Disponible à / Available at permalink :

202  Download (0)

Texte intégral

(1)

- - -

- - -

Dépôt Institutionnel de l’Université libre de Bruxelles / Université libre de Bruxelles Institutional Repository

Thèse de doctorat/ PhD Thesis Citation APA:

Didier, D. (2009). Functionalized analogues of Tröger's base: synthesis, enantioseparation, and application as a chiral scaffold in organocatalysis (Unpublished doctoral dissertation). Université libre de Bruxelles, Faculté des Sciences – Chimie, Bruxelles.

Disponible à / Available at permalink : https://dipot.ulb.ac.be/dspace/bitstream/2013/210277/6/9fb1990d-9010-4de1-ab67-6f9a662f4e5e.txt

(English version below)

Cette thèse de doctorat a été numérisée par l’Université libre de Bruxelles. L’auteur qui s’opposerait à sa mise en ligne dans DI-fusion est invité à prendre contact avec l’Université (di-fusion@ulb.ac.be).

Dans le cas où une version électronique native de la thèse existe, l’Université ne peut garantir que la présente version numérisée soit identique à la version électronique native, ni qu’elle soit la version officielle définitive de la thèse.

DI-fusion, le Dépôt Institutionnel de l’Université libre de Bruxelles, recueille la production scientifique de l’Université, mise à disposition en libre accès autant que possible. Les œuvres accessibles dans DI-fusion sont protégées par la législation belge relative aux droits d'auteur et aux droits voisins. Toute personne peut, sans avoir à demander l’autorisation de l’auteur ou de l’ayant-droit, à des fins d’usage privé ou à des fins d’illustration de l’enseignement ou de recherche scientifique, dans la mesure justifiée par le but non lucratif poursuivi, lire, télécharger ou reproduire sur papier ou sur tout autre support, les articles ou des fragments d’autres œuvres, disponibles dans DI-fusion, pour autant que :

Le nom des auteurs, le titre et la référence bibliographique complète soient cités;

L’identifiant unique attribué aux métadonnées dans DI-fusion (permalink) soit indiqué;

Le contenu ne soit pas modifié.

L’œuvre ne peut être stockée dans une autre base de données dans le but d’y donner accès ; l’identifiant unique (permalink) indiqué ci-dessus doit toujours être utilisé pour donner accès à l’œuvre. Toute autre utilisation non mentionnée ci-dessus nécessite l’autorisation de l’auteur de l’œuvre ou de l’ayant droit.

--- English Version ---

This Ph.D. thesis has been digitized by Université libre de Bruxelles. The author who would disagree on its online availability in DI-fusion is invited to contact the University (di-fusion@ulb.ac.be).

If a native electronic version of the thesis exists, the University can guarantee neither that the present digitized version is identical to the native electronic version, nor that it is the definitive official version of the thesis.

DI-fusion is the Institutional Repository of Université libre de Bruxelles; it collects the research output of the University, available on open access as much as possible. The works included in DI-fusion are protected by the Belgian legislation relating to authors’ rights and neighbouring rights.

Any user may, without prior permission from the authors or copyright owners, for private usage or for educational or scientific research purposes, to the extent justified by the non-profit activity, read, download or reproduce on paper or on any other media, the articles or fragments of other works, available in DI-fusion, provided:

The authors, title and full bibliographic details are credited in any copy;

The unique identifier (permalink) for the original metadata page in DI-fusion is indicated;

The content is not changed in any way.

It is not permitted to store the work in another database in order to provide access to it; the unique identifier (permalink) indicated above must always be used to provide access to the work. Any other use not mentioned above requires the authors’ or copyright owners’ permission.

(2)

D 03676

LIBRE DE BRUXELLES

Faculté des Sciences

Laboratoire de Chimie des Polymères

O oc Z3

UJ

a

<UJ t-

QC Ul

>

Z

:d

F unctionalized analogues of T rôger ' s base

SYNTHESIS, ENANTIOSEPARATION,

AND APPLICATION AS A CHIRAL SCAFFOLD IN ORGANOCATALYSIS

Delphine Didier

Thèse présentée en vue de L'obtention du grade académique de

Docteur en Sciences

Août 2009

Thèse réalisée sous la supervision du

Professeur Yves Geerts et du Docteur Sergey Sergeyev

Université

Libre de .Bruxelles

(3)

■'... . ifrlo6/<p^

A ma famille,

(4)

Qu'est-ce donc que la vérité?

C'est l'équilibre fragile qui naît du choc des antagonismes.

C'est la blanche écume des vagues.

C'est le parfum, synthèse de tous les ingrédients qui mijotent dans la marmite.

La vérité n'est point monolithique.

Elle est enrichissement réciproque dans le respect des contraintes.

Irénée Guilane Dioh

Tout ce que nous voyons n'est qu'une ombre projetée par les choses que nous ne voyons pas.

Martin Luther King

(5)

Some parts of this thesis hâve aireadv been published:

1. D. Didier, S. Sergeyev. Analogues of Trôger's base: recent developments and controversies.

Targets in Heterocydie Systems, Eds. A. Ottanasi, D. Spinelli, vol. 11, 2007, 258-283.

2. D. Didier, S. Sergeyev. Synthesis of symmetrical amino and aminomethyl dérivatives of Trôger's base via Pd-catalyzed C-C and C-N bond formation. Tetrahedron, 2007, 63, 3864-3869.

3. D. Didier, S. Sergeyev. Bromination and iodination of 6H,12H-5,ll-methanodibenzo [^>»/][l»5]diazocine: a convenient entry to unsymmetrical analogs of Trôger's Base. fur. J. Org.

Chem, 2007, 3905-3910.

4. V. Lemaur, J. Cornil, D. Didier, A. Mujawase, S. Sergeyev. A joint theoretical and experimental insight into the electronic structure of chromophores derived from 6H,12H-5,11- methanodibenzo[b,f][l,5] diazocine. Helv. Chim. Acta, 2007, 2087-2095.

5. D. Didier, S. Sergeyev. Analogues of Trôgers base: recent developments and controversies. Chimie Nouvelle, 2007, 92-97.

6. C. M. L. Vande Velde, D. Didier, F.BIockhuys and S. Sergeyev. Racemic 1,2,3,4,7,8,9,10-octafluoro- 6H,12H-5,ll-methanodibenzo[b,f][l,5]-diazocine: an octafluorinated analogueof Trôger's base.

Acta Crystallogr., Sect. E: Struct. Rep. Online., 2008, E64, o538

7. D. Didier, B. Tylleman, N. Lambert, C. M. L. Vande Velde, F. Blockhuys, A. Collas, S. Sergeyev.

Functionalized analogues of Trôger's base: scope and limitations of a general synthetic procedure and facile, predictable method for the séparation of enantiomers. Tetrahedron, 2008, 64, 6252- 6262.

(6)

Some parts of this thesis were presented:

Oral présentations

1. Synthesis of symmetrical and unsymmetrical dérivatives of Trôger's base and their synthetic applications, Sème édition des journées de Rencontres des Jeunes Chimistes, Herbeumont-sur- semois, Belgium, March 8-9 2007.

2. Synthesis of symmetrical and unsymmetrical dérivatives of Trôger's base and their synthetic applications. Journée de l'École Doctorale Thématique CHIM, Université Libre de Bruxelles, Bruxelles, April 17 2007.

3. Functional analogues of Trôger's base : New synthetic methods and facile, predictable chromatographie means for the séparation of enantiomers, 12''’ Sigma-Aldrich Organic Synthesis Meeting, Sol Cress, Spa, December 4-5 2008.

4. Functional analogues of Trôger's base: new synthetic methods and facile, predictable chromatographie means for the séparation of enantiomers. Journée de rencontres des Jeunes Chimistes, 6ème édition. Leonardo Hôtel, Namur, March 19-20 2009.

