solution photolysis of acid 37. This yieldeddiastereomers 106 and 107 in the ratio 1:1. The 11.1-NMR spectra of diastereomers 106and 107 are shown in Figure 2-54. The signals for the proton on the cyclopropane ringsare at 8 = 4.68 ppm and 8 = 4.72 ppm for compounds 106 and 107 respectively. Thesesignals are well resolved. 1H-NMR spectra of compound 106 formed by solid statephotolysis of (S,S)-(+)-pseudoephedrine salt 105a showed no sign of diastereomer 107.This experiment confirms that photolysis of pseudoephedrine salts 105 yields product 52in over 95% enantiomeric excess.85Chapter 2 Results and Discussion11-1-NMR Spectra of 1:1 Mixture of Diastereomers 106 and 1077.0 ppm o.o7.0 ppm 0.0Figure 2-54 11-I-NMR Spectra of Diastereomers 106 and 10786Chapter 2 Results and DiscussionThe results in Table 2-6 illustrate that the regioselectivity of the di-n-methanerearrangement of acid 37 is not affected by the salt formation. The only observedphotoproduct is compound 52. The regioselectivity of the rearrangement of acid 37 canbe explained in terms of the initial radical forming adjacent to the acid group which is abetter radical stabilizer and a more electron withdrawing group, than the methylfunctionality. It is interesting that salt formation does not affect the di-n-methanerearrangement, although a carboxylate anion is a poorer electron withdrawing group thana carboxylic acid.No enantioselectivity was observed for the di-it-methane rearrangement of salts ofacid 37 in solution, presumably because the salts have dissociated. In contrast the solidstate photolyses led to optical activity in the photoproduct 52. The enantiomeric excessobserved was in the range of 26% to more than 95%. The enantioselectivity of the di-it-methane rearrangement depends on the chiral base used in each instance. In general,these results corroborate those observed for the salts of ester-acid 36. The solid statephotolyses could be carried out to 40% conversion without effecting the enantiomericexcess in the photoproduct.87Chapter 2 Results and Discussion233 Photolyses of Salts of Dimethyl 9-Carboxy-9,10-dihydro-9,10-ethanoanthracene-11,12-dicarboxylate (38)Optically active salts of acid 38 (Figure 2-55) were prepared by mixing equimolarquantities of acid and base in diethyl ether and filtering the resulting precipitate. Thesesalts were shown to be simple 1:1 complexes by infrared and 11-1-NM11 spectroscopy,mass spectrometry and elemental analysis. Elemental and X-ray structure analyses ofsalts 108 and 109 demonstrated that they crystallize with half an equivalent of water.Table 2-7^Photoproduct Mixture Composition for Salts 108 and 109Optically ActiveAmineSalt Medium Work-up with CH2N2 Work-up with CH3CHN,67 68 69 70Yield(%)aeea Yield(%)aYield(%)aeea Yield(%)aeea108 Acetone 814110nil 19•P13NCH3^\H(R,S)-(-)-Ephedrine Solid state 10 - 90 10 24 90 -?(-)-95b109 Acetone 90 nil 104*^SH N/cH30.6^NH(S,S)-(+)-Pseudoephedrioe Solid state 80 16 20 18a The estimated error is ± 5%. b The sign of the rotation of the predominant enantiomer is shown inparentheses.88110MOH111IChapter 2 Results and DiscussionSalts 108 and 109 were photolyzed in acetone solutions at room temperature. Thereaction mixtures were acidified, treated with excess diazomethane and subjected to silicagel column chromatography to yield the photoproducts. Solution photolyses of salts 108and 109 gave compound 67 as the major product with small amounts of product 68(Table 2-7, see Figure 2-55 for structures).38: X = H108:X = (R,S)-(-)-Ephedrine109:X = (S,S)-(+)-Pseudoephedrine1.1v_______►2. H +6768 70Figure 2-55 Photolysis and Workup of Salts 108 and 109(R,S)-Ephedrine salt 108 was photolyzed in the solid state and the photoproductswere isolated following the same procedure as was used for the isolation of the products89Chapter 2 Results and Discussionof solution photolyses of salts 108 and 109. As described earlier, the possibledibenzocyclooctatetraene product cannot be isolated using this method and therefore onlythe di-re-methane rearrangement products of (R,S)-ephedrine salt 108 were isolated. Solidstate photolysis of salt (R,S)-ephedrine 108 yielded photoproducts 67 and 68 in the ratio1:9. The regioselectivity of the di-ir-methane rearrangement has been completelyreversed in the solid state compared to solutions. Although product 68 is achiral, itsprecursor, photoproduct 110, is chiral (Figure 2-55). The solid state photoreactions ofsalts 108 and 109 were repeated and worked up with diazoethane instead ofdiazomethane. The product ratios of the solid state photolyses were determined by 400MHz 11-I-NMR spectroscopy. For enantiomeric excess determination, use was made ofthe chiral shift reagent (+)-Eu(hfc) 3 . The signals monitored were the methyl singlets at 8= 3.4 ppm and 8 = 3.5 ppm for compounds 69 and 70, respectively. Large chemical shiftnonequivalences were observed for the enantiomeric signals (Figure 2-56). In additionthe signs of rotation of the predominant enantiomer were measured at the sodium D lineby polarimetry. The results of these experiments are summarized in Table 2-7.Several conclusions can be drawn from the results summarized in Table 2-7. Nooptical activity was observed in the photoproducts of photolysis of salts 108 and 109 inacetone solutions. The product ratios are the same as observed for the photolysis of acid38 in acetone. This can be attributed to dissociation of the salts in solution.90••••••■•■••■•,..■■■•■Chapter 2 Results and Discussion4.0 1.0LO^ 4.011-I-NMR Spectra of Photoproduct 70 in C 6D6 after Addition of (+)-Eu(hfc)3ILO^ s.0^ 0.0111-NMR Spectra of Photoproduct 70 in C6D6Figure 2-56 11-I-NMR Spectra of Product 7091Chapter 2 Results and DiscussionThe product ratios for the photolysis of salts 108 and 109 in the solid state gavedifferent results (Table 2-7). The product ratio for the solid state irradiation of (R,S)-(-)-ephedrine salt 108 is reversed from the product ratio in solution reactions, whereas theproduct ratio for photolysis of (S,S)-(+)-pseudoephedrine salt 109 in solution and thesolid state is the same. It can be concluded that the crystal lattice controls the reactivityof salt 108. The enantiomeric excess in the photoproducts of salts of acid 38 depended onthe nature of the chiral handle and was poor for salt 109 but good for salt 108. The solidstate photolyses all stopped before 25% conversion. Within this range, theenantioselectivity of these reactions was not affected by the degree of conversion.In order to explain the regioselectivity of the photoreaction of salt 108, an X-raystructure analysis was obtained (Figure 2-57). This confirmed that salt 108 crystallizeswith half of an equivalent of water. Furthermore it illustrated that acid 38 exists as itsanionic form, whereas the amine group in the ephedrine counterion is protonated. Thereis a hydrogen bond between the carboxylic anion oxygen (02) of acid 38 and thehydroxy group of the ephedrine moiety. There are also five other intermolecularhydrogen bonds. As expected, the ester group on the vinyl bond next to the bridgeheadsubstituent in acid 38 is out of conjugation, whereas the other ester carbonyl isconjugated to the vinyl bond. This favors formation of the observed product 110 (Figure2-55).As mentioned earlier the motions associated with the ester group attached to thevinyl bond are considered to be the most important determinants of regioselectivity in thesolid state di-it-methane rearrangement. 42 Both carbonyl groups are involved in hydrogenbonding which should affect their capability to stabilize an adjacent radical similarly.92C9CC1311C19Chapter 2 Results and DiscussionFigure 2-57 Crystal Structure of (S,S)-Ephedrine Salt 10893Chapter 2 Results and DiscussionWhen the intermolecular features of the crystal structure are analyzed it becomesclear that the movement of the ester group on C12 is hindered because of the aromaticring on an adjacent molecule in the crystal lattice. The carbonyl oxygen, 06, is weaklyhydrogen bonded to one of the hydrogens on this aromatic ring. For the other ester groupon C11 there is close contact between 03 and a methyl group on an adjacent molecule. Itcan be suggested that the initial vinyl benzo bridging takes place between C9a and C11,which requires the ester group to swing away from the methyl group on the adjacentmolecule. The space around the carbonyl oxygen 04 is free of any intermolecularcontacts except to a water molecule, which is hydrogen bonded to 04. However, it can beexpected that the water molecule is better able to move around in the crystal lattice,presumably because of its small size.There are few known examples in the literature of co-crystallized water moleculesaffecting solid state reactivity. As discussed earlier, Jones et al. 21 studied the [4+4]photodimerization of the acridizinium salt 11 (Figure 1-5). They observed that for saltsthat co-crystallize with water molecules, the majority of reactions were topotactic. Incontrast, for the same salts crystallized without water molecules the reactions stopped at80% conversion. Although the source of this effect is still unclear, apparently theincorporation of water molecules affects reactivity.Similarly Hasegawa et al. 87 found that photodimerization of p-formylcinnamicacid crystals with incorporated water gives much higher yields of the photodimer thananhydrous crystals.94Chapter 2 Results and Discussion2.3.4 Photolyses of Salts of Dimethyl 9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)Hydrochloride salt 112 was prepared by mixing a solution of amine 39 in ethanolwith concentrated hydrochloric acid (Figure 2-58). The crystals that formed were shownto be a simple 1:1 salt by III-NMR spectroscopy, infrared spectroscopy and elementalanalysis. Salt 112 was irradiated in acetonitrile and in the solid state. The reactionmixtures were made basic and the samples analyzed by GC, which showed thatphotoproduct 60 was the only product formed.The results represent a striking reversal of the regioselectivity of the di-n-methanerearrangement of the hydrochloride salt 112 compared with amine 39. As describedearlier, irradiation of amine 39 gave predominantly the photoproduct 61. Paddick et al. 63explained the regioselectivity of this reaction by suggesting that intramolecular hydrogenbonding between the amine group and its nearest ester group renders the ester group lesscapable of stabilizing an adjacent radical and therefore product 61 is favored overproduct 60 (Figure 2-58). The regioselectivity of the di-n-methane rearrangement of salt112 is not affected by the reaction medium; the major photoproduct is 60 both in solutionand in the solid state. Since the ammonium ion is more electronegative than the aminogroup, we propose that the electronegativity of the ammonium ion destabilizesintermediate 63 more than the intramolecular hydrogen bonding destabilizes intermediate64, and consequently product 60 is favored for irradiation of the hydrochloride salt 112.8895Chapter 2 Results and DiscussionhvPath AhvPath B6439: X = NH113: X = NH.2 C1113: X = NAc63161: X= N1-12115: x = NHAc60: X= N1-12114: X= NHAcFigure 2-58 Photolysis of Compounds 39, 112 and 113It is interesting to compare the above results with the regioselectivity of the di-n-methane rearrangement of acetamide 113 (Figure 2-58). Paddick et a.63 found thatsolution photolysis of acetamide 113 yields a mixture of di-n—methane rearrangementproducts 114 and 115 in the ratio 70:30. The authors suggested that the electronegativityeffect in acetamide 113 outweighs the hydrogen bonding effect and consequently product114 is favored. The regioselectivity is the same as observed for the di-n-methanerearrangement of hydrochloride salt 112 but reversed compared with amine 39.Chiral salts of amine 39 were prepared by mixing equimolar quantities of baseand acid in a suitable solvent and filtering off the crystals that formed upon standing. Thesalts 116a, 116b and 117 were shown to be simple 1:1 salts by infrared and 'H-NMRspectroscopy, mass spectrometry and elemental analysis. Camphorsulfonic acid salts96Chapter 2 Results and Discussion116a and 116b co-crystallized with one equivalent of H2O as shown by elementalanalysis. This was confirmed by X-ray structure analysis of salt 116a. Elemental analysisand FAB-MS indicated that crystals of salt 117 contained one equivalent of HC1.Attempts to prepare salts of amine 39 with various optically active carboxylic acids werenot successful.Table 2-8^Enantiomeric Excess in Photoproduct 60 from Salts 116a to 117Acid Salt Solid StateeeaSolution Phaseeea5tHR-(-)-Camphorsulfonic acid116a,0(-)-68 nilN OSO3 H^—S-(+)-Camphorsulfonic acid116b (-0-64 nilSQ3HO(-)-3-Bromocarnphor-8-sulfonic acid117 (+MP nila The estimated accuracy in these values is ±5%; The sign of the rotation of the predominant enantiomeris shown in parentheses. b Only in the case of salt 117 was a small amount of photoproduct 61 detectable(ca. 15%).97Chapter 2 Results and Discussion111-NMR Spectra of Photoproduct 60 inCDC13 after Addition of (+)-Eu(hfc)3 3.97 PPm^6^5^4^3^2^1^0111-N'MR Spectra ofPhotoproduct 60 in CDC13.1.■■••■1^7 ppm 5^4^3^2^1Figure 2-59 1H-NMR Spectra of Photoproduct 6098Chapter 2 Results and DiscussionSalts 116a, 116b and 117 were irradiated in solution at room temperature and inthe solid state at -40°C. After photolysis the samples were dissolved in ethyl acetate,washed thoroughly with 10% aqueous sodium hydroxide, dried, and analyzed by GC.This revealed that compound 60 was the sole product formed, except in the case of solidstate irradiation of salt 117, where a small amount of photoproduct 61 was detected (ca.15%). The photoproducts of the chiral salts 116a, 116b and 117 were isolated by silicagel chromatography and analyzed by polarimetry and 111-NMR chiral shift reagentanalysis: for the latter, use was made of the chiral shift reagent (+)-Eu(hfc)3. The signalmonitored was the methyl singlet at 8 = 3.83 ppm, which showed almost full baselineseparation of the enantiomers (Figure 2-59). The results are listed in Table 2-8.The results in Table 2-8 show that good enantiomeric excess can be achieved forthe di-n-methane rearrangement of salts of amine 39 in the solid state, but that in solutionnegligible asymmetric induction is observed. The degree of asymmetric inductiondepends on the nature of the chiral acid. The fact that the sign of rotation of thephotoproduct can be reversed by using the optical antipode of the chiral induction agentindicates that the system is well behaved. The solid state photolyses could be carried outto 60% conversion without affecting the enantiomeric excess in the photoproduct.This demonstrates that asymmetric induction in the solid state photochemistry ofamines with optically active acids works as well as the opposite approach, i.e.,asymmetric induction in the solid state photochemistry of carboxylic acids complexedwith optically active amines.99Chapter 2 Results and Discussion2.4 Absolute Steric Course of Solid State Di -x-methane RearrangementsIn salts of dibenzobarrelene derivatives 36, 37 and 39, the electronic effectscontrol the regioselectivity of the di-re-methane rearrangement or at least the electroniceffects and the crystal lattice reinforce each other. In contrast, the regioselectivity of thedi-rt-methane rearrangement of salts of acid 38 was affected by the crystal lattice, anddifferent regioselectivity was observed in solution compared with the solid state.Photolyses of optically active salts of the four dibenzobarrelene derivatives investigatedyielded low to good enantiomeric excesses in the photoproducts. The results depended onthe nature of the chiral counterion, however a counterion leading to high enantiomericexcess in the photoproducts of one dibenzobarrelene derivative does not necessarily givegood results for the others. The chiral counterions ensure chiral crystals, but the crystallattice alone is accountable for asymmetric induction and salts of differentdibenzobarrelene derivatives crystallize in different crystal packing arrangements.As described earlier, there are four different di-n-methane systems in thedibenzobarrelene skeleton and each of these four systems is associated with a pathwaythat leads to a different product (Figure 1-14). That is, two pathways lead to the sameregioisomer and the other two to a different regioisomer. However, pathways that lead tothe same regioisomer yield opposite enantiomers of the product. In dibenzobarrelenederivatives for which the di-n-methane rearrangement is both regio- and enantioselective,it is possible to specify which of these pathways is followed by comparing the absoluteconfigurations of the starting material and the photoproduct.For each dibenzobarrelene derivative the absolute configuration of the chiral saltwhich gave the best results and its photoproduct were determined. The solid state100Chapter 2 Results and Discussionreaction pathways were determined by comparing the absolute configuration of thestarting material with that of its product. Knowing which pathway is followed makes itpossible in principle to identify the crystal forces that control the enantioselectivity of thephotorearrangement in the solid state. Salts of dibenzobarrelene derivatives 36, 37 and39, were investigated. Furthermore, by studying salts of different dibenzobarrelenederivatives, the crystal forces in different systems can be compared in order to propose ageneral explanation of factors that control asymmetric induction in dibenzobarrelenesystems.2.4.1 Absolute Configurations of S-(-)-Proline tert-Butyl Ester Salt 96 andPhotoproduct 32aPhotolysis of S-(-)-proline tert-butyl ester salt 96 yielded photoproduct 32a inover 95% enantiomeric excess. The absolute configuration of salt 96 can be easilyobtained since the absolute configuration of the proline moiety is known. The absoluteconfiguration of S-(-)-proline tert-butyl ester salt 96 was obtained by X-ray structureanalysis89 which demonstrated that the dibenzobarrelene moiety in salt 96 has theabsolute configuration 11M, 12M (Figure 2-60). The designation 11M, 12M focuses onthe conformational dissymmetry conferred on the molecule by the carboxylic groups onC11 and C12. The absolute configuration was obtained by determining the smallesttorsion angle between the groups of highest priority, or fiduciary groups, attached to eachend of the single bond for which the conformation is specified.45 If two of the groups areidentical, that which provides the smallest angle is fiducial. A positive torsion angle(clockwise rotation) is designated P (plus) and a negative torsion angle (counter-clockwise rotation) is designated M (minus). The torsion angles 04-C14-C12-C11 and101Chapter 2 Results and Discussion01-C13-C11-C12 of S-(-)-proline ten-butyl ester salt 96 were analyzed and both werefound to be negative.Figure 2-60 Crystal Structure of Proline ten-Butyl Ester Salt 961021C9*CIS02Cs'C4' .02'ea'CS'C16C1301.`C14CIIm19Cs04O.^•4.1 VI•C4  %%4 04C3CIOACSC2CltCIOC9"C3. 02.ca .Cs'(Alb t, :le Ol e•02 ClsNleC111C13^01MC14C11 s.^111:11v•CitC101 C9AC90403'41• a;Oa •:".011PW •41 OC2 •C3 C4 %be 4 •CrC6C4'CS'C10/ICSChapter 2 Results and DiscussionCrFigure 2-60 Crystal Structure of S-(-)-Proline tort-Butyl Ester Salt 96 (continued)The crystal structure of S-(-)-proline tent-butyl ester salt 96 revealed that theamine group of the proline moiety is protonated whereas the carboxylic acid functionalityin the dibenzobarrelene moiety is in its anion form. The oxygen of the carboxylic anionare hydrogen bonded to different proline moieties. It was suggested earlier that theregioselectivity of the di-it-methane rearrangement of salt 96 is controlled by electronic103(CH3)2CHOOC COOCH(CH3)2(CH3)2CHOHChapter 2 Results and Discussioneffects. There is considerably more steric influence from the crystal lattice around thecarboxylate anion than in the vicinity of the ester group, mainly because of the hydrogenbonding of the carboxylate anion to the proline moiety. The intramolecular packingarrangement of the dibenzobarrelene reinforces the electronic effects since the oxygensof the carboxylate anion are almost completely out of conjugation with the vinyl bondwhereas the ester carbonyl is nearly fully conjugated to the vinyl bond. This favorsformation of the observed product 32a.