Synthesis and photochemistry of chiral bichromophoriccompoundsCitation for published version (APA):Miesen, F. W. A. M. (1994). Synthesis and photochemistry of chiral bichromophoric compounds. Eindhoven:Technische Universiteit Eindhoven. https://doi.org/10.6100/IR413691

DOI:10.6100/IR413691

Document status and date:Published: 01/01/1994

Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:www.tue.nl/taverne

Take down policyIf you believe that this document breaches copyright please contact us at:openaccess@tue.nlproviding details and we will investigate your claim.

Download date: 28. May. 2020

SYNTHESIS AND PHOTOCHEMISTRY

OF CHIRAL BICHROMOPHORIC

COMPOUNDS

PROEFSCHRIFr

ter vcrkrijging van de graad van doctor aan de

Technische Universiteit Eindhoven, op gezag van

de Rector Magnifictl$, prof. dr. J.H. van Lint,

voor een commissie aangewezen door het College

van Dekanen in het openbaar te vcrdedigen op

dinsdag 29 maan 1994 om 16_00 uur

door

Franciscus Wilhelmus Antonius Maria Miesen

geborcn te Maastricht

(i~uk: wit.)r0 dl~~or!i:1I~odrukkorlj, hQlmond.

Dit proefschrift is goedgekeurd door de prornotoren:

proLdr. E.W. Meijer

Cn

prof.dr. J.W. Verhoeven

Copromotor: dr,iL R.A.J. Janssen

The work described in this Ph.D. thesis ha..~ been financially

supported by the Netherlands Foundation for Chemical

Research (SON) with financial aid from the Netherlands Or­

ganization for Scientific Research (NWO) and by the

Eindhoven University of Technology.

o/OOT Jofande, 9'{swmi "" 'TUllO

Contents

Contents

1 Mechanistic PbotocbemJstry aod Chil"ality

1.1 Introduction 1.2 Mechanistic organic photochemistry

1.2.1 Photoisomerizations

1.2.2 Photochemical sigmatropic shifts

1.2.3 Photofragmentations 1.2.4 Photoadditions

1.3 Interaction of chiral molecules with radiation

1.4 Chiroptica1 techniques

1.5 Asymmetric photochemistry

1.6 Scope of the thesis

References

2 Asymmetric Syuthesis of (2S,411)~ and (2R~4R)-

1,2,3 ,4-tetrabydro-4--etbyl-l, 1 ,4-trimetbyl-(3Z)­

etbyUdene-2-naphthalenolo;

2.1 InbUduction

2.2 Synthesis

2.2.1 Synthesis of the optically activc precursor

2.2.2 Synthesis of the optically pure tetraalkyl naphthalenones

2.2.3 Characterisation of the naphthalenones

2.2.4 Synthesis and characterisation of the naphthalenols

2.3 Concluding Remarks

2.4 Experimental Section References

3 Photocbemistry and Photooxygenation of (2S,4R)- aDd

(2R,4R)-l,2,3,4-tetrabydro-4-etbyl-l t1 t4-trimetbyl-(3Z)­

etbylidene-2-napbtbalenols

3.1 Introduction

3.1.1 Brief review On the photochemistry of all y1ic alcohols

3.1.2 Interaction of oxygen with olefins

3.1.3 Formation of epoxides in photooxygenation.

9

9

10 12

13

14

14

15

17

19 20 22

27 27

30

30

30

33

36

38

39 44

47

48 50

52 55

5

6

3.2 Photochemistry

3.2.1 Results photochemistry of 2-naphthalenols

3.2.2 Fluorescence studies

3.2.3 Conformational studies

3.2.4 Discussion photochemisty of 2-naphthalenols

3.3 Photooxygenation

3.3.1 Results photooxygenation of 2-naphthalcnols

3.3.2 Discussion photooxygenation of 2-naphthalenols

3.4 Concluding Remarks

3.5 Experimental Section

3.5.1 UVand fluorescence measurements

3.5.2 Irradiation procedure

3.5.3 Characterisation of photoproducts

References

4 Intramolec:ular Hydrogen-Transfer and Cycllr

Addition Photochemistry of Cyclic l,3-Dienes

4.1

4.2

Introduction

Results and Discussion

4.2.1 Synthesis, purification and identification

4.2.2 Photochemistry

4.3 Conformational studies

4.4 Concluding Remarks

4.5 experimental Section

4.5. I General procedures

4.5.2 Irradiation procedure

4.5.3 Characterisation of photoproducts

References

Chapter 5 Excited State chirality. Synthesis and CPr, of

Optically Pure 3-(ln ... *}-(l S,6R)-Bkydo[4.4.0Jdecane--3,8-diooe

5.1 Introduction

5.1.1 Circular polarb:ation in luminescence

5.1.2 Chemiexcitation via themolysis of 1,2-dioxetancs

5.1.3 Dye-sensitized photooxygcnation

5.2 Synthesis of 1,2-dioxetanes

Contents

56

56

59

60

61

63

63

64

67

68

68

69 70

72

75

76

78 78 79

88 91

91

91

94 95

98

101

102

104 !O5 108

llO

Contents

5.2.1 Synthesis of optically active (1 S,6R,8S)-8-hydroxy­

bicyclo[4.4.0]decan-3-one

5.2.2 Synthesis of (lS,6R)-3-(E,Z-methoxymethylene)­

bicyc1o[ 4.4. O]decan-8-one

5.2.3 Photooxygenation of model enol ethers

5.2.4 Synthesis of optically active 1,2-dioxetanes

5.3 Characterisation of the l,2-dioxetanes

5.3.1 Decomposition of the 1,2-dioxetanes

5.4 Circular polarization in chemiluminescence 5.5 Structur.a.l aspects. of centrosymmetric and meso diketones

5.5.1 Crystal structure

5.5.2 Modelling diketones with semiempirical PM3 calculations

5.6 Discussion

5.6.1 Discussion on photooxygenation

5.6.2 Discussion on the thermolysis of 1,2-dioxetanes 5.6.3 Discussion on circular polarization in luminescence

5.7 Concluding Remarks

5.8 Experimental Section

5.8.1 General procedures

5.8.2 General procedure for the synthesis of 1,2-dioxetanes

derived from enol ethers

References

Summary

Samenvatting

Curriculum Vitae

Dankwoonl

7

02

113 113

116 117

117

120

122

122 125

125 125

126

126

127

128

128

134

138

143

146

149

1SO

1 Mechanistic Photochemistry and Chirality

1.1 Introduction

Photochemical processes l under the influence of high-energy Solar radiation (A < 200 nm) have taken place for probably more than a billion years and have definitely