Posters

1. Halogénation of 6H,12H-5,ll-methanodibenzo[ô,f][l,5]-diazocine: convenient entry to unsymmetrical dérivatives of Trôger's base. Organic Chemistry: Présent and Future, International Symposium in honour of Prof. Léon Ghosez, Louvain-la-Neuve, Belgium, April 10-13, 2007.

2. Functional dérivatives of Trôger base: towards novel bifunctional organocatalysts and chromophores, ISth European Symposium on Organic Chemistry, Dublin, Ireland, July 08-13, 2007.

3. Synthesis of symmetrical and unsymmetrical dérivatives of Trôger's base and their synthetic applications, ISth European Symposium on Organic Chemistry, Dublin, Ireland, July 08-13, 2007.

4. Synthesis of symmetrical and unsymmetrical dérivatives of Trôger's base and their synthetic applications, 23"' European Colloquium on Heterocyclic Chemistry, Antwerp, Belgium, September 9- 13, 2008.

5. Synthesis of symmetrical and unsymmetrical dérivatives of Trôger's base and their synthetic applications, 2"'' EuCheMS Chemistry Congress, Torino, Italy, September 16-20 2008.

(7)

Acknowledgments

Je remercie les Professeurs François Reniers, Ivan Jabin, Michel Luhmer et Jean-Claude Braekman ainsi que le Professeur Yves Geerts et le Docteur Sergey Sergeyev d'avoir accepté de faire partie du jury de cette thèse.

La réalisation d'une thèse de doctorat peut être comparée à la formation d'une jeune recrue des forces d'intervention spéciales. On en ressort plus fort à bien des niveaux mais elle ne pourrait se réaliser sans l'aide de nombreuses personnes.

C'est pourquoi je tiens tout particulièrement à remercier François Reniers. D'abord poiu"

avoir fait basculer mon choix pour des études de chimie en février 2000 quand j'ai assisté à un de ses coirrs. Ensuite, pour la confiance qu'il m'a accordée en me recrutant dans son équipe pédagogique à la fin de ma licence. J'ai adoré mon poste d'assistante. Merci, enfin, pour son soutien, tous ses bons conseils et son humour...

Ce travail n'aiu-ait jamais vu le jour sans Yves Geerts. Je tiens à le remercier pour m'avoir permis de réaliser une thèse de doctorat dans son laboratoire mais aussi pour avoir toujours réussi à tempérer avec tact et psychologie l'ardeur de mon tempérament. J'ai beaucoup appris tant au niveau scientifique qu'au niveau personnel. Je le remercie aussi pour son enthousiasme, ses nombreux encouragements, ainsi que de m'avoir permis de bénéficier d'un équipement de pointe tout au long de ma thèse (surtout à Noël, merci Père Noël!).

Then I would like to thank Sergey Sergeyev. I am grateful for his advices, and for the fruitful collaboration that we hâve developed during my PhD. Thank you for your help and support.

Even if we did not always agréé, I did learn a lot in many fields. Thank you!

Je voudrais également remercier Rita D'Orazio pour le relevé de nombreux spectres RMN (haute température, 10000 scans, merci !), ainsi que le Professeur Michel Luhmer pour ses nombreux conseils scientifiques.

I am grateful to Professor Frank Blockhuys, Professor Matthias Zeller and Professor Luc van Meervelt for the collection of experimental data sets for single crystal XRD analysis, and to Dr. Christophe Vande Velde for the refinement of crystal structures.

I would like to thank the FNRS for the grant for chiral stationary phases.

Je remercie, le service de « structure et fonction des membranes biologiques » pour m'avoir permis d'employer leur spectropolarimètre, ainsi que le service de « pharmacognosie, bromatologie et nutrition humaine » pour m'avoir permis d'utiliser leur polarimètre.

Ensuite, je tiens tout particulièrement à remercier Adrierme Remacle pour m'avoir appris à

dompter THPLC, m'avoir aidée dans la mesure de nombreux OR, spectres UV et CD et poiu-

avoir réalisé la séparation préparative de nombreux racémats. Je la remercie également pour

son efficacité dans la gestion pratique du laboratoire, sa bonne humeur, son café le matin,

son soutien moral et j'en passe. Merci Adrienne !

(8)

Au cours de ma thèse de doctorat j'ai eu la chance d'encadrer de nombreux étudiants pendant les TP ainsi que pendant les vacances d'été (merci Yves !). Je voudrais leur exprimer toute ma gratitude pour avoir contribué par leur aide technique à la réalisation de cette thèse de doctorat. Merci à Aline, Mihaly, Natacha, Jean-François et Bob ! Je remercie aussi Saloua qui m'a permis de créer des stocks d'intermédiaires, et, last but not least, je tiens à souligner la contribution de Benoit Tylleman. Il a permis l'isolation de nombreux intermédiaires, réalisé de nombreuses synthèses mais également permis par ses bonnes ondes, la croissance de monocristaux enantiopures. Sans sa patience, j'attendrais encore !

Ik zou ook graag Cris bedanken voor al de X-straal metingen en voor de bepaling van de absolute configuraties, die in deze thesis gepresenteerd worden. Ik bedank hem ook voor al de uitleg die hij me gegeven heeft en voor de verklaring van het principe van de Flack parameter.

Je profite de l'occasion pour remercier tous mes collègues du labo : OU, Betty, JY, Flo, Cricri, Claire, Gabin, Sara, Julie, Françoise, Saloua, Adrienne, Natacha, Véronique et les petits (Guillaume et Nicolas). Il s'en est passé des choses en 5 ans ! Sans eux tous ça aurait été beaucoup moins agréable. Alors merci pour tous ces bons moments passés ensemble au et en dehors du labo! Kasteel, Orval, Rochefort, ... n'ont qu'à bien se tenir ! Merci à Cricri pour ses nombreux conseils et (longues ;)) discussions scientifiques. Et puis Claire pour ses gâteaux, délicieux repas, et nos petites virées...

Je voudrais également remercier tous les membres de l'équipe pédagogique et plus particulièrement Françoise, Yannick, Nicolas, Epiphanie, Pauline, Anne et Thierry! Merci de m'avoir aidée à mes débuts et de m'avoir toujours prodigué de bons conseils.

11 est aussi de ces persormes sans qui la vie a moins de sens : nos amis. Alors je profite de l'instant pour remercier Sandrine et Laurence d'avoir toujours été là pour moi. Aaah les pavés ... les fourchettes du bar à Tapas... l'effet du mélange Kir/antitussif... les chips/Chavroux et puis de les nombreuses visites dans des infirmeries (même à Walibi et oui !), les soirées Alias, les rendez-vous sur Skype... il s'en passe des choses en 9 ans !

J'en arrive à ma famille, merci papa, merci maman et merci Cécé de m'avoir toujours soutenue, poussée dans les études, et avoir toujours été fiers de moi ! J'ai eu beaucoup de chance de grandir dans une famille aussi soudée qui a toujours su faire face aux aléas de la vie. Je suis arrivée au bout grâce à vous. Merci maman d'avoir pris le temps de lire toute ma thèse! J'en profite pour te dire encore une fois merci de m'avoir appris l'anglais, c'est un cadeau extraordinaire. Merci aussi papa, c'est grâce à toi que j'ai eu le goût des sciences quand déjà toute petite j'allais avec toi à l'école... Et puis merci p'tit frère d'avoir participé à mes premières expérimentations scientifiques et d'être toujours fière de moi.

Enfin, je voudrais te remercier. Benoit, pour m'avoir encouragée dans la dernière ligne droite, au jour le jour ! Mon coach moral c'est toi!

Merci !

(9)
(10)

Summary

Until recently, catalysts for asymmetric synthesis hâve been restricted to enzymes and transition- metal-complexes. However, asymmetric organocatalysis has lately emerged as a third approach to enantiopure organic compounds. Organocatalysts are small molecular weight molécules mainly composed of carbon, hydrogen, nitrogen, sulfur and phosphorus. They are often easier to préparé, more stable, cheaper, and less-toxic compared to enzymes and transition-métal complexes.