(+)-32a^ (+)-35(4bR, 8bR, 8cR, 8dR)^ (4bR, 8bR, 8cR, 8dR)Figure 2-61 Transesterification of Di-Ester 32a into Di-Ester 35Photoproduct 32a from the solid state photolysis of S-(-)-proline tert-butyl estersalt 96 was isolated as described before. The circular dichroism spectrum of 32a is shownin Figure 2-62; the sign of the optical rotation was positive at the sodium D line (589nm). Photoproduct 32a was refluxed in sec-butanol to yield the (+)-enantiomer of di-ester 35 (Figure 2-61). Scheffer et al 42 showed that the (+)-enantiomer of di-ester 35 hasthe absolute configuration (4bR, 8bR, 8cR, 8dR) and therefore the (+)-enantiomer ofproduct 32a must have the same absolute configuration. The circular dichroism spectrumfor di-ester 35 is shown in Figure 2-62; as expected the circular dichroism spectra for di-esters 32a and 35 are almost identical."104Chapter 2 Results and DiscussionDOAbsorbenceFigure 2-62 Circular Dichroism Spectra for Compounds 32a and 35Comparison of the absolute configurations of S-(-)-proline tert butyl ester salt 96and its photoproduct (+)-32a demonstrates that the initial benzo-vinyl bridging of thesolid state di-it-methane rearrangement in salt 96 is formed between C8a and Cl 1 . Inorder to identify the factors that favor initial vinyl bonding between C8a to C11 over C9ato C11, the intramolecular arrangement of the dibenzobarrelene system was investigated.The vinyl bond substituents are frozen in non-C2„ orientation which gives the molecule achiral conformation. The motions associated with the vinyl substituents are considered tobe most important in determining the regio- and enantioselectivity of the di-x-methanerearrangement.42 The carbonyl group in the ester is almost completely conjugated withthe vinyl bond (04-C14-C12-C11 = -171°) whereas the oxygens of the carboxylate anionare almost completely out of conjugation with the double bond (01-C13-C11-C12 = -81°,02-C13-C11-C12 = +104°). This suggests that moving C11 towards C8a would causesimilar steric interactions between the vinyl substituents, as moving C11 towards C9a,that is neither initial benzo-vinyl bridging between C11 and C8a or C11 and C9a shouldbe favored. However, when the -9° torsion angle between C13-C11-C12-C14 (Figure 2-1054^ Torsion Angles0201-C13-C11-C12 = _ 81002-C13-C11-C12 =+ 104°03-C14-C12-C11 = + 6°04-C14-C12-C11 = - 171°C13-C11-C12-C14 = - 9°Chapter 2 Results and Discussion63) is taken into account it can be suggested that the initial benzo vinyl bridging betweenC8a and C11 should be favored because it involves movement of the Cl 1-C13 bondaway from the C12-C14 bond. In contrast, initial benzo-vinyl bonding between C9a andC11 requires eclipsing of the C11-C13 bond with the C12-C14 bond, which brings thesubstituents on C11 and C12 closer together causing steric interactions between them andwould thus be expected to be unfavorable.The intermolecular distance between C8a and C11, 2.44(8) A, is the same as thedistance between C9a and C11, 2.44(7) A. The torsion angle between C13-C11-C12-C14suggests that the p orbital on C11 is better aligned to form a bond with C8a than C9a(Figure 2-63). In other words if the initial excited state has a similar conformation as thestarting material it requires less movement of C11 to form a bond with C8a than C9because the p orbital on C11 is tilted towards C8a.Figure 2-63 Some Torsion Angles of S-(-)-Proline tent-Butyl Ester Salt 96Analyzing the intermolecular packing arrangement of salt 96 reveals thatcarboxylate anion 01 is hydrogen bonded to a proline moiety above it in the crystallattice whereas carboxylate anion 02 is hydrogen bonded to a proline moiety beside it inthe crystal lattice (Figure 2-60). These intermolecular hydrogen bonds result inintermolecular steric crowding between the carboxylate anion and the proline moieties.However, it can be proposed that the intermolecular arrangement of the dibenzobarrelene106(CH2)n (CH2)n Chapter 2 Results and Discussionmoiety in salt 96 does not favor either formation of the (+) or the (-)-enantiomer ofproduct 32a. This suggests that the intramolecular packing arrangement of salt 96controls the enantioselectivity in the di-rr-methane rearrangement.Scheffer et al 9 1 studied type II photoreactions of macrocyclic di-ketones in thesolid state. Irradiation of these ketones yielded either cis or trans-cyclobutanols. Theauthors explained the regioselectivity of the type II reaction from the X-ray crystalstructure of the starting materials. They proposed that the cis or trans-cyclobutanol whichwas favored required less movement of the 1,4 biradical intermediate for ring closure.That is the p orbitals in the biradical intermediate were better aligned for ring closure toform the favored cyclobutanol.hv(CHO W^ (CH2) 2^(CH2)n^(CH2)n.20^ 0n = 5, 7^85% 13%n = 6, 8 3% 88%Figure 2-64 Type II Reaction of Macrocyclic Di-ketonesAs described earlier, Scheffer et al. 43 explained the enantioselectivity of the di-ir-methane rearrangement of di-ester 34 by suggesting that paths I and II shown in Figure1-14 were favored because they caused the ester groups to move away from each other inthe initial stage of the reaction. Paths III and IV, on the other hand, were thought to leadto severe steric interactions between the ester groups. Both the ester carbonyls are107C20C IC16'Cl)C9C8aC8C20C9AI?:.00 4rskC4^C8C3C7CIC19• t.,111 :04 ..e1111C15C I C18VP/C16^C13C16'C11C8aC3Chapter 2 Results and Discussionpartially conjugated with the vinyl bond (Figure 2-65) which makes it easier to visualizean interaction between the two ester groups in the initial step of the di-it-methanerearrangement. The torsion angle between the ester substituents on the vinyl bond is +5°(C13-C11-C12-C17), which suggests that pathways I and II in Figure 1-14 are favored asthey will not cause eclipsing of the C11-C13 bond with the C12-C17 bond. Thishypothesis does not discriminate between the two pathways which both lead to the (-)-enantiomer of product 35.Figure 2-65 Crystal Structure of Di-ester 34108Chapter 2 Results and Discussion2.4.2 Absolute Configurations of Salts 103, 105a and Photoproduct 52Photolysis of (S,S)-(+)-pseudoephedrine salt 105a yielded the (+)-enantiomer ofphotoproduct 52 in over 95% enantiomeric excess. The absolute configuration of thedibenzobarrelene moiety in the (S,S)-(+)-pseudoephedrine salt 105a was determined byX-ray crystal structure analysis and shown to be 12P (Figure 2-66). The X-ray structurerevealed that acid 37 is in its anionic form whereas the amine group in thepseudoephedrine moiety is protonated. Both oxygens in the dibenzobarrelene moiety areinvolved in hydrogen bonding. The space around the methyl group on the vinyl bond isrelatively free from interaction with the neighboring molecules, whereas the spacearound the carboxylate anion is very crowded due to hydrogen bonding with thepseudoephedrine moiety.Figure 2-66 Crystal Structure of (S,S)-(+)-Pseudoephedrine Salt 105a.109Chapter 2 Results and DiscussionFigure 2-66 Crystal Structure of (S,S)-(+)-Pseudoephedrine Salt 105a (continued)The photoproduct 52 (the (+)-enantiomer) of the solid state irradiation of (S,S)-(+)-Pseudoepheclrine salt 105a was worked up as before and hydrolyzed with K2CO3 toyield acid 115. Refluxing acid 115 with oxalyl chloride gave acid chloride 116 whichwas refluxed with S±)-a-methylbenzylamine to yield compound 106 (Figure 2-67).110Chapter 2 Results and Discussion1. K2W3,2. H*52^115S- }phCIACH3)N1-12 116^106Figure 2-67 Synthesis of Compound 106The absolute configuration of compound 106 was determined by X-ray analysis(Figure 2-68). There are two independent molecules in the unit cell and both have theabsolute configuration (C9R, ClOS, Cl1S, C12S, C15S) (Figure 2-68, Thecrystallographer used a different numbering system). Consequently it can be concludedthat the (+)-enantiomer of product 52 has the absolute configuration (4bS, 8bR, 8cS,8dS).111C113C6 C3CI9C2201CI•^•Cl2CIICacaC2 C7CI310'^-;414114 C6C9Aea'^rr.C4C20CalC5^It ^111.'• ,'-' .1°llgelg 16' S' *_.„ talisliC 9C9AOrC10^•.'•^aii• .0^CI1,.., C 1 OA C4^CT^ •eir^C5^ lb. 1Pg Ca C3 •C2Chapter 2 Results and DiscussionFigure 2-68 Crystal Structure of Compound 106112Torsion Angles01-C14-C12-C11 = + 12°02-C14-C12-C11 = - 167°C13-C11-C12-C14 = + 8°O2Chapter 2 Results and DiscussionA comparison of the absolute configurations of (S,S)-(+)-pseudoephedrine salt105a and its photoproduct 52 illustrates that the initial vinyl benzo bridging takes placesbetween C4a and C11. The hypothesis used to explain the enantioselectivity of S-(-)-proline tert-butyl ester salt 96 can also be applied the enantioselectivity of salt 105a. Thetorsion angle between the acid and methyl substituents (C13-C11-C12-C14) is +8° and isshown along with the torsion angles between the carboxylic oxygen and the vinyl bondin Figure 2-69. This suggests that the initial vinyl benzo bridging is favored between C4aand C11 as this does not involve the eclipsing of the C11-C13 bond with the C12-C14bond. It can be suggested, as before, that the p orbital on C11 is better aligned for bondformation with C4a than C9a.+8°13^ C " C14Figure 2-69 Some Torsion Angles in (S,S)-(+)-Pseudoephedrine Salt 105aThe packing diagram of salt 105a reveals that the space around the methyl groupon the vinyl bond is relatively free from interaction with the neighbbring molecules. Thissuggests that the intermolecular packing arrangement of salt 105a does not favor eitherenantiomer of product 52.To further test this hypothesis, the X-ray structure of S-(+)-prolinol salt 103 wasobtained. The X-ray structure, which is shown in Figure 2-70, reveals that theasymmetric unit contains two independent molecules, 103A and 103B, with the absoluteconfigurations 12P and 12M, respectively. The enantioselectivity of the photoreaction of11303Molecule 1014Molecule 103BH198H1313CIOACS-CI^CO\Chapter 2 Results and DiscussionS-(+)-prolinol salt 103 is 38% and the (+)-enantiomer of product 52 is favored. There aresteric interactions between the crystal lattice and the acid groups in molecule 103A and103B, mainly as the acid groups are hydrogen bonded to the prolinol moieties. Themethyl group attached to the vinyl bond is relatively free of interaction with the crystallattice in both molecules 103A and 103B.Figure 2-70 Crystal Structure of S-(+)-Prolinol Salt 103114Chapter 2 Results and DiscussionThe carboxylate anion is in conjugation with the vinyl bond in both molecules103A and 103B (Figure 2-71). The torsion angle between C13-C11-C12-C14 is +1° in103A and +5° in 103B. Since the torsion angle between the vinyl substituents inmolecule 103B is greater, it can be proposed that molecule 103B reactsenantioselectively to yield mainly the (+)-enantiomer of product 52, that is via initialvinyl-benzo bridging between C11 and C9a. In contrast, it can be suggested thatmolecule 103A reacts without extensive enantioselectivity because the conformation of103A does not strongly favor either initial vinyl benzo bridging between C4a and C11 orC9a and C 1 1 . The torsion angle between C13-C11-C12-C14 is very small and formationof either the (+) or the (-) enantiomer of product 52 will move the Cl 1-C 13 bond awayfrom C12-C14 bond.13^ 1301103ATorsion Angles01-C14-C12-C11 =+ 173 °02-C14-C12-C12 = - 6 0C13-C11-C12-C14 = + 1 0910313Torsion Angles01-C14-C12-C11 = - 169 002-C14-C12-C11 = + 14 °C13-C11-C12-C14 = + 5 °Figure 2-71 Some Torsion Angles in S-(+)-Prolinol Salt 103The packing diagram of salt 103 shows that the space around the methyl group onthe vinyl bond in both molecules 103A and 103B is relatively free from interaction with115Chapter 2 Results and Discussionthe neighboring molecules. Consequently, it can be suggested that the intermolecularpacking arrangement of salt 103 does not favor either enantiomer of product 52.Yang et al.92 studied the photochemistry of salt 117 which exists in dimorphicforms: needles and plates (Figure 2-72). Irradiation of the needle dimorph of salt 117 inthe solid state yielded the major product 118 in over 95% enantiomeric excess with asmall amount of the minor product 119. In contrast irradiation of the plates gave themajor product 118 in 12% enantiomeric excess. The authors explained these results fromX-ray structure analyses of both dimorphs of salt 117. In the needle-like crystals thecarboxylate anions adopt a single homochiral conformation whereas in the plate formthere are two independent carboxylate anions in the asymmetric unit that have oppositeabsolute configurations. It was suggested that the needles react stereospecifically to formone stereoisomer of the photoproduct, whereas in the plates there are competingstereospecific photoreactions that yield photoproducts of low overall enantiomericexcess.119Figure 2-72 Photochemistry of Salt 117116Chapter 2 Results and Discussion2.43 Absolute Configurations of R-O-Camphorsulfonic Acid Salt 116a andPhotoproduct 60Photolysis of R-(-)-camphorsulfonic acid salt 116a in the solid state yielded thephotoproduct 60 in 70% enantiomeric excess. The absolute configuration of salt 116awas obtained by X-ray structure analysis (Figure 2-73). The dibenzobarrelene moietywas shown to have the absolute configuration 11P, 12P. It was suggested earlier that theregioselectivity of the solid state photoreaction of salt 116a was controlled by electroniceffects. The crystal packing arrangement reinforces the electronic effect since the estergroup on C12 is almost out of conjugation with the vinyl bond whereas the ester groupon C11 is in conjugation with the vinyl bond. The packing arrangement favors formationof the observed product 60.117Chapter 2 Results and DiscussionFigure 2-73 Crystal Structure of Rf)-Camphorsulfonic Acid Salt 116a118Chapter 2 Results and DiscussionThe circular dichroism spectrum of the (-)-enantiomer of photoproduct 60 of thesolid state irradiation of R±)-camphorsulfonic acid salt 116a is shown in Figure 2-74.Photoproduct 60 was formed with 70% excess of the (-)-enantiomer.-30 F^ 1.1■•••■•■••Absorbance190.0^ML Ens)^260.0Figure 2-74 Circular Dichroism Spectrum of Photoproduct 60This spectrum is very similar to the circular dichroism spectra of compounds (+)-32a and(+)-34 (Figure 2-74). It can be suggested that the (-)-enantiomer of product 60 has thesame absolute configuration as the (+)-enantiomers of compounds 32a and 34. A chiralhandle was introduced into product 60 allowing confirmation of the absoluteconfiguration by X-ray structure analysis. Refluxing a racemic mixture of product 60 andR-(-)-a-methoxyphenylacetic acid chloride yielded diastereomers 120 and 121 in theratio 1:1. Diastereomers 120 and 121 were separated by HPLC chromatography.Diastereomer 120 was crystallized from an acetone and n-hexane solution to give crystalscontaining one equivalent of acetone. X-ray structure analysis of these crystals showedthat compound 120 has the absolute configuration (9R, 10S, 11S, 12R, 18R).119C17141C10C5 ClOaChapter 2 Results and DiscussionFigure 2-75 Crystal Structure of Compound 120Photoproduct 60 from the solid state irradiation of salt 116a was refluxed with R-(-)-a-methoxyphenylacetic acid chloride. GC analysis of the reaction mixturedemonstrated that diastereomers 120 and 121 were formed in the ratio 86:14. It can beconcluded that the (-)-enantiomer of product 60 has the absolute configuration (4bS,8bR, 8cR, 8dS) as shown in Figure 2-76 (Crystallographer used a different numbering120cH300c coocH38b 8dCH300C coccH31160(4bS, 8bR, 8cR, 8dS)Chapter 2 Results and Discussionsystem). Furthermore, this does agree with the absolute configuration suggested afteranalyzing the circular dichroism spectrum of product 60.120(9R, 10S, 11S, 12R, 18R)Figure 2-76 Absolute Configuration of (-)-Enantiomer of.Product 60 and Compound120Comparison of the absolute configurations of salt 116a with the (-)-enantiomer ofproduct 60 reveals that initial vinyl benzo bridging between C 1 Oa and C12 is favored.The torsion angles between the vinyl substituents, C13-C11-C12-C15, is +1° and isshown in Figure 2-77 along with the torsion angles between the ester groups and thevinyl bond. According to our hypothesis that the torsion angle between the vinylsubstituents controls the enantioselectivity of the di-n-methane rearrangement, salt 116ais not expected to react enantioselectively, since the torsion angle between the vinylgroup substituents is very small. This theory favors formation of the (+)-enantiomerproduct of 60 (initial bonding between C12 and C8a) and is opposed to what is observed.121Chapter 2 Results and Discussion3 Torsion Angles01-C13-C11-C12 = - 156°02-C13-C11-C12 = + 22 003-C15-C12-C11 = - 102°04-C15-C12-C11 = + 84 0C13-C11-C12-C14 = + 1°iFigure 2-77 Some Torsion Angles in R-(-)-Camphorsulfonic Acid Salt 116aConsidering that there is an intermolecular steric interaction between 04 and anadjacent aromatic ring which hinders the movement of the ester group on C12 towardsC8a, it can be proposed that the intramolecular and the intermolecular arrangements ofsalt 116a affect the asymmetric induction in opposite ways. That is to say, theintramolecular arrangement slightly favors formation of an initial vinyl benzo-bridgebetween C8a and C12. In contrast the crystal lattice favors formation of an initial benzovinyl bridge between ClOa and C12 since steric interaction between 04 and an adjacentaromatic ring disfavors formation of an initial vinyl benzo bridging between C8a andC12.2.5 Solid State Photochromism of Dibenzobarrelene Derivatives 38, 43 and 44Photochromism is a phenomenon whereby a substance undergoes a color changeupon absorption of light. By definition, this process must be reversible either thermallyor photochemically.93 Photochromic compounds have been studied in a sporadic manner12238:R=H43: R = CH344: R = CH2CH3Chapter 2 Results and Discussionsince the end of the last century. 94 Photochromism has been observed for variouscompounds in different media and some of these systems have been studied intensively.Several compounds that are photochromic in the solid state have been reported. 8jmFigure 2-78 Dibenzobarrelene Derivatives 38, 43 and 44 which Display Solid StatePhotochromismCompounds 38, 43 and 44 (Figure 2-78) exist as white crystals. Upon irradiationthe crystals of these compounds turn dark blue or blue-green. Tri-ester 44 exists asdimorphs and both forms show very similar photochromism. Acid 38 is trimorphic andall the crystal forms show a color change to green-blue upon irradiation. Salts 108 and109, ephedrine and pseudoephedrine salts of acid 38, were photochromic at lowtemperatures (-40°C); they turned a pink color. No such color change was observed forthese salts at room temperature. None of these compounds are photochromic in solution.The intensity of color developed in these compounds was approximately proportional tothe irradiation time. The photolyzed crystals lose their color after a few hours to a fewdays in the dark. Heating the colored crystals or dissolving them removed the color muchmore effectively. GC analysis of the colored crystals showed no sign of reaction. Thecoloring and bleaching process could be repeated several times without damaging thecrystals. Prolonged irradiation of the colored crystals did eventually turn them colorlessand photoreaction was observed. Exposure of the crystals to UV light under aerobicconditions or pure oxygen has no deleterious effect on the photochromism.123Chapter 2 Results and DiscussionTri-ester 431:Before irradiation.2: After irradiation.2200^ 500^ 800Nanometers(nm)Acid 38.1:Before irradiation.2: After irradiation.21U500Nanometers(nm)0Figure 2-79 Solid State UV-Visible Absorption Spectra for Compounds 38 and 43124Chapter 2 Results and DiscussionThe solid state UV-visible absorption spectra95 for the colored crystals ofdibenzobarrelenes 38 and 43 are shown in Figure 2-79 along with the spectra of thecrystals before irradiation. The spectra of the irradiated crystals display absorption in the500-800 nm range which is absent before irradiation and was not detected after the colorhad faded.The ESR spectnun95 of colored crystals of tri-ester 43 is shown in Figure 2-80.The g-value indicates that the radical species is organic in nature 9 6 The ESR signal fadeswith time and is absent when the crystals have returned to their original color. The solidstate UV and ESR spectra of tri-ester 43 suggest that a radical species which absorbs inthe 400-800 nm range is responsible for the photochromic behavior..0.04140044fti•VAPIvA3340^ 3380^ 3420[G]Figure 2-80 ESR Spectra of Irradiated Crystals of Tri-ester 43125Chapter 2 Results and DiscussionIn our laboratory, similar solid state photochromic behavior has been found forother dibenzobarrelene derivatives. Dibenzobarrelene 12271 is dimorphic and both crystalforms turn purple when irradiated at room temperature. Dibenzobarrelene 123 51 istrimorphic and two of the crystal forms exhibit blue photochromism at roomtemperature, whereas the other crystal form showed a color change to pink only at lowtemperatures (-40°C). The UV and ESR spectra of the blue colored crystals ofdibenzobarrelene 122 and 123 are very similar to the ones recorded for tri-ester 43.Compound R 1 R2 Photochromic39 NH2 H no46 CHO H at low T80 CH3 CH3 at low T122 Cl Cl at RT123 CH2C1 H at RT124 CH3 H no125 Cl H noFigure 2-81 Bridgehead Substituted Dibenzobarrelene Derivatives and theirPhotochromismDibenzobarrelenes 46 and 80 74 (Figure 2-81) turn green and pink, respectivelywhen irradiated at low temperatures (-40°C). No such color change is observed forirradiations at room temperature.Generally, only dibenzobarrelene-11,12-diester derivatives with either one or twobridgehead substituents have shown photochromism. It should be mentioned that not allcompounds that have this structure are photochromic. For example, dibenzobarrelene126Chapter 2 Results and Discussionderivatives 39, 12497 and 12597 are not photochromic. No photochromism has beenobserved for dibenzobarrelene-11,12-diesters lacking bridgehead substituents. Thephotochromism is not observed in solution or polymer films and therefore it can beproposed that the crystal lattice must play a key factor in the photochromism.Observations of different photochromism for different dimorphs of the same compoundalso support this theory. X-ray structures have been obtained for most of thesebridgehead substituted dibenzobarrelene derivatives, both the photochromic and the non-photochromic ones. Correlation of crystal packing arrangements and photochromism hasbeen attempted, however no final conclusions have been drawn. Some possibleexplanations for this unusual solid state phenomenon are presented below, but thesetheories are speculative. Much more information about the photochromic compounds isneeded before the nature of colored state is known.126^127^ 128Figure 2-82 Radical Ions 126-128The first theory suggests that a radical ion species, such as structures 126-128(Figure 2-82) which are formed via either an intra- or intermolecular photoinducedelectron-transfer process, might be responsible for the observed photochromism. Thefeasibility of photochemical electron transfer in polar solvents can be predicted on thebasis of the Weller98 equation, in which the free-energy change for the electron transfer iscalculated from the redox potentials and the excitation energy. Our estimation of AG127Chapter 2 Results and Discussionobtained from this equation gives 1G as negative. This is based on the known oxidationand reduction potentials for dimethylbenzene and maleic anhydride which can serve as amodel for the photochromic dibenzobarrelenes. The oxidation potential fordimethylbenzene is 44 kcal/mol and the reduction potential for maleic anhydride is -21kcal/mo1.98b The excitation energy of the photochromic dibenzobarrelenes can beestimated to be approximately 100 kcal/mol.by^-.