played an essential role in the generation of primitive life on Earth- When the principal

constituents of the primitive Earth's atmosphere (mixtures of methane and ammonia, or

nitrogen and water) are subjected to UV or ionizing radiation, many important biological

compounds like sugars, amino acids and purines may be formed. lb •d•2

The natural origin of optical activity in biomolecules has fascinated scientists since

the time of Pasteur's original observation that chiral molecules (chiral in the meaning of

handedness) exist as two mirror images, and that one of them predominates in most

organic natural products and all living organisms. 3 This prevalence of only one of the two

mirror-image enantiomers among natural organic molecules led to the proposal of several

hypotheses for the transition from achiral geochemistry to chiral biochemistry.4 One that

is still tenable is the direct asymmetric photochemistry with circularly polarized light.

Based on the preferential ~citation of a particular enantiomcr from a racemic substrate

(Circular Dichroism), and the fact that natural sunlight exhibits circular polarization as a

result of reflections at the Earth's surface,Jb the enrichment of a racemic mixture upon

irradiation with circularly polarhed light might explain the formation of optically active

substances in nature. The optical purity of the product mixture is determined by Kuhn's

anisotropy factor,s and the speed of interconversion of enantiomers at different stages of

the photochemical processes in which chiral centres are preserved, created, destroyed or

asymmetrically fixed (see subsection 1.5).

"Photoohemistry" is a term rather loosely applied and is generally restricted to the

9

10 MECHANISTIC PHOTOCHEMISTRY AND CHIRALITY

processes where light interacts with matter in sLlch a way that chemical conversions may

be initiated. I However, a number of physical processes that do not involve any overall

chemical change also lie within the province of the photochemist: for instance

f1uorescence or phosphorescence (in which light is emitted from a spt..'Cies that has

absorbed radiation) or chemiluminescence (in which light is emitted as a result of a

chemical reaction). 1,6

Many photochemical reactions have been known for almost as long as "chemistry"

exists. The first systematic contributions were made by Ciamidan and Silbcr who

reported about a widc range of reactions in thc period of 1900-1915.1 BaseO on their

pioneering work, some very important discoveries, dealing with dimerl7.ation, hydrogen

abstraction and cycloaddihons were made. Following this, the development of quantum

theory led to the reali7.ation that upon absorption of rddiation a particle or a part of a

molecule (chromophore) is converted to onC of its excited statcs for each quantum of

radiation absorbed (transition). All this led to the development of the basic concepts in

modern organic photochemistry.

Since many photochemical processes are strongly dependent on conformation and

configuration of ground-state substrates, thc use of chiral molecules within this area of

chemistry is increasing rapidly. fn this dissertation, a variety of stereochemical aspects in

photochcmistry is described. In this introductory chapter, the most important issues with

respect 10 mechanistic photochemistry and chirality arC given.

1.2 Mechanistic Organic Photochemistry

Light is manifesting itself as a unique reagent, and a thorough understanding of

photochemical processes (such as absorption and emission of light, intersystem crossing,

internal conversion, energy transfer and chemical reactions) is of great value to a wide

area of applications in photochemistry.8 These applications may vary from vision9 and

photosynthesis,IO to photoimaging, II photochromism,12 and solar energy storaget3 or

even photorlegradation of polymers and synthetic utW ... .ations.

Whereas classical chemistry involves reactions of molecules in their lowcst electronic

energy (ground state), photochemical reactions originate from states with higher electronic

energy (excited states). Since excited states have a changed electronic structure,

photochemical reactions are often strikingly different from reactions proceeding in the

] . 2 Mechanistic organ.ic photochemistry 11

ground state, where molecules are thermally activated. Relevant parameters of ground.

state reactivity, such as the energy and geometry of all participating molecules, are often

not sufficient to describe a photochemical reaction. The presence of avoided crossings and

funnels through which an excited molecule may fall back to its ground state

(photodynamics) are also of great importance in order to arrive at a full mechanistic

description. ]4

Generally an overall photochemical reaction can be thought of as composed of

photophysical and strictly photochemical processes. lnitiaJly, the interaction of light with

a molecule leads to the absorption of a photon. The electronically excited molecule will

relax its geometry to a minimum on the excited state potential energy surface (e.g. the

singlet excited state S]), from which a number of processes are possible. The first one is

the emission of a photon (fluorescence) after which the molecule relaxes to the ground

state (So> equilibrium geometry, so that no net reaction has occurred. Secondly,

intersystem crossing leads to the formation of the triplet excited state T l' From the latter,

phosphorescence can be observed upon emission of a photon. Furthermore, from both S]

and T] radiation less transitions can occur. As an alternative, from both excited states a

photochemical reaction may take place.

* 'I'

R/)3et31'1! Pholophysics

, ,

h h 'Hot· Gro\.Jnd~

C'O$$ing Intermedi!llt! State Reaction (dlabslit;)

+l'1u

FIClUR!> 1.1 Schematic potential energy representation of various photoprocesses,

representing adiabatic, diabatic and 'hot' groundstate reaction pathways.