Beyond the plethora of organocatalysts developed each year, only few examples are based on new chiral scaffolds. This is quite surprising since the évaluation of catalysts based on completely new and highiy chiral scaffold may give rise to the discovery of new substrate combinations and new highiy stereoselective catalytic reactions. In addition, there is aiways interest to develop new molecular scaffolds from industry since it may provide "freedom to operate". Therefore, we proposed to design a new sériés of thiourea bifunctional organocatalysts (±)-12-15 based on Trôger's base ((±)-la) scaffold (Figure S.l).

Unfortunately, a major stumbling block in the application of enantiopure analogues of Trôger's base is their poor availability. Consequently, the resolution of thiourea ((±)-12-15) was foreseen as an additional challenge.

Therefore, we decided to develop a practical and predictable method for the enantioseparation of Trôger's base and its analogues by chromatography on the chiral stationary phase (CSP) Wheik 01. As the thorough évaluation of such a method necessitated a relatively large library of racemic compounds, we prepared a sériés of racemic analogues of Trôger's base

Scope and limitation of the condensation of anilines 4-6 with paraformaldéhyde in CF3COOH were established providing a large number of symmetrical analogues of Trôger's base (±)-l-3 (Scheme S.l).

(±)-la

(±)-12; n=l, Ar=Ph (±)-13: n=l, Ar=3,5-(CF3)jPh (±)-14: n=l, Ar=3-pyridyl (±)-15: n=0, Ar=3,5-(CF3)2Ph

H H

Figure S.l. Trôger's base ((±)-la)(left); new thiourea catalysts (±)-12-15 (right).

R'*

4-6 (5 mmol) (±)-l-3

Scheme S.l. Symmetrical analogues of Trôger's base prepared by the condensation of the corresponding aniline with paraformaldéhyde in trifluoroacetic acid.

(11)

As the usage of reactive formaldéhyde and acid catalysis clearly pose a problem of functional group compatibility, other synthetic methods were investigated. Thus, amino (±)-lu, (±)-2h and cyano (±)- Is, (±)-2i analogues of Trôger's base were prepared using metal-catalyzed cross-coupling reactions {Figure S.2).

Figure S.2. Symmetrical analogues of Trôger's base prepared by Pd-catalyzed C-N and C-C bond forming reactions

Then, we concentrated our effort on the synthesis of unsymmetrical halogen analogues of Trôger's base. We demonstrated that the previousiy published general method for the synthesis of unsymmetrical analogues of Trôger's base bearing different substituents in position 2 and 8 was impractical for the préparation of these intermediates. Therefore, we developed an alternative, shorter and more practical synthesis based on the transformation of 6W,12H-5,11- methanodibenzo[ô,/][l,5]diazocine ((±)-lb) into unsymmetrical bromo (±)-7 and iodo (±)-8 analogues (Figure S.3).

(±)-9

Figure S.3. (±)-lb and unsymmetrical halogen analogues of Trôger's base (±)-7-9

In our opinion, this method is up to now the most usefui and experimentally simple approach to these intermediates. Indeed, our procedure allows the easy préparation of (±)-7 and (±)-8 as a very compétitive alternative to the tedious desymmetrization of dibromide (±)-ln and diiodide (±)-lo.

Moreover, this convenient method was successfully used for the préparation of (±)-9 bearing an I- atom on one ring and a Br-atom on the other. This intermediate is inaccessible by any other route.

Besides, these halogénations are the first examples of electrophilic substitutions in the 6H,12H,-5,11- methano dibenzo[ô,/][l,5]diazocine System. Although our study highiighted its low reactivity compared to A/,A/-dialkylanilines, we aiso demonstrated the feasibility of a perfectiy regioselective electrophilic substitution in the poro-position to the nitrogen atom.

(12)

Finally, we préparée! a sériés of ethylene-bridged analogues of Trôger's base (±)-ll by the reaction between (±)-l and 1,2-dibromoethane {Figure S.4). The scope and limitations of this previousiy reported method were studied and we demonstrated that this "bridge replacement" has a direct impact on the values of pKa of Trôger's base tertiary amine functions.

U2CO3 (4.5 eq) Br-CHjCHj-Br (2 eq)

DMF, 100 °C, 96 h

(±)-l

Figure S.4. Ethylene-bridged analogues of Trôger's base.

Subsequently, we systematically studied the séparation of our library of analogues of Trôger's base on the commercially available CSP Wheik 01. Structure-enantioselectivity relationships were established in terms of the substitution pattern of the aromatic rings, of the electronic nature of substituents and of the geometry. The complementarity with the CSP RegisCell was aiso demonstrated. It was then possible to predict whether or not the enantioseparation of a given, perhaps yet unknown dérivative of Trôger's base would be feasible with our method.

Next, a sériés of compounds (±)-l and (±)-ll were preparatively separated by chiral HPLC according to our predictable method on the CSP Wheik 01. The absolute configurations of compounds (+)-l and (-)-l were assigned by X-ray diffraction, chiroptic corrélation and Chemical corrélations. We found that for 2,8-disubstituted analogues of Trôger's base (±)-l, the absolute configuration of the first eluted peak on CSP (/?,/?)-Whelk 01 was aiways (S,S) and the second peak {R,R). In the ethylene- bridge sériés 11, absolute configurations were assigned by chiroptic corrélations with the unambiguousiy known configuration of lia. It appeared that the order of elution was again in perfect corrélation with the absolute configuration of compounds (±)-ll.

Our newiy designed predictable method is probably superior to any other method of resolution previousiy described for analogues of Trôger's base for a small-scale séparation. In addition, it gave the absolute configuration of the analogues of Trôger's base directly from the elution order.

Finally, we investigated the catalytic activity of thiourea dérivatives of Trôger's base (±)-12-15 in Michael additions to nitroolefins (Scheme S.2).

3.2, 3.5: R= CH(COOEt)2 3.3, 3.6: R= CHICNjj 3.4, 3.7: R=

Me Me

°X°

Scheme S.2. Michael addition af various malonic dérivatives to trans-^nitrostyrene.

The efficiency of racemic thiourea dérivatives of Trôger's base (±)-12-15 was first investigated in the catalytic conjugated addition of various malonate dérivatives to tra/is-y0-nitrostyrene. A bifunctional mechanism of our catalytic System, based on the complementary actions of tertiary amine and thiourea, was demonstrated. However, due to the low basicity of Trôger's base, the outcome of the addition reactions was strongly dépendent on the pKa of the nucleophiles.

(13)

Our new predictable method for the resolution of the analogues of Trôger's base on the CSP Wheik 01 was successfully applied to the resolution of thiourea catalysts (±)-12,13. Based on the comparison of their chiroptical properties and their order of elution on Wheik 01, their absolute configuration was tentatively assigned as (-)-(S,S)-12 and (-)-(S,S)-13 (Figure S.5).

(-)-(S,S)-12 H-(5,S)-13

Figure S.5. Tentative assignaient ofthe absolute configuration of catalysts 12 and 13.

Using enantiopure catalysts (+)-12, (-)-12, (+)-13 and (-)-13 in the catalytic addition of diethylmalonate (3.2) and CH2(CN)2 (3.3) to trans-/9-nitrostyrene, no stereoselectivity was observed.

The lack of selectivity may in our opinion be attributed to the unfavorable spatial arrangement of the two catalytic moieties.