^tA + D --0- A + DAG = Eox - Ered - EexcitEox: OxidiationPotentialEN'd: ReductionPotentialBoo: Excitation energy(Weller equation)This is similar to what Sakaguchi et al. 99 found when they studied the solid statephotochromism of styrylpyridinium tetraphenylborate 129 (Figure 2-83). Theydiscovered that the yellow salt turned blue upon irradiation. The photochromism wasassigned to photoinduced electron transfer from the tetraphenylborate anion to thestyrylpyridinium cation in the solid state to form a styrylpyridinium radical.129Figure 2-83 Styrylpyridinium Tetraphenylborate Complex, 129128Chapter 2 Results and DiscussionReturning to the dibenzobarrelene photochromism, it is possible that breaking ofthe C9-C 12 bond of the dibenzobarrelene ring results in the formation of a zwitterion 130(Figure 2-84), thus leading to the observed photochromism. Zwitterion 130 is thenstabilized through resonance. This is consistent with the fact that only bridgehead-substituted dibenzobarrelenes are photochromic, since R1 should stabilize the cation atC9 (Figure 2-82). The electron transfer theory fails to account for the effect of thebridgehead substituents.tw 130Figure 2-84 Zwitterion 130Another possible explanation involves the biradical intermediate 131 produced byinitial benzo-vinyl bridging during the photorearrangement (Figure 2-85). The biradicalspecies resembles structure 132, which is known to absorb at 560 nm in the visibleregion.00 However, radical 132 shows a weak absorption in the visible region whereasthe photochromic dibenzobarrelene derivatives show a strong broad absorption in thesame region. Another drawback of this theory is that it does not take into account thatonly bridgehead substituted dibenzobarrelenes are photochromic.129Chapter 2 Results and Discussionhv A1131^132Figure 2-85 Biradicals 131 and 132It is interesting to compare the photochromism of the dibenzobarrelenes with thesolid state photochromism of compound 133 (Figure 2-86). Wudl et al. 101 discovered thatcolorless crystals of compound 133 turn pink when irradiated in the solid state with UV-light or X-rays. However, no color change was produced when solutions of thesecompounds were irradiated. Heating of the colored crystals returned them to theiroriginal white color. ESR studies of the pink crystals of 133 indicated the presence oftwo radical species, one with an unpaired electron localized on a sulfur atom and theother with an electron localized on a carbon atom. The X-ray crystal structure revealedintramolecular phenyl-thiobenzene contacts as well as intermolecular phenyl-phenylcontacts. The authors were unable to identify the nature of the colored state. Howeverthey proposed three hypotheses to explain the photochromism. They suggested thatintermolecular charge transfer between thiobenzenes or intermolecular charge transferbetween thiobenzene and phenyl rings are possible explanations for the pink crystals.Alternatively, homolytic cleavage of benzyl-sulfur bonds can also explain thephotochromism.130Chapter 2 Results and Discussion133Figure 2-86 Compound 1332.5.1 7—Ray Irradiation of Salt 134A notable property of organic salt crystals is the strong lattice forces which bindthem and translate into high melting points. Consequently the likelihood of the crystalmelting during reaction is reduced and the probability of observing a topotactic reactionis increased. It was decided to look for a single crystal-to-crystal reaction of a salt of adibenzobarrelene derivative, and acid 4058 was selected for this study. The piperidine salt(134) of acid 40 was prepared (Figure 2-88). 102 Salt 134 was photolyzed in acetone. Thereaction mixture was made acidic and treated with excess diazomethane to yield product135.58 Crystals of the piperidine salt 134 were irradiated through a Pyrex filter. GCanalysis of single crystals indicated 60% conversion to product 135. The crystals hadturned yellow and were no longer transparent.131)COO \134Chapter 3 Experimental1. hv2. CH2N2135Figure 2-88 Photolysis of Piperidine Salt 134There are examples in the literature in which solid state y-ray irradiation yieldscleaner reactions than UV irradiation. 103 An understanding of the nature of the interactionof y-rays with organic molecules in crystals is very limited, however. 104Crystals of piperidine salt 134 were irradiated with y-rays from a 60Co source forone week. The crystals looked undamaged, but slightly yellow, and they were almostindistinguishable from their non-irradiated counterparts. GC analysis indicated 10-25%conversion to product 135 for single crystals. This was repeated with an another sampleof piperidine salt 134, but this time the crystals were irradiated for a month. As beforethe crystals looked undamaged after irradiation but slightly yellow. GC analysis of singlecrystals showed conversion in the range 10-25% as before. No further attempts weremade to achieve topotactic reaction for piperidine salt 134.Although topotactic reactions were not observed for piperidine salt 134, it wouldbe most interesting to study y-irraditions as well as solid state reactions induced by UVirradiation.132Chapter 3 ExperimentalCHAPTER 3 EXPERIMENTAL3.1 GeneralMelting Points (MP). Melting points were determined on a Fisher-Johns meltingpoint apparatus and are not corrected.Infrared Spectra (IR). A Perkin Elmer 1710 Fourier transform infraredspectrometer was used for obtaining infrared spectra. The positions of the absorptionmaxima are reported in cm- 1 . The spectra of liquid samples were recorded withoutsolvent as thin films between two sodium chloride pellets. Solid samples (2-5 mg) wereground in KBr (100-200 mg) and pelleted in an evacuated die (Perkin-Elmer 186-0002)with a laboratory press (Carver, model B) at 15,000 psi.Mass Spectra (MS). Low and high resolution electron ionization (EI) massspectra were obtained on a Kratos MS 50 mass spectrometer. Coupled gaschromatography-mass spectral (EI) analysis was performed on a Kratos MS 80spectrometer attached to a Carlo-Erba chromatogram. Fast atom bombardment (FAB)mass spectra were registered on an AEI MS 9 mass spectrometer.Nuclear Magnetic Resonance Spectra. Proton nuclear magnetic resonancespectra ( 1H-NMR) were recorded on Bruker AC-200 (200 MHz), Varian XL-300 (300MHz) and Bruker WP-400 (400 MHz) spectrometers. Signal positions are reported as133Chapter 3 Experimentalchemical shifts (8) in parts per million (ppm) with tetramethyl silane (TMS) as aninternal reference. The multiplicity of the signals, number of protons, coupling constants(J) in Hz and assignments are given in parentheses following the chemical shifts.Carbon nuclear magnetic resonance spectra ( 13C-NMR) were recorded at 100.6Hz on Bruker Am-400 spectrometer, at 75.4 MHz on Varian XL-300 and at 50.3 MHz onBruker AC-200 spectrometers. Chemical shifts (8) are reported under broad band protondecoupling in ppm and are followed by their assignments, which were determined in partby the attached proton test (APT) experiment.Ultraviolet Spectra (UV). Ultraviolet spectra were recorded on a Perkin ElmerLambda-4B UV/Vis spectrometer. The wavelength (X) in nanometers (nm) and theextinction coefficient (E (1/mol/cm)) of each absorption maximum are given.Elemental Analysis. All elemental analyses reported were performed by Mr. P.Borda, Department of Chemistry, University of British Colombia.Chromatography. Gas liquid chromatography (GC) analyses were performed ona Hewlett Packard 5890A gas chromatograph fitted with a flame ionization detector, andthe instrument was equipped with a Hewlett Packard 3392A integrator. All thechromatographic analyses were carried out on a 15 m x 0.25 mm DB-1 and a 15 m x 0.25mm DB-17 columns from J&W Scientific Inc.Gravity column chromatographic separations were carried out by using 230-400mesh gel (E. Merck) with a suitable solvent or solvent combinations.Thin layer chromatographic analyses were performed on pre-coated silica gelplates (type 5554 from E. Merck).High performance liquid chromatography (HPLC) was performed on a Waters600E system equipped with a Waters 486 UV detector and a Waters fraction collector.134Chapter 3 ExperimentalRadial-Pak cartridge (II Porasil, particle size 10 p.), with 8 mm x 100 mm from Millipore(cat# 85720) was used for analytical studies. For preparative scale work, a similarcolumn with an internal diameter of 25 mm (cat # 38504) was used.Optical rotations and Circular dichroism. Circular dichroism spectra wereobtained on a Jasco-J710/ORD-M polarimeter. Specific ellipticity, [► I , was calculatedby the following equation:[w] . =11 cWhere^t = the temperature at which the optical rotation was measuredX = the wavelengthv = the angle of ellipticity in degrees1= the path length in decimetersc = the concentration of the sample solution in milligrams per 10 mLOptical rotations were measured on a Jasco-J710/ORD-M and a Perkin Elmer 141polarimeter at the sodium D line (589 tun) at room temperature. Specific rotation, [at,was calculated by the following equation:[a]D = Tac 100Where^t = the temperature at which the optical rotation was measuredD= the sodium D line at 589 nma = the recorded optical rotation in degrees1= the path length in decimetersc = the concentration of the sample solution in milligrams per 10 mL135Chapter 3 ExperimentalSolvents and Reagents: Unless otherwise specified, all the solvents and reagentswere used directly without any further purification. When further purification wasneeded, known methods and procedures were followed in each case.'° 5Crystallographic Analyses: All crystal structures were determined on a Rigaku4-circle diffractometer by the following people M. Kaftory, A.D. GuOmundsdOttir, W.Li, S.J. Rettig, J. Trotter of the University of British Columbia Chemistry Department.3.2 Preparation of Starting Materials3.2.1 Synthesis of Starting MaterialsDimethyl 9,10-Dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (45)46In a round bottom flask equipped with a condenser, a mixture of anthracene (7 g,39 mmol, Eastman) and dimethyl acetylenedicarboxylate (6 mL, 49 mmol, Aldrich) washeated to 190°C. After 30 min of heating, the mixture was cooled and the resulting darksolid recrystallized three times from a chloroform-ethanol solution to yield diester 45 (18g, 56 mmol, 72% yield).MP: 160-161°C (lit a6 160-161°C).IR (KBr) vmax: 1729, 1713 (C=0), 1632 (C=C), 1273 (C-0) cm- 1 .MS ink (relative intensity): 320 (M+, 22), 261 (62), 202 (25), 178 (83), 28 (100).Exact mass calculated for C201-11604: 320.1049. Found: 320.1041136Chapter 3 Experimental11I-NMR (400 MHz, CDC13) 8 7.4 -7.3 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 5.48 (s, 2H, bridgehead H), 3.76 (s, 6H, CO2CEI3) ppm.13C-NMR (50 MHz, CDC13) 8 165.9 (C=O), 147.0 (vinyl C), 143.8 (aromatic C), 125.4,123.8 (aromatic C-H), 52.5, 52.3 (CO,CH3 and bridgehead C-H) ppm.9,10-Dihydro-9,10-ethenoanthracene-11,12-dicarboxylic acid (47) 50,106Diester 45 (18 g, 56 mmol) was dissolved in ethanol (65 mL) and aqueous NaOH(30%, 110 mL) added. This solution was refluxed for 24 h, cooled to room temperatureand washed two times with diethyl ether to remove all unreacted starting material. Theaqueous fraction was cooled to 0°C, conc. HC1 added until acidic and then extractedtwice with diethyl ether. The combined organic fractions were dried over MgSO4 andevaporated to dryness. The resulting powder was recrystallized from n-hexane-acetonesolution to yield di-acid 47 (15 g, 52 mmol, 92%).MP: 214-215°C (lit. 106 215-216°C).IR (KBr) vmax : 3600-2400 (OH and CH), 1696 (C=O) cm -1 .MS Li* (relative intensity): 292 (M+, 0.8), 274 (21), 248 (33), 230 (30), 202 (100), 178(40).Exact mass calculated for C18H1204: 292.0736. Found 292.0737.1H-NMR (300 MHz, CDC13) 6: 7.5 -7.4 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 6.03 (s, 2H, bridgehead H) ppm, no 0-H signals detectable.13C-NMR (100 MHz, CD3OD) 6 167.1 (C=0), 147.7 (vinyl C), 143.7 (aromatic C),124.6, 123.0 (aromatic C-H), 52.4 (bridgehead C-H) ppm.137Chapter 3 Experimental9,10-Dihydro-9,10-ethenoanthracene-11,12-dicarboxylic acid anhydride (48) 46A mixture of di-acid 47 (6 g, 20 mmol) and oxalyl chloride (3.5 mL, 40 mmol) inanhydrous methylene chloride (60 mL) was refluxed for 24 h. The excess oxalyl chlorideand solvent were removed under vacuum. The resulting solid was recrystallized from achloroform-ethyl acetate solution to yield anhydride 48 (4.5 g, 16 mmol, 82%).MP: 252-254°C (lit.46 247°C).IR (KBr) vex: 1843 (C=0 asym.), 1787, 1767 (C=0 sym.), 1636 (C=C) cm- 1 .MS m/e (relative intensity): 274 (W, 15), 230 (22), 202 (100), 178 (17).Exact mass calculated for C 1 8111003: 274.0630. Found 274.0633.1H-NMR (400 MHz, CDC1 3) 8 7.5-7.0 (m, 8H, aromatic H), 5.55 ppm (s, 2H, bridge-head H) ppm."C-NMR (50 MHz, CDC1 3) 6 160.3, 159.5 (C=0 and vinyl C), 142.8 (aromatic C),126.1 124.9 (aromatic C-H). 47.9 (bridgehead C-H) ppm.Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic acid (36)A solution of anhydride 48 (4.5 g, 16 mmol) in ethanol (75 mL) was refluxed for6 h, after which the solvent was removed under vacuum and the remaining solidrecrystallized from acetonitrile to give acid 36 (4.0 g, 13 mmol, 76% yield).MP: 224-226°C.IR (KBr) va.: 3500-3400 (OH), 1734, 1678 (C=0), 1629 (C=C), 1282 (C-0) cm-I.MS m/e (relative intensity): 320 (M+, 3.5), 276 (22), 247 (23), 230 (22), 202 (100), 178(32).138Chapter 3 ExperimentalExact mass calculated for C20 H1604: 320.1049. Found: 320.1041.1H-N1VIR (400 MHz, CDC13) 6 7.5-7.4 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 6.19 (s, 1H, bridgehead H), 5.83 (s, 1H, bridgehead H), 4.45 (q, 2H, J=7 Hz,CO7CH2CH3), 1.48 (t, 3H, J=7 Hz, CO2CH7C113) ppm, no 0-H signal detectable.13C-NMR (75 MHz, CDC13) 8 168.2, 163.0 (C=0), 153.0, 145.9 (vinyl C), 143.7, 143.1(aromatic C), 126.0, 125.1, 124.5, 123.4 (aromatic C-H), 64.3 (CO7CH,CH3), 54.3, 52.8(bridgehead C-H), 14.0 (CO2CH2CH3) ppm.UV (Acetonitrile) A...: 212 (E 19,000), 278 (e 1,600) nm.Anal. calculated for C20111604: C, 74.99; H, 5.03. Found: C, 74.79; H, 5.11.Methyl 12-Methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylate (41)A Carius tube containing methyl 2-butynoate (5 mL, 4.9 g, 50 mmol, Aldrich)and anthracene (5 g, 28 mmol) was sealed and heated at 175°C for 60 h. The resultingyellow oil was purified by column chromatography (silica gel, petroleum ether:ethylacetate, 80:20) and then recrystallized twice from ethanol to yield ester 41 (6.8 g, 25mmol, 88%).MP: 135-137°C (lit." 136.9-137.3°C) with immediate resolidification and remelting at145-146°C (lit." 146.1-146.5°C).IR (KBr) v„,„„: Crystals from ethanol; 1704 (C=0), 1633 (C=C), 1225 (C-0) cm-1, crys-tals from melt; 1699 (C--0), 1625 (C--C), 1228 (C-0) cm- 1 .MS ink (relative intensity): 276 (M+, 49), 217 (100), 202 (32), 178 (42).Exact mass calculated for C19111602: 276.1151. Found: 276.1152.139Chapter 3 Experimental1H-NMR (400 MHz, CDC13) 8 7.4-7.3 (m, 4H, aromatic H), 7.1-6.9 (m, 4H, aromaticH), 5.68 (s, 1H, bridgehead H), 4.90 (s, 1H, bridgehead H), 3.75 (s, 3H, CO,CH3) 2.45(s, 3H, vinyl CH3) ppm.13C-NMR (50 MHz, CDC13) 8 166.4 (C:-.)), 162.4 (vinyl C), 145.4, 144.0 (aromatic C),135.3 (vinyl C), 125.3, 124.9, 123.32, 123.28 (aromatic C-H), 60.0 (bridgehead C-H),51.5, 51.4 (bridgehead C-H and CO2CH3), 19.56 (vinyl CH3) ppm.UV (Ethanol) kn.: 214 (e 10,000), 271 (E 1,200), 279 (e 1,300) nm.12-Methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid (37) 50The procedure used for the hydrolysis of ester 45 was applied to ester 41 (1.44 g,5.35 mmol), yielding acid 37 (0.85 g, 3.23 mmol, 62%), which was recrystallized fromethanol to give prism like crystals.MP: 278-280°C (lit." 279.8-280.0°C).IR (KBr) v..: 3400-3000 (OH), 1679 (C=O), 1638 (C=C), 1193 (C-0) cm -1 .MS ink (relative intensity): 262 (M+, 61), 217 (100), 202 (87), 178 (48).Exact mass calculated for C18141402: 262.0994. Found: 262.0996.III-NMR (400 MHz, CDC13) 6 7.4-7.3 (m, 4H, aromatic H), 7.0-6.9 (m, 4H, aromaticH), 5.68 (s, 1H, bridgehead H), 4.94 (s, 1H, bridgehead H), 2.42 (s, 3H, vinyl CH 3) ppm,no 0-H signal detectable.13C-NMR (50 MHz, CDC13) 6 171.6 (C=0), 165.7 (vinyl C), 145.2, 143.5 (aromatic C),134.7 (vinyl C), 125.4, 124.8, 123.31, 123.28 (aromatic C-H), 60.2, 51.0 (bridgehead C-H), 19.8 (vinyl CH3) ppm.UV (Ethanol) XT.: 219 (c 9,800), 271 (E 2,300), 279 (e 2,600) nm.140Chapter 3 ExperimentalEthyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate (42) 5°A Carius tube containing anthracene (5.0 g, 28 mmol) and ethyl propiolate (2.8mL, 2.7 g, 28 mmol, Aldrich) was sealed and heated at 175°C for 2.5 h. Unreactedstarting materials were removed from the Diels Alder adduct by column chromatographyon silica gel eluted with 20% ethyl acetate in petroleum ether. The crude product wasrecrystallized from ethanol to yield crystals of ester 42 (3.0 g, 11 mmol, 40%).MP: 103-104°C (lit." 111.5-112.5°C).lR (KBr) vmax : 1699 (C=O), 1619 (C=C) cm-1.MS m/e (relative intensity): 275 (M+, 38), 203 (100), 178 (47), 89 (39).Exact mass calculated for C19l-11602: 276.1151. Found: 276.1151.111-NMR (400 MHz, CDC13) 8 7.86 (dd, 1H, J=2 Hz and J=7 Hz, vinyl C-H), 7.4-7.3 (m,4H, aromatic H), 6.9-7.0 (m, 4H, aromatic H), 5.68 (d, 1H, J=2 Hz, bridgehead H), 5.21(d, 1H, J=7 Hz, bridgehead H), 4.18 (q, 2H, J=7 Hz, CO2_CH2CH 3), 1.25 (t, 3H, J=7 Hz,CO2CH7CF13) ppm.13C-NMR (50 MHz, CDC1 3) 8 164.8 (C=0), 149.3 (vinyl C-H), 145.4 (aromatic C),144.8 (vinyl C), 144.5 (aromatic C), 125.1, 124.9, 123.8, 123.5 (aromatic C-H), 60.7(CO,CH,CH3) 51.6, 50.4 (bridgehead C-H), 14.3 (CO2CH,CH 3) ppm.UV (Ethanol) X.: 222 (c 6,400), 271 (e 2,400), 279 (c 2,800) nm.9,10-Dihydro-9,10-ethenoanthracene-11 -carboxylic acid (40)50The same procedure was followed as in the saponification of ester 45. Startingwith ester 42 (2.8 g, 10 mmol), acid 40 (0.7 g, 2.8 mmol, 28% yield) was obtained andrecrystallized from acetonitrile to give transparent prisms.141Chapter 3 ExperimentalMP: 253-254°C (lit." 249.2-250.0°C.).IR (KBr) vmax: 3400-2600 (OH), 1675 (C=O), 1611 (C=C), 1233 (C-0) cm- 1 .MS ink (relative intensity): 248 (M+, 53), 203 (100), 178 (11), 101 (21).Exact mass calculated for C1411202: 248.0838. Found: 248.0838.11I-NMR (400 MHz, DMSO-d6) 8 12.48 (s, 1H, 1)20 exchangeable), 8.84 (dd, 1H, J=2Hz and J=7 Hz, vinyl C-H), 7.5-7.4 (m, 4H, aromatic H), 7.0-6.9 (m, 4H, aromatic H),5.63 (d, 1H, J=2 Hz, bridgehead H), 5.42 (d, 1H, J=7 Hz, bridgehead H) ppm.13C-NMR (50 MHz, DMSO-d6) 8 165.8 (C=0), 149.4 (vinyl C-H), 145.6, 144.9(aromatic C), 144.0 (vinyl C), 124.84, 124.77, 123.76, 123.6 (aromatic C-H), 50.7, 49.9(bridgehead C-H) ppm.UV (Ethanol)^224 (e 5,400), 271 (e 1,700), 279 (c 2,100) nm.9-Aminoanthracene (136)47A suspension of 9-nitroanthracene (2.0 g, 8.7 mmol, Aldrich) and SnC1 2•H20 (14g, 62 mmol) in conc. HCl (14 mL) was refluxed for 1 h. The solution was cooled to 0°C,made basic by the addition of 60% aqueous NaOH and extracted three times with diethylether. The combined organic fractions were washed with saturated NaCI solution anddried over MgSO4. The solvent was removed under vacuum to yield 9-aminoanthracene(1.5 g, 7.9 mmol, 88%).MP: 140-145°C (•m 148-51°C).