As an example of the competition between the photochemical and photophysical

processes, photochemical reactions such as the cis-trans isomerizations in biologically

relevant molecules require an efficient radiationless transition to prevent further

photochemical reactions and to account for a high quantum efficiency. On the other hand,

12 MECHANISTIC PIiOTOCHEMIsTRY AND CHIRALITY

a small non-adiabatic interaction is a necessity for the occurrence of photochemical

sigmatropic s.hifts. 15 However a close examination of photodynamics requires the detailed

knowledge of potential energy surfaces, which have to be derived from qualitative

arguments or quantumchemieal calculations_ 16 Despite the fact that extensive calculations

of electronic structure and dynamics are of great relevance in the mechanisms of

photochemical reactions, it is generally the experiment that has to act as the final arbiter.

Although both singlet and triplet excited species can undergo chemical reactions,

they are mOre common for triplets, due to their much longer lifetimes. Excited singlet species more often undergo physical processes before any reaction has a chance to take

place. To establish (he mechanism of the photochemical behaviour of various

chromophores, bearing the theoretical criteria in mind, it is of general use to design

molecules from which one may eliminate those mechanisms. that are inconsistent with the

product types or stereochemical pathways_ Since intermediates involved are often difficult

to trace, tools as labelling, and the combination of stereochemistry and photochemistry

are of great value.

Especially of mechanistic interest in relation to stereochemistry and this thesis are the

reactions of ketones and olefins which are very briefly listed below.

1. 2. J PhQtoisomerizatiollS

The most common reaction is definitely the cis-trans isomerization of many olefins

by both direct and sensitized irradiation,17 Since the direct irradiation is difficult to

achieve in the case of the simple olefins, the cis-trans isomerization may generally be

o cis R-trans S-trans

effected by photosensitization (i.e. a photochemical or photophysical alteration in the

olefin as a result of initial absorption by an other molecular entity called a

photosensitizer.) and may serve as an indicator of triplet energy. Often as a result of

photoinduced cis-trans i~omerization to highly reactive transoid alkenes, subsequent

additions, cycloadditions or cIectrocycIic rearrangements may take place.

In the singlet excited state of myrcene, electrocyclic ringcIosure occurs, (a 1,3-

1.2.2 Photochemical sigmatropic shifts 13

diene reaction) in addition to the normal [2+2] cycloaddilion. 1S In the triplet state, a

cross cycloaddition product is formed. This reaction follows the empirical rule of five)9

which states that if triplet cycli~tion can lead to rings of different size, the one formed

by 1,5 addition is preferred kinetically.

Sifl 1&1 dil9d:

T" 11:1

aenalllZed

1.2.2 Photochemical sigmatropic shifts

+

An [1 J] sigmatropic shift is a thermally or photochemically induced intramolecular

pericyclic reaction, which involves a migration of a (I'-bond along a ~-system from atom 1

to atom j either suprafacially (at the same face of the ~-system) or antarafacially (at the

opposite face of the 'i'r"system). The process occurs in a concerted manner via a cyclic

transition state. The orbital symmetry (Woodward and Hoffmann) rules show that there

4n electrons must be involved in photochemical suprafacial shifts (and 4n+2 in an

antarafacial shift).20 The migrating substituent is. usual a hydrogen or an alkyl group. If

the carbon atom of the alkyl group is chiral, a suprafacial shift can either take place with

retention (4n) or inversion (4n +2).

An interesting case with respect to the stereochemical fate of the photochemical

H~ ~ 1 3 eN Ph ,

H CH3 CN

hv

[1.3l~ift 254 nm Etl'l8nol

H.~ ~ 3 1 CH3 Ph

CN CN H

rearrangement of an [l,3]-C shift, is studied by Cookson and Shanna.21 They observed

that the photochemical [I,3]-benzylic shift in the dicyanomethylene cyclohexane, depicted

14 MECI1ANISTIC PHOTOCf-I/i:MISTRY AND CI1IRALITY

above, occurs with retention of configuration of the migrating benzylic carbon atom.

1.2.3 PhOlojragmentations

Carbonyl compounds undergo fragmentation reactions by cleavage of a (I-bond to furnish

biradicals (l,2-biradical for Norrish type 122 and 1,4-biradical for Nomsh type U2J). This

is due to UV absorption in the accessible 280 nm region, which corresponds to the n~1f*

transition, i.e. promotion of One of the non-bonded electrons On oxygen from an orbital

localized in the plane of the carbonyl group jnto the higher energy antibonding 'IT. orbital

delocalized over the carbonyl group:

OH

J.: :CH2

Ph"--CI • .:>t-t-R RI R2 R F 9 RI

Ph-C-fH R2

:~=f:t-R RI R

R H

+H R H

Nornsh II:

R" H. R1= R2 = CH,:

R= CH3• R 1 = R2 = H: 1::550 tilTKJS mora doallDgo

1.2 A PhoroadtiitiollS

Many photochemical reactions may be classified as overall linear additions across

unsaturated linkages. Both acyclic and cycloadditions are possible. In general, these

additions do not occur in a single step but proceed via biradical or lwitterionic

intermediates. Direct or sensitized irradiation of cyclic alkenes (C6• C7) in protle solvents

causes protonation of the highly strained alkene with the formation of a cycloaJkane

carbocalion, stabilized by a subsequent rearrangement of the double bond, hydr'dtion, or

alkoxylation. 24 A photoaddition of great importan~ in our work is the photo-oxygenation

of double bonds with singlet oxygcn,6 leading in general I.(J hydroperoxides in an ene

reaction: 25

1.3 Interaction of chiral molecules with radiation 15

Alkenes with electron rich double bonds, or those that do not possess an allylic

hydrogen can undergo a [2 + 2] cycloaddition to form a 1,2'dioxetane. Dioxetanes can be

quite stable, but generally they undergo chemiluminescent (see chapter 5) thennal

decomlX'sition to produce two carbonyl fragments, one of which is produced in the

excited state.