(14)
(15)

Content

List of abbreviations... 1

Compounds numbering...3

Introduction... 4

1. Chirality... 4

Optical activity... 7

2. Assignment of the absolute configuration...8

Chemical corrélations...8

Non-empirical methods for the assignment of absolute configuration...8

Empirical methods for the assignment of absolute configuration...12

3. Chiral discrimination... 14

4. Préparation of chiral enantiopure compounds...16

General methods for the resolution of enantiomers...17

Asymmetric synthesis...24

Critical comparison between the most important méthodologies for the préparation of chiral, enantiomerically pure compounds...28

5. Organocatalysis - a new tool for asymmetric catalysis...30

Choice of the catalytic System: H-Bonding... 30

6. Trôger'sbase... 35

Synthesis of symmetrical analogues of Trôger's bases... 36

Synthesis of unsymmetrical analogues of Trôger's base... 41

The reactivity of Trôger's bases...44

Resolution of Trôger's base and its analogues... 47

Assignment of the absolute configuration of Trôger's base...53

Selected applications of Trôger's base and its analogues... 53

Aim of the work... 58

Chapter 1: Synthesis of analogues of Trôger's base...60

1. Synthesis of symmetrical analogues of Trôger's base... 60

Synthesis of analogues of Trôger's base bearing moderately electron-donating alkyl substituents ...62

Synthesis of analogues of Trôger's base bearing strong electron-donating substituents... 62

Synthesis of analogues of Trôger's base bearing hydroxyl functional groups... 64

Synthesis of analogues of Trôger's base bearing moderately electron-withdrawing halogen atoms... 65

(16)

Synthesis of analogues of Troger's base bearing strongly electron-withdrawing substituents ... 67

Synthesis of analogues of Troger's base from 4-unsubstituted anilines... 69

Mechanistic rationalization...70

2. Synthesis of the symmetrical amino analogues (±)-lu, (±)-2h of Troger's base...74

3. Synthesis of the symmetrical cyano (±)-ls, (±)-2i and aminomethyl (±)-lv, (±)-2j analogues of Troger's base...76

4. Synthesis of unsymmetrical halogen analogues of Troger's base... 78

5. Synthesis of ethylene-bridged analogues of Troger's base... 88

6. Synthesis of other analogues of Troger's base not accessible by direct condensation... 94

7. Synthesis of thiourea dérivatives of Troger's base... 95

8. Conclusions of chapter 1...96

Chapter 2: Predictable enantioseparation of analogues of Troger's base... 97

1. Analytical enantioseparation of analogues of Troger's base on the chiral stationary phase WheIkOl...97

The chiral stationary phase (CSP) and the chiral récognition mechanism... 97

Analytical enantioseparation of analogues of Troger's base on the chiral stationary phase Wheik 01... 99

2. Chiral HPLC: a new tool for the détermination of the absolute configuration of analogues of Troger's base...111

3. Conclusions of chapter 2... 122

Chapter 3: Troger's base: a new chiral scaffold in organocatalysis...123

1. Racemic thiourea dérivatives of Troger's base as organocatalysts for the conjugate addition of malonates to nitrostyrene... 125

2. Resolution of the thiourea catalysts...130

3. Evaluation of the enantioselectivity of thiourea dérivatives of Troger's base 12 and 13... 133

4. Conclusions of chapter 3...136

Conclusions and perspectives...137

Experimental part... 140

1. General...140

2. Synthesis... 142

General procedure for the synthesis of analogues of Troger's base (±)-l-3 from anilines 4-6. 142 (±)-2,8-Dimethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-la... 142

(±)-6H,12H-5,ll-Methanodibenzo[b,f][l,5]diazocine (±)-lb... 143

(±)-2,8-Diethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-lc...143

(±)-2,8-Di(tert-butyl)-6H,12H-5,ll-methanodibenzo[b,f|[l,5]diazocine (±)-le... 143

(17)

(±)-2,8-Dimethoxy-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-lg...144

(±)-2,8-Bis(methylsulfanyl)-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-lh...144

(±)-[(6H,12H-5,ll-Methanodibenzo[b,f][l,5]diazocin-2,8-diyl)bis(4,l-phenylene)]dicarbonitrile (±)-lr... 145

(±)-4;10-Dibromo-2,8-dimethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine...145

(±)-2b... 145

(±)-4,10-Dibromo-2,8-bis(trifluoromethyl)-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)- 2c... 146

(±)-4,10-Dimethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-3b...146

(±)-3A9/10-Tetramethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-3c...147

(±)-lA7,10-Tetramethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-3e...147

(±)-N,N'-Dibenzhydrylidene-6H,12H-541-methanodibenzo[b,f][l,5]diazocine-2,8-diamine (±)- 1.7... 148

(±)-6H,12H-5,ll-Methanodibenzo[b,f][l,5]diazocine-2,8-diamine (±)-lu... 148

(±)-N,N'-Dibenzhydrylidene-2,8-dimethyl-6H,12H-5,ll-methanodibenzo-[b,f][l,5]diazocine- 4,10-diamine (±)-1.8... 149

(±)-2,8-Dimethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine-4,10-diamine (±)-2h...149

(±)-6H,12H-5,ll-Methanodibenzo[b,f][l,5]diazocine-2,8-dicarbonitrile (±)-ls... 150

(±)-2,8-Dimethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine-4,10-dicarbonitrile (±)-2i. 150 (±)-C-(8-Aminomethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocin-2-yl)-methylamine ((+)- Iv)... 151

(±)-(10-Aminomethyl-2,8-dimethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocin-4-yl)- methylamine (±)-2j... 151

(±)-2-Bromo-6H,12H-5,ll-methanodibenzo[b,f](l,5]diazocine (±)-7... 152

(±)-2-lodo-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-8... 152

(±)-2,8-Diiodo-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-lo...153

(±)-8-Bromo-2-iodo-6H,12H-5,ll-methanodibenzo[b,fI[l,5]diazocine (±)-9... 153

(±)-8-Bromo-4,10-dimethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-10... 154

(±)-2,8-Dibromo-4,10-dimethyl-6H,12H-5,ll-inethanodibenzo[b,f][l,5]diazocine (±)-2m... 154

General method forthe synthesis of ethylene-bridged analogues of Trôger's base:...155

(±)-2,8-Dimethyl-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-lla...155

(±)-2,8-Di(methylsulfanyl)-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-llb...155

(±)-2,8-Diethyl-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-llc... 156

(±)-2,8-Di(tert-butyl)-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-lld (±)-2,8-Dihexyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocine (±)-lf... 144

156

(18)

157

(±)-2,8-Dichloro-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-llf... 157

(±)-2,8-Dibromo-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-llg... 157

(±)-2,8-Di(ethoxycarbonyl)-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-llh...158

(±)-2,8-Di(methylsulfanyl)-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-lli...158

General procedure forthe synthesis of thiourea catalysts:...159

(±)-l-{2,8-Dimethyl-10-[(3-phenylthioureido)methyl]-6H,12H-5,ll- methanodibenzo[b,f][l,5]diazocin-4-ylmethyl}-3-phenylthiourea (±)-12... 159

(±)-l-(3,5-Bistrifluoromethylphenyl)-3-{2,8-dimethyl-10-[(3-(3,5 bistrifluoromethylphenyl)thioureido)methyl]-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocin-4- ylmethyl}thiourea (±)-13... 160

(±)-l-{2,8-Dimethyl-10-[(3-pyridin-3-ylthioureido)methyl]-6H,12H-5,ll- methanodibenzo[b,f][l,5]diazocin-4-ylmethyl}-3-pyridin-3-ylthiourea (±)-14...161

(±)-l-[3,5-Bis(trifluoromethyl)phenyl]-3-{10-(3-[3,5-bis(trifluoromethyl)phenyl]thioureido)-2,8- dimethyl-6H,12H-5,ll-methanodibenzo[b,f][l,5]diazocin-4-yl}-thiourea (±)-15...162

Bibliography... 163

Selected NMRspectra... 169 (±)-2,8-Dihexyl-6H,12H-5,ll-ethanodibenzo[b,f][ l,5]diazocine (±)-lle.