(KBr)^3600-3400 (N-H), 1623 (N-H) cm-1.MS ink (relative intensity): 193 (M+, 100), 179 (39), 165 (55), 152 (44).Exact mass calculated for C14Hi IN: 193.0892. Found: 193.0884.1H-NMR (200 MHz, CDC1 3) 8 7.5-7.0 (m, 9H, aromatic C-H) ppm, no detectable N-H.142Chapter 3 Experimental13C-NMR (50 MHz, CDC13) 8 137.9, 132.1 (aromatic C), 128.4, 125.2, 123.8, 121.1(aromatic C-H) 118.2 (aromatic C), 116.3 (aromatic C-H) ppm.Dimethyl 9 -Amino-9,10-dihydro-9,10-ethenoanthracene -11,12-dicarboxylate (39)63An excess of dimethyl acetylenedicarboxylate (2 mL, 16 mmol) was added to asolution of 9-aminoanthracene (1.5 g, 7.8 mmol) in benzene at 0°C (15 mL). Theresulting white precipitate was filtered and recrystallized from ethanol to give needles ofcompound 39 (1.8 g, 5.3 mmol, 68% yield).MP: 220-222°C (lit.63 220-222°C).ER (KBr) vmax: 3382, 3324 (N-H), 1747, 1738, 1719, 1698 (C=O), 1638 (C=C), 1297(C-0) cm- 1.MS m/e (relative intensity): 335 (M+, 15), 276 (34), 244 (12), 216 (21), 193 (100).Exact mass calculated for C20H17N04: 335.1158. Found: 335.1166.311-NMR (400 MHz, CDC13) 6 7.5-7.3 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 5.62 (s, 1H, bridgehead H), 3.83 (s, 3H, CO203_3), 3.76 (s, 3H, CO2_CH_3), 2.45 (br. s,2H, 1320 exchangeable, N-1-12) ppm.13C-NMR (50 MHz, CDC1 3) 6 166.9, 163.8 (C=0), 155.4 (vinyl C), 145.3, 144.0(aromatic C), 141.0 (vinyl C), 125.6, 125.2, 123.6, 119.8 (aromatic C-H), 66.4(bridgehead C), 52.7, 52.4 (CO2CH 3), 49.6 (bridgehead C-H) ppm.UV (Acetonitrile)^219 (E 16,000), 270 (c 2,600), 278 (c 2,800) nm.143Chapter 3 ExperimentalMethyl 9-Anthraceneearboxylate (137)48A suspension of 9-anthracenecarboxylic acid (7.0 g, 32 mmol, Aldrich) andSOC12 (10 mL) in anhydrous chloroform (100 mL) was refluxed for 0.5 h. The solventand excess SOC12 were removed under vacuum to yield a yellow solid assumed to be thecorresponding acyl chloride. The acyl chloride was dissolved in anhydrous chloroform(100 mL) and anhydrous methanol (30 mL) was added. This solution was refluxed for 20h and the solvent removed under vacuum. The residue was dissolved in diethyl ether,washed three times with 10% aqueous NaOH, two times with a saturated aqueous NaCIsolution and dried over MgSO4 . Evaporation of the solvent resulted in a yellow solidwhich was recrystallized from ethanol to yield ester 137 (6.2 g, 26 mmol, 84%).MP: 110-112°C (lit." 112-113°C).IR (KBr) vex : 1729 (C=O), 1625 (C=C), 1209 (C-0) cm-1MS m/e (relative intensity): 236 (M+, 83), 205 (100), 177 (86).Exact mass calculated for C16141202: 236.0838. Found: 236.0846.1H-NMR (400 MHz, CDC13) 8 8.52 (m, 1H, aromatic H), 8.1-8.0 (m, 4H, aromatic H),7.6-7.4 (m, 4H, aromatic H), 4.18 (s, 3H, CO2013) ppm.13C-1sIMR (50 MHz, CDC13) 8 170.1 (C=O), 131.0 (aromatic C), 129.5, 128.6 (aromaticC-H), 128.5, 127.8 (aromatic C), 127.0, 125.5, 125.0 (aromatic C-H), 52.6 (CO2CHa)PPIn•144Chapter 3 ExperimentalTrimethyl 9,10-Dihydro-9,10-ethenoanthracene-9,11,12-tricarboxylate (43) andTrimethyl 1,4-Dihydro-1,4-ethenoanthracene-9,11,12-tricarboxylate (138)In a round bottom flask equipped with a condenser, ester 137 (2.0 g, 8.5 mmol)and dimethyl acetylenecarboxylate (3 mL, 3.5 g, 24 mmol) were refluxed for 15 min.The resulting brown oil was chromatographed on silica gel eluted with 30% ethyl acetatein petroleum ether. Two crude isomeric Diels-Alder adducts were obtained from theseparated fractions; 959 mg (25 mmol, 30% yield) of compound 43 and 106 mg (2.8mmol, 3% yield) of compound 138. Both isomers were purified further byrecrystallization from ethanol and characterized as follows:Compound 43 was characterized as trimethyl 9,10-dihydro-9,10-ethenoanthracene-9,11,12-tricarboxylate. This was further supported with X-ray crystal structure.67MP: 174-176°CIR (KBr) v.: 1737, 1712 (C=O), 1629 (C=C), 1286, 1263 (C-0)MS m/e (relative intensity): 378 (M+, 26), 260 (100), 201 (85), 177 (55).Exact mass calculated for C22111806: 378.1103. Found: 378.1112.111-1s1MR (400 MHz, CDC1 3) 8 7.6-7.4 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 5.62 (s, 1H, bridgehead H), 4.08 (s, 3H, CO7CH3), 3.85 (s, 3H, CO2C113), 3.75 (s,3H, CO7CH3) ppm.13C-NMR (50 MHz, CDC1 3) 8 168.8, 167.1, 163.8 (C=0), 150.34 (vinyl C), 144.1(aromatic C), 143.6 (vinyl C), 142.0 (aromatic C), 126.0, 125.4, 123.8, 123.4 (aromaticC-H), 64.1 (bridgehead C), 52.5, 52.5, 52.4 (CO2CH3), 51.1 (bridgehead C-H) ppm.UV (Ethanol) X.: 219 (c 14,000), 277 (c 2,800) nm.Anal. calculated for C22111806: C, 69.84; H, 4.80. Found C, 69.75; H, 4.70.145Chapter 3 ExperimentalCompound 138 was characterized as trimethyl 1,4-dihydro-1,4-ethenoanthracene-9,11,12-dicarboxylate.MP: 117-118°C.1R (KBr) vmax: 1729, 1713 (C=O), 1643 (C=C), 1276, 1249 (C-0) cm-1.MS ink (relative intensity): 378 (M+, 49), 346 (37), 318 (100), 287 (78), 177 (21).Exact mass calculated for C221-11806: 378.1103. Found: 378.1097.1H-NMR (400 MHz, CDC13) 8 8.0-7.9 (in, 2H, aromatic H), 7.8-7.7 (m, 2H, aromaticH), 7.5-7.4 (m, 2H, aromatic H), 7.0-6.9 (m, 2H, vinyl H), 5.65 (m, 1H, bridgehead H),5.30 (m, 1H, bridgehead H), 4.10 (s, 3H, CO2CH3), 3.81 (s, 3H, CO2CH3), 3.80 (s, 3H,CO2CH3) ppm.13C-NMR (100 MHz, CDC13) 8 168.3 165.9, 165.4 (C=O), 147.2, 145.5 140.7, 140.4(aromatic C-H and or vinyl C), 138.8, 137.8 (vinyl C-H), 131.0 (aromatic C or vinyl C),128.0 (aromatic C-H), 127.8 (aromatic C or vinyl C), 127.0, 126.5, 125.22, 125.16,123.8 (aromatic C-H), 52.41, 52.37, 52.3 (CO2CH3), 49.7, 47.4 (bridgehead C-H) ppm.Anal. calculated for C22}11806: C,69.84; H, 4.80 Found; C, 69.68; H, 4.82.Ethyl 9-Anthracenecarboxylate (139)49The procedure used to prepare methyl 9-anthracenecarboxylate (137) wasmodified by using anhydrous ethanol instead of anhydrous methanol. Starting with 9-anthracenecarboxylic acid (3.0 g, 14 mmol), crude ethyl 9-anthracenecarboxylate (139)was obtained in 91% yield (3.2g, 13 mmol) and purified further by recrystallization fromethanol.MP: 112-114°C (lit.49 102°C).IR (KBr) vinax: 1714 (C=O), 1625 (C=C), 1217 (C=O) cm-1146Chapter 3 ExperimentalMS m/e (relative intensity): 250 (M+, 92), 222 (41), 205 (100), 177 (89).Exact mass calculated for C1411402: 250.0994. Found: 250.0994.III-NMR (400 MHz, CDC1 3) 8 8.50 (s, 1H, aromatic H), 8.1-8.0 (m, 4H, aromatic H),7.6-7.4 (m, 4H, aromatic H), 4.70 (q, 2H, J=7 Hz, CO2CH 2CH3), 1.50 (t, 3H, J=7 Hz,CO2CH,CH3) ppm.13C-NMR (50 MHz, CDC13) 8 169.7 (C=0), 131.0 (aromatic C) 129.2, 128.6 (aromaticC-H), 128.7, 128.4 (aromatic C), 126.9, 125.4, 125.0 (aromatic C-H), 61.8(CO,CH,CH3), 14.5 (CO2CH2CH3) ppm.Dimethyl^9-ethoxycarbony1-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (44)In a round bottom flask fitted with a condenser, ester 139 (3.0 g, 14 mmol) anddimethyl acetylenecarboxylate (2 mL, 2.3 g, 16 mmol) were refluxed for 15 min. Theresulting yellow oil was crystallized from ethanol three times to yield ester 44 (3.0 g, 7.7mmol, 67%). Crystallization from ethanol gave two forms of crystals; prisms andneedles.MP: Prisms 147-149°C, needles 145-149°C.IR (KBr) v„,„„: Needles 1746, 1735, 1714 (C=0), 1630 (C=C), 1279, 1256 (C-0); prisms1732, 1703 (C--0), 1627 (C=C), 1290, 1262 (C-0) cm-1.MS m/e (relative intensity): 392 (M+, 65), 364 (40), 287 (43), 260 (100), 177 (23).Exact mass calculated for C23H2o06: 392.1260. Found: 392.1256.1H-NMR (400 MHz, CDC1 3) 6 7.6-7.4 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 5.64 (s, 1H, bridgehead H), 4.62 (q, 2H, J=7 Hz, CO2a12CH 3), 3.82 (s, 3H,CO7CF13), 3.78 (s, 3H, CO2CH3), 1.48 (t, 3H, J=7 Hz, CO2CH2CH3) ppm.147Chapter 3 Experimental13C-N'MR (50 MHz, CDC13) 8 168.3, 167.1, 163.9 (C=0), 150.5 (vinyl C), 144.2(aromatic C), 143.6 (vinyl C), 142.2 (aromatic C), 125.9, 125.4, 123.8, 123.5 (aromaticC-H), 63.8 (bridgehead C), 62.0 (CO7CH,CH3), 52.5, 52.3 (CO 20:13), 51.1 (bridgeheadC-H), 14.2 (CO2CH,CH3) ppm.UV (Ethanol)^216 (e 17,800) nm.Anal. calculated for C23H2006: C, 70.40; H, 5.14. Found: for plates C, 70.19; H, 5.17; forneedles C, 70.36; H, 5.05.Dimethyl 9-Formy1-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (46) 51A mixture of 9-anthraldehyde (5.0 g, 24 mmol, Aldrich) and dimethyl acety-lenedicarboxylate (4 mL, 4.4 g, 32 mmol) was refluxed for 1.5 h. The resulting yellowoil was purified on silica gel eluted with 25% ethyl acetate in petroleum ether andrecrystallized from an ether-ethanol solution to yield aldehyde 39 (5.0 g, 14 mmol, 60%yield).MP: 171-172°C (lit. 31 171-173°C).IR (KBr) vmax: 2755 (CH=O), 1728 (C=O), 1630 (C=C) cm-1 .MS rn/e (relative intensity): 348 (M+, 31), 316 (19), 290 (17), 260 (100), 202 (67).Exact mass calculated for C21111605: 348.0998. Found: 348.0991.1H-N1VIR (200 MHz, CDC13) 8 10.93 (s, 1H, CH=O), 7.5-7.0 (m, 8H, aromatic H), 5.64(s, 1H, bridgehead H), 3.85 (s, 3H, CO2Ch3), 3.80 (s, 3H, CO 2CH3) ppm.13C-NMR (100 MHz, CDC1 3) 8 197.3 (CH=O), 166.39, 163.7 (C=0), 148.6 (vinyl C),144.4 (aromatic C), 144.2 (vinyl C), 144.0 (aromatic C), 126.1, 125.3, 124.4, 122.2(aromatic C-H), 64.1 (bridgehead C), 52.5, 51.7 (CO,CH3), 50.9 (bridgehead C-H) ppm.148Chapter 3 ExperimentalDimethyl^9-Carboxy-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylatemonAldehyde 46 (1.1 g, 3.2 nunol) and resorcinol (1.5 g, 14 mmol) were dissolved intert-butanol (40 mL). A solution of NaC1O2 (1.1 g, 14 mmol) and Na 2HPO4 .7H20 (1.1 g,4 mmol) in water (10 mL) was added. This was allowed to stand for 24 h and then theaqueous layer was removed. The organic layer was extracted three times with 5%aqueous NaHCO3, the combined aqueous fractions were made acidic with conc. HC1 andextracted with diethyl ether three times. The organic fractions were washed twice withsaturated NaCI solution, dried over MgSO4 and the solvent was removed under vacuum.The residue was recrystallized from ethanol to give crystals of acid 38 (964 mg, 2.35mmol, 74%). 111-NMR and "C-NMR showed that these crystals contained a 1:1 ratio ofacid 38 and ethanol. Recrystallization from acetonitrile gave crystals which containedsolvent molecules in the crystal lattice, while recrystallization from ethyl acetate and sec-butanol gave crystals without any solvent molecules. The structure of the crystalsrecrystallized from ethanol was further supported by X-ray analysis. 79MP: Crystals from ethyl acetate 210-212°C, crystals from acetonitrile 208-212°C,crystals from ethanol 210-212 °C.IR (KBr) vn.: Crystals from ethyl acetate 3600-3200 (0-H), 1751, 1716, 1699 (C=0),1625 (C=C), 1296 (C-0) cm- 1 .Crystals from acetonitrile 3200-2600 (0-H), 1724, 1702 (C=0), 1624 (C=C), 1290, 1252(C-0) cm-1•Crystals from ethanol 3500-2500 (0-H and C-H), 1703 (C=0), 1621 (C=C), 1296, 1261(C-0) cm-1.MS m/e (relative intensity): 364 (M+, 17), 332 (34), 260 (100), 229 (42), 201 (43).Exact mass calculated for C21111606: 364.0947. Found: 364.0938.149Chapter 3 Experimental1H-NMR (200 MHz, CDC13) Crystals from ethyl acetate 6 8.8 (br s, 1H, D20exchangeable) 7.8-7.7 (m, 2H, aromatic H), 7.5-7.4 (m, 2H, aromatic H), 7.2-7.0 (m, 4H,aromatic H), 5.65 (s, 1H, bridgehead H), 3.82 (s, 3H, CO2CH3), 3.78 (s, 3H, CO2CH3)ppm.Crystals from acetonitrile 6 7.7-7.6 (m, 2H, aromatic H), 7.4-7.3 (m, 2H, aromatic H),5.50 (s, 1H, bridgehead H), 5.40 (br s, 1320 exchangeable), 3.78 (s, 3H, CO2CH3), 3.70(s, 3H, CO2CH3), 1.98 (s, 3H, CNC_H3) ppm.Crystals from ethanol; 6 7.8-7.7 (m, 2H, aromatic H), 7.5-7.4 (m, 2H, aromatic H), 7.1-7.0 (m, 4H, aromatic H), 6.15 (br s, D20 exchangeable), 5.64 (s, 1H, bridgehead H), 3.82(s, 3H, CO2CH3), 3.80 (q, 2H, J=7 Hz, CH3CH2OH), 3.78 (s, 3H, CO2CH3), 1.38 (t, 3H,J=7 Hz, CR3CH2OH) ppm.13C-NMR (50 MHz, CDC1 3) Crystals from ethyl acetate 6 173.3, 167.4, 163.9 (C=0),150.1 (vinyl C), 144.0 (aromatic C), 143.8 (vinyl C), 141.4 (aromatic C), 126.2, 125.6,124.0 123.6 (aromatic C-H), 64.2 (bridgehead C), 52.7, 52.6 (CO2CH3), 51.1(bridgehead C-H) ppm.UV (Ethanol) Xmax: 216 (e 39,900), 277 (c 4,800) nm.Anal. calculated for crystals recrystallized from ethyl acetate C211-11606: C, 69.23; H,4.43. Found: C, 69.47; H, 4.39.150Chapter 3 Experimental3.2.2 Salt Formation of Starting Materials3.2.2.1 Salt Formation of Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic acid (36)Sodium salt of acid 36 (85)To a solution of acid 36 (339 mg, 1.06 mmol) in ethanol (9 mL) was addedNaOH (54 mg, 1.06 mmol) in water (1 mL). This solution was stirred until clear, thesolvent removed under vacuum and the resulting white solid dried at 100°C at 10- 2 nunHg.MP: Does not melt below 300°C.IR (KBr) vmax : 1708 (C=O ester), 1615 (COO - asym.), 1391 (COO- sym.) cm-'.Calcium salt of acid 36 (86)Acid 36 (333 mg, 1.04 mmol) was dissolved in ethanol (10 mL) and CaO (27.9mg, 0.50 mmol) added. This solution was stirred for 10 min. The white precipitate thatformed was filtered (230 mg, 0.339 mmol, 65% yield) and dried at 100°C under vacuum.MP: does not melt below 300°C.IR (KBr) vex : 1719 (C=O ester), 1579 (COO- asym.), 1402 (COO- sym.) cm -1 .151Chapter 3 ExperimentalDiethylamine salt of acid 36 (87)Acid 36 (242 mg, 0.756 mmol) was dissolved in 10 mL of diethyl ether and 5 mL(3.5 g, 48 mmol) of diethylamine added. The resulting precipitate was filtered andrecrystallized from acetonitrile to yield colorless needles of salt 87 (299 mg 0.761 mmol,99%).MP: 168-173°C.IR (KBr)^3100-2400 (N-H, ), 1711 (C=0 ester), 1631 (COO' asym.), 1386 (COO -sym.) cm- 1 .MS FAB: 394 (M+1).1H-NMR (400 MHz, CDC1 3) 8 7.4-7.3 (m, 4H, aromatic H), 7.0-6.9 (m, 4H, aromaticH), 5.54 (s, 1H, bridgehead H), 5.30 (s, 1H, bridgehead H), 4.15 (q, 2H, J=7 Hz,CO7CH2CH3), 2.16 (q, 4H, J=7 Hz, HN(CH,CH3)2), 1.26 (t, 3H, J=7 Hz, CO2CH7CH3),1.12 (t, 6H, J=7 Hz, HN(CH2H3)2) ppm, no 0-H or N-H signals detectable.UV (Ethanol) knax: 217 (e 13,000), 272 (e 2,200), 279 (8 2,500) nm.Anal. calculated for C24H27N04: C, 73.26; H, 6.92; N, 3.56. Found: C, 73.32; H, 7.06;N, 3.65.Pyrrolidine salt of acid 36 (88)Pyrrolidine (0.5 mL, 0.43 g, 6.0 mmol) was added to a solution of acid 36 (168mg, 0.53 mmol) in diethyl ether (15 mL). The precipitate which formed was filtered andrecrystallized from acetonitrile to yield colorless plates of salt 88 (152 mg 0.390 mmol,73%). The elemental analysis and infrared spectra of salt 88 suggest that it crystallizeswith one equivalent of H2O.152Chapter 3 ExperimentalMP: 134-140°C.IR (KBr) v..: 3600-2800 (N-H, C-H, 0-H), 1696 (C=0 ester), 1621 (COO' asym.),1386 (COO- sym.) cm- 1 .MS FAB: 392 (M+1).111-NMR (400 MHz, CDC13) 8 7.4-7.3 (m, 4H, aromatic H), 7.0-6.9 (m, 4H, aromaticH), 5.49 (s, 1H, bridgehead H), 5.29 (s, 1H, bridgehead H), 4.16 (q, 2H, J=7 Hz,CO,CH2CH3), 2.89 (m, 4H, -N(CH2)2-), 1.58 (m, 4H, -N(CH2CH7)2-), 1.28 (t, 3H, J=7Hz, CO2CH2CH3) ppm, no 0-H or N-H signals were detectable.UV (Ethanol) X,„,„: 221 (e 13,000), 272 (e 3.500), 279 (e 4,100) nm.Anal. calculated for C24H25N04•H20: C, 70.40; H, 6.65; N, 3.42. Found: C, 70.60; H,6.60; N 3.42.Piperidine salt of acid 36 (89)To a solution of acid 36 (0.43 g, 1.3 mmol) in diethyl ether (45 mL) was addedpiperidine (1.75 mL, 1.5 g, 18 mmol) and the solution stirred for 20 min. The resultingprecipitate was filtered and recrystallized from an acetonitrile-acetone solution to yieldcolorless plates of salt 89 (0.42 g, 1.0 mmol, 77%).MP: 160-162°C.IR (KBr) v.,: 3100-2400 (N-H, C-H), 1708 (C=0 ester), 1626 (C=C), 1581 (COO-asym.), 1386 (COO- sym.), 1217 (C-0) cm-1.MS FAB: 406 (M+1).1H-NMR (400 MHz, CDC13) 8 7.4-7.3 (m, 4H, aromatic H), 7.0-6.9 (m, 4H, aromaticH), 5.52 (s, 1H, bridgehead H), 5.35 (s, 1H, bridgehead H), 4.17 (q, 2H, J=7 Hz,CO,CH2CH3), 2.66 (t, 4H, J=6 Hz, -N(CH2)2-), 1.38 (m, 4H, -N(CH2CH02-), 1.18 (m,153Chapter 3 Experimental2H, -N(CH2CH2)7CH2), 1.18 (t, 3H, J=7 Hz, CO2CH2CH3) ppm, no 0-H or N-H signalsdetectable.UV (Ethanol) X.: 223 (e 12,000), 272 (e 3,700), 279 (e 4,200) nm.Anal. calculated for C251127N04: C, 74.05; H, 6.71; N, 3.45. Found: C, 73.89; H, 6.78;N, 3.33.S-(-)-Proline complex of acid 36 (90a)Acid 36 (275 mg, 0.858 mmol) and S-(-)-proline (98.2 mg, 0.854 mmol, Sigma)were dissolved in 50 mL of ethanol. Approximately half of the solvent was removedunder vacuum and the remaining solution was stirred for 0.5 h, yielding a whiteprecipitate of complex 90a (299 mg, 0.687, 80%) which was filtered and dried over P2O 5(100°C, 10-2 mmHg). All attempts to recrystallize complex 90a from different solventsfailed.MP: 173-175°C.IR (KBr) vex: 3400-2400 (0-H, N-H, C-H), 1706 (C=0 ester), 1633 (C=C), 1579(COO' asym.), 1391 (COO - sym.), 1289, 1224 (C-0) cm- 1 .MS FAB: 436 (M+1).III-NMR (400 MHz, CDC13) 6 7.5-7.4 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 6.18 (s, 1H, bridgehead H), 5.79 (s, 1H, bridgehead H), 4.44 (q, 2H, J=7 Hz,CO,CH,CH3), 4.2-4.1 (m, 1H, -NCH(CO2)-), 3.4-3.3 (m, 2H, (-NCH2-), 2.4-2.2 (m, 1H),2.2-2.0 (m, 1H), 2.0-1.9 (m, 2H), 1.45 (t, 3H, J=7 Hz, CO2CH2CHa) ppm, no 0-H or N-H signals were detectable.UV (Ethanol) Amax: 214 (c 13,000), 271 (e 1,500), 279 (e 1,600) nm.154Chapter 3 ExperimentalAnal. calculated for C25H25N06: C, 68.95; H, 5.79; N, 3.22. Found: C, 68.82; H,5.83; N, 3.29.R-(+)-Proline complex of acid 36 (90b)Complex 90b was prepared in 80% yield (267 mg, 0.613 mmol) from acid 36(246 mg, 0.768 mmol) and R-(+)-proline (88.9 mg, 0.773 mmol, Sigma) exactly asdescribed above; complexes 90a and 90b exhibited identical spectroscopic properties.MP: 173-174°C.Anal. calculated for C25H25N06: C, 68.95; H, 5.79; N, 3.22. Found C, 68.62; H, 5.87; N,3.26.(S,R)-(±)-Proline complex of acid 36 (90c)Complex 90c was formed in 72% yield from acid 36 (140 mg, 0.436 mmol) and(±)-proline (51.6 mg, 0.448 mmol, Sigma) exactly as described for complex 90a.MP: 183-186°C.IR (KBr) vim : 3200-2400 (0-H, N-H, C-H), 1704 (C=O ester), 1633 (C--C), 1582(COO- asym.), 1380 (COO' sym.), 1289, 1222 (C-0) cm-1 .MS FAB: 436 (M+1).111-NMR (400 MHz, CDC13) 6 7.5-7.4 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 6.05 (s, 1H, bridgehead H), 5.77 (s, 1H, bridgehead H), 4.42 (q, 2H, J=7 Hz,155Chapter 3 ExperimentalCO7CH2CH3), 4.1-4.0 (m, 1H, -N-), 3.4-3.1 (m, 2H, -NCH,-), 2.3-2.2 (m, 1H), 2.15-2.05 (m, 1H), 1.80 (m, 2H), 1.45 (t, 3H, CO2CH,CH3) ppm, no 0-H or N-H detectable.UV (Ethanol) X.: 214 (c 12,000), 271 (E 2,000), 279 (e 2,100) nm.Anal. calculated for C25H25N06: C, 68.95; H, 5.79; N, 3.22. Found: C, 69.13; H, 5.85;N, 3.00.S-(-)-Proline methyl ester salt of acid 36 (92)A solution of acid 36 (245 mg, 0.764 mmol), the hydrochloride salt of S-(-)-proline methyl ester (127 mg, 0.765 mmol, Aldrich) and KOH (43.9 mg, 0.782 mmol) inanhydrous ethanol (30 mL) was stirred for 0.5 h and then centrifuged for 20 min. Thewhite precipitate which formed was discarded and the solvent removed under vacuum.The residue was solidified in 2 mL of diethyl ether, filtered and recrystallized fromethanol to yield colorless prisms of salt 92 (205 mg 0.456 mmol, 60%).MP: 126-130°C.IR (103r) vmas : 3000-2000 (C-H, N-H), 1746, 1723, 1708 (C=O), 1630 (COO - asym.),1387 (COO- sym.), 1204 (C-0) cm-1.MS DCI: 450 (M+1).1H-NMR (400 MHz, CDC1 3) 8 7.4-7.3 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromaticH), 5.80 (s, 1H, bridgehead H), 5.68 (s, 1H, bridgehead H), 4.34 (q, 2H, J=7 Hz,CO2CH,CH3), 4.16 (m, 1H, -Nal-), 3.76 (s, 3H, CO2CH3), 3.11 (m, 2H, -NCH2-), 2.25(m, 1H), 2.0-1.8 (m, 3H), 1.38 (t, 3H, J=7 Hz, CO2CHICE.3) ppm, no 0-H or N-Hsignals detectable.UV (Ethanol) X.: 217 (e 13,000), 271 (e 2,500), 279 (E 2,800) nm.156Chapter 3 ExperimentalAnal. calculated for C261-127N06: C, 69.47; H, 6.05; N, 3.12. Found C, 69.30; H, 5.93; N,2.90.S-(+)-Prolinol salt of acid 36 (93)Acid 36 (255 mg, 0.798 mmol) was dissolved in diethyl ether (30 mL) and S-(+)-prolinol (0.5 mL, 513 mg, 5.08 mmol, Aldrich) was added. The precipitate whichformed was filtered and recrystallized from ethanol to yield colorless prisms of salt 93(296 mg 0.673 mmol, 84%). The elemental analysis and infrared spectra of salt 93suggested that it crystallizes with half an equivalent of water.MP: 158-165°C.IR (KBr) v.: 3600-2400 (0-H, N-H, C-H), 1699 (C=0 ester), 1625 (C=C), 1609(COO- asym.), 1387 (COO' sym.), 1289, 1225, 1207 (C-0) cm- 1 .MS FAB: 422 (M+1).III-NMR (400 MHz, CDC13) 8 7.4-7.3 (m, 4H, aromatic H), 7.0-6.9 (m, 4H, aromaticH), 5.50 (s, 1H, bridgehead H), 5.30 (s, 1H, bridgehead H), 4.20 (q, 2H, J=7 Hz,CO2CH2CH3), 3.70 (dd, 1H, J=1 and J=10 Hz), 3.45 (dd, 1H, J=6 and J=10 Hz), 3.25 (m,1H), 2.93 (m, 1H), 2.80 (m, 1H), 1.8-1.4 (m, 4H), 1.30 (t, 3H, CO2CH,CH3) ppm, no O-H or N-H signals detectable.UV (Ethanol) Xi.: 215 (E 10,000), 272 (E 1,400), 279 (e 1,600) nm.Anal. calculated for C25H27N05•1/2H20: C, 69.75; H, 6.56; N,3.25. Found: C, 69.67; H,6.55; N, 3.50.157Chapter 3 ExperimentalS-(-)-Prolineamide salt of acid 36 (94)S-(-)-Prolineamide (64.3 mg, 0.564 mmol) was dissolved in ethanol (5 mL) andadded to a solution of acid 36 (179 mg, 0.559 mmol) in diethyl ether (20 mL). Theresulting white precipitate was filtered and recrystallized from acetonitrile to yield flakesof salt 94 (220 mg 0.506 mmol, 91%).MP: 200-204°C.JR (KBr) v.: 3400-2800 (N-H, C-H), 1694, 1674 (C=0), 1623 (C=C), 1583 (C00 -asym.), 1382 (COO - sym.), 1288, 1226, 1207 (C-0) cm- 1 .MS FAB: 435 (M+1).1H-NMR (400 MHz, D20) 8 7.5-7.4 (m, 4H, aromatic H), 7.1-7.0 (m, 4H, aromatic H),5.66 (s, 1H, bridgehead H), 5.38 (s, 3H, bridgehead H), 4.35 (m, 1H, -NCH-), 4.16 (q,2H, J=7 Hz, CO2 CH,CH3), 3.4-3.3 (m, 2H), 2.42 (m, 1H), 2.12 (m 3H), 1.20 (t, 3H, J=7Hz, CO2CH2Cli1) ppm.UV (Ethanol) X.: 218 (e 13,000), 271 (e 2,500), 279 (e 2,800) nm.Anal. calculated for C25H26N205: C, 69.11; H, 6.03; N, 6.45. Found: C, 68.88; H, 5.90;N, 6.56.S-(+)-2-(Methoxymethyl)pyrrolidine salt of acid 36 (95)Acid 36 (523 mg, 1.63 mmol) was dissolved in diethyl ether (50 mL) and S-(+)-2-(methoxymethyl)pyrrolidine (200 pL, 1.69 mmol, Aldrich) added. This solution wasstirred for 10 min and the white precipitate which formed was filtered and recrystallizedfrom an acetone-petroleum ether solution to yield prisms of salt 95 (615 mg, 1.41 mmol,86%).158Chapter 3 ExperimentalMP: 164-170°C.IR (KBr) v.: 3200-2400 (N-H, C-H) 1708 (C=O ester), 1620 (C=C), 1578 (C00 -asym.), 1377 (COO- sym.), 1287, 1212 (C-0) cm- 1 .MS FAB. 436 (M+1).11 1-NMR (400 MHz, CDC13) 5 7.4-7.3 (m, 4H, aromatic H), 6.9-7.0 (m, 4H, aromaticH), 5.52 (s, 1H, bridgehead H), 5.40 (s, 1H, bridgehead H), 4.22 (q, 2H, J=7 Hz,CO,CH,CH3), 3.50 (m, 1H), 3.42 (m, 2H), 3.16 (s, 3H, CH2OCH3), 3.05 (m, 2H), 1.9-1.6 (m, 3H), 1.58 (m, 1H), 1.28 (t, 3H, J=7 Hz, CO2CH7CHO ppm, no 0-H or N-Hsignals detectable.UV (Ethanol) X.: 219 (E 10,000), 252 (e 3,200), 272 (e 2,200), 279 (e 2,500) nm.Anal. calculated for C 26H29N05 : C, 71.70; H, 6.71; N, 3.21. Found: C, 71.44; H, 6.75;N, 3.32.S-(-)-Proline tert-Butyl ester salt of acid 36 (96)Acid 36 (658 mg, 2.06 mmol) was dissolved in diethyl ether (50 mL) and S-(-)-proline tert-butyl ester (372 mg, 2.17 mmol, Sigma) was added and the solution wasstirred for 10 min. The resulting precipitate was filtered and recrystallized from acetoneto give colorless needles of salt 96 (872 mg, 1.78 mmol, 86%). The structure of this saltwas further studied by X-ray diffraction analysis. 