~oc~ .. ~O. ~ \ ·CI-\OCOH ~

1.3 Interaction of chiral molecules with radiation26

Absorption of light by a molecule is related experimentally to the extinction

coefficient € and is connected to a classical theoretical quantity t the oscillator strength J, by integrating €(jI) over the whole absorption band in cgs units: 27 ,2S

f = 4.3 X 10-9 /Ii} f ~(JJ~dJJ The oscillator strength is a dimensionless quantity that measures the probability of an

electric dipole transition, compared to that of a free electron oscillating in three

dimensions. When light is absorbed, a photon interacts with an atom or group of atoms

and promotes transitions between quantized states. This interaction can be described as

being due to an electric component of the electromagnetic wave with an oscillating

electric dipole moment during the transition, and is the result of a perturbation by the

outside fieJd, by changing the electron from state "'i to a final state "'j ; vij is the

wavenumber of the transition. According to classical theory, the oscillator strength is

connected to the fundamental molecular property, the electric dipole"transition moment "'ij

for a given transition by:

16 MECHANIstIC PHOTOCHEMISTRY AND CHIRALITY

The dipole strength J) is just the square of ftij and may either be obtained from

experimental measurements or from quantumchemical calculations. The transition dipole

moment is given by:

and:

It = -e 1: 'i + e EZs Rs i .f

Combining both equations for f, and substituting the average extinction of right (€R) and

left (€J circularly polari1ffii light for the extinction coefficient ~. we get for the dipole

strength:

D = 92.0x 10-4°1 €L +tR. dy 2 11

Optical activity of a chiral substance is usually characterized In terms of the

fundamental quantity, the rotational strength R. Like thc oscillator strength it may be

obtained from theoretical calculations and experimental values of the circular dichroism or

from measurements of thc optical activity. In egs units the rotatory strength is related to

the area under the circular dichroism band by:

Quanlumchemically the rotational strength characterizes a transition from a lower to

a higher state and is defined as the imaginary part of the scalar product of the molecular

transition dipole moment with the transition magnetic moment for the transition involved:

in which the transition magnetic moment is imaginary:

m -i 92.7x 10-22 -; EliXVi i

1.4 Chiroptical techniques 17

On reflection or inversion the rotational strength changes sign and specifies the structural

asymmetry required for an optically active molecule: only molecules not superimposable

on their mirror images are optically active. The two components of the non­

superimposable mirror image pair are called enantiomers (cnantiomorphs). If the molecule

has a centre of symmetry or a plane of symmetry, the scalar product of the integrals will

be r.ero for aU states i and j and R;j = O. Molecules with either a plane of symmetry or a

centre of inversion (more precise: with a rotary-reflection axis) are therefore optically

inactive. The overall symmetry property of a molecule therefore determines its optical

activity; a phenomenon already known since its discovery by Pasteur.

1.4 ChiropticaI techniques29

Any material that rotates the plane of linearly polarized light is said to be optically

active. 30 A linearly polari~ beam may be resolved into two circular components,

rotating in opposite directions with equal amplitudes and identical wavelengths (in phase).

When left and right circularly polarized light passes through an optically active medium,

the speed of both rotating components is no longer the same and the phase difference of

the transmitted beam sets up a new plane of polarization. Since the speed of light depends

on the refractive index n, the material becomes circularly birefringent. The rotation of the

plane of polarization is called the optical rotation a;

Generally the optical rotation is reported as the specific rotation [a]20o = (a/c/l) x

100% in which c is the concentration in g/loo mL, and I the path length in dm, although

sometimes other units may be used. Since the refractive index is a function of the

oscillator strength and the wavelength, the optical activity Or optical rotation also depends

on the wavelength of the incident light. This dependence is called optical rotatory

dispersion (ORD).

Similarly to the circular birefringence, both components of the light may also be

absorbed differentially, i.e. the amplitude of the electric vector Er. is no longer equal to

~ and their resultant oscillates now along an ellipse. As a result of a difference in the

absorption index k, the optically active medium exhibits circular dichroism (CD), which

is governed by the elipticity if! {radians per unit length];

18 MECHANISTIC P~ltrroCHEMISTRY AND CHIRALITY

By analogy with the specific rotation the molar elipticity [I.I} is given by: 3300 (€L - ~R )

in which the absorption indices are replaced by the molar extinction coefficients ~L and €R

uSing the relation k = (In 10) E C, with C in moles per litre. ORD and CD spectra

contain the same information and can be interconverted by the Kronig " Kramcrs

relation. 31

Alternatively, the optical activity may be given in terms of electronic transitions;

since in the eIipticity the strength of excitation is not taken into account, the anisotropy

factor g has been introduced by Kuhn;5

4R D

Provided the absorption and band shapes are identical, g = 4RID, in which R represents

the rotational strength and D the dipole moment (vide supra).2S, 31

The chiroptical techniques described are well known and have been used for quite a long time. However thc emission analogue of CD, i.e. the circular polarizal.ion in

luminescence (CPL) is more recent.32 This phenomenon refers to the differential emission

(defined as .6.I "" IL - IR ; and r as the mean intensity) of light by chiml luminescent

species and reflects the chirality and thus the structural features of molecular emitting

states. As a result optical activity in luminescence has been defined as the dissymmetry

factor in luminescence or emission glum or ge. Provided the total luminescence and CPL

band shapes are identical:

The fact that the optical activity can be governed by thc excited state in CPL might

implicate that a molecule, which is optically inactive in the ground state, may become

optically active in the excited state as a result of the structural change, provided no

racemic mixture of excited state molecules is present. Since unpolarized exciting light

produces equal concentrations of optically active enantiomers, the net effect, however will

be the same as in CD: no optical activity will be observed. However, if circularly

polarized exciting light or enantioselective chcmiexcitation is used, excited state chirality

can bc observed.