(19)
(20)

List of abbreviations

Abbreviation Full name Structure / molecular formula

AC 9-BBN

BINAP

Absolute configuration 9-Borabicyclo[3.3.1]nonane

(±)-2,2'-Bis(diphenylphosphino)-l,l'-binaphthalene

Bn Benzyl

Boc rert-butoxycarbonyl

.â:v*

Me

CD Circular dichroism

CSD Cambridge Structural Database CSP Chiral stationary phases

Cy Cyclohexyl

0“’

DCC A/,A/'-dicyclohexylcarbodiimide

dba Dibenzylideneacetone 0

DMA A/,/V-dimethylacetamide 0

.Me Me^ N

Me

DMAP 4-Dimethylaminopyridine /=\

Nv />—N w

DMEDA /V,/V'-dimethylethylenediamine Me

y—N H HN—'

Me

DMF A/,A/-dimethylformamide 0

JL ^Me

H N

Me

DMSO Dimethylsulfoxyde 0

II

Me Me

1

(21)

dppf l,l'-Bis(diphenylphosphino)ferrocene Ph

Ph Fe

I

P—Ph

\ Ph

ECD Electronic circular dichroism

ee Enantiomeric excess

Et Ethyl CH3CH2-

HMT Hexamethylenetetramine

N

rf'i

i-Pr /so-propyl Me

/■-PrOH /so-propyl alcohol Me

WoH

Me'

Me M ethyl CH3-

Mp Melting point

MSH 0-mesitylenesulfonylhydroxylamine Me

Me

Py

r.t.

TEAA

Tf THF

[Pd(PPh3)4] Tetrakis(triphenYlphosphine)Palladium (0)

Ph Phenyl

Pyridine

Room température

Triethylammonium acetate

Trifluoromethylsulfonyl Tetrahydrofurane

Q

e.A

\ -NH1^ ^

^■'^0 'I "^0 CF3SO2

.0.

Me

O

Ts Tosyl

Me

VCD Vibrational circular dichroism

(22)

Compounds numbering

Literature examples and intermediates are numbered as following e.g.

Me1 Me

Me^

fl

^OMe

MeO^ 'Me

Me Me

i.l7

"i" for introduction - 17 is the number of appearance

Analogues of Trôger's base prepared in this work are numbered as following e.g.

(23)
(24)

Introduction

1. Chirality

In organic chemistry, the term referring to the relations between molécules of identical molecular formula is called isomerism. By définition, Chemical species that hâve the same number and kind of atoms but differ in physical and/or Chemical properties because of différence in structure e.g.

constitution and/or configuration and/or conformation, are called isomers. They can be classified as described in Figure 1.1.

Isomers Same molecular formula

Constitution isomers Stereoisomers

Different connectivity Same connectivity

Enantiomers Diastereomers

Mirror image one of each other Not mirror image one of each other Figure 1.1. Relations between isomers: isomerism.

Stereoisomers are isomers of identical constitution but differing in the arrangement of their atoms in space. Subclasses are enantiomers and diastereomers. Diastereomers are stereoisomers not related as mirror image while enantiomers are pairs of molecular species that are mirror images of each other and not superposable.[1]

The property of being non superposable to its image in a mirror (the spécular image) is called chirality.

Thus, an object is called chiral if it is not superposable to its image in a mirror. The word chiral cornes from the Greek word xevp meaning "hand". Indeed, hands are the most straightforward example of chiral objects (Figure 1.2). They are mirror image of each other and not superposable. In practice, one may easily distinguish a chiral from an achiral object considering that: chirality implies absence of an S„ axis including a plane ((T=Si} or center (I = SJ of symmetry.[l,2]

Figure 1.2. My hands (on the left) are chiral as well as the molécule oflimonene (on the right).

Introduction

4

(25)

The notion of chirality relies on two distinct concepts: the chirotopicity and the stereogenicity.

The chirotopicity defines the chirality of a molécule. This concept is purely géométrie and based on the symmetry properties of the molécule. By définition, an atom or set of atoms lying in a chiral environment is said chirotopic; while an atom or set of atoms lying in an achiral environment is said achirotopic.

The stereogenicity is the property of an atom connected to four different substituents. An atom is called a stereogenic center when the interconversion of two substituents conducts to another stereoisomer (Figure 1.3). If the inversion conducts to an identical molécule id., the atom is called non stereogenic. A compound with n stereogenic centers may only hâve 2" different stereoisomers.[l]

Figure 1.3. Two enantiomers with one stereogenic carbon atom.

Stereogenic centers are unambiguousiy distinguished using stereochemical descriptors in the CIP (Cahn-Ingold-Prelog) System: R [rectus] and S (sinister) (Figure l.4).[l]

Figure 1.4. Application ofthe CIP System with increasing priorityfor D<C<B<A. D is set at the back.

When more than one stereogenic center is présent in a molécule, only stereoisomers with opposite absolute configuration for each stereogenic centers are enantiomers. Other combinations are pairs of diastereomers (Figure 1.5).

Figure 1.5. Stereochemical relationships between the 4 stereoisomers of 2-bromo-3-chlorobutane.

(26)

However, if these stereogenic centers are identically substituted, the number of stereoisomers is reduced. Considering 2,3-dichlorobutane (Figure 1.6), the stereoisomer (R,S) and its specular image (S,R) are superposable, thus identical. Hence, the diastereoisomer (S,R) is achiral and called a meso compound because it has a plane of symmetry. Thus, the presence of more than one stereogenic center may lead to an achiral molécule.

Figure 1.6. Stereochemical relationship between stereoisomers of 2,3-dichlorobutane.

Mixtures of stereoisomers are characterized by the enantiomeric excess (ee) for a pair of enantiomers (R and S) or the diastereomeric excess (de) for a pair of diastereomers (A and B).

Enantiomers

wl ^ J (S)

ee —

de =--- * 100 % A-F B

A compound characterized by 100 % ee is called enantiomerically pure.

Introduction

(27)

Optical activity

In an achiral environment, enantiomers hâve the same Chemical and physical properties except that they will interact differently with polarized light. The rotation of the plane of linearly-polarized light, generally measured in a polarimeter, caused by the presence of chiral molécules is called the optical rotation {Figure 1.7).

Source

a Angle of rotation

Non polarized light polarizer Oscillâtes in ail directions

1 ;

light

» (

11

Cell with an enantiopure

/ \

sample

Analyzer

Figure 1.7. Measurement of optical rotation a with a polarimeter.

The angle of rotation for the two enantiomers is identical but of opposite sign. If the plane is turned clockwise as seen by an observer towards whom the light travels, the angle of rotation is positive, and the enantiomer is called dextrogyre. If the plane is turned counterclockwise, the angle of rotation is négative, and the enantiomer is called levogyre.[l]

The spécifie optical rotation [a] is fonction of the observed rotation a, the cell length (/ in dm), the température (T), the solvent (and pH), the concentration (c in g/100 mL), the pressure and the wavelength (A,). For a solution [a] is expressed as follows:

[«]! = a.lOOl.c

However, there is no general corrélation between the sign of the optical rotation and the absolute configuration, although for a sériés of structurally similar dérivatives, such corrélation may, in some instances, be established.[l]

A racemic compound or racemate is a composite of equimolar quantifies of two enantiomeric species. It is devoid of optical activity. In the Chemical name or formula it may be distinguished from the individuel enantiomers by the prefix (±)- or rac- or by the descriptor RS.

(28)

2. Assignment of the absolute configuration Chemical corrélations

Chemical corrélation may be achieved between a compound of known absolute configuration (AC) and a compound of unknown AC when configuration of the stereogenic center remains unmodified or when inversion of the configuration is unambiguousiy known (Figure 1.8).

Figure 1.8. Examples of Chemical corrélation; top: the stereogenic center is unmodified, apparent inversion of configuration is due to inversion ofCIP priorities; bottom: the inversion of stereogenic

center by a known 5,^2 reaction.

Chemical corrélation has to be used with a certain caution since rétention or inversion of configuration is not aiways obvious and racemization can aiso occur under reaction conditions.