89MP: 154-169°C.IR (KBr) v.: 3200-3000 (N-H, C-H), 1728 (C=O tert-butyl ester), 1698 (C=O ethylester), 1626 (C=C), 1589 (COO - asym.), 1372 (COO - sym.), 1252, 1155 (C-0) cm -1 .MS FAB: 492 (M+1).159Chapter 3 Experimental1H-NMR (400 MHz, CDC13) 6 8.0 (s, 2H, 1320 exchangeable), 7.4-7.3 (m, 4H, aromaticH), 7.0-6.9 (m, 4H, aromatic H), 5.55 (s, 1H, bridgehead H), 5.50 (s, 1H, bridgehead H),4.22 (q, 2H, J=7 Hz, CO2CH?CH3), 4.10 (m, 1H, -NCH-), 3.18 (m, 1H), 3.08 (m, 1H),2.22 (m, 1H), 1.88 (m, 2H), 1.72 (m, 1H), 1.46 (s, 9H, CO2C(0:31 )3), 1.28 (t, 3H, J=7Hz, CO2CH,CH3) ppm.UV (Ethanol) X.: 222 (E 19,000), 252 (e 9,600), 271 (e 4,700), 279 (E 4,800) nm.(S,S)-(+)-Pseudoephedrine salt of acid 36 (97)Acid 36 (339 mg, 1.06 mmol) was dissolved in diethyl ether (40 mL) and (S,S)-(+)-pseudoephedrine (175 mg, 1.06 mmol, Sigma) was added. The resulting solution wasstirred for 10 min and the white precipitate which formed was filtered to give salt 97(471 mg, 0.972 mmol, 92%). All attempts to recrystallize this compound from varioussolvents failed.MP: 153-169°C.ER (KBr) vim: 3600-2400 (0-H, N-H, C-H), 1703 (C=0 ester), 1626 (C=C), 1587(COO- asym.), 1387 (COO' sym.), 1292, 1224 (C-0) cm -1 .MS FAB: 486 (M+1).1H-NMR (400 MHz, CDC13) 8 7.4-7.2 (m, 9H, aromatic H), 7.0-6.9 (m, 4H, aromaticH), 5.52 (s, 1H, bridgehead H), 5.38 (s, 1H, bridgehead H), 4.78 (d, 1H, J=8 Hz,Phil-), 4.16 (q, 2H, J=7 Hz, CO7CH2CH3), 3.05 (m, 1H, PhCH(OH)CH-), 2.38 (s, 3H,-NCH3), 1.25 (t, 3H, J=7 Hz, CO2CH2CH3), 0.78 (d, 3H, J=7 Hz, PhCH(OH)CH(CH3)-)ppm, no 0-H or N-H signals were detectable.UV (Ethanol) A: 210 (e 16,000), 272 (e 1,700), 279 (e 1,900) nm.160Chapter 3 ExperimentalAnal. calculated for C301131N05: C, 74.21; H, 6.44; N, 2.88. Found: C, 73.86; H, 6.45;N, 2.88.(-) Strychnine salt of acid 36 (98)Acid 36 (325 mg, 1.02 mmol) and strychnine (340 mg, 1.02 mmol, Aldrich) weredissolved in diethyl ether (15 mL) and stirred for 30 min. The resulting precipitate wasfiltered, affording salt 98 (474 mg, 0.725 mmol, 73%). All attempts to recrystallize salt98 from various solvents failed. The elemental analysis and infrared spectra of salt 98suggest that it crystallizes with one equivalent of H2O.MP: 142-150°C.IR (KBr) v„.: 3400-3200 (0-H), 1700 (C=0 ester), 1673 (C=0 amide), 1633 (C=C),1598 (COO- asym.), 1391 (COO- sym.), 1288, 1219 (C-0) cm- 1 .MS FAB: 655 (M+1).1H-NMR (400 MHz, CDCI 3) 6 8.08 (m, 1H, aromatic H), 7.5-6.9 (m, 11H, aromatic H),6.26 (m, 1H), 5.63 (s, 1H, bridgehead H), 5.61 (s, 1H, bridgehead H), 4.63 (s, 1H), 4.8-4.0 (m, 6H), 3.97 (m, 1H), 3.82 (m, 1H), 3.28 (m, 1H), 3.2-3.0 (m, 3H), 2.68 (m, 1H),2.55 (m, 1H), 2.25 (m, 1H), 2.15 (m, 1H), 1.63 (m, 1H), 1.37 (m, 4H) ppm, no 0-Hsignal was detectable.UV (Ethanol) A.,„„„: 219 (26,000), 252 (17,000), 279 (7,400) nm.Anal. calculated for C41H3814206•H20: C, 73.18; H, 5.99; N, 4.29. Found: C, 72.98; H,6.10; N, 4.30.161Chapter 3 Experimental3.2.2.2 Salt Formation of 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylic acid(40)Piperidine salt of acid 40 (134)102To a solution of acid 40 (457 mg, 1.84 mmol) in diethyl ether (50 mL) was added2.4 mL (1.80 mmol) of piperidine. This solution was stirred for 20 min and the resultingprecipitate was filtered and recrystallized from acetonitrile to give colorless prisms of salt134 (600 mg, 1.80 mmol, 98%).MP: 195-208°C (lit. 102 197-208°C).IR (KBr) v..: 3200-2400 (N-H, C-H), 1634 (C--C), 1548 (COO - asym.), 1373 (COO-sym.) cm-1 .MS DCI: 334 (M+1).1H-NMR (400 MHz, CDC13) 5 8.8 (br s, 2H, D20 exchangeable), 7.6-7.5 (m, 1H, vinylH), 7.4-7.2 (m, 4H, aromatic H), 7.0-6.9 (m, 4H, aromatic H), 5.64 (m, 1H, bridgeheadH), 5.16 (m, 1H, bridgehead .H), 3.0-2.8 (m, 4H, -N(CH2)2-), 1.7-1.5 (m, 4H, -N(CH7C117)2-), 1.4-1.3 (m, 2H, -N(CH2CH2)2CH,) ppm.UV (Ethanol) knax: 223 (e 13,000), 272 (e 4,100), 280 (e 5,000) nm.Anal. calculated for C22H23N04: C, 79.24; H, 6.95; N, 4.28. Found: C, 79.19; H, 6.85;N, 4.12.162Chapter 3 Experimental3.2.23 Salt Formation of 12-Methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid (37)S-(+)-Prolinol salt of acid 37 (103)Acid 37 (307 mg, 1.17 mmol) was dissolved in diethyl ether (30 mL) and S-(+)-prolinol (133 mg, 1.32 mmol) was added. The precipitate which formed was filtered andrecrystallized from acetonitrile to give colorless prisms of salt 103 (395 mg, 1.09 mmol,93%). X-ray diffraction analysis 67 was performed on these crystals, which further provedthe salt formation.MP: 193-208°C.IR (KBr) v„,,,,: 3400-2200 (0-H, N-H, C-H), 1636 (C=C), 1527 (COO - asym.), 1386(COO' sym.) cm -1 .MS FAB: 364 (M+1).1H-NMR (400 MHz, CDC13) 8 7.4-6.9 (m, 8H, aromatic H), 5.58 (s, 1H, bridgehead H),4.80 (s, 1H, bridgehead H), 3.65 (m, 1H), 3.55 (m, 1H), 3.45 (m, 1H), 3.00 (m, 2H), 2.36(s, 3H, vinyl CH3), 1.75 (m, 2H), 1.60 (m, 2H) ppm, no 0-H or N-H detectable.UV (Ethanol) X.: 215 (e 11,400); 272 (c 1,800); 280 (c 2,100) nm.Anal. calculated for C23}125NO3: C, 76.04; H, 6.94; N, 3.86. Found C, 76.12; H, 6.91; N,3.76.163Chapter 3 ExperimentalS-(-)-Proline tert-butyl ester salt of acid 37 (104)Salt 104 was prepared by dissolving acid 37 (231 mg, 0.882 mmol) in diethylether (40 mL) and adding S-(-)-proline tert-butyl ester (182 mg, 1.07 mmol). Theresulting precipitate was filtered and dried to afford salt 104 (284 mg, 0.656 mmol,74%). Crystallization from various solvents gave only an amorphous solid.MP: 153-165°C.1R (KBr)^3200-2200 (N-H, C-H), 1737 (C=O ester), 1688 (NH2+), 1631 (C=C),1541 (COO' asym.), 1370 (COO - sym.), 1236 (C-0) cm-1.MS FAB: 434 (M+1).1H-NMR (400 MHz, CDC1 3) 8 7.6 (br s, 2H, D20 exchangeable), 7.4-7.3 (m, 4H,aromatic H), 7.0-6.9 (m, 4H, aromatic H), 5.68 (s, 1H, bridgehead H), 4.90 (s, 1H,bridgehead H), 3.96 (m, 1H), 3.12 (m, 2H), 2.40 (s, 3H vinyl CH3), 2.22 (m, 1H), 1.94(m, 2H), 1.75 (m, 1H), 1.45 (s, 9H, CO2C(CH3)3) ppm.UV (Methanol) Xmax: 223 (c 13,400); 271 (e 4,700); 279 (5,500) nm.(S,S)-(+)-Pseudoephedrine salt of acid 37 (105a)Salt 105a was prepared by dissolving acid 37 (362 mg, 1.38 mmol) and (S,S)-(+)-Pseudoephedrine (229 mg, 1.39 mmol) in diethyl ether (30 mL). The resulting whiteprecipitate was filtered and recrystallized from acetonitrile to afford colorless prisms ofsalt 105a (480 mg, 1.12 mmol, 81%). The structure of salt 105a was studied in detail byX-ray diffraction analysis.67MP: 174-183°C.164Chapter 3 ExperimentalIR (KBr) v.: 3600-2400 (0-H, N-H, C-H), 1629 (C=C), 1548 (COO - asym.), 1376(COO' sym.) cm-1.MS FAB: 428 (M+1).1H-NMR (400 MHz, CDC13) 8 7.4-7.3 (m, 9H, aromatic H), 7.0-6.9 (m, 4H, aromaticH), 6.85 (br s, 2H, 1320 exchangeable), 5.68 (s, 1H, bridgehead H), 4.80 (s, 1H,bridgehead H), 4.42 (d, J=8 Hz, PhCli-), 2.94 (m, 1H, PhCH(OH)C11-), 2.45 (s, 3H,vinyl CH3), 2.30 (s, 3H, -NCH3), 0.85 (d, 3H, J=7 Hz, PhCH(OH)CHC33-) ppm.UV (Ethanol) X.: 220 (E 25,900); 258 (e 6,000); 271 (e 7,600); 280 (c 8,700) nm.Anal. calculated for C28H29NO3: C, 78.66; H, 6.84; N, 3.28. Found: C, 78.94; H, 6.78;N, 3.05.(R,R)-(-)-Pseudoephedrine salt of acid 37 (105b)Salt 105b was prepared in 69% yield (235 mg, 0.550 mmol) from acid 37 (210mg, 0.802 mmol) and (R,R)-(-)-pseudoephedrine (132 mg, 0.800 mmol) exactly asdescribed above. Salts 105a and 105b exhibited identical spectroscopic properties.MP: 174-183°C.Anal. calculated for C28H29NO3: C, 78.66; H, 6.84; N, 3.28. Found: C, 78.62; H, 7.00;N, 3.21.165Chapter 3 Experimental3.2.2.4 Salt Formation of 9-Carboxy-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (38)(R, S)-(-)-Ephedrine salt of acid 38 (108)Acid 38 (145 mg, 0.333 mmol) and (R,S)-(-)-ephedrine (80.2 mg, 0.486 mmol)were dissolved in diethyl ether (30 mL) and the solution stirred. The resulting precipitatewas filtered and recrystallized from acetonitrile to yield colorless flakes of salt 108 (167mg, 0.316 mmol, 89%). X-ray diffraction analysis revealed that salt 108 crystallizes withhalf an equivalent of H20. 69MP: 155-160°CIR (KBr) v.: 3600-2400 (0-H, N-H, C-H), 1708 (C=O), 1631 (COO - asym), 1392(COO' sym.), 1287 (C-0) cm-1.MS FAB: 530 (M+1).1H-NMR (400 MHz, CDC13) 8 8.04 (m, 1H, aromatic H), 7.90 (m, 1H, aromatic H),7.5-7.0 (m, 11H, aromatic H), 5.63 (s, 1H, bridgehead H), 5.55 (s, 1H, PhCH-), 3.82 (s,3H, CO2_CH3), 3.80 (s, 3H, CO2 3), 3.48 (m, 1H, PhCH(OH)H-), 2.42 (s, 3H, Nat),1.10 ppm (d, 3H, J=7 Hz, PhCH(OH)CHCH3-) ppm, no 0-H or N-H signals detectable.UV (Ethanol) ?: 217 (E 16,300); 278 (e 2,500) nm.Anal. calculated for C3 11-1311■107•1/2•H20: C, 69.13; H, 5.99; N, 2.60. Found C, 68.89; H,6.08; N, 2.67.166Chapter 3 Experimental(S,S)-(+)-Pseudoephedrine salt of acid 38 (109)Salt 109 was prepared by dissolving acid 38 (115 mg, 0.281 mmol) and (S,S)-(+)-pseudoephedrine (47.1 mg, 0.286 mmol) in diethyl ether (50 mL), this solution wasstirred for 24 h. The resulting precipitate was filtered and recrystallized from diethylether to afford fine needles of salt 109 (138 mg, 0.261 mmol, 93%). Elemental analysisrevealed that salt 109 crystallizes with half an equivalent of H2O.MP: 164-168°C.IR (KBr) v.: 3600-2600 (0-H, N-H, C-H), 1713 (C=0), 1611 (COO - asym.), 1386(COO' sym.), 1275 (C-0) cm- 1 .MS FAB: 530 (M+1).1H-NMR (400 MHz, CDC13) 8 8.0-7.9 (m, 2H, aromatic H), 7.5-7.3 (m, 7H, aromaticH), 7.1-7.0 (m, 4H, aromatic H), 5.62 (s, 1H, bridgehead H), 4.88 (d, 1H, J=8 Hz,PhE-), 3.75 (s, 3H, CO2a13), 3.74 (s, 3H, CO2CH3), 3.40 (m, 1H, PhCH(OH)CLI-),2.70 (s, 3H, NCH3), 1.10 (d, 3H, J=7 Hz, PhCH(OH)CHCH3-) ppm, no 0-H or N-Hsignals detectable.UV (Ethanol)^215 (c 15,300); 278 (c 1,300) nm.Anal. calculated for C311-131NO7•Y2.H20: C, 69.13; H, 5.99; N, 2.60. Found: C, 68.87; H,5.93; N, 2.56.167Chapter 3 Experimental3.2.2.5 Salt Formation of Dimethyl 9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)Hydrochloride salt of amine 39 (112)The hydrochloride salt 112 was prepared by mixing a solution of amine 39 (512mg, 1.53 mmol) in ethanol (10 mL) with conc. hydrochloric acid (10 mL). Uponevaporation of the solvent, colorless crystals of salt 112 (495 mg, 87%) were formed.Recrystallization from ethanol-water gave platelets.MP: 194-212°C111 (KBr) v.: 3600-2200 (N-H, C-H), 1718 (C=O), 1632 (C=C), 1584, 1528 (NH 3+),1275, 1224 (C-0) cm -1 .11-1-NMR (400 MHz, acetone-d6) 8 7.65 (m, 4H, aromatic H), 7.25 (m, 4H, aromatic H),5.80 (s, 1H, bridgehead H), 3.82 (s, 3H, CO7CH3), 3.80 (s, 3H, CO2CH3) ppm, no N-Hsignals detectable.UV (Acetonitrile) Amax : 213 (E 8,000) nm.Anal. calculated for C201-118NO4C1: C, 64.61; H, 4.87; N, 3.77. Found C, 64.38; H, 5.12;N, 3.99.R-(-)-10-Camphorsulfonic acid salt of amine 39 (116a)Salt 116a was prepared by dissolving a 1:1 mixture of R-(-)-10-camphorsulfonicacid (343 mg, 1.48 mmol, Aldrich) and amine 39 (503 mg, 1.50 nunol) in hot ethyl168Chapter 3 Experimentalacetate (55 mL) and allowing the solution to stand for 2 days. The resulting solid wasfiltered and dried to afford crystals of salt 116a (721 mg, 1.27 mmol, 86%). X-raydiffraction analysis67 of salt 116a revealed that it crystallizes with inclusion of oneequivalent of H2O.MP: 183-184°C.IR (KBr) v...: 3200-2200 (N-H, C-H), 1737 (C=0), 1634 (C=C), 1558 (NH3+), 1275(C-0), 1177, 1043 (S=0).MS FAB: 568 (M+1).111-NMR (400 MHz, CDC13) 8 8.40 (m, 1H, aromatic H), 8.10 (m, 1H, aromatic H), 7.5-7.0 (m, 6H, aromatic H), 5.60 (s, 1H, bridgehead H), 4.05 (s, 3H, CO7CH3), 3.80 (s, 3H,CO7CH3), 3.35 (m, 1H), 2.50 (m, 1H), 2.35 (m, 1H), 2.10 (m, 1H), 1.8-1.5 (m, 3H),1.25 (m, 1H), 1.0 (br m, 1H), 0.35 (br s, 3H), 0.10 (br s, 3H) ppm, no 0-H or N-Hsignals detectable.UV (Acetonitrile) knax : 215 (e 20,000) nm.Anal. calculated for dried powder C301-133NO8S: C, 63.48; H, 5.86; N, 2.47. Found: C,63.19; H, 6.03; N, 2.39. Anal. calculated for crystals C30}135N09S•H20: C, 61.52; H,6.02; N, 2.39. Found: C, 61.58; H, 6.06; N, 2.40.S-(+)-10-Camphorsulfonic acid salt of amine 39 (116b)Salt 116b was prepared in 92% yield (407 mg, 0.717 mmol) from S-(+)-10-camphorsulfonic acid (181 mg, 0.780 mmol, Aldrich) and amine 39 (261 mg, 0.779mmol) exactly as described above. Salts 116a and 11613 exhibited identical spectroscopicproperties.169Chapter 3 ExperimentalMP: 183-185°C.Anal. calculated for dried crystals C3 0H33NO8S: C, 63.48; H, 5.86; N, 2.47. Found: C,63.46; H, 6.00; N, 2.29.(-)-3-Bromocamphor-8-sulfonic acid salt of amine 39 (117)Salt 117 was prepared by dissolving a mixture of hydrochloride salt 112 (252 mg,0.678 mmol) and ammonium (-)-3-bromocamphor-8-sulfonate (223 mg, 0.680 mmol,Aldrich) in hot ethanol (50 mL) and allowing the solution to stand at room temperaturefor 2 days. The resulting solid was filtered and dried to afford crystalline salt 117 (374mg (0.648 mmol, 81%) which, according to mass spectra and elemental analysis,crystallizes with one equivalent of HCI.MP: 210-212°C.IR (KBr) v.: 3200-2200 (N-H, C-H), 1755, 1718 (C=0), 1630 (C=C), 1573 (NH 3+),1291 (C-0), 1178, 1039 (S=O) cm-1.MS FAB: 684 (M+1).1H-NMR (400 MHz, CDC13) 6 10.6 (br s, 1H, D20 exchangeable), 8.60 (m, 1H,aromatic H), 8.25 (m, 1H, aromatic H), 7.50 (m, 2H, aromatic H), 7.25-7.10 (m, 4H,aromatic H), 5.70 (s, 1H, bridgehead H), 4.10 (m, 1H), 3.95 (s, 3H, CO2(113), 3.80 (s,3H, CO2CH3), 2.6-2.5 (m, 3H), 1.85 (br s, 2H, D20 exchangeable, N-H), 1.25 (m, 1H),0.8 (m, 3H), 0.75 (br s, 311), 0.70 (br s, 3H) ppm.UV (Acetonitrile) A.: 212 (e 18,000)Anal. calculated for C30H32NO8SBr•HC1: C, 52.72; H, 4.87; N, 2.05. Found: C, 52.40; H,4.85; N, 1.94.170Chapter 3 Experimental3.3 Photochemical studies33.1 General ProceduresAll analytical photolyses, both in solution and in the solid state, were carried outusing Pyrex (X>290 nm) or quartz (b.200 nm) filtered light from a 450W Hanoviamedium pressure mercury lamp at room temperature unless otherwise stated. Spectralgrade solvents (BDH) were used for solution phase photolyses and the concentration ofthe sample solution was kept at ca. 0.01 M. The samples were degassed by three freeze-pump-thaw cycles and sealed with paraffin film under a nitrogen atmosphere in each caseprior to irradiation. Photolyses in the solid state were conducted either on single crystalsin Pyrex or quartz tubes sealed under nitrogen or on powders sandwiched between twoPyrex or quartz glass plates. The photolyzed samples were normally analyzed by GC,GC-MS and 'H-NMR.For preparative scale photolyses in solution, 50 mg to 500 mg of the appropriatecompound was dissolved in spectral grade solvent and put into a preparative photolysisapparatus.'°8 Oxygen was purged by passing a steady flow of nitrogen through thesolution for 30 min prior to and during irradiation. The light source used was a 450 WHanovia medium pressure mercury lamp and the desired output wavelength was achievedusing a Pyrex glass filter.Samples were prepared for solid state photolysis by crushing crystals betweentwo Pyrex or quartz glass plates and sliding the top and bottom plates back and forth soas to distribute the crystals over the surface in a thin, even layer. The sample plates were171Chapter 3 Experimentalthen taped together at the top and bottom ends, placed in polyethylene bags andthoroughly degassed with nitrogen. The bags were sealed under a positive pressure ofnitrogen with a heat-sealing device and irradiated with the output from a 450 W Hanoviamedium pressure mercury lamp equipped with Pyrex or quartz glass filters.Low temperature solid state photolyses were carried out by maintaining thesample in a cooling bath controlled by the Cryocool CC-100-II immersion coolingsystem from Neslab Instruments Inc. The temperature was kept within ± 2°C and thesamples irradiated with a Hanovia medium pressure mercury lamp. The desired outputwavelength was achieved using a Pyrex glass filter.3.3.2 Diazomethane WorkupAfter irradiation of dibenzobarrelene acid derivatives 36, 37, 38 and 40, thereaction mixtures were treated with an ethereal solution of diazomethane to esterify thestarting materials as well as the photoproducts. The resulting solutions were left in thefume hood until the yellow diazomethane color had disappeared. At that time the methylester of dibenzobarrelene derivatives had undergone further derivatization to formpyrazoline cycloaddition products with diazomethane, while the corresponding methylesters of the dibenzosemibullvalene photoproducts did not react further. 57. 109 Thissimplified the chromatographic separation of the starting materials and thephotoproducts, because the pyrazoline derivatives were found to elute much faster fromsilica gel than the dibenzosemibullvalene compounds.The disadvantage of this method is that the dibenzocyclooctatetraenes formedfrom unsensitized irradiation of acid 38 undergo cycloaddition with diazomethane andcan not be separated from the derivatized starting material.172Chapter 3 Experimental333 Photolyses of Starting MaterialsPhotolysis of Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic acid (36)A solution of acid 36 (80.6 mg, 0.252 mmol) in anhydrous benzene (50 mL) wasphotolyzed through a Pyrex filter for 12 h. The solvent was removed under vacuum andthe remaining oil was treated with ethereal diazomethane. GC (DB-17) showed formationof photoproduct 32a with no remaining starting material. The solvent was removed undervacuum and the residue purified on silica gel eluted with 10% ethyl acetate in petroleumether. The resulting oil was recrystallized twice from ethanol-petroleum ether to givephotoproduct 32a (60.4 mg, 0.183 mmol, 73%).Compound^32a^was^characterized^as^4b,8b,8c,8d-tetrahydro-dibenzo[a,ficyclopropa[c,d]pentalene-8b,8c-dicarboxylic acid, 8b-ethyl 8c-methyl ester.The 1 H-NMR spectrum is identical to that reported by Garcia-Garibay. 57MP: 86-88°C.1R (IMO v..: 1733, 1717 (C=O), 1246 (C-0) cm -1 .MS m/e (relative intensity): 334 (M+, 47), 274 (100), 261 (94), 202 (42).Exact mass calculated for C21111804: 334.1205. Found: 334.1204.1H-NMR (400 MHz, CDC13) 6 7.4-7.0 (m, 8H, aromatic H), 5.06 (s, 1H, benzylic H),4.69 (s, 1H, cyclopropyl H), 4.18 (q, 2H, J=7 Hz, CO2CH7CH3), 3.87 (s, 3H, CO2CH3),1.26 (t, 3H, J=7 Hz, CO2CH2CH3) PPm.13C-NMR (50 MHz, CDC1 3) 8 169.0, 168.6 (C=O), 150.04, 149.96, 134.89, 133.6(aromatic C), 127.9, 127.8, 127.6, 126.9, 125.9, 125.8, 125.5, 121.4 (aromatic C-H),173Chapter 3 Experimental67.4 (cyclopropyl), 61.3 (CO 2CH)CH3), 57.5 (cyclopropyl C), 55.7 (benzylic C-H), 52.7(CO2C113), 49.2 (cyclopropyl C-H), 14.2 (CO2CH7CH3) ppm.Crystals of acid 36 (252 mg, 0.788 mmol) were crushed between fourmicroscopic plates and irradiated through a Pyrex filter for 36 h. The crystals, which hadturned brown and sticky, were dissolved in diethyl ether and treated with excessdiazomethane. CG (DB-17) indicated 20% conversion with formation of photoproducts32a and 33a in the ratio 1:9. The solvent was removed under vacuum and the residuechromatographed on silica gel eluted with 20% ethyl acetate in petroleum ether to yieldcompounds 32a and 33a (52.4 mg, 0.157 mmol, 20%) in the ratio 1:9. No productseparation could be achieved and compound 33a was therefore analyzed in the mixture.Photoproduct^33a^was^characterized^as^4b,8b,8c,8d-tetrahydro-dibenzo[a,ficyclopropa[c,d]pentalene-8b,8c-dicarboxylic acid, 8b-methyl 8c-ethyl ester.The 'H-NMR spectrum is identical to that reported by Garcia-Garibay.57IR of mixture (CDC13) v.: 1733 (C=O), 1247 (C-0) cm- '.MS m/e (relative intensity): 334 (M+, 30), 275 (24), 260 (43), 233 (21), 202 (69), 155(35), 100 (20), 91 (100).Exact mass calculated for C21111804: 334.1205. Found: 334.1197.111-NMR (400 MHz, CDC13) 8 7.4-7.0 (m, 8H, aromatic H), 5.05 (s, 1H, benzylic H),4.67 (s, 1H, cyclopropyl H), 4.35 (q, 2H, J=7 Hz, CO2CH2CH3), 3.73 (s, 3H, CO2013),1.34 (t, 3H, J=7 Hz, CO2CH,CH3) ppm.13C-NMR (50 MHz, CDC13) 6 169.4, 168.0 (C=O), 150.0, 149.9, 134.8, 133.6 (aromaticC), 127.9, 127.8, 127.6, 126.9, 125.9, 125.8, 125.5, 121.3 (aromatic C-H), 65.6(cyclopropyl C), 61.8 (CO2CH2CH3), 57.6 (cyclopropyl C), 55.8 (benzylic C-H), 52.2(CO2CR3), 49.2 (cyclopropyl C-H), 14.3 (CO2CH2CH3) ppm.174Chapter 3 ExperimentalThe regioselectivity of the di-n-methane rearrangement of acid 36 was studied indifferent solvents. Samples of acid 36 were irradiated through a Pyrex filter andafterwards the solvent was removed under vacuum. The residue was treated with etherealdiazomethane and analyzed by GC (DB-17). The results are listed in Table 3-9.Table 3-9^Medium Dependent Photochemistry of Ester-Acid 36Solvent Concentration (M) Photoproduct ratioa32a:33aBenzene 0.052 100:0II 0.010 100:0II 0.00054 100:0Acetone 0.0095 60:40Acetonitrile 0.012 57:43DMSO 0.0091 41:59Methanol 0.0090 46:59NaHCO3 (aq) 0.0083 100:0Crystalsb 0:100a The estimated error in the GC analyses is ±5%. b Conversion was kept below 10%.Photolysis of Methyl 12-Methy1-9,10-dihydro-9,10-ethenoanthracene-11-carboxylate(41)A solution of ester 41 (237 mg, 8.59 mmol) in acetone (350 mL) was purged withnitrogen for 25 min and then irradiated for 1.5 h through a Pyrex filter. GC (DB-17)indicated formation of photoproduct 52 with little remaining starting material. The175Chapter 3 Experimentalsolvent was removed under vacuum and the residue chromatographed on a silica gelcolumn eluted with 20% ethyl acetate in petroleum ether. The resulting oil wasrecrystallized from ethanol to yield crystals of photoproduct 52 (187 mg, 6.83 mmol,80%).Compound 52 was characterized as methyl 8b-methy1-4b,8b,8c,8d-tetrahydro-dibenzo[aAcyclopropa[c,d]pentalene-8c-carboxylate.MP: 92-94°C.IR (KBr) vim: 1713 (C---0), 1231 (C-0) cm -1 .MS m/e (relative intensity): 276 (M+, 32), 217 (100), 202 (69).Exact mas calculated for C19111602: 276.1151. Found: 276.1150.111-NMR (400 MHz, CDC13) 8 7.3-6.9 (m, 8H, aromatic H), 5.00 (s, 1H, benzylic H),3.72 (s, 3H, CO 2CH3), 3.62 (s, 1H, cyclopropyl H), 1.90 (s, 3H, cyclopropyl CH 3) ppm.13C-NMR (50 MHz, CDC13) 8 171.3 (C---0), 150.0, 149.1, 139.0, 136.8 (aromatic C),127.2, 126.8, 126.7, 126.5 124.7, 123.9, 121.2, 121.1 (aromatic C-H), 66.6 (cyclopropylC), 54.9 (cyclopropyl or benzylic C-H), 53.7 (cyclopropyl C), 51.8, 51.6 (cyclopropyl orbenzylic C-H and CO2CH3), 16.3 (cyclopropyl CH3) ppm.Anal. calculated for C19141602: C, 82.58; H, 5.84. Found: C, 82.58; H, 5.90.Irradiation of ester 41 in anhydrous CH2C12 through a Pyrex filter gave 52 as theonly observed photoproduct.Photolyses of ester 41 in the solid state were performed on both the low meltingcrystals, crystallized from ethanol, and the high melting crystals from the melt. Thereaction was found to be very slow when a Pyrex filter was used, so a quartz filter wassubstituted. After 18 h of irradiation, GC (DB-17) showed 30% conversion for the highmelting crystals and 9% for the low melting crystals. The only photoproduct observed inboth cases was compound 52.176Chapter 3 ExperimentalPhotolysis of 12-Methy1-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic acid(37)Analytical solution photolyses of acid 37 were carried out in acetone andchloroform. The solutions were irradiated through a Pyrex filter. The solvents wereremoved under vacuum, the residues were dissolved in diethyl ether and treated with oneequivalent of diazomethane. GC (DB-17) analyses of these solutions showed thatcompound 52 was the only photoproduct formed.Crystals of acid 37 were photolyzed through a Pyrex filter for 20 h. Thephotolyzed crystals were dissolved in diethyl ether and one equivalent of diazomethanewas added. GC (DB-17) indicated 11% conversion to form photoproduct 52.Photolysis of Dimethyl 9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)A solution of amine 39 (214 mg, 0.639 nunol) in acetone (300 mL) was purgedwith nitrogen for 30 min and irradiated. After 2 h, GC (DB-17) indicated the formationof photoproducts 60 and 61 in a 3:7 ratio with little starting material remaining. Removalof the solvent under vacuum afforded an oil which was crystallized from diethyl ether toafford photoproduct 61 (60 mg, 33%).The mother liquor from above was chromatographed on silica gel with ethanol-petroleum ether (10:90) as the eluting solvent. The first band contained photoproduct 60(45 mg, 21%) which was recrystallized from ethanol to give colorless crystals.The second chromatographic fraction contained a mixture of compounds 61 and62 (85 mg, 40%). Rechromatography of this mixture afforded the rearranged product 62e(50 mg, 23%).177Chapter 3 ExperimentalPhotoproduct 60 was characterized as dimethyl 4b-amino-4b,8b,8c,8d-tetrahydro-dibenzo[a,ficyclopropa[c,d]pentalene-8b,8c-dicarboxylate.MP: 131-132°C.1R (KBr) vi„„„: 3397, 3327 (N-H), 1731, 1714 (C=0), 1237, 1208 (C-0) cm-1 .MS m/e (relative intensity): 335 (M+, 6), 276 (100), 244 (54), 216 (95), 193 (43), 189(37).Exact mass calculated for C20H17N04: 335.1158. Found: 335.1162.111-NMR (400 MHz, CDC13) 8 7.55-7.05 (m, 8H, aromatic H), 4.40 (s, 1H, cyclopropylH), 3.83 (s, 3H, CO2CH3), 1.70 (s, 2H, D20 exchangeable, NH2) ppm.13C-NMR (100 MHz, CDC1 3) 8 168.8, 167.5 (C=O), 152.0, 151.6, 132.8, 160.6(aromatic C), 127.81, 127.77, 127.73, 127.65, 126.2, 125.3, 119.91, 119.88 (aromatic C-H), 76.9, 73.5, 54.2 (benzylic C and cyclopropyl C), 52.5, 52.1 (CO 2CH3), 49.3(cyclopropyl C-H) ppm.Anal. calculated for C201-1 17N04: C, 71.63; H, 5.11; N, 4.17. Found: C, 71.55; H, 5.17;N, 4.19.Photoproduct 61 was characterized as dimethyl 8d-amino-4b,8b,8c,8d-tetrahydro-dibenzo[a,ficyclopropa[c,d]pentalene-8b,8c-dicalboxylate.MP: 138-141°C.IR (KBr) v.: 3263 (N-H), 1730, 1713 (C--0), 1274, 1216 (C-0) cm -1 .MS m/e (relative intensity): 335 (M+, 27), 276 (100), 244 (83), 216 (92), 189 (31).Exact mass calculated for C201 -117N04: 335.1158. Found: 335.1153.111-NMR (400 MHz, CDC13) 8 7.8-7.1 (m, 8H, aromatic H), 5.45 (s, 1H, benzylic H),3.75 (s, 6H, CO2CH3), 1.60 (s, 2H, D20 exchangeable, NH2) ppm.13C-NMR (50 MHz, CDC13) 8 179.3, 172.5 (C=0), 169.9, 151.3, 143.2 (aromatic C),133.4, 128.9, 128.1, 127.8, 125.1, 124.92, 124.85, 123.3 (aromatic C-H), 69.8, 65.8(cyclopropyl C), 55.6 (benzylic C-H), 53.0, 52.4 (CO2CH3) ppm.178Chapter 3 ExperimentalAnal. calculated for C20}117N04: C, 71.63; H, 5.11; N, 4.17. Found: C, 71.44; H, 5.21;N, 4.14.Compound 62 was characterized as dimethyl 4b,10-dihydro-10-oxo-ideno[1,2-a]indene-9,9a(9H)-dicarboxylate. All the spectra are identical to those reported by Richards et al."MP: 177-179 (lit. 64 185°C).IR. (KBr) 1739, 1722 (C---0), 1240 (C-0) cm-1.MS m/e (relative intensity): 336 (M+, 7), 304 (48), 276 (86), 249 (30), 234 (23), 217(100), 189 (80).Exact mass calculated for C2011 1605 : 336.0998. Found: 336.0990.1H-NMR (400 MHz, CDC13) 6 7.9-7.8 (m, 1H, aromatic H), 7.7-7.6 (m, 2H, aromaticH), 7.4-7.1 (m, 5H, aromatic H), 5.64 (s, 1H, benzylic H), 4.81 (s, 1H), 3.71 (s, 3H,CO2CH3), 3.70 (s, 3H, CO2013) ppm.13C-NMR (75 MHz, CDC13) 6 199.1, 172.3, 167.64 (C--0), 156.9, 143.0, 137.5(aromatic C), 136.3 (aromatic C-H), 133.2 (aromatic C), 128.9, 128.4, 128., 125.5,125.4, 125.3, 125.0 (aromatic C-H), 72.7 (tertiary C), 54.8, 53.9, 53.1, 52.5 (CO2CH 3and benzylic C-H) ppm.Crystals of amine 39 were crushed between glass plates and photolyzed through aPyrex filter. The reaction was found to be very slow when Pyrex plates were used, soquartz plates were used instead. After 24 h of irradiation, GC (DB-17) showed 20%conversion and a 60:61 ratio of 14:86; no rearrangement product 62 could be detected.179Chapter 3 ExperimentalPhotolysis of Trimethyl 9,10-dihydro-9,10-ethenoanthracene-9,11,12-tricarboxylate(43)A solution of ester 43 (120 mg, 0.317 mmol) in acetone (250 mL) was purgedwith nitrogen for 30 min and irradiated through a Pyrex filter for 0.5 h. GC (DB-17)indicated the formation of photoproduct 67 with little starting material remaining. Thesolvent was removed under vacuum and the residue was chromatographed on silica geleluted with 20% ethyl acetate in petroleum ether. The resulting oil was crystallized fromethanol to yield photoproduct 67 (94.4 mg, 80%).Compound 67 was characterized as trimethyl 4b,8b,8c,8d-tetrahydro-dibenzo[a,f]cyclopropa[c,d]pentalene-4b,8b,8c-tricarboxylate.MP: 126-127°CIR (KBr) vmax: 1743 with a shoulder at 1730 (C=0), 1274 (C-0) cm- 1 .MS m/e (relative intensity): 378 (M+, 51), 319 (74), 291 (100), 232 (35), 216 (27), 193(56).Exact mass calculated for C22H1806: 378.1103. Found: 378.1100.111-NMR (400 MHz, CDC1 3) 6 7.5-7.0 (m, 8H, aromatic H), 4.48 (s, 1H cyclopropyl H),3.88 (s, 6H CO,CH3), 3.70 (s, 3H, CO,CH3) ppm.13C-NMR (50 MHz, CDC13) 6 169.2, 168.2, 167.1 (C=0), 144.1, 148.3, 134.0, 132.2(aromatic C), 128.1, 128.0, 127.74, 127.66, 126.5, 125.7, 121.7, 121.2 (aromatic C-H),69.9, 69.6, 54.9 (cyclopropyl and benzylic C), 52.8, 52.6, 52.5 (CO,CH 3), 49.1(cyclopropyl C-H) ppm.Anal. calculated for C22111806: C, 69.83; H, 4.80. Found: C, 69.77; H, 4.86.A solution of ester 43 (360 mg, 0.953 mmol) in anhydrous CHC1 3 (250 mL) wasdegassed for 0.5 h and irradiated for 0.5 h. GC (DB-17) indicated 63% conversion to180Chapter 3 Experimentalphotoproducts 67 and 71 in the ratio 70:30 respectively. The solvent was removed undervacuum and the residue chromatographed on silica gel eluted with 20% ethyl acetate inpetroleum ether. Two fractions were collected: The first one contained a 1:1 mixture ofthe starting material and photoproduct 67 (223 mg, 62%). The second fraction containeda mixture of compounds 43:67:71 (120 mg, 33%) in the ratio 38:42:21. Compound 71was found to be somewhat unstable on silica gel, but after the second fraction had beenrechromatographed several times, compound 71 (9.3 mg, 3%) was isolated as an oil.Photoproduct 71 was characterized as trimethyl dibenzo[a,e]cyclooctene-5,6,11-tricarboxylate.IR (CHC13) vin.: 1722 (C=0), 1634 (C=C), 1250 (C-0) cm- 1 .MS m/e (relative intensity): 378 (M+, 10), 318 (100), 287 (83), 201 (90).Exact mas calculated for C22111806: 378.1103. Found: 378.1103.111-NMR (400 MHz, CDC13) 8 8.08 (s, 1H, vinyl H), 7.4-7.1 (m, 8H, aromatic H), 3.82(s, 3H, CO2CH3), 3.75 (s, 6H, CO,CH3) ppm.Photolyses of ester 43 in solutions were repeated and analyzed on a different GCcolumn, DB-1, which revealed formation of an additional minor photoproduct. Thisphotoproduct was later characterized as compound 68. Photoproducts 67 and 68 werefound to have the same retention time on GC column DB-17, while they were wellseparated on a column of the DB-1 type. GC (DB-1) analyses of photolyses of ester 43 inacetone indicated formation of photoproducts 67 and 68 in the ratio 91:9, whileirradiation of ester 43 in chloroform showed formation of photoproducts 67:68:71 in theratio 60:6:34.Crystals of ester 43 (227 mg, 0.601 nunol) were crushed between four Pyrexplates and irradiated for 48 h. The photolyzed solid was recrystallized from ethanol to181Chapter 3 Experimentalyield unreacted starting material (150 mg, 66%). The mother liquor waschromatographed on silica gel eluted with 20% ethyl acetate in petroleum ether. The firstband contained the starting material (24 mg, 11%). The second band contained a mixtureof the starting material and three photoproducts (52.4 mg, 23%) in the ratio 28:72. Thethree photoproducts were compounds 67:68:71, which were formed in the ratio 36:38:26as determined by 11 1-1-NMR.Photolysis of Dimethyl 9-Carboxy-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (38)Acid 38 (49.8 mg, 0.137 mmol) was dissolved in acetone (15 mL) and irradiatedfor 2 h. The solvent was removed under vacuum and the resulting oil was dissolved indiethyl ether and treated with excess diazomethane. GC (DB-17) indicated 80%conversion to photoproduct 67. The solvent was removed under vacuum and the residuechromatographed on silica gel with 20% ethyl acetate in petroleum ether as the elutingsolvent. The resulting oil (41.6 mg, 0.1101 mmol, 80%) was analyzed by 1 1-I-NMR,which showed the only photoproduct formed was compound 67.Crystals of acid 38 (514 mg, 1.412 mmol) which had been recrystallized fromacetonitrile were crushed between two quartz plates and photolyzed for 4 days. The solidwas dissolved in diethyl ether and treated with excess diazomethane. CG (DB-1)indicated 10% conversion to form photoproducts 67 and 68 in the ratio 1:9 respectively.The solvent was removed under vacuum and the residue chromatographed on silica geleluted with 15% ethyl acetate in petroleum ether. The resulting oil (33.8 mg, 0.0904mmol, 7%) was recrystallized from ethanol to yield crystals of photoproduct 68 (28.6mg, 5%).182Chapter 3 ExperimentalCompound 68 was characterized as trimethyl 4b,8b,8c,8d-tetrahydro-dibenzo[a,ficyclopropa[c,d]pentalene-8b,8c,8d-tricarboxylate.MP: 184-186°C.IR (KBr) v.: 1737 (C=O), 1229 (C-0) cm -1 .MS m/e (relative intensity): 378 (M+, 6), 350 (4), 318 (10), 260 (23), 202 (47), 57 (100).Exact mass calculated for C22111806: 378.1103. Found: 378.1106.111-NMR (400 MHz, CDC1 3) 8 7.45 (m, 2H, aromatic H), 7.2-7.1 (m, 6H, aromatic H),5.07 (s, 1H, benzylic H), 3.92 (s, 6H, CO2CH3), 3.78 (s, 3H, CO20.13) ppm.13C-NMR (100 MHz, CDC1 3) 8 167.4, 1672 (C--0), 150.2, 132.8 (aromatic C), 128.4,127.0, 126.6, 121.2 (aromatic C-H), 67.5, 58.6 (cyclopropyl C), 58.0 (benzylic C-H),52.8, 52.6 (CO2CH3) ppm.Anal. calculated for C221-11806: C, 69.83; H, 4.80. Found: C, 69.92, H, 4.80.Crystals of acid 38 crystallized from ethyl acetate were photostable and did notreact when irradiated through a quartz filter.Crystals of acid 38 (95.1 mg, 0.26 mmol) which had been recrystallized fromethanol were crushed between two Pyrex plates and irradiated for 25 h. Afterwards thesolid was dissolved in diethyl ether and excess diazomethane was added. The solvent wasremoved under vacuum and the residue chromatographed on silica gel eluted with 40%ethyl acetate in petroleum ether, yielding photoproducts 67 and 68 (4.0 mg, 0.0106mmol, 4%). 1H-NMR revealed that compounds 67 and 68 were formed in the ratio 40:60.183Chapter 3 ExperimentalPhotolysis of Dimethyl 9-Carboxyethy1-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (44)A solution of ester 44 (384 mg, 0.797 mmol) in acetone (350 mL) was purgedwith nitrogen for 20 min and irradiated for 0.5 h. GC (DB-17) indicated formation ofphotoproducts 69 and 70 in the ratio 93:7 respectively with little remaining startingmaterial. The solvent was removed under vacuum and the residue chromatographed onsilica gel eluted with 15% ethyl acetate in petroleum ether. The first band containedcompound 69 (86 mg, 22%), which was recrystallized from ethanol to give colorlessprisms. The second fraction contained a mixture of photoproducts 69 and 70 (201.5 mg,54%).Compound^69^was^characterized^as^4b,8b,8c,8d-tetrahydro-dibenzo-[a,f]cyclopropa[c,d]pentalene-4b,8b,8c-tricarboxylic acid, 4b-ethyl-8b-methyl-8c-methylester.MP: 134-135°C.ilt (KB r) v ,„,, x : 1746, 1723, 1718 (C=O), 1298, 1274, 1184 (C-0) cm- 1 .MS m/e (relative intensity): 392 (M+, 100), 365 (10), 319 (36), 291 (22), 260 (38).Exact mass calculated for C23H2006: 392.1260. Found: 392.1257.1H-NMR (400 MHz, CDC13) 5 7.5-7.1 (m, 8H, aromatic H), 4.48 (s, 1H, cyclopropyl H),4.36 (q, 2H, J=7 Hz, CO2CH,2CH3), 3.88 (s, 3H, CO2CH3), 3.70 (s, 3H, CO20:13), 1.32(t, 3H, J=7 Hz, CO2CH,CH3) ppm.13C-NMR (75 MHz, CDC13) 8 168.7, 168.4, 167.0 (C=O), 149.4, 148.3, 134.1, 132.0(aromatic C), 128.0, 128.0, 127.68, 127.60 126.63 125.7, 121.8, 121.0 (aromatic C-H),70.0, 69.5 (benzylic C and cyclopropyl C), 61.7 (CO7CH2CH3), 54.4 (cyclopropyl C),52.8, 52.3 (CO,CH3), 49.2 (cyclopropyl C-H), 14.2 (CO2CH,CH3) ppm.Anal. calculated for C23H2006: C, 70.40; H, 5.14. Found: C, 70.34; H, 5.16.184Chapter 3 ExperimentalA solution of ester 44 (330 mg, 0.843 mmol) in chloroform (350 mL) was purgedwith nitrogen for 20 min and irradiated for 4 h. GC (DB-1) showed 80% conversion andformation of photoproducts 69 and 70 plus an additional peak in the ratio 35:9:56,respectively. The solvent was removed under vacuum and the resulting oilchromatographed on silica gel eluted with 15% ethyl acetate in petroleum ether. The firstfraction contained the starting material (17 mg, 5%). The second band consisted of amixture of the starting material and photoproducts 69 and 70 (118 mg, 36%). The thirdfraction contained a colorless oil (105 mg, 32%), which showed up as a single peak onthe GC (DB-17). This oil was crystallized from ethanol to yield colorless prisms, whichwere characterized as photoproduct 72 (19 mg, 6%). The mother liquor was crystallizedto give a compound (41.2 mg, 12%) with the same retention time on the GC (DB-17) ascompound 72 but which was characterized as photoproduct 73. 1 11-NMR revealed thatthe remaining mother liquor was a mixture of compounds 72 and 73 in the ratio 40:60.The structures of photoproducts 72 and 73 were further supported by X-ray analyses ofcrystals of both compounds.67 .69Photoproduct 72 was characterized as dibenzo[a,e]cyclooctene-5,6,12-tricarboxylic acid,5-ethyl 6-methyl 12-methyl ester.MP: 174-175°C.IR (KBr) v,„„„: 1731, 1714 (C=0), 1638 (C=C), 1284, 1225, 1208 (C-0) cm- 1 .MS ink (relative intensity): 392 (M+, 5), 364 (20), 287 (46), 260 (50), 229 (87), 201(100).Exact mass calculated for C23H2006: 392.1260. Found: 392.1262.111-NMR (400 MHz, CDC13) 8 8.08 (s, 1H, vinyl H), 7.4-7.1 (m, 8H, aromatic H), 4.23(q, 2H, J=7 Hz, CO2(Z2CH3), 3.82 (s, 3H, CO7CH3), 3.76 (s, 3H, CO 7CH3), 1.25 (t, 3H,J=7 Hz, CO2CH7CH3) ppm.185Chapter 3 Experimental13C-NMR (100 MHz, CDC1 3) 8 166.9, 166.8 (C=O), 141.8 (vinyl C-H), 140.2, 137.9,136.0, 135.3, 134.5, 134.22, 133.5 (aromatic C), 129.4, 128.5, 128.2, 127.9, 127.8,127.6, 127.5 127.2 (aromatic C-H), 61.7 (CO2CH,CH 3), 52.6, 52.4 (CO2 .3), 13.9(CO2CH2C113) ppm.Photoproduct 73 was characterized as dibenzo[a,e]cyclooctene-5,11,12-tricarboxylicacid, 5-ethyl 11-methyl 12-methyl ester.MP: 123-125°C.IR (KBr) va.: 1733, 1718 (C=O), 1635 (C=C), 1266, 1240, 1196 (C-0) cm-I.MS ink (relative intensity): 392 (M+, 5), 364 (23), 318 (28), 287 (44), 260 (54), 229(78), 201 (100).Exact mass calculated for C23H2006: 392.1260. Found: 392.1259.111-NMR (400 MHz, CDC13) 8 8.09 (s, 1H, vinyl H), 7.4-7.1 (m, 4H, aromatic H), 4.23(q, 2H, J=7 Hz, CO7CH2CH3), 3.83 (s, 3H, CO2CH3), 3.76 (s, 3H, CO2CH3) 1.25 (t, 3H,J=7 Hz, CO2CH2CH_3) ppm.13C-NMR (100 MHz, CDC13) 8 167.0, 166.6, 166.5 (C--0), 141.9 (vinyl C-H), 139.0,135.9, 135.3 134.4, 134.3, 133.5 (aromatic C and vinyl C), 129.4, 129.3, 128.29, 128.1,127.8, 127.6, 127.5, 127.2 (aromatic C-H), 61.7 (CO2CH,CH3), 52.5, 52.3 (CO 2CH3),13.9 (CO2C112CH3) ppm.Crystals of the prism form of ester 46 (465 mg, 1.18 mmol) were crushedbetween 6 quartz plates and irradiated for 4 days. GC (DB-17) showed 18% conversionto photoproducts 69:70:(72+73) in the ratio 29:32:39. Compounds 72 and 73 wereformed in the ratio 22:77 as determined by 1H-NMR. The photolyzed crystals wererecrystallized from ethanol to yield the starting material (238 mg, 58%). The motherliquor was chromatographed on silica gel eluted with 20% ethyl acetate in petroleumether. The first band contained (40.2 mg, 7%) of the starting material. The second one186Chapter 3 Experimentalcontained a mixture of the photoproducts and the starting material (58.6 mg, 13%),which was recrystallized from ethanol to yield crystals of compounds 43 and 69. Themother liquor, which contained 60% of photoproduct 70, was treated with excessdiazomethane and rechromatographed to yield 80% pure compound 70 (16 mg, 4%).This compound was further purified by recrystallization from ethanol, yielding crystalsof pure compound 70 (12 mg, 3%).Photoproduct^70^was^characterized^as^4b,8b,8c,8d-tetrahydro-dibenzo-[aAcyclopropa[c,d]pentalene-8b,8c,8d-tricarboxylic acid, 8b-methyl 8c-methyl 8d-ethylester.MP: 120-126°C.IR (KBr) v..: 1745, 1731, 1714 (C=0), 1277, 1224, 1200 (C-0) cm- 1 .MS m/e (relative intensity): 392 (M+, 12), 364 (3), 332 (12), 318 (12), 291 (12), 260(19), 217 (19), 201 (20), 149 (100).Exact mass calculated for C23H2006: 392.1260. Found: 392.1256.1H-NMR (400 MHz, CDC1 3) 6 7.5 (m, 2H, aromatic H), 7.2-7.1 (m, 6H, aromatic H),5.06 (s, 1H, benzyl H), 4.47 (q, 2H, J=7 Hz, CO2CH2CH3), 3.91 (s, 3H, CO2CH3), 3.88(s, 3H, CO2CH3), 1.44 (t, 2H, J=7 Hz, CO2CH2C113) ppm.13C-N1VIR (100 MHz, CDCI3) 6 167.5, 167.2, 166.7 (C=O), 150.3, 150.2, 133.0, 133.0(aromatic C), 128.5, 128.4, 127.7, 127.6, 127.1, 127.0, 121.2, 121.0 (aromatic C-H),67.6 (cyclopropyl C), 62.0 (CO2CH,CH3), 62.0, 61.7 (cyclopropyl C), 58.1 (benzyl C-H), 52.8, 52.5 (CO,CH3), 14.0(CO2CH2CHO ppm.Anal. calculated for C23H2o06: C,70.40; H, 5.14. Found: C, 70.60; H, 5.05.Needle-like crystals of ester 44 (57.0 mg, 0.145 mmol) were crushed between twoquartz plates and irradiated for 8 h. GC (DB-1) indicated 30% conversion and formation187Chapter 3 Experimentalof photoproducts 69:70:(72+73) in the ratio 26:42:32. GC analysis on DB-1 showed thatphotoproducts 72 and 73 were formed in the ratio 33:66 respectively.3.3.4 Photolyses of Salts33.4.1 Photolyses of Salts Formed with Ethyl 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylate-12-carboxylic Acid (36)Photolysis of Salts 85 to 89Salts 85 to 89 were photolyzed through a Pyrex filter in solution and in the solidstate. After photolysis the solvent was removed under vacuum and the residue dissolvedin diethyl ether. The resulting solution was washed thoroughly with 15% aqueous HCI,dried over MgSO4, treated with ethereal diazomethane to produce the correspondingmethyl esters and analyzed by GC (DB-17) to determine product ratios and the extent ofconversion. The results are listed in Table 3-10.188Chapter 3 ExperimentalTable 3-10 Photoproduct Mixture Composition of Salts 85 to 89 as a Function of thePhotolysis MediumSalt of acid 36with:Medium Concentration(M)Photoproduct ratio32a:33aaSodium (85) Crystals 100:0H2O 0.0063 100:0Calcium (86) Crystals 100:0Acetone 0.020 100:0Benzene 0.020 100:0Methanol 0.022 100:0Diethylamine (87) Crystals 100:0Acetone 0.013 75:25Benzene 0.011 100:0Methanol 0.013 100:0Pyrrolidine (88) Crystals 100:0Acetone 0.011 68:32Benzene 0.0093 100:0Methanol 0.0098 100:0Piperidine (89) Crystals 100:0Acetone 0.0076 70:30Benzene 0.013 100:0Methanol 0.010 100:0a As determined by GC after correction for unreacted starting material. The estimated error in the GCanalyses is ±5%.189Chapter 3 ExperimentalPhotolyses of S-(-)-Proline Salt 90aS-(-)-Proline salt 90a (116 mg, 0.267 mmol) was irradiated for 12 h at roomtemperature as a powder sandwiched between two Pyrex microscopic slides. The reactionmixture was dissolved in diethyl ether and extracted three times with 15% aqueous HC1and twice with saturated aqueous NaC1 solution. The organic layer was dried overMgSO4 and treated with excess diazomethane to produce the corresponding methylesters. The solvent was removed under vacuum and the residue subjected to silica gelcolumn chromatography eluted with 15% ethyl acetate in petroleum ether. This affordedthe known photoproducts 32a and 33a (7.2 mg, 0.0216 mmol, 8%) as a mixture. Theregioisomeric and enantiomeric compositions were determined by 400 MHz 111-NMR.For enantiomeric excess determination, use was made of the chiral shift reagent (+)-Eu(hfc)3 (Aldrich). The signals monitored were the methyl singlets at 8 = 3.9 ppm and 8= 3.7 ppm for compounds 32a