J.5 Asymmetric photochemistry 19

1.5 Asymmetric photochemistry

Although optical activity has stimulated the chemist's interest for many years, the

combination of stcrco- and enantiodifferentiation into reactions of electronically excited

states was not extensively explored until the past three decades_:33 Based on the

phenomenon of Circular Dichroism (CD), asymmetric photochemical reactions have been

initially restricted to the preferential excitation of a. particular enantiomer from a racemic

substrate by circularly polari:red light (see subsection 1.1). However, its rela.tionship to

the origin of optical activity, its small enantiomeric differentiation (however, upon double

quantum excitation the enantiomeric differentiation may bc cnhanced considerably34), and

the fact that activation energy of racemization is much smaller in thc excited state than in

the ground state has made circularly polarized light not very suitable for the purpose of

asymmetric synthesis. For instance, in the asymmetric cyclization reactions leading to

helicenes only low (c_c_ < 5%) optical purities were obtained.3S In this type of

asymmetric photochemistry the enantiomcric ratio of the product mixture is always

derennined by the anisotropy factor in absorption but depends also on the stage at which

the enantiomers intcrconvert, and chiral centres are preserved, created, destroyed or asymmetrically fixed _ 33

One of the highest optical purities obtained is the direct asymmetric destruction by CPL

of a chira! ketone with a large anisotropy factor ga (0.24 at 313 nm):36

H

Extw ttl deetl'lJttlM 99'l(,

Optical purity 30%

Although the use of circularly polarized light may still be of great value in certain

cases (Optical Recording,37 Second Harmonic Generation38) its synthetic value is modest.

The most frequently employed techniques to synthesize enantiomerlcally pure compounds,

involve chiral $ubstituents close to the prochiral substrate as a chiral auxiliary, or a chiral

unit built in the molecule to serve as a readily available reactant for fonnation of the final

product. Asymmetrical photoreactions with high levels of optical yields can be achieved

when the diastereodifferentiation is induced by an intramolecular chira! substituent. In

20 MECHANJSTIC PHOTOCHEMISTRY AND CI'IIRALITY

these cases factors as chemical asymmetric induction, sterie interactions, and hydrogen

bonding interactions in excited and ground states arC responsible for the high level of

diastereosdcctivity and regioselectivity_ These reaction systems are of practical

importance for the elucidation of reaction mechanisms, and require asymmetric synthesis

which is often dependent on available biological enzymes or other common chiral sources.

Chiral induction or transfcr of chirality is often governed by motional freedom_

Restriction of translational and rotational motion of reactant molecules as well a.~ that of

the reactive intermediates (derived from the reactant molecules) often tends to be reJlected

in the product distribution_ As an example the efficiency of hydrogen abstraction and

fOrmation of the 1,4·biradical is markedly dependent on the preferential ground state,

particularly when the excited state is shortlived_ 23a This influence of the conformation is

beautifully expressed in the reaction of t -mcthyl-cyclohexyl-phcnyl ketone23.:o where two

kinetically distinguishable excited triplet n,7* states of the equatorial and axial isomers

lead to different reaction types: a-cleavagc or r-abstraction and cycIization, respectively.

For an efficient ,a-cleavage in the Norrish type II process shown in subsection 1.2_3

the biradicaJ requires a conformation in which the singly occupied p orbitals overlap

significantly with the bonds to be broken. Any feature which inhibits the attainment of the

parallel geometry will reduce the efficiency of the j'J-cIeavage. Although 1,4-biradicals

generally prefer to cleave rather than undergo cydization, increased substitution on the Ci

carbon and conformational effects can be extremely important and even favour the

cyclization step. Conformational control in the excited state and ground state thus plays an

important role in many photochemical reactions. Beautiful examples of chiral induction

are e)Cprcssed in photochemical transformations in the solid state or chiral host guest

lattices?9 A very nice example has been reported by Zimmerman for the photochemical

4,5,5-triphenylcyclohe:xenone rearnmgement in which the formation of a bicyclo[3.3.1}

structure proceeds with inverted stereochemistry in the crystal compared with solution. 40

1.6 Scope of the thesis

The work in this thesis is based on the role of enantio- and diastereodifferentiating

chemical and physical processes in reactions of electronically excited states or reactive

intermediates therefrom. Whereas our primary concem dealt with the mechanistic organic

photochemistry of a chira! unit built into the substrate skeleton of a 2-propenol unit, we

were also intrigued by one of the most fascinating challenges in the study of optical

1. 6 Scope 0/ the thesis 21

activity, i.c. to synthesize a molecule whose chirality is solely due to the presence of a

localized excited state, a phenomenon belonging prior to this research to the realm of

fantasy.

Considering the mechanistic photochemistry, In the past an interesting

enantiodifferentiating rcarmngement was found in the 2-propcnol unit of racemic·2-

naphthalenol derivatives.41 ,42 In this molecule a remarkable {l,3]-OH shift was claimed to

occur, in accordance with quantumchemical and experimental studies on 2-propenol and

hydroxy gennacrenc B..n ,44 Based on the observed stereoselectivity, a non-Woodward­

Hoffmann reaction path for the migrating hydroxyl group was postulated, in which the

reactive alkene moiety underwent a 90¢ twist aftcr excitation. This twist was assumed to

be accompanied by a complete separation of charge, and thus a singlet state. 10 this

molecule the chiral centre next to the prochiral substrate is able to induce a preferential

twisting, and therefore the stereochemical outCOme of the photoproducts led to the

proposal that the migration of the hydroxyl group occurs within the plane of the exocyclic

double bond_

Despite the fact that the enantiomcric differentiation seems in agreement with a

planar [1,3]-OB shift, some questions. remained unresolved and motivated us to

reinvestigate the photochemistry of the 2-naphthalenol derivatives. Since the product

distribution of the photoproducts is of essential importance, we synthesized thc 2-

naphthalenols in thcir optically pure form. For this purpose, a new synthetic route has

been developed, as is described in chapter 2. All compounds were fully characterized

with one- and two-dimensional NMR spectroscopy.

In chapter 3 the reinvestigation of the photochemistry of the optically pure 2-

naphthaJenols is described. The negative conclusion in this chapter with respect to the

feasibility and actual occurrenCe of the (l,3].QH shift and the fact that photooxygenation

of chiml alcohols attained increasing interest has prompted us investigate the

photochemistry of the optically active naphthaJenols in the presence of oxygen_ We have

found that the double bond is rapidly oxygenated and leads to stereoselective fonnation of

new products. The observed stereospecificity is explained by the selective coordination of

the chiral hydroxyl group in the 2-propen-l-ol unit.