Non-empirical methods for the assignment of absolute configuration

The assignment of absolute configuration remains a thorny problem even with simple molécules and was solved for the first time in 1951 by Bijvoet with the aid of "anomalous X-ray diffraction".[3,4]

There are numerous methods that allow the reliable assignment of absolute configurations. They can be distinguished in non-empirical (or absolute) and empirical (or relative). Non-empirical methods do not need any reference substance while empirical methods rely on some kind of reference or standard.[5]

Electronic circuler dichroism (ECD), vibrational circuler dichroism (VCD) and "anomalous" XRD are the only non-empirical methods available today for the assignment of AC.

ECD and VCD in combination with ab initia calculations hâve manifested themselves as reliable non- empirical methods for the assignment of AC. ECD, often termed simpiy CD, and VCD both rely on différences in refractive indices of left and right circularly polarized electromagnetic radiation in a chiral medium. ECD opérâtes in the UV-Vis région, where electronic transitions take place, while VCD makes use of vibrational transitions in the IR région. Both methods possess certain advantages and drawbacks; a brief OverView is given in Table /.l.[6-8]

Electronic circular dichroism (ECD) and vibrational circuîar dichroism FVCD)

Introduction

8

(29)

Table 1.1. OverView of advantages and drawbacks ofECD and VCD.

ECD VCD

Limitation by molecular structure:

chromophore must be présent

No limitation on molecular structure No or little solubility limitation. Only a dilute

solution of CO. lO'^-lO'^ M is needed.

Similar to UV-vis spectroscopy, a number of solvents fully transparent in UV-Vis région are available.

9 9

Very concentrated solutions are needed (co. 1 M) and this often poses serious solubility problems.

Ail solvents possess more or less intense absorption bands in IR range;

choice of solvent is narrower.

Few transitions are observed in ECD spectra of most molécules. Thus, reliability of calculations is reduced.

r~ —1

« a'-- A large number of transitions are typically observed in VCD spectra.

Calculation of ECD spectra requires the knowledge of excited State of the molécule.

This reduces reliability of calculations.

• • \

Vibrational transitions occur in the ground State, no knowledge of elusive excited State is necessary. This éliminâtes important source of errors.

Equipment is unsophisticated and is widely available; it is easy to operate by a non- specialist. Measurements are relatively quick.

9 9 '

Equipment is more sophisticated and not yet widely spread. It requires more trained personnel. Measurements are comparatively long.

"Anomalous" XRD

A complété description of this method may be found in reference[9].

Information such as the molecular geometry, bond distances, angles and the packing of the molécules in the crystal are obtained from XRD experiments. Moreover, these data in principle allow distinction between the enantiomorphs of a chiral crystal structure and the enantiomers of a chiral molécule. However, some restrictions remain.

|F(H)| ^ |F(-H)|

0(F(H)) ^ 0(F(-H))

Figure 1.9. Différences in intensity due ta résonant effects: Fo(H)/Fo(-H) are sums ofatomic structure factors resuiting from elastic scattering only; Fa'/Fb are sums of the reai parts of the anomalous

scattering vector; Fa'/Ft," are sums ofthe imaginary parts ofthe anomalous scattering vector;

F(H)/F(-H) are sums of structure factors ofa real crystal, with one or more anomalousiy scattering atoms.

(30)

Since AC assignment is obtained from the crystal-structure, it dépends on the ability to identify very snnall diffraction intensity différences between two crystal-structure models of opposite chirality.

However, with compounds containing only light atoms, the différence is often not even significant.

Indeed, when working with visible light, it is easy to see the différence between a right and a left hands, thus between objects of opposite chirality. The major différence in the diffraction pattern of the right and the left hand occurs in the (non-observable) phases of the inversion-related reflections whereas their (observable) amplitudes are identical in the absence of any résonant effects (Figure 1.9). Différences in intensifies only occur if a résonant frequency of the diffracting object is close to that of the incident electromagnetic radiation. This is where the major difficulties for the distinction of chirality using XRD lie. Only a small portion of the électrons that cause diffraction is résonant. The intensity différence between inversion-related reflections is thus small. Furthermore, the résonance frequency of light atoms occurs at long X-ray wavelengths which are difficult to access experimentally.[10]

The Flack Parameter

The assignment of the AC of noncentrosymmetric crystal structures relies on the résonant scattering phenomenon and is correlated to the Flack parameter. The Flack parameter indicates the correctness of the handedness of the model of a non-centrosymmetric crystal structure, from which subsequently one may deduce the AC of the chiral molécules forming the crystal.

The Flack parameter relies on the physical model of a crystal twinned by inversion and composed of distinguishable domains. Thus, the macroscopie crystal is formed of two types of homogeneous and perfectiy-oriented domains, and the relationship between the two domain types is inversion. A simple way to represent a crystal twinned by inversion is to imagine a racemic conglomerate in which the crystals hâve stuck together and grown in a perfectiy oriented manner, giving a completely superposable diffraction patterns, that is indistinguishable from that of a single crystal.

If X represents the model crystal structure as given by its cell dimensions, space group, and atomic coordinates and X its image inverted through a point. Then, the macroscopie crystal may be represented as C = (1 - x) X -F x X (C being the concentration), for which the Flack parameter % measures respectively the mole fractions (1 — x) and xof the two types of domains X and X respectively:

H When X = there is only one domain in the crystal: the model X,

H When X = 1/ there is only one domain in the crystal: the inverted model X,

tf When X = 0.3 both types of domains are présent in the crystal: 70 % of X to 30 % of X.

H When X = 0.5, both types of domains are equally présent in the crystal (racemate).

Thus, the physically meaningfui values of x are within range 0 < x ^ 1- However, experimental values may lie a little outside of this range.

Then a crystal structure model of an enantiomerically pure compound with the correct AC has a value of the Flack parameter equal to or close to zéro.

The Flack parameter indicates the correctness of the absolute structure model employed, in reference to the inverted model. If the Flack parameter is 0, the absolute structure model employed

Introduction

(31)

is the correct one, and the absolute configuration of the chiral molécules that compose the structure can be correctiy deduced.

It is important to stress the différence between:

H the absolute configuration = the spatial arrangement of the atoms of a physically identified chiral molecular entity and its stereochemical description e.g., (R) or (S) and,

H the absolute structure = the spatial arrangement of the atoms of a physically identified noncentrosymmetric crystal and its description by way of unit-cell dimensions, space group, and représentative coordinates of ail atoms.

Absolute structure détermination

Certain conditions should be respected to assume that an absolute structure was correctiy assigned.

First, the absolute structure détermination should be sufficiently précisé (the standard uncertainty u of the Flack parameter %(u) should be less than 0.04). Then, the value of the Flack parameter itself should be close to zéro within a région of three standard uncertainties (i.e. u < 0.04 and -^< 3.0).Irl

Absolute-Configuration Détermination

When an absolute structure is properly assigned, the absolute configuration may be deduced.

However, one should make sure that ail occurrences of the chiral molécule in the crystal structure are the same enantiomer.

There are a number of further practical limitations that may prevent the use of Flack parameter for the assignment of absolute configuration of chiral molécules.[11]

First and foremost, single crystals of sufficiently high quality must be available. This can be very difficult in case of substances which are chemically unstable or hâve melting points below ambient température. In addition, many organic materials tend to form amorphous solids, liquid crystalline or highiy disordered crystalline mesophases, "difficult" crystals with strong tendency to ID or 2D growth (fibres or thin sheets), or solvatés which quickly desolvate.

Next, the Flack parameter is related to the wavelength of the incident X-rays and to the électron density of atoms in the crystal lattice {Figure 1.10). It can easily be seen that résonant scattering is practically absent for atoms with small atomic numbers (Z < 5) and quickly grows with increasing Z.

For the same Z, the shorter the wavelength, the smaller the résonance scattering.

Figure l.l0.Anomalous scattering factors for iight atoms. Figure reproduced from ref[ll].