and 33a respectively. The NMR revealed that compound32a was formed in 83% yield and 42% enantiomeric excess, whereas compound 33a wasformed in 17% yield with no observed enantiomeric excess. In addition, the sign ofrotation of the predominant enantiomer was measured by polarimeter and found to bepositive.The solid state photolyses of S-(-)-proline salt 90a were repeated as describedabove but at varying temperatures. The conversion was kept below 20% to minimizemelting and each experiment was repeated at least once. The results are listed in Table 3-11.190Chapter 3 ExperimentalTable 3-11^Solid State Photoproduct Mixture Composition for S-(-)-Proline Salt 90aas a Function of TemperatureTemperature Photoproduct 32a Photoproduct 33aYield*(%)eeb Yield*(%)eeb20°C 83 (+)-42 17 nil0°C 90 (+)-55 10 nil-25°C 90 (+)-66 10 nil-40°C 94 (+)-76 4 nila As determined by NMR after correction for unreacted starting material; b The estimated accuracy inthese values is t5%; the sign of rotation of the predominant enantiomer is shown in parentheses.S-(-)-Proline salt 90a (69.8 mg, 0.160 mmol) was dissolved in ethanol (250 mL)and nitrogen was bubbled through the solution for 25 min. The resulting solution wasirradiated for 4 h. The photoproducts were isolated following the same procedure usedfor isolating the photoproducts in the solid state photolysis of S-(-)-proline salt 90a. Thisyielded a mixture containing photoproducts 32a and 33a (15.4 mg, 0.047 mmol, 30%) inthe ratio 47:53 respectively with no enantiomeric excess as determined by 111-NMR.Photolysis of Salts 90b to 98Salts 90b, 90c, 92, 93, 94, 95, 96, 97 and 98 were irradiated in the solid state at -40°C. The regioisomeric and enantiomeric composition of the photoproducts for each191Chapter 3 Experimentalphotolysis was studied in the same way as for the solid state photolysis of S-(-)-prolinesalt 90a. The results of these experiments are summarized in Table 3-12.Table 3-12 Solid State Photoproduct Mixture Composition for Photolyses of Salts90b, 90c and 92 to 98Optically Active Amine Photoproduct 32a Photoproduct 33aYielda(%)eeb Yielda(%)eebR-(+)-Proline (90b) 96 (-)-80 4 nil(±)-Proline (90c)c 84 nil 16 nilS-(-)-Proline methyl ester (92) 93 (+)-58 7 nilS-(+)-2-Prolinol (93) 100 (-)-37 - -S-(-)-Prolineamide (94) 100 (-)-24 - -S-(42-(Methoxymethyl)-pyrrolidine (95) 100 (-)-16 - -S-(-)-Proline tert-butyl ester (96) 100 >(+)-95 - -(S,S)-(+)-Pseudoephedrine (97) 87 (-)-87 13 73(-)-Strychnine (98) 65 (+)-14 35 nilAs determined by 1 H-NMR; bThe estimated accuracy in these values is t 5%; the sign of rotation of thepredominant enantiomer is shown in parentheses. C Photolyses carried out at -20°C.Salts 90b, 90c, 92, 96 and 98 were irradiated in solution. The regioisomeric andenantiomeric composition of the photoproducts was studied in the same way as for thephotolysis of S-(-)-proline salt 90a in acetone. The results of these experiments aresummarized in Table 3-13.192Chapter 3 ExperimentalTable 3-13 Solution Photoproduct Mixture Composition for Salts 90a, 90b, 90c, 92and 98Optically Active Amine Solvent Photoproduct 32a Photoproduct 33aYielda(%)ee Yielda(%)eeR-(+)-Proline (90b) Ethanol 50 nil 50 nilR,S-(±)-Proline (90c) Ethanol 50 nil 50 nilS-(-)-Proline methyl ester (92) Benzene 95 nil 5 nilS-(-)-Proline tert-butyl ester (96) Acetone 50 nil 50 nil(-)-Strychnine (98) Ethanol 60 nil 40 nila As determined by IFI-NMR after correction for the starting material. The estimated accuracy in thesevalues is ±5%.33.4.2 Photolyses of Salts Formed with 12-Methyl-9,10-dihydro-9,10-ethenoanthracene-11-carboxylic Acid (37)Photolyses of (S,S)-(+)-Pseudoephedrine Salt 105a(S,S)-(+)-Pseudoephedrine salt 105a (328 mg, 0.767 mmol) was irradiatedthrough a Pyrex filter at -40°C for 3 days. The reaction mixture was dissolved in diethylether and extracted three times with 15% aqueous HC1 and twice with saturated aqueousNaCI solution. The organic fraction was dried over MgSO4 and excess diazomethane was193Chapter 3 Experimentaladded to produce the corresponding methyl esters. The solvent was removed undervacuum and the residue chromatographed on silica gel eluted with 10% ethyl acetate inpetroleum ether. The resulting oil was recrystallized from ethanol to yield photoproduct52 (52.1 mg, 0.189 mmol, 25% yield). The enantiomeric composition of thephotoproduct was determined by 400 MHz 1H-NMR using the chiral shift reagentEu(hfc)3 . The signal monitored was the methyl singlet at 5 = 1.90 ppm which showedthat compound 52 was formed in greater than 95% enantiomeric excess. The opticalrotation was measured and found to be positive ([a] i3= (+)-59 (Chloroform, C=5.2)).A mixture of (S,S)-(+)-pseudoephedrine (19.1 mg, 0.116 mmol) and acid 37(27.0 mg, 0.102 mmol) was dissolved in acetone (20 mL) and irradiated through a Pyrexfilter for 3 h. The photoproduct was isolated following the same procedure as used forthe isolation of the photoproduct in the solid state photolyses of (S,S)-(+)-pseudoephalrine salt 105a. This yielded photoproduct 52 (25.8 mg, 0.0935 mmol, 92%)which was formed without any enantiomeric excess as determined by polarimetry.Photolysis of Salts 103, 104 and 105bSalts 103, 104 and 105b were irradiated in the solid state at -40°C. Theenantiomeric composition of the photoproduct was studied in the same way as for solidstate photolysis of (S,S)-(+)-pseudoephedrine salt 105a. The photolyses were stoppedbefore 40% conversion. The results are summarized in Table 3-14.194Chapter 3 ExperimentalTable 3-14 Solid State Photoproduct Mixture Composition for Salts 103, 104 and105bOptically Active Amine Photoproduct 32aeeaS-(+)-2-Prolinol (103) (+)-38S-(-)-Proline tert-butyl ester (104) (+)-26(R,R)-(-)-Pseudoephedrine (105b) ?..(-)-95'The estimated accuracy in these values is ± 10%; the sign of the rotation of predominant enantiomer isshown in parentheses.The enantiomeric composition of the photoproduct of the solution photolyses ofsalts 103, 104 and 105b was studied in the same way as for the solution photolysis of(S,S)-(+)-pseudoephedrine salt 105a. The results of these experiments are summarized inTable 3-15.Table 3-15 Solution Photoproduct Mixture Composition for Salts 103, 104 and 105bOptically Active Amine Solution Photoproduct 52eeaS-(+)-2-Prolinol (103) Acetone nilS-(-)-Proline tert-butyl ester (104) Acetone nil(R,R)-(-)-Pseudoephedrine (105b) Acetone nila The estimated accuracy in these values is ±5%.195Chapter 3 Experimental3.3.4.3 Photolyses of Salts Formed with 9,10-Dihydro-9,10-ethenoanthracene-11-carboxylic Acid (40)Photolyses of Piperidine Salt 134A solution of piperidine salt 134 (77.5 mg, 0.232 mmol) in acetone (250 mL) waspurged with nitrogen for 30 min and irradiated through a Pyrex filter for 4 h. The solventwas removed under vacuum and the residue dissolved in diethyl ether and washed threetimes with 15% aqueous HCl and two times with saturated aqueous NaC1 solution. Theorganic fraction was dried over MgSO4 and treated with excess diazomethane to producethe corresponding methyl ester. GC (DB-17) indicated formation of photoproduct 135with little remaining starting material. The solvent was removed under vacuum and theresidue chromatographed on silica gel eluted with 10% ethyl acetate in hexane. Theresulting oil was crystallized from ethanol to yield compound 135 (47.5 mg, 0.181 mmol,78% yield).Product 135 was characterized as methyl 4b,8b,8c,8d-tetrahydro-dibenzo-[a,f]cyclopropa[c,d]pentalene-8c-carboxylate. The spectra are identical to those reportedby Ciganek.58MP: 165-168°C (11 58 169.5-170.5°C).IR (KBr) v.: 1715 (C=O), 1252 (C-0) cm-1•MS ink (relative intensity): 262 (M+, 31), 219 (6), 203 (100).Exact mass calculated for C18111402: 262.0994. Found: 262.0998.196Chapter 3 Experimental1H-NMR (400 MHz, CDC13) 8 7.3-7.0 (m, 8H, aromatic H), 4.98 (s, 1H, benzylic H),3.78 (s, 2H, cyclopropyl H), 3.72 (s, 3H, CO2CH3) ppm.'IC-NAIR (50 MHz, CDC1 3) 8 172.39 (C=O), 150.18, 135.47 (aromatic c), 127.00,126.64, 124.76, 121.32 (aromatic C-H), 62.08 (cyclopropyl C), 53.95, 51.98 (benzylic C-H and CO2CH3), 47.04 (cyclopropyl C-H) ppm.Crystals of piperidine salt 134 (16.7 mg, 0.0501 mmol) were irradiated through aPyrex filter for 12 h. The crystals turned yellow and cloudy upon irradiation. Afterwardsthey were dissolved in diethyl ether and the resulting solution was washed thoroughlywith 15% aqueous HCl solution, dried over MgSO 4 and excess diazomethane was added.GC (DB-17) analyses indicated 62% conversion to photoproduct 135.Crystals of piperidine salt 134 were irradiated with y-rays from a 60Co source(Gammacell 220 by AECL, 200Ci) for a month. The crystals looked undamaged butslightly yellow. Individual crystals were dissolved in diethyl ether, acidified, treated withexcess diazomethane and analyzed by GC(DB-17) to reveal formation of photoproduct135 with conversion varying from 10 to 25%.197Chapter 3 Experimental33.4.4 Photolyses of Salts Formed with Dimethyl 9-Amino-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (39)Photolyses of Hydrochloride Salt 112Hydrochloride salt 112 (145 mg, 0.391 mmol) was dissolved in acetonitrile (300mL) and the resulting solution was purged with nitrogen for 0.5 h and irradiated for 4h.GC (DB-17) indicated formation of photoproduct 60 with little remaining startingmaterial. The solvent was removed under vacuum. The residue was dissolved in ethylacetate and extracted with three portions of 10% aqueous sodium hydroxide solutionfollowed by two portions of aqueous saturated sodium chloride solution. The organiclayer was dried over MgSO4 and concentrated under vacuum. The residue was subjectedto a silica gel column eluted with 10% ethanol in petroleum ether to yield photoproduct60 (77.9 mg, 0.233 mmol, 60%).Crystals of hydrochloride salt 112 (20 mg, 0.0539 mmol) were irradiated througha Pyrex filter for 10 h. The irradiated crystals were dissolved in ethyl acetate, washedthoroughly with 10% aqueous sodium hydroxide solution, dried over MgSO4 andanalyzed by GC (DB-17) which indicated 10% conversion to form photoproduct 60.Photolyses of R-0-10-Camphorsulfonic Acid Salt 116aR-0-10-Camphorsulfonic acid salt 116a (233 mg, 0.398 mmol) was crushedbetween Pyrex plates and irradiated at -40°C for 24 h. The photolyzed solid wasdissolved in ethyl acetate and the resulting solution extracted with three portions of 10%198Chapter 3 Experimentalaqueous sodium hydroxide solution followed by two portions of aqueous saturatedsodium chloride solution. The organic layer was dried over MgSO 4 and concentratedunder vacuum. The photoproduct was purified on silica gel eluted with 10% ethanol inpetroleum ether to yield photoproduct 60 (18.0 mg, 0.0537 mmol, 14%). Compound 60was analyzed for enantiomeric excess by 400 MHz 1H-NMR using (+)-Eu(hfc)3 as thechiral shift reagent. The signal monitored was the methyl ester singlet at 8 = 3.83 ppmwhich revealed that compound 60 was formed in 68% enantiomeric excess. In addition,the sign of the rotation of the predominant enantiomer was measured by polarimeter andfound to be positive.A mixture of amine 39 (191 mg, 0.570 mmol) and R-(-)-10-camphorsulfonic acid(136 mg, 0.587 mmol) was dissolved in acetone (300 mL). This solution was purgedwith nitrogen for 0.5 h and irradiated for 1.5 h. GC (DB-17) showed 90% conversion tocompound 60. The photoproduct was purified following the same procedure as for thephotoproduct in the solid state photolysis of R-(-)-10-camphorsulfonic acid salt 116a.Compound 60 (163 mg, 0.487 mmol, 86%) was formed with no enantiomeric excess asdetermined by 1 11-NMR.Photolyses of Salts 116b and 117Salts 116b and 117 were irradiated in the solid state at -40°C and in acetonesolutions at room temperature. The solid state photolyses were stopped before 60%conversion. The regioisomeric and enantiomeric ratios of the photoproducts for eachphotolysis were studied in the same manner, as for the photolyses of R-(-)-10-camphorsulfonic acid salt 116a. Isolation of the photoproducts revealed that regioisomer60 was the sole compound formed, except in the case of the solid state irradiation of salt199Chapter 3 Experimental117 where a small amount of photoproduct 61 was detectable (ca. 15%). The results arelisted in Table 3-16.Table 3-16 Enantiomeric Excess in Photoproduct 60 as a Function of PhotolysisMedium and Sulfonic Acid StructureOptically Active Sulfonic Acid Solid State ee Solution Phase ee(%)a (%)aR-(-)-10-Camphorsulfonic acid (116a) (-)-68 nilS-(+)-10-Camphorsulfonic acid (116b) (+)-64 nil(-)-3-Bromocamphor-8-sulfonate (117) (+301' nila The estimated accuracy in these values is t 5%; the sign of rotation of the predominant enantiomer isshown in parentheses. b Only in the case of salt 117 was a small amount of photoproduct 61 detectable(ca 15%).33.4.5 Photolyses of Salts Formed with Dimethyl 9-Carboxy-9,10-dihydro-9,10-ethenoanthracene-11,12-dicarboxylate (38)Photolyses of (R,S)-(-)-Ephedrine Salt 108(R,S)-(-)-Ephedrine salt 108 (70.7 mg, 0.131 mmol) was crushed between Pyrexplates and irradiated at -40°C for 3 days. The photolyzed crystals were dissolved indiethyl ether, extracted three times with 15% aqueous HC1 and twice with saturatedaqueous NaCl. The organic fraction was dried over MgSO4 and excess diazomethane was200Chapter 3 Experimentaladded. The solvent was removed under vacuum and the residue chromatographed on asilica gel column eluted with 15% ethyl acetate in petroleum ether. This yielded amixture of the achiral compound 68 and the chiral compound 67 in the ratio 9:1,respectively (8.1 mg, 0.207 mmol, 16%).A solution of acid 38 (51.3 mg, 0.3125 mmol) and (R,S)-(-)-ephedrine (20.7 mg,0.125 mmol) in acetone (20 mL) was irradiated through a Pyrex filter for 2 hours. Thephotoproducts were isolated following the same procedure as used for the isolation of thephotoproduct in the solid state photolysis of (R,S)-(-)-ephedrine salt 108. This yielded amixture of photoproducts 67 and 68 (26.7 mg, 0.0681 mmol, 55%) in the ratio 81:19 asdetermined by GC (DB-1). No optical activity was observed when the photoproductmixture was subjected to polarimetry.Photolysis of (S,S)-(+)-Pseudoephedrine Salt 109A mixture of acid 38 (27.0 mg, 0.102 mmol) and (S,S)-(+)-pseudoephedrine(17.3 mg, 0.105 mmol) in acetone (20 mL) was irradiated for 2 h through a Pyrex filter.The regioisomeric and enantiomeric composition of the photoproducts was studied in thesame way as for the photolysis of the (R,S)-(-)-ephedrine salt 108 in acetone. Thesestudies showed that photoproducts 67 and 68 (25.8 mg, 0.0683 mmol, 67%) were formedin the ratio 9:1 and exhibited no optical activity.Photolysis of Salts 108 and 109 in the Solid State, Followed by Diazoethane WorkupCrystals of salts 108 and 109 were irradiated at -40°C. The photolyses werestopped before 25% conversion. The reactions were worked up as before, but201Chapter 3 Experimentaldiazoethane was substituted for diazomethane. This afforded the known photoproducts69 and 70 as a mixture. The regioisomeric and enantiomeric composition of thesemixtures was determined by 400 MHz 1H-NMR. For the enantiomeric excessdetermination, use was made of the chiral shift reagent (+)-Eu(hfc)3. The signalsmonitored were the methyl singlets at 8 3.4 and 3.5 for compounds 69 and 70respectively. In addition, the sign of rotation of the predominant enantiomer wasmeasured by polarimetry. The results of these experiments are summarized in Table 3-17.Table 3-17 Solid State Photoproduct Mixture Composition for Salts 108 and 109,after Diazoethane WorkupOptically Active Amine Photoproduct 69 Photoproduct 70Yield (%) eea Yield C .%) eea(R,S)-(-)-Ephedrine (108) 10 24 90 --(-)-95(S,S)-(+)-Pseudoephedrine (109) 80 16 20 18a The estimated accuracy in these values is t 5%; the sign of rotation of the predominant enantiomer isshown in parentheses.202Chapter 3 Experimental33.5 Absolute Configuration of Some Photoproducts33.5.1 Absolute Configuration of 4b,8b,8c,8d-tetrahydrodibenzo[a,f]cyclo-propa[c,d]pentalene-8b,8c-dicarboxylic acid, 8c-methyl 8b-ethyl ester (32a)(S)-(-)-Proline tert-butyl ester salt 96 (189 mg, 0.287 mmol) was photolyzed at -40°C for 5 days. The photoproduct was isolated following the same method as for thesolid state photolysis of S-(-)-proline salt 90a. This yielded the colorless oil ofphotoproduct 32a (16.6 mg, 0.0497 mmol, 17%). The optical rotation was measured bypolarimetry and the specific rotation was calculated assuming quantitative optical yield;bah, (+)-27° (chloroform, c= 0.02).The absolute configuration of photoproduct 32a was obtained by transforming itinto the di-isopropyl ester derivative 35, which has a known absolute configuration.43This was done in the following manner: a mixture of photoproduct 32a (16.6 mg, 0.0497mmol) from the photolysis described above, anhydrous isopropanol (25 mL) and H2SO4(0.5 mL) was refluxed for 3 weeks. GC (DB-17) showed formation of compound 35 withlittle remaining starting material. The solvent was removed under vacuum and theresidue dissolved in diethyl ether, washed with saturated aqueous NaCI solution anddried over MgSO4. After filtration, the solvent was removed under vacuum and theresulting oil chromatographed on silica gel eluted with 20% ethyl acetate in petroleumether to give, as a colorless oil, compound 35 (11.2 mg, 0.0298 mmol, 60%). The opticalrotation was measured and the specific rotation was calculated assuming quantitativeoptical yield, [a]D= (+)-26° (chloroform, C=0.01). Garcia-Garibay et a143 showed that203Chapter 3 Experimentalthe (+)-enantiomer of compound 35 has the absolute configuration (R,R,R,R) andtherefore it can be concluded that compound 32a has the same absolute configuration:"Compound 35 was characterized as diisopropyl 4b,8b,8c,8d-(R,R,R,R)-tetrahydrodibenzo[aAcyclopropa[c,d]pentalene-8b,8c-dicarboxylate. All the spectrawere identical to those reported by Garcia Garibay.57112 (CDC13) v.: 1723 (C=0), 1249 (C-0) cm-1.MS ink (relative intensity): 376 (M+, 15), 316 (13), 289 (28), 275 (16), 247 (100), 202(71).Exact mass calculated for C24H2404: 376.1675. Found: 376.1674.1H-NMR (400 MHz, CDC13) 5 7.4-7.0 (m, 8H, aromatic H), 5.22 (m, 1H,CO2CH(CH3)2), 5.05 (m, 1H, CO2CMCH3)2), 5.00 (s, 1H, benzylic H), 4.45 (s, 1H,cyclopropyl H), 1.5-1.2 (m, 12H, CO2CH(CH 3)2) ppm.33.5.2 Absolute Configuration of Methyl 8b-Methy1-4b,8b,8c,8d-tetrahydro-dibenzo[aAcyclopropa[c,d]pentalene-8c-carboxylate (52)Crystals of (S,S)-(+)-pseudoephedrine salt 105a were photolyzed at -40°C for 4days. The photoproduct was isolated following the same procedure as described before,yielding crystals of photoproduct 52 (95.1 mg, 0.350 mmol, 35%). The optical rotationof compound 52, was measured by polarimetry and the specific rotation was calculated tobe [a]p =(+)-59° (chloroform, c= 0.095) by assuming quantitative optical yield.A resolved chiral handle was attached to product 52 allowing the determination ofabsolute configuration by X-ray structure analysis. This was done in the following way;photoproduct 52 (64.2 mg, 0.34 mmol) from the photolysis described above washydrolyzed by dissolving in methanol (15 mL) with the addition of K 2CO3 (40.7 mg,204Chapter 3 Experimental0.295 mmol). The resulting solution was refluxed for 14 h. Afterwards the solvent wasremoved under vacuum and the residue dissolved in saturated aqueous NaHCO3 andwashed with ethyl acetate to remove unreacted starting material. The aqueous fractionwas made acidic with conc. HC1 and extracted with ethyl acetate. The organic layer waswashed with saturated aqueous NaCl, dried over MgSO4 and the solvent removed undervacuum. The resulting oil was recrystallized from ethanol to yield crystals of compound115 (49.8 mg, 0.191 mmol, 82%). The optical rotation was measured and the specificrotation calculated assuming quantitative optical yield, [a]D= (+)-73° (chloroform, C =0.050).Compound 115 was characterized as 8b-Methy1-4b,8b,8c,8d-tetrahydro-dibenzo [a,f] cyclopropa[c,d]pentalene-8c-carboxylic acid.MP: 223-224°C.IR (KBr)^3400-2400 (0-H, C-H), 1678 (C=0), 1257 (C-0) cm-t.MS Ink (relative intensity): 262 (M+, 4), 216 (100), 189 (12).Exact mass calculated for CI8H1402: 262.0994. Found: 262.0995.1H-NMR (400 MHz, CDC1 3) 8 7.3-7.0 (m, 8H, aromatic H), 4.98 (s, 1H, benzylic H),3.64 (s, 1H, cyclopropyl H), 1.94 (s, 3H, cyclopropyl CH3) ppm, no 0-H signaldetectable.13C-NMR (50 MHz, CDC13) 8 177.5 (C--0), 149.9, 149.0, 138.7, 138.4 (aromatic C),127.3, 126.9, 126.7, 126.5, 124.6, 123.9, 121.2, 121.2 (aromatic C-H), 66.1, 54.7(cyclopropyl C), 54.5, 52.4 (cyclopropyl C-H and benzylic C-H), 16.5 (cyclopropyl CH 3)ppm.Anal. calculated for CI8H1402: C, 82.42; H, 5.38. Found: C, 82.53; H, 5.29.A resolved chiral handle was introduced into compound 115 in the followingway: a solution of compound 115 (33.3 mg, 0.127 mmol) and oxalyl chloride (1 mL) in205Chapter 3 Experimentalanhydrous chloroform (6 mL) was refluxed for 30 min. The solvent and excess oxalylchloride were removed under vacuum to yield a yellow oil assumed to be thecorresponding acid chloride. The acid chloride was dissolved in anhydrous chloroform(10 mL), S-(-)-a-methylbenzylamine (1 mL, 0.94 g, 7.8 mmol) was added and theresulting solution was refluxed for 30 min. The reaction mixture was washed three timeswith saturated aqueous NaHCO3 solution and twice with saturated aqueous NaClsolution. The solvent was removed under vacuum and the residue chromatographed onsilica gel eluted with 30% ethyl acetate in petroleum ether. This yielded a colorless oilwhich was crystallized from ethanol to give compound 106 (42.0 mg, 0.110 mmol,92%). The optical rotation was measured with a polarimeter and the specific rotationcalculated assuming quantitative optical yield, [a]p = (+)-81° (chloroform, C = 0.016).The absolute configuration of compound 106 was obtained by an X-ray crystal structureanalysis.69Compound 106 was characterized as N-(S-1-phenylethyl) 8b-Methy1-4b,8b,8c,8d-(S,R,S,S)-tetrahydrodibenzo[a,flcyclopropa[c,d]pentalene-8c-carboxylic amide.MP: 174-175°C.IR (KBr) vmax: 3600-3200 (N-H and C-H), 1634 (C=0), 1538 (N-H) cm-1.MS ink (relative intensity): 365 (M+, 1), 217 (100), 202 (47), 105 (26).Exact mass calculated for C26H23NO: 365.1781. Found: 365.1786.1H-NMR (400 MHz, CDC13) 6 7.4-7.0 (m, 13H, aromatic H), 5.82 (m, 1H, NH), 5.20(m, 1H, HNCH), 4.68 (s, 1H benzylic H), 3.55 (s, 1H, cyclopropyl H), 1.80 (s, 3H,cyclopropyl CH3), 1.52 (m, 3H, HNCHCH3) ppm.13C-NMR (100 MHz, CDC13) 168.2 (C=0), 149.5, 148.5, 143.0, 139.5, 137.1 (aromaticC), 128.7, 128.7, 127.4, 127.0, 126.9, 126.6, 126.5, 126.3, 126.2, 126.1, 124.9, 124.1,121.1, 120.9 (aromatic C-H), 68.32 (cyclopropyl C), 55.71 (benzylic C-H), 52.24206Chapter 3 Experimental(cyclopropyl C-H), 49.2, 49.1 (cyclopropyl C-H and HNCE), 21.62, 15.99 (cyclopropylCH3 and HNCHC113) ppm.33.5.3 Absolute Configurations of Dimethyl 8d-Amino-4b,86,8c,8d-tetrahydro-dibenzo[a,ficyclopropa[c,d]pentalene-8b,8c-dicarboxylate (60)In order to obtain the absolute configuration of photoproduct 60, a resolved chiralhandle was introduced into a racemic mixture of compound 60. The two diastereomersthat formed were separated and an X-ray structure analysis of one of the diastereomerswas used to determine its absolute configuration. This was performed in the followingway; R-(-)-a-methoxyphenylacetic acid (104 mg, 0.627 mmol) and oxalyl chloride (5mL) were dissolved in anhydrous chloroform (30 mL) and refluxed for 30 min. Thesolvent and excess oxalyl chloride were removed under vacuum to yield a yellow oilwhich was assumed to be the corresponding acid chloride. The acid chloride wasdissolved in anhydrous chloroform (50 mL) and compound 60 (205 mg, 0.611 mmol)was added. The resulting solution was refluxed for 12h. GC (DB-1) indicated formationof compounds 120 and 121 in the ratio 1:1 with little remaining starting material. Thesolvent was washed thoroughly with saturated aqueous NaHCO3 solution, dried overMgSO4 and removed under vacuum to yield a mixture of compounds 120 and 121 (227mg, 0.470 mmol, 77%). Diastereomers 120 and 121 were separated on a preparativeHPLC column eluted with 30% ethyl acetate in n-hexane. All attempts to crystallizecompound 121 from various solvents failed, while diastereomers 120 was recrystallizedfrom an acetone-n-hexane solution to give prism-like crystals. X-ray analysis ofcompound 120 showed that it crystallizes with one equivalent of acetone and that is hasthe (R,R,S,S,R) configuration.69207Chapter 3 ExperimentalCompound 120 was characterized as dimethyl N-(R-(a)-methoxyphenyl aceticacid)-4b-amino-4b,8b,8c,8d-(R,S,S,R)-tetrahydrodibenzo[a,ficyclopropa[c,d]pentalene-8b,8c-dicarboxylate.