Chapter 4 describes the photochemistry of 9·eodo-hydroxy-9-exo-vinylbicyclo[4.2.1]

nona-2,4-diene analogues. This study was performed on the basis of deuterium labelling

experiments, conformational analysis~ and One· and two-dimensional NMR spectroscopy.

The results revealed that the hydroxyl proton of the 2-propen-l-01 unit was transferred

regiospecificallyand intramolecularly towards the endo side of the diene moiety.

Chapter 5 deals with the synthesis of a compound in which the optical activity is

22 MECHANISTIC PHOTOCHEMISTRY AND CHIRALITY

solely due to its excited state. The route to the optically active excited diketone is based

on the chemicxcitation via an optically active 1,2-dioxetane.45 Chemicxcitation has been

performed by photochemical sensitation of an electron-rich olefin with singlet oxygen.

The optical activity was established by determining the circular polarization in the

chemiluminescence% upon thermolytic decomposition of the obtained optically pure 1,2-

dioxetane.

References

1. For general reading in organic photochemistry see: (a) J. G. Galven and 1. N. Pitts, Jr. In Photochemistry, John Wiley, New York, 1966. (b) D. O. Cowan and R. L. Drisko, In Elements or organic photochemistry, Plenum Press, New York, 1976. (e) N. J. Turro, in Modem molecular photochemistry, Benjamin/Cummings, Menlo Park, 1978. (d) R. P. Wayne, In Principles and applications of photochemistry, Oxford university Press, Oxford, 1988. (e) J. C. Scaiano, ed. CRe Handbook of Organic photochemistry, CRC press, Boca Raton, Florida, Vols I and 2, 1989. (f) A. Gi/ben and J. Baggott, In Essentials of molecular photochemistry, In eRC press, Boca Raton, Florida, 1991. (g) J. Kopecky In Organic photochemistry, VCR publishers, Inc. New York, 1991. (h) W. HOT.rpool and D. Annesto, In Organic photochemistry, Ellis Horwood Limited, Chlchsester, 1992.

2. (a) R. M. Lemmon, Chern. Rev. 69, 95 (1969). (b) R. P. Wayne, Chem. Ber. 24, 225 (1988).

3. (a) L. Pasteur, C. R. Hebd. Seances Acad. Sci. 26, 535, (1848); ibid. 27, 401 (1848); Ann. Chim. Phys. 24, 442 (1848). (b) S. F. Mason, Chern. Soc. Rev. 17, 347 (1988) and references therein. (c) R. A. Hegstrom and D. K. Kondepudi, Sci. Am. january, (1990).

4. (a) W. Elias, J. Chern. Educ. 49, 448 (1972). (b) W. A. Bonner, In Exobiology, CX. PQnnamperura, Ed., North-Holland PubJishing Co., 170·234 (1972). (c) L. D. Barron, Chern. Soc. Rev. 15, 189, (1986). (d) S. F. Mason, Nature, 311, (1984). (e) C A. A. van Boeckd, G. M. Visser, R. A. Hegstrom and J. H. van Boom, J. Mol. Evol. 25, 100 (1987). (t) M. Quack, Angew. Chern. Int. Ed. EngL 28, 571 (1989) and references therein.

5. (a) W. Kuhn, In Stereochemie, K. Freudenberg, FAI., Verlag Franz Deutike, Leipzig und Wicn, 366 (1979). (b) W. Kuhn, Trans. Faraday Soc. 26, 293, (1930).

6. (a) A. A. Frimer, Ed., In Singlet oxygen, eRe Publishing, Boca Raton, Florida, Vols 1-4, (1985). (b) This thesis, chapter 5 and references therein.

Riferences 23

7. (a) G. Ciamician and p, Silber, :aer. 35, 4129 (1902); ibid. 36, 4266 (1903); ibid. 33,291 t (1909). (b) H. D. Roth, Angew. Chem. 101, 1220 (1989)_

8, For terms recommended by lUPAC for use in photochemistry, see S. E, Braslavsky and K. N Houk, Pure Appl, Chern. 61, 1055 (1988).

9. (a) L. Stryer, Sci. Am. 256 (I), 32 (1987)_ (b) Photobiochemistry and Photo­biophysics 13, december (1986).

10_ G. Porter, Pure Appl, Chern, 58, 1171 (1986) and references therein.

11. M. G. Clark, Chem. Ind. 258, (1985).

12. H. Durr, Angew. Chern. 101, 427 (1989).

13. (a) R. Memming, Top. Curro Chern. 143, 79 (1988)_ (b) A_ Harrima1lll, J­Photochem. 25, Part I, (1984).

14. 1. Michl and V. Bonabc-KoutCckj, In Electronic aspects of organic photochemistry, John Wiley, New York, 1990.

15. G. J. M. Domums, PhD Thesis, Eindhoven University of Technology (1987).

16_ (a) W. G, Dauben, L. Salem and N. J. Turro, Acc. Chern. Res. 8, 41 (1975)_ (b) L. Salem, Science, 191, 822 (1976). (c) L. Salem, J, Am, Chern. Soc. 96, 3486 (1974). (d) L. Salem, C. I.eforeslier, G- Segal and R. Wetmore, J. Am. Chern. Soc. 97, 479 (1975).

17. (a) 1. Saltiel, 1. D'Agostino, E. D. Megarity, L_ Ml!tts, '- M. Neuberger, M. Wrighton and O. C. ZafriJou, arg. Photochcm_ 3, 1 (1973). (b) J. Salliel, D. W. L. Chang, E D_ Megarity, A_ D_ Rous.l'l!Qu, P. T, Shannon, B. 11wnws, and A. K. Uriarte, Pure AppL Chern, 41, 559, (1975). (c) H. Meier, In Methoden der organischen chemie (Houben Weyl), G. Thieme, Stuttgart, VoL 4/5a, Photochemie, 189 (1975).