(32)

For technical reasons, very few monochromatic X-ray sources of sufficient intensity are available, the most popular being Mo and Cu radiation. As can be seen from Figure 1.10, for Mo radiation (X = 0.71 Â) anomalous scattering becomes non-zero for Z > 10, that is, éléments of the third or higher rows of the periodic System hâve to be présent (in case of organic molécules. Si (Z = 14) or heavier). For Cu radiation

(X,

= 1.54 Â), sufficient résonant scattering can be observed aiready for oxygen (Z = 8). In the absence of oxygen or heavier atoms, there is no easily available technical possibility to use the Flack parameter for the assignment of the absolute configuration. Notably, this means that the assignment of absolute configuration of chiral amines (containing only C, H, N atoms) by this method is not feasible. Chemists may try to circumvent this by synthesizing dérivatives with heavier atoms, e.g., by transforming amines into halide salts, amides of haloacetic or halobenzoic acids or sulphonamides.

Finally, there must be proof that the composition of a crystal used for the measurement is représentative of the composition of the bulk sample. The sample should ideally be enantiomerically pure. In case crystals are obtained from non-racemic samples with ee different from 100 %, evidence for the enantiomeric composition of a single crystal is particularly important. It is easy to illustrate how deceptive the use of the Flack parameter could be in case of spontaneous resolution. Indeed, crystallization of a racemic sample will produce enantiomerically pure crystals, for any of which the absolute structure and hence the absolute configuration can be obtained. By doing so, with an arbitrarily chosen crystal, a chemist would corne (in the absence of independent proof for the enantiomeric composition of the bulk sample) to the conclusion that he is dealing with an enantiomerically pure substance and would assign a certain absolute configuration to it. However, talking about absolute configuration of a roce/n/c substance is a total nonsense.

Empihcal methods for the assignment of absolute configuration

Circular dichroism

Circular dichroism (CD) is the différentiel absorption (Ae) of left and right circularly polarized light by a nonracemic sample. It is characterized by a signed absorption band in the UV, vis or IR région of the spectrum (= Cotton effect). Transitions in UV-vis-near IR, aiso termed ECD-electronic CD, are associated with electronic transitions and are most commonly used. Transitions in the vibrational range (IR) can aIso be used in VCD (vibrational CD). However, VCD is an emerging, but by now, less widely spread method and will not be discussed here. [12]

The magnitude of Cotton effect can be expressed as:

As — Si — Sr in cm'^.mmof^

As — ---y/.M C. 1.32982

Where, y is the measured ellipticity angle, C the concentration (in mmol/ml), M the molar mass (g/mol) and l the cell length (cm).

The CD curves of a pair of enantiomers are opposite in sign (Figure 1.11). Typically, the CD absorption maximum corresponds to the UV maximum of absorption.[1]

Introduction

12

(33)

Figure 1.11. Typical CD curves of opposite enantiomers.

In sériés of analogues, corrélations between the sign of Cotton effect in CD curves can be used for the assignment of AC if a reference molécule with unambiguousiy known AC is available. It must be kept in mind that "similarity" of CD spectra for structurally similar molécules is relative. Hence, such assignments are not infallible and must be treated with great care. The sign of optical rotation can aiso be used for assignment of AC, but in this case corrélation is even less good, and even more care is advised.[l]

(34)

3. Chiral discrimination

Our hands and feet are chiral, as gloves and shoes are: we can easily put the right foot in the right shoe but not the left foot in the right shoe. It is a simple démonstration of chiral discrimination. This is aiso true for an organic receptor, which is aIso chiral, interacting with a chiral drug or aroma. It will interact differently with each enantiomer and therefore induce a different biological activity. This phenomenon called biodiscrimination can be described by the three-point interaction model {Figure 1.12). When a chiral substrate interacts with a chiral receptor, an optimal response/activity is obtained when the agonist substrate (the eutomer = the active enantiomer) interacts simultaneousiy at three complementary binding sites with the receptor. A stereoisomer of the agonist (the distomer = the inactive enantiomer) will only be able to bind at two binding sites of the receptor and thus induce a decreased or completely vanished response/activity (Figure I.12).[l]

Eutomer Distomer

Figure 1.12. Three-point interaction model.

For example, due to chiral receptors involved in the sense of smell, our nose can easily distinguish the smell of spearmint from the smell of caraway seeds which are due to left- and right-handed version of carvone, respectively {Figure I.13).[2]

R-(-)-carvone S-(+)-carvone spearmint caraway seeds

Figure 1.13. The two enantiomers of carvone.

Moreover, synthetic foodstuffs were specially designed due to their inability of being metabolized by enzymes.[13] For example l-hexoses which are as sweet as D-hexoses, were patented as nonnutritive sweeteners {Figure /.14).[14]

H HO H H

D-glucose L-glucose

Figure 1.14. The two enantiomers of glucose.

Introduction

14

(35)

Different behaviors may be expected with chiral compounds. At first, both enantiomers may présent (almost) identical pharmacological activity or, on the contrary, the desired biological activity may be attributed to one enantiomer (the eutomer) while the other one is essentially inactive (the distomer);

e.g. only (-)-morphine (i.l) possesses the analgésie activity and only natural (-)-cocaine (i.2) is psychoactive and thus a controlled substance. Next, the activity of both enantiomers may be qualitatively identical but quantitatively different; e.g. (-)-nicotine (i.3) is much more toxic than the dextrogyre enantiomer. Finally, the activity of each enantiomer can be qualitatively different. For example, l-methyl-5-phenyl-5-propylbarbituric acid enantiomers (i.4) exhibit opposite effects on the central nervous System: the levogyre enantiomer is a strong sédative while the dextrogyre enantiomer is a convulsant (Figure /.I5).[l,15]

Figure 1.15. Chiral pharmacological active compounds.

Besides, the infamous case of thalidomide (i.5) perfectiy illustrâtes the importance of chirality in the field of médicinal chemistry and pharmacology (Figure 1.16). Indeed, racemic thalidomide was sold from 1957 to 1961 in Europe as a sédative and antinausea agent for use during early pregnancy.

s-(-)-i.5 R-(+)-i.5

Figure 1.16. The two enantiomers of thalidomide.

Unfortunately, many years after its release, this drug was found to be a very potent teratogen causing fêtai abnormalities, and many children born from mothers who used thalidomide had a high incidence of deformity. The teratogenicity was first attributed to the (5)-(-)-enantiomer by Blaschke et al. in 1980 while no fêtai toxicity was evidenced for the (/î)-(+)-enantiomer.[16] However, recent studies demonstrated that putative différences in therapeutic or adverse effect between (/?)-(+)-and S)-(-)- thalidomide would to a large extent be abolished by rapid interconversion In vivo of separately administered enantiomers.[17] Currently, thalidomide is used for the treatment of severe immunological manifestations in erythrema rodosum leprosum, skin and mucous membrane disorders and for the prévention and treatment of graft-vs-host diseases after bone marrow transplantation.[17]

(36)

4. Préparation of chiral enantiopure compounds

When a chiral enantiopure compound is needed, it can be prepared following three general méthodologies.

The first one relies on the isolation of the necessary material from the natural source. As a complément to it, the modification of the natural material w/t/iout création of new chiral éléments can be used. This method is so-called "chiral pool"strategy.

As mentioned earlier, many natural products are chiral and thus aiready include one or more stereogenic centers. A very economical way of producing compounds as single enantiomers is to use pure natural products as starting material. Amino acids and sugars are efficiently used for this strategy. For example, 2-deoxy-D-ribose i.6, was used for the synthesis of one enantiomer of the ambrosia beetle aggregation pheromone (R)-sulcatol i.7{Scheme 1.1).

MeOH MsO Rgney Ni

HO^ HO^ Msé V

I.D

OMe HjO (yie

TT

OH , OH

Me^^'-^^^CHO

Me

Ph3P=«(

CHO Me

OH I

i.7

Schéma Ll.Synthesis of (R)-sulcatol 1.7from 2-deoxy-D-ribose 1.8.