[a]D = (+)-17° (chloroform, C = 0.005, assuming quantitative optical yield).)MP: 100-103°C.IR (KBr) v.,: 3406 (N-H), 1724, 1700 (C=O), 1504 (N-H), 1294, 1248, 1223 (C-0)cm-1.MS m/e (relative intensity): 483 (M+, 4), 424 (9), 392 (10), 302 (9), 260 (9), 217 (8),121 (100).Exact mass calculated for C29H25N06: 483.1683. Found 483.1684.1H-NMR (400 MHz, CDC13) 8 7.8-7.1 (m, 14H, aromatic H and N-H), 4.78 (s, 1H,CH3OCH), 4.42 (s, 1H, cyclopropyl H), 3.82 (s, 3H, COSH3), 3.55 (s, 3H, CO2CH3),3.19 (s, 3H, OCH3) ppm.13C-NMR (75 MHz, CDC13) 8 170.8, 168.4, 166.0 (C=O), 149.3, 148.2, 136.6, 131.8,130.7 (aromatic C), 128.2, 128.1, 127.9, 127.8, 127.7, 127.2, 127.1, 125.9, 119.0, 118.6(aromatic C-H), 83.7 (COCH 3), 74.8, 71.2 (benzylic C and cyclopropyl C), 57.6(CHOCH3), 53.7 (cyclopropyl C), 52.5, 51.8 (CO7CH3), 49.02 (cyclopropyl C-H) ppm.Compound 121 was characterized as dimethyl N-(R-(a)-methoxyphenyl aceticacid)-4b-amino-4b,8b,8c,8d-(S,R,R,S)-tetrahydrodibenzo[af]cyclopropa[c,d]pentalene-8b,8c-dicarboxylate.[a]D= (+)-53° (chloroform, c =0.023, assuming quantitative optical yield).IR (CHC13) v.: 3375 (N-H), 1738, 1731, 1698 (C=O), 1289, 1221 (C-0) cm-1.MS m/e (relative intensity): 483 (M+, 2), 424 (3), 392 (4), 260 (14), 217 (13), 149 (24),121 (100).Exact mass calculated for C29H25N06: 483.1683. Found 483.1688.208Chapter 3 Experimental1H-NMR (400 MHz, CDC13) 8 7.7-6.7 (m, 14H, aromatic H and N-H), 4.70 (s, 1H,CH3OCH), 4.44(s, 1H, cyclopropyl H), 3.82 (s, 3H, CO2CH3), 3.50 (s, 6H, CO,CH 3 andCH3OCH) ppm.13C-NMR (100 MHz, CDC13) 8 171.1, 168.3, 166.5 (C=0), 149.2, 148.1, 136.8, 131.6,130.7 (aromatic C), 128.4, 128.3, 128.2, 127.8, 127.7, 127.6, 126.8, 126.6, 125.7, 118.8,118.4 (aromatic C-H), 83.9 (CI110CH), 74.7, 70.7 (benzylic C and cyclopropyl C), 58.3(CH3OCH), 54.2, 52.6 (CO,CH 3), 48.7 (cyclopropyl C-H) ppm.The absolute configurations of the photoproduct 60 and the starting material, R-(-)-10-camphorsulfonic acid salt 116a were correlated as follows: crystals of R-(-)-10-camphorsulfonic acid salt 116a (136.1 mg, 0.253 mmol) were photolyzed at -40°C for 2days. The photoproduct was isolated following the same procedure as described before,yielding crystals of photoproduct 60 (38.9 mg, 0.116 mmol, 50%), which were formed in70% enantiomeric excess as determined by 400 MHz 1 H-NIVIR using chiral shift reagent,(+)-Eu(hfc)3 . The optical rotation was measured by polarimeter and the specific rotationcalculated as [a]D= (-)-27° (chloroform, C = 0.039).A portion of photoproduct 60 (11.8 mg, 0.0352 mmol) from the photolysisdescribed above was dissolved in anhydrous chloroform (10 mL) and R±)-a-methoxyphenyl acetic acid chloride (0.275 mmol) was added. This solution was refluxedfor 30 min, washed thoroughly with water and dried over MgSO4 . The solvent wasremoved under vacuum to yield a mixture of compounds 120 and 121. GC (DB-1)analysis showed formation of compounds 120 and 121 in the ratio 86:14, respectively.The product ratio was further supported by a 111-NMR study. It may therefore beconcluded that (-)-enantiomer of 60 which is the major photoproduct from solid stateirradiations of R-(-)-10-camphorsulfonic acid salt 116a, has the absolute configuration(4bS, 8bR, 8cR, 8dS).209REFERENCESREFERENCES^1 a^Morrison, J.D.; Mosher, H.S.; in Asymmetric Organic Reactions, AmericanChemical Society, Washington, D.C., 1976^b^Asymmetric Synthesis, Morrison, J.D.; Ed., Academic Press Inc., New York,1983, Vol. 1-5.c^Kagan, H.B.; Fiaud, J.C.; Top. Stereochem., 1978, 10, 175.2^Addadi, L.; Lahav, M.; Pure & Appl. Chem., 1979, 51, 1269.3 a^Green, B.S.; Lahav, M.; J. Mol. Evol., 1975, 6, 99.b^Addadi, L.; Lahav, M.; in Origins of Optical Activity in Nature, Walker, D.C.,Ed., New York, 1979, Chapter 14.c^Bonner, W.A.; Top. Stereochem., 1988, 18, 1.d McBride, J.M.; Carter, R.L.; Angew. Chem. Int. Ed. Engl., 1991, 30, 293.^4 a^Rau, H.; Chem. Rev., 1983, 83, 535.b^Inoue, Y.; Chem. Rev., 1992, 92, 741.c^Inoue, Y.; Yamasaki, N.; Yokoyama, T.; Tai, A.; J. Org. Chem., 1992, 57, 1332.^5 a^Green, B.S.; Lahav, M.; Rabinovich, D.; Acc. Chem. Res., 1979, 12, 191.b^Ramamurthy, V.; Tetrahedron, 1986, 42, 5753.c^Scheffer, J.R.; Garcia-Garibay, M.A.; in Studies in Surfaces Science andCatalysis, Anpo, M.; Matsuura. T.; Eds.; Elsevier, Amsterdam, 1989, Vol. 47,501.d Hollingsworth, M.D.; McBride, J.M.; in Advances in Photochemistry, Volman,D.; Hammond, G.; Gollnick, K.; Eds., Interscience, New York, 1990, Vol. 15,279.e Vadia, M.; Popovitz-Biro, R.; Leiserowitz, L.; Lahav, M.; in Photochemistry inOrganized and Constrained Media, Rarnamurthy, V., Ed.; 1991, Chapter 6.^6 a^Dunitz, J.D.; X-ray Analysis and the Structure of Organic Molecules, CornellUniversity Press, Ithaca, New York, 1979, p 312.210REFERENCESb^Buckert, U.; Allinger, N.L.; Molecular Mechanics, Am. Chem. Soc. Monograph,Washington, D.C., 1982, p 177.7 a^Riiber, C.N.; Chem. Ber., 1902, 35, 2411.b^Ciamician, G.; Silber, P.; Chem. Ber., 1902, 35, 4128.8 The following is a list of some of the review articles on the subject of chemical studiesin organic crystals in recent years.a^Gavezzotti, A.; Simonetta, M.; Chem. Rev., 1982, 82,1.b^Hasegawa, M.; Chem. Rev., 1983, 83, 507.c^McBride, J.M.; Acc. Chem. Res., 1983, 16, 304.d Trotter, J.; Acta Crystallogr., Sect B, 1983, B39, 373.e Green, B.S.; Arad-Yellin, R.; Cohen, M.D.; Top. Stereochem., 1986, 16, 131.f^Ramamurthy, V.; Venkatesan, K.; Chem. Rev., 1987, 87, 433.g Organic Solid State Chemistry, Desiraju, G.R., Ed.; Elsevier, Amsterdam, 1987.h Organic Chemistry in Anisotropy Media, Scheffer, J.R.; Turro, N.J .;Ramamurthy, V.; Eds., Tetrahedron Symposia-in-Print, Number 29,Tetrahedron, 1987.i^Cohen, M.D.; Tetrahedron, 1987, 43, 1211.j^Scheffer, J.R.; Garcia-Garibay, M.; Nalamasu, 0.; in Organic Photochemistry,Padwa, A.; Ed., Marcel Dekker, New York, 1987, Vol. 2, Part 2, Chapter 20.k^^Photochemistry on Solid Surfaces, Anpo, M.; Matsuura, T.; Eds., Elsevier,Amsterdam, 1989.1^Photochemistry in Organized and Constrained Media, Ramamurthy, V., Ed.;VCH, New York, 1991.m^Chen, J.; Scheffer, J.R.; Trotter, J.; Tetrahedron, 1992, 48, 3251.9^Kohlshutter, H.W.; Z. Anorg. Allg. Chem., 1918, 105, 121.10 a^Cohen, M.D.; Schmidt, G.M.J.; Sonntag, F.I.; J. Chem. Soc., 1964, 2000.b^Schmidt, G.M.J.; J. Chem. Soc., 1964, 2014.11 a^Cohen, M.D.; Schmidt, G.M.J.; J. Chem. Soc., 1964, 1996.b^Schmidt, G.M.J.; Pure AppL Chem., 1971, 27, 647.12 a^Cohen, M.D.; Angew. Chem., Int. Ed. Engl., 1975, 14, 386.b^Cohen, M.D.; Mol. Cryst. Liq. Cryst., 1979, 50,1.211REFERENCES13 Ramamurthy,^V.;^Weiss,^R.G.;^Hammond,^G.S.^in^Advantages^inPhotochemistry, Hammond, G.S.; Neckers, D.C.; Volman, D.; Eds., Interscience,New York, in press.14 Wegner, G.; Pure Appl. Chem., 1977, 49, 443.15 a Jones, W.; Nakanishi, H.; Theocharis, C.R.; Thomas, J.M.; J. Chem. Soc.,Chem., Commun., 1980, 610.b Nakanishi, H.; Jones, W.; Thomas, J.M.; Hursthouse, M.B.; Motevalli, M.; J.Chem. Soc., Chem. Commun., 1980, 611.c Nakanishi, H.; Jones, W.; Thomas, J.M.; Hursthouse, M.B.; Motevalli, M.; J.Phys. Chem.,1981, 85, 3636.16 a Ariel, S.; Askari, S.; Scheffer, J.R.; Trotter, J.; Walsh, L.; J. Am. Chem. Soc.,1984, 106, 5726.b Ariel, S.; Askari, S.; Evans, S.V.; Hwang, C.; Jay, J.; Scheffer, J.R.; Trotter, J.;Walsh, L.; Tetrahedron, 1987, 43, 1253.c Scheffer J.R.; Trotter, J. in The Chemistry of Quinonoid Compounds, Patai, S.;Rappoport, Z.; Eds.; John Wiley & Sons, 1988, Vol. 2, Chapter 20.17 a Appel, W.K.; Jiang, Z.Q.; Scheffer, J.R.; Walsh, L.; J. Am. Chem. Soc., 1983,105, 5354.b Scheffer, J.R.; Trotter, J.; GuOmundsdOttir, A.D.; in Handbook of OrganicPhotochemistry and Photobiology, Horspool, W.M.; Ed., CRC Press, in press.18 GuOmundsd6ttir, A.D.; Scheffer, J.R.; Tetrahedron Lett., 1989, 30, 423.19 Gavezzotti, A.; J. Am. Chem. Soc., 1983, 105, 5220.20 Gavezzotti, A.; Tetrahedron, 1987, 43, 1241.21 Wang, W.N.; Jones, W.; Tetrahedron, 1987, 43, 1273.22 Yap, M.P.; Ph.D. Thesis, University of British Columbia, 1992.212REFERENCES23 Jacques, J.; Collet, A.; Wilen, S.H.; Enantiomers, Racemates and Resolutions,Wiley Interscience, New York, 1983.24 a Buerger, MJ.; Elementary Crystallography, Wiley, New York, 1963, 199.b Hahn, T.; Klapper H.;, in International Tables for Crystallography, Hahn, T.;Ed., Reidel, Dordrecht, Holland, 1983, Vol. A, Chapter 10.25 For discussion of Pasteur's work on sodium ammonium tartrate, see Fieser, M.; inAdvanced Organic Chemistry, Reinhold, New York, 1961, p 69.26 a Pincock, R.E.; Wilson, K.R.; J. Am. Chem. Soc., 1971, 93, 1291.b Pincock, R.E.; Perkins, R.R.; Ma, A.S.; Wilson, K.R.; Science, 1971, 174, 1018.27 Desiraju, G.R.; Crystal Engineering: The Design of Organic Solids, Elsevier,New York, 1989.28 Green, B.S.; Lahav, M.; Schmidt, G.M.J.; Mol. Cwt. Liq. Cryst., 1975, 29, 187.29 Elgavi, A.; Green, B.S.; Schmidt, G.M.J.; J. Am. Chem. Soc., 1973, 95, 2058.30 Rabinovich, D.; Shakked, Z.; Acta Crystallogr., Sect. B, 1975, B31, 819.31 Warrshel, A.; Shakked, Z.; J. Am. Chem. Soc., 1975, 97, 5679.32 Hasegawa, M.; in Organic Solid State Chemistry, Desiraju, G.R.; Ed., Elsevier,New York, 1987, p 153.33 a Addadi, L.; Lahav, M.; J. Am. Chem. Soc., 1978, 100, 2838.b Addadi, L.; Lahav, M.; J. Am. Chem. Soc., 1979, 101, 2152.c Addadi, L.; van Mil, J.; Lahav, M.; J. Am. Chem. Soc., 1982, 104, 3422.34 a Hasegawa, M.; Kunita, A.; Chung C.; Hayashi, K.; Sato, S.; Chem. Lett, 1989,641.b Hasegawa, M.; Chung, C.M.; Muro, N.; Maekawa, Y.; J. Am. Chem. Soc., 1990,112, 5676.213REFERENCES^35 a^Evans, S.V.; Garcia-Garibay, M.; Omkaram, N.; Scheffer, J.R.; Trotter, J.;Wireko, F.; J. Am. Chem. Soc.,1986, 108, 5648.b^Evans, S.V.; Trotter, J.; Acta Crystallogr., Sect. B, 1989, B45, 500.36 a^Toda, F.; Yagi, M.; Soda, S.; J. Chem. Soc., Chem. Commun., 1987, 1413.b^Sekine, A.; Hori, K.; Ohashi, Y.; Yagi, M.; Toda, F.; J. Am. Chem. Soc., 1989,111, 697.^37 a^Feng, X.W.; McBride, J.M.; J. Am. Chem. Soc., 1990, 112, 6152.b^McBride, J.M.; Bertman, S.B.; Cioffi, D.Z.; Segmuller, B.E.; Weber, B.A.; Mol.Cryst. Liq. Cryst. Inc. Nonlin. Opt., 1988, 161,1.^38 a^Holland, H.L.; Richardson, M.F.; MoL Cryst. Liq. Cryst., 1980, 58, 311.b^Chenchaiah, P.C.; Holland, H.L.; Richardson, M.F.; J. Chem. Soc., Chem.Commun., 1982, 436.39 a^Hixon, S.S.; Mariano, P.S.; Zimmerman, H.E.; Chem. Rev., 1973, 73, 531b^Zimmerman, H.E.; in Rearrangements in Ground and Excited States, de Mayo,P.; Ed., Wiley Interscience, New York, 1980, Vol. 3. Chapter. 16.40 a^Zimmerman, H.E.; Grunewald, G.L; J. Am. Chem. Soc.; 1966, 88, 183.b^Zimmerman, H.E.; Binkley, R.W.; Givens, R.S.; Sherwin, M.A.; J. Am. Chem.Soc., 1967, 89, 3932.41^De Lucchi, 0.; Adam, W.; in Comprehensive Organic Synthesis, Trost, B.M.;Fleming, I.; Paquette, L.A.; Eds, Pergamon Press, Oxford, 1991, Vol. 5.42 a^Scheffer, J.R.; Trotter, J.; Garcia-Garibay, M.; Wireko, F.C.; MoL Cryst. Liq.Cryst. Inc. Nonlin. Opt.,1988, 156, 63.b^Garcia-Garibay, M.; Scheffer, J.R.; Trotter, J.; Wireko, F.C.; Acta Crystallogr.,Sect. B, 1990, B46, 79.c^Trotter, J.; Wireko, F.C.; Acta Crystallogr., Sect. C, 1991, C47, 793.43 a^Garcia-Garibay, M.; Scheffer, J.R.; Trotter, J.; Wireko, F.; J. Am. Chem. Soc.;1989, 111, 4985.214REFERENCESb^Garcia-Garibay, M.; Scheffer, J.R.; Trotter, J.; Wireko, F.; Acta Crystallogr.,Sect. B, 1990, B46, 431.44^Bijvoet, J.M.; Peerdeman, A.F.; Van Bommel, J.A.; Nature, 1951, 168, 271.45 a^Chan, R.S.; Ingold, C.; Prelog, V.; J. Chem. Soc., London, 1951, 612.b^Chan, R.S.; Ingold, C.; Prelog, V.; Experientia, 1956, 12, 81.c^Chan, R.S.; Ingold, C.; Prelog, V.; Angew. Chem., Int. Ed. Engl., 1966, 5, 385.46^Diels, O.; Alder, K.; Justus Liebigs Ann. Chem., 1931, 486, 191.47 a^Morgan, G.T.; Harrison, H.A.; J. Soc. Chem. Ind.; London, Trans. Commun.,1930, 49, 143.b^Bartlett, P.D.; Cohen, S.G.; J. Am. Chem. Soc., 1940, 62, 1183.48^Bartlett, P.D.; Greene, F.D.; J. Am. Chem. Soc., 1954, 76, 1088.49^Adam-Briers, M.; Fierens, P.J.C.; Maritims, R.H.; Hely. Chim. Acta, 1955, 38,2009.50^Vaughan, W.R.; Milton, K.M.; J. Am. Chem. Soc., 1952, 74, 5623.51^Chen, J.; Ph.D. Thesis, University of British Columbia, 1991.52^Lindgren, B.O. Nilsson, T.; Acta Chem. Scand., 1973, 27, 888.53^Zimmerman, H.E.; in Organic Photochemistry, Ed. Padwa, A.; Marcel Dekker,New York, 1991, Vol. 10, 1.54 a^Garcia-Garibay, M.; Scheffer, J.R.; Watson, D.G.; J. Chem. Soc., 1989, 600.b^Garcia-Garibay, M.; Scheffer, J.R.; Watson, D.G.; J. Org. Chem., 1992, 26, 241.55 a^Cristol, S.J.; Parungo, F.P.; Plorde, D.E.; Schwarzenbach, K.; J. Am. Chem. Soc.,1965, 87, 2879.b^Cristol, S.J.; Parungo, F.P.; Plorde, D.E.; J. Am. Chem. Soc., 1965, 87, 2870.c^Cristol, S.J.; Bopp, R.J.; Johnson, A.E.; J. Org. Chem., 1969, 11, 3574.215REFERENCES56^Silverstein, R.B.; Bassler, G.C.; Morill, T.C.; Spectrometric Identification ofOrganic Compounds; John Wiley & Sons, 1981, Chapter 3.57^Garcia-Garibay, M.; Ph.D. Thesis, University of British Columbia, 1988.58^Ciganek, E.: J. Am. Chem. Soc., 1966, 88, 2882.59^Rattray, G.; Yang, J.; Gu8mundsdOttir, A. D, Scheffer, J.R.; Tetrahedron Lett.,1993, 34, 45.60^Bordwell, F.; Lynch, T.Y.; J. Am. Chem. Soc., 1989, 111, 7558.61^Ceppi, E.; Eckhardt, W.; Grob, C.A.; Tetrahedron Lett., 1973, 37, 3627.62^Cristol, S.J.; Kaufman, R.L.; Opitz, S.M.; Szalecki, W.; Bindel, T.H.; J. Am.Chem. Soc., 1983, 105, 3226.63^Paddick, R.G.; Richards, K.E.; Wright, G.J.; Aust. J. Chem., 1976, 29, 1005.64^Richards, K.E.; Tillman, R.W.; Wright, G.J.; Aust. J. Chem., 1975, 28, 1289.65^Iwamura, M.; Tukada, H.; Iwamura, H.; Tetrahedron Lett., 1980, 21, 4865.66 a^Gunther, H.; Tetrahedron Lett., 1970, 52, 5173.b^Hoffman, R.; Tetrahedron Lett., 1970, 52, 2907.c^Hoffman, R.; Stohrer, W-D.; J. Am. Chem. Soc., 1971, 93, 6941.d^Gunther, H.; Wehner, R.; J. Am. Chem. Soc., 1975, 97, 923.67^X-ray crystal structures done by Gu8mundsdattir, A.D.; Chemistry Department,University of British Columbia.68^Demuth, M.; Amrein, W.; Bender, C.0; Braslaysky, S.E.; Burger, U.; George,M.V.; Lemmer, D.; Schaffner, K.; Tetrahedron, 1981, 37, 3245.216REFERENCES69^X-ray crystal structures done by Rettig, S.J.; Chemistry Department, University ofBritish Columbia.70^Paquette, L.A.; Bay, E.; J. Am. Chem. Soc., 1984, 106, 6693.71^Pokkuluri, P.R.; Ph.D. Thesis, University of British Columbia, 1990.72^Zimmerman, H.E.; Givens, R.S.; Pagni, R.M.; J. Am. Chem. Soc., 1968, 90, 6090.73^Zimmerman, H.E, Bender, C.O.; J. Am. Chem. Soc., 1970, 92, 4366.74^Pokkuluri, P.R.; Scheffer, J.R.; Trotter, J.; J. Am. Chem. Soc.; 1990, 112, 3676.75^Asokan, C.V.; Kumar, S.A.; Das, S.; Rath, N.P.; George, M.V.; J. Org. Chem.;1991, 56, 5890.76^Zimmerman, H.E.; Zuraw, M.J.; J. Am. Chem. Soc., 1989, 111, 7974.77^Pokkuluri, P.R.; Scheffer, J.R.; Trotter, J.; Tetrahedron Lett.; 1989, 30, 1061.78^Kumar, C.V.; Murty, B.A.R.C.; Lahiri, S.; Chakachery, E.; Scaiano, J.C.; George,M.V.; J. Org. Chem.; 1984, 49, 4923.79^X-ray crystal structure done by Kaftory, M.; Chemistry Department, University ofBritish Columbia.80^GuamundsdOttir, A.D.; Scheffer, J.R.; Tetrahedron Lett., 1990, 31, 6807.81^Morrison, R.T.; Boyd R.N.; Organic Chemistry, 4th ed, Allyn and Bacon, 1983.82^Scheffer, J.R.; Fu, T.; Trotter, J.; Unpublished results.83^Parker, D.; Chem. Rev.; 1991, 91, 1441.217REFERENCES84^Crans, D.C.; Whitesides, G.M.; J. Am. Chem. Soc.; 1985, 107, 7019.85^Meresse, A.; Courseille, C.; Lerory, F.; Chanh, N.B.; Acta Crystallogr., Sect. B,1979, B35, 2087.86^Chen, J.X.; Garcia-Garibay, M.A.; Scheffer, J.R.; Tetrahedron Lett., 1989, 30,6125.87^Nakanishi, F.; Nakanishi, H.; Tasai, T.; Suzuki, Y.; Hasegawa, M.; ChemistryLett., 1974, 525.88^Gu6mundsd6ttir, A.D.; Scheffer, J.R.; Photochemistry and Photobiology; 1991, 4,535.89^X-ray crystal structure done by Li, W. and Trotter, J., Chemistry Department,University of British Colombia.90 a^Barrett, G.C.; in Techniques of Chemistry, Bentley, K.W; Kirby, G.W.; Eds.;1963, Vol. 4 , Chapter 8.b^Purdie, N.; Swallows, K.A.; Analytical Chemistry, 1989, 61, 77A.91^Lewis, T.J.; Rettig, S.J.; Scheffer, J.R.; Trotter, J.; Wireko, F.; J. Am. Chem. Soc.,1991, 112, 3679.92^Jones, R.; Scheffer, J.R.; Trotter, J.; Yang, J.; Tetrahedron Lett., 1992, 33, 5484.93^Dun, H.; Angew. Chem., Int. Ed. Engl., 1989, 28, 413.94^Exelby, R.; Grinter, R.; Chem. Rev.; 1965, 65, 247.95^The UV-visible spectra and the ESR spectra were obtained by Ramamurthy, V.;Central Research and Development, the Dupont Company, Delaware.96^Thompson, C.; in Electron Spin Resonance: A Specialist Periodical Report, TheChemical Society, London, 1973, Vol. 1, Chapter 1.218REFERENCES97 Pokkuluri, P.R.; M.Sc. Thesis, University of British Columbia, 1987.98 a Rehm, D.; Weller, A.; Isr. J. Chem.; 1970, 8, 259.b Mariano, P.S.; Stavinoha, J.L.; Synthetic Organic Photochemistry, Horspool,W.M., Ed.; Plenum Press, 1984, Chapter 3.c Mattes, S.L.; Farid, S.; in Organic Photochemistry, Padwa. A., Ed.; MarcelDekker, New York, 1983, Vol. 6, Chapter 4.d Eberson, L.; Rees, C.W.; Electron Transfer Reactions in Organic Chemistry,Springer Verlag, Berlin, 1987.99 Sakaguchi, H.; Nagamura, T.; Matsuo, T.; J. Chem. Soc., Chem. Comm., 1992,209.100 Effio, A.; Griller, D.; Ingold, K.U.; Scaiano, J.C., Sheng, S.J, J. Am. Chem. Soc.,1980, 102, 6063.101 Cox, S.D.; Dirk, C.W.; Moraes, F.; Wellman, D.E.; Wudl, F.; Solits, M.; Strouse,C.; J. Am. Chem. Soc., 1984, 106, 7131.102 Scheffer J.R.; Watson, D.; unpublished results.103 a Diaz de Delgado, G.C.; Wheeler, K.A.; Snider, B.B.; Foxman, B.M.; Angew.Chem. Int. Ed. Engl., 1991, 30, 420.b Restaino, A.J.; Mesrobian, R.B.; Morawetz, H.; Ballantine, D.S.; Dienes, G.J.;Metz, D.J.; J. Am. Chem. Soc.; 1956, 78, 2939.c Katchalsky, A.; Blauer, G.; Trans. Faraday. Soc., 1951, 47, 1360.d Morawetz, H.; J. Polym. Sci. (C), 1966, 79.104 Radiations Chemistry, Principles and Applications, J.E. Willard in Farhataziz,Roger, M.A.J., Eds.; VCH, Weinheim, 1987, 395.105 Perrin, D.D.; Armarego, W.L.F.; Perrin, D.R.; Purification of LaboratoryChemicals, 2nd ed., Pergamon Press, Oxford, 1980.106 Figeys, H.P.; Dralants, A.; Tetrahedron, 1972, 28, 3031.219REFERENCES107^Mead, J.F.; Rapport, M.M.; Senear, A.E.; Maynard, J.T; Koepfli, J.B.; J. Biol.Chem., 1946, 163, 465.108 a^Murov, S.H.; Handbook of Photochemistry, Marcel Dekker, Inc., New York,1973.b^Horspool, W.M.; Synthetic Organic Photochemistry, Horspool, W. Ed.; PlenumPress, 1984, Chapter 9.109^Hemetzberger, H.; Holstein, W; Werres, F.; Tetrahedron, 1983, 39, 1151.220

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059590/manifest

Comment

Related Items