18. (a) K. 1. Crowley, Tetrahedron 21, 1001 (1965). (b) J. Saltiel and O. C. Zajitious, Mol. Photochem. 1, 1 (1969).

19. (a) R. Srinivasan and K. H. Carlough, J_ Am, Chern. Soc. 89, 4832 (1967). (b) W. C. Agosta and S. Wolf, J. Org. Chern. 49, 3139 (1980).

20. R. B. Woodward and R. Hojfrnann. In The conservation of orbital symmetry, Verlag Chernie International, Deerfreld, Florida (1970)_

21. (a) M. SJwmw., J. Am. Chern. Soc_ 97, 1153 (1975); :see also ref. If, p. 244. (b) R. C. Cookson, J. Hudec and M. Shanna, J. Chem. Soc., Chern. Commun. 107, 108 (1975). (c) R. C. Cookson, Quart. Rev. Chern. Soc. 22, 423 (1968).

24 MECHANIstiC l'HOTOCH~MISTRY AND CHIRALITY

22. N. J. Thrro, J. C. Dalton, K Dawes, G. Farrington, R. Hautala, D. Morton, M. Nie.mc<:yk and N. Schore, Acc. Chern. Res. 5, 92. (b) J. C. Dalton and N. J. Turro Ann. Rev. Phys. Chem. 21, 499 (1970).

2:1. (a) P. J. Wagner, Ace. Chern. Res. 4, 168 (1971). (b) J. Scaiant), Ace. Chern. Res. 38,819 (1982). (c) P.1. Wagner, Ace. Chem. Res. 16,461 (1983).

24. P. J. Kropp, Org. Photochem. 4, 1 (1979). (b) P. 1. Kropp and H. J. Kraus, LAm. Chem. Soc. 89, 5199 (1967).(c) 1. A. Marshiill, Acc. Chern. Res. 2, 33 (1969).

25. (a) C S. Foote, Acc. Chern. Res. 1, 104 (1968). (b) C. S. Foote and G. Uhde, Org Photochem. Synth. 1, 60 (1971).

26. E. Charney, In The molecular basis of molecular optical activity, John Wiley, New York, 1979.

27. One cgs unit for a dipole moment is defined as 10+ 18 Debeye, and 1 Debeye is defined as 3.33564 x 10-30 C m.

28. H. P. 1. M. Dekkers and L E. Closs, J. Am. Chem. Soc. 98, 2210 (1976).

29. F. Giardelli and P. Sa/vadori, In Fundamental aspects in recent developments in ORD and CD, Eds, London (1971).

30. J. March, In Advanced organic chemistry, John Wiley and Sons, New York, 4th edition 1992, Chapter 4 and references therein.

31. A. Moscowilz, Adv. Chern. Phys. 4, 67 (1962).

32. (a) '- P. Riehl and F. S Richardson, Chern. Rev. 86, 1 (1986). (b) F. S. Richardson and J. P. Riehl, Chem. Rev. 77, 773 (1977).

33. (a) Y Inoue, Chern. Rev. 92, 741 (1992). (b) H Rau, Chern. Rev. 83,535 (1983). (c) H. B. Kagan and 1. C. Fiaud, Top. Stcreochern. 10, 175 (1978).

34. D. N. Nikcgosyan, Y. A. Repeyev, E. V. KlwrQshilova, I. V. Kryukov, E. V. Klwroshilov and A. V. Sharkoy, Chem. Phys. 147,437 (1990).

35. O. Buchardt, Angew. Chem. 86, 222 (1975)

36. H. B. Kagan and J. C Fiaud, TOp. Stereochcm. 18, 249 (1988).

37. B. L. Feringa, W. P. Jager and B. de Lange, Tetrahedron, 49, 8267 (1993). (b) B. De Lange, PhD Thesis, University of Groningen, 1993.

38. E. W. Meije.r, E. E. lJavinga and G. L. J. A. Rikken, Phys. Rev. LeU. 65, 37 (1990). (b) E. W. Meijer and E. E. Havinga, Synth. Metals, 57,4010 (1993).

References 25

39, (a) V, Ramamurthy, Ed., In Photochemistry in organized and constrained media, VCR Publishers, New York, 1991. (b) J, R. Scheffer and M. Garcia-Garibay, In Photochemistry on solid surfaces, M. Anpo and T. Matsuura, Eds_, Elsevier, Amsterdam, Chapter 9, 1989.

40. H. E. Zimmerman and M. Zuraw, 1. Am. Chem. Soc. Ill, 2358 (1989)_

41_ (a) W. J. G. M. Peijnenburg and H M. Buck, Tetrahedron 44, 4821 (1988). (b) W. J. G. M. Peijnenburg, PhD Thesis, Eindhoven University of Technology (1988).

42. W. 1. G. M_ Peijnenburg and H. M. Buck, Tetrahedron 44, 4927 (1988).

43. (a) H_ R_ Fransen and H. M. Buck, J. Chem. Soc., Chem_ Commun. 786 (1982). (b) H. R. Fransen, G. J. M. Dannans, O. J. 8ezemer and H. M. Buck, ReeL Trav, Chim. Pays-Bas 103, 115 (1984).

44_ (a) O. J, M. Donnans, H. R. Fransen and H. M. Buck, 1. Am. Chem. Soc. 106, 1213 (1984). (b) G. J. M. Dormans. W. 1. G. M. Peijnenhurg and H M. Buck, ], MoL Struct. (Theochern) 119, 367, (1985). (e) G- J. M. Donnans. G. C. Groenen­boom and H. M. Buck, J. Chern. Phys_ 86, 4895 (1987).

45. (a) E. W. Meijer and H. "ynberg, Angew_ Chem_ 100, 1004 (1988); Angew. Chem. Int. Ed. Eng!. 27, 975 (1988)_ (b) E. W. Meijer and H. "ynberg, J. Am_ Chem_ Soc. 104, 1145 (1982).