There is a major drawback in the chiral pool approach; because of the natural origin of the starting material, often only one enantiomer is available.[2]

The préparation of an enantiopure compound from achiral starting material may be achieved following two other procedures. On the one hand, the product may be synthesized as a racemate which is resolved in the latest step of the synthesis. On the other hand, the enantiopure target may be prepared by asymmetricsynthesis (Figure /.I7).[1,18]

Achiral ] Synthesis

Racemic starting materiall * Products

Resolution

|-)-enantiomer| 50%

(+)-enantiomerl 50%

Achiral 1 Asymmetric synthesis Enantiopure 1

starting materiall product J

Figure 1.17. Préparation of enantiopure compounds.

These two méthodologies are complementary. Each of them has important advantages and drawbacks, which will be critically compared later in this section. First, we will give a brief overview of the most important techniques for resolution and asymmetric synthesis.

Introduction

(37)

General methodsfor the resolution of racemates

The séparation of racemates into enantiomers is called resolution. Since the starting points of these methods are racemates, the maximai yield of each enantiomer is 50 %. Resolution methods may involve either physical processes or Chemical reactions, thermodynamic or kinetic control. The most general methods of resolution are described below.

Spontaneous crvstaUization

Spontaneous crystallization of enantiomers from solutions of racemates in the absence of a resolving agent is one of the oldest and most fascinating methods. This can only occur if the crystallization of a racemate leads to the formation of a conglomerate (only 5-10 % of ail organic solids). In other words, the crystallization of the racemate under equilibrium conditions should lead to the formation of both enantiomers of the substance as enantiomorphous crystals in equal quantities rather than crystals containing equal amounts of both enantiomers within one and the same crystal (Figure 1.18).

Conglomerate crystals are then manually sorted renderingthis method quite unpractical.

Figure 1.18. Schematic représentation of racemic solid and racemic conglomerate.

If a racemate does not crystallize in a conglomerate it can in some instances be converted into a dérivative that does. Tartaric acid, the first substance ever resolved, is a very nice example of this approach. Indeed, as such tartaric acid crystallizes as a racemic compound, while Pasteur's sait (racemic NaNH4 tartrate-4H20) is a conglomerate below 28 °C and résolves spontaneously.[19,20]

Diastereomer-mediated resolution

Diastereomer-mediated resolution is a well-known and widely used method for the séparation of enantiomers. This method involves the formation and séparation of diastereomers. Hence, the racemic mixture is treated with one enantiomer of a chiral substrate (called resolving agent and often from the chiral pool). The resulting diastereomers demonstrate different physical properties and can be separated (by crystallization or chromatography). The newiy formed diastereomer pairs can either be a diastereomeric sait, a covalent solid, a charge transfer complex, or an inclusion compound. The first example of such a resolution was described by Pasteur in 1853. When rac- tartaric acid (TT) was treated with (+)-cinchotoxine (C) only (-)-TT (+)-C précipitâtes while (+)-TT (+)-C remains in solution (Figure I.19).[19]

(38)

Figure 1.19. Thefirst resolution via diastereomers described by Pasteur in 1853.

With this method, a number of compounds including acids, lactones, bases, amino acids, alcohols, thiols, aldéhydes and ketones can be resolved by treatment with an appropriate resolving agent (Figure 1.20).

(S)-(+)-mandelic acid (S)-(-)-Amphetamine

Figure 1.20. Examples of resolving agents.

Kinetic resolution

Kinetic resolution occurs when a Chemical reaction between an enantiopure reagent and a racemic substrate (rac)-A forms one diastereomer B quicker than the other. Différences in reaction rate arise from the différences in activation energy (Ea) of the two transition State of the reaction (Figure l.21).[21]

kp > k;

(ffoc)-A --- «► B + (S)-A Enantiopure reagent

(R)-A

Figure 1.21. Principle of kinetic resolution.

The efficiency of kinetic resolution dépends on the relative rate (s = kR / ks) of reaction of the two enantiomers (s is the stereoselectivity factor, C is the conversion and kR or ksare the relative kinetic constant of both enantiomers). The efficiency is measured by the ee of the substrate A during the course of the resolution. 5 is expressed as following for a first order reaction with respect to A:

In [(l-O(l-ee)]

In [(1 - C)(l + ee)] [S] - [ft] = 0.5 (e-'^s-‘ - e-'^«-‘)

Pasteur was the first to describe such a method in 1858 with the resolution of tartaric acid by fermenting yeast. In fact, this resolution, catalyzed by an enzyme, represents the most studied kind of kinetic resolution: enzymatic resolution.[1] Advantage of enzymatic resolution is the unique enantioselectivity of enzymes, which may offer access to products with extremely high ee.

Introduction

(39)

Resolution bv chromatoaraphv on chiral stationarv phases

Chromatography on chiral stationary phases (CSP) has received a growing interest from scientists.

The development of gas chromatography (GC) and of high performance liquid chromatography (HPLC) allowed the analysis of enantiomer mixtures. Since the early âge of chromatography, many enantioselective stationary phases hâve been developed for the analysis and enantioseparation of various analytes (Table 1.2).[1,22]

Most of chiral stationary phases (CSP) can be classified according to their structure either as polymeric phases (type I) or brush-type phases (type II). Because of their structure, each type of CSP interacts differently with the analyte.

Table 1.2. Examples ofcommercially avallable chiral stationary phases for HPLC (Table adapted from ref[22]).

CSP Chiral selector CSP Type

Cellulose and amylase dérivatives

Chiralcel OB™ Cellulose tribenzoate Ib

Chiralcel OC™ Cellulose tris(phenylcarbamate) Ib

Chiralcel OD™ Cellulose tris(3,5-dimethylphenylcarbamate) Ib

Chiralcel OJ™ Cellulose tris(4-methylbenzoate) Ib

Chiralcel OF™ Cellulose tris(4-chorophenylcarbamate) Ib

Chiralcel OG™ Cellulose tris(4-methylphenylcarbamate) Ib

Chiralcel OK™ Cellulose tricinnamate Ib

CTA-I™ Cellulose triacetate la

ChiralPak IB™ Cellulose tris(3,5-dimethylphenylcarbamate) le ChiralPak IC™ Cellulose tris(3,5-dichlorophenylcarbamate) le RegisCell™ Cellulose tris-(3,5-dimethylphenylcarbamate) Ib Chiralpak AD™ Amylose tris(3,5-dimethylphenylcarbamate) Ib Chiralpak AS™ Amylose tris[(S)-a-methylbenzylcarbamate] Ib

RegisPack™ Amylose tris-(3,5-dimethylphenylcarbamate) Ib

ChiralPak IA™ Amylose tris(3,5-dimethylphenylcarbamate) le Synthetic polymers

CHI-DMB™ Cross-linked 0-3,5-dimethylbenzoyl tartramide Ib

CHI-TTB™ Cross-linked 0-4-tert-dimethylbenzoyl tartramide Ib

NEA™ (R or S) Poly(A/-metacryloylnaphtylethylamine) Ib

Cyclodextrin selectors

Cyclobond™ a-, P-, y-Cyclodextrin II

ChirDex™ P-Cyclodextrin II

ChiralPrep CD ST/PM™ P-Cyclodextrin II

Brush type

DNBLeu™/Leucine™ 3,5-Dinitrobenzoylleucine II

DNBPG™/Phenylglycine™ 3,5-Dinitrobenzoylphenylglycine II

Whelk-01™ 3,5-Dinitrobenzoyl tetrahydrophenantrene amine II

DACH-DNB™ Diaminocyclohexane 3,5-dinitrobenzamide II

ULMO™ Diphenylethylene diamine 3,5-dinitrobenzamide II

Chiris-QN™ Quinine II

Chiris-QD™ Quinidine II

Figure

Updating...

Sujets connexes :