46. (a) P. H. Schippers and H. P. 1. M. Dekkers, 1. Am. Chern. Soc. 105, 145 (1983)_ (b) P. 8. Schippers. PhD Thesis, University of Leiden, 1982. (c) /I. Numan, PhD Thesis, University of Groningen, 1978_

2 Abstract

Asymmetric Synthesis of (2S,4R)- and

(2RAR)-1 ~2~3 ,4-tetrahydro-4-ethyl-1 ,1,4-

trimethyl-(3Z)-ethylidene-2-naphlhalenols#

The synthesis and characterisation of optically pure (2S,4R)· and (2R,4R)-1,2,3,4-tetrahydro-

4--ethyl-l, 1 ,4-trimethyl-(3Z)-ethylidene-2-naphthalenols lOa and lla are described_ Naphthalenols

lOa and lla are prepared in seven steps from 2-phcnyl-butane nitrile. In the preparative procedure

the optically pure (2R)-2-methyl-2-phenyl-butanoic acid bas been employed as a key starting

material_ Re..~olution of the racemic acid has been achieved in :> 99% enantiomeric excess, by

fractional crystallization of the quinine salts from aqueous ethanol. The next steps in synthesis are

based upon acylation of isobutene by the optically active 2-methyl-2-phenyl-butanoyl chloride S, a

Friedel Craft cyclization and a Wittig reaction, while the ethylidene naphthalenones 8a and !}a are

separated by HPLC. After reduction the target moll;)cules lOa and 118 are obtained_ Using one·

and two-dimensional NMR spectroscopy, including INADEQUATE and 13C.1H correlation, the

structures are unambiguously characterized.

2.1 Introduction

Quantumchemical calculations concerning the photochemical cis-trans isomerization

of olefins, based on the pioneering work of Mulliken,l reveal that excitation of the double

bond results in a 900 twisted diradicaloid structure. 2 In biradicals a small polarizing

perturbation causes a complete polariution of the excited singlet states, a phenomenon

often referred to as sudden polarization_ J

Although the phenomenon of sudden polarization does not seem to have any

#F_W_A.M_ Miesen, l_L.l_ van Oonsen and E_W, Meijer, part of II paper accepted for pUblication in ltee!. Trav, Chim, Pays-BIIB.

27

28 ASYMMETRIC SYNTHESIS OF CHIRAl.. NAPHTHALENOLS

consequence for the cis-trans isomerism as. such, it may lead to reactions such as. an

attack by a polar solvent on the highly polarized alkene. Studies of Marsha1l4 and KroppS

have already implicated the presence of polar intermediates in the photochemical addition

of alcohols to alkenes. (n these additions cis-trans isomerization of the olefin is followed

by protonation of the transoid intermediate. However, more re.:::ently it has been con­

cluded that photoaddition of water and alcohols onto 3-nitrostyrenes occurS via a planar

polarized triplet state. 6 Twisting of the alkene moiety in T I of 3-nitrostyrenes results in

deactivation, only to the ground state.

Although the photochemistry of alkenes in both inert and hydroxylic solvents is well

documented, photochemistry of bichromophoric compounds with suitably positioned

hydroxyl groups has enjoyed only limited study.7.S.9 A remarkable photochemical

rearrangement has been claimed to occur in the 2.propenol units of both 8-

hydroxygermacrene B (germacrol)IO 3 and 1,2,3,4-tetrahydro-4-ethyl-l,1,4-trimethyl­

(3Z)-alkylidene-2-naphthaleool compoundslO,ll 2-1a (scheme 2.1). Upon UV-irradiation

of a racemic rnixturel2 of (2R,4R)- and (2S,4S)-1 ,2,3,4-tetrahydro-4-ethyl-l, 1 ,4-

trimethyl-(3Z)-ethylidene-2-naphthalenol, la, not only a fast E-Z isomeri7,ation has been

found to take place yielding a photostationary state for E:Z of approximately 1; I, but also

a slow stereospecific formation in low yields, of photoprwuct 2a is reported. ll

The experimental data presented. suggest that the photoreactivity is controlled by

the preferred conformation in the ground state,H,14,15 a phenomenon well-known in

photochemistry. It has been reported that the diastereomeric alcohols Ib (a racemic

mixture of 2S,4R and 2R.4S) do not give 2 due to a difference in ground-state confor­

mation, Thc reported stereospecificity in the photoreactivity of la is interesting. It has

been postulated that differentiation of reactivity originates from a preferential sense of

twisting of the polarized singlet excited ene moiety, dictated by the sterie interactions of

stcreocentres_ It should be noted, however, that direct excitation of the olefinic

chromophore in I into its S I state, must be considered unlikely due to the fact that simple

alkenes exhibit poor absorption in the spectral regions accessible to the photochemical

equipment employed. Actually, the alkyl benzene chromophore in 1 might be expected to

be primarily excited. Such an aromatic chromophore is known to be able to sensitize the

triplet state of alkyl ethylenes (Er "" 84 kcallmol for benzene, and 81 kcallmol for

tetrahydronaphthaiene and that for ethem; 79 kcal/mol).16 As a consequence, this would

rather produce the triplet than the singlet excited state of the exocyclic double bond in 1.

Another unre~()lved difference in the comparison between germacrol 3 and la,b is found

in the results of the photoreactivity in methanol. For germacrol an efficient intramolecular

reaction is found, but for the naphthalenol derivatives methanol addition has been

observed, Moreover, in polar non·hydroxylic solvents a strong inhibition of reactivity has

2.1 Introduction 29

been found_ 17 The conflicting experimental results of la in comparison with Ib and 3 and

increasing mechanistical interest in stereochemistry combined with photochemistry of

hydroxylic bichromophoric compounds, has motivated us to reinvestigate the photochemistry of the 2-naphthalenol compounds. Hence, in this chapter we present, the

synthesis and characterization of the optically pure starting materials la and Ib (referred to as 11a and lOa respectively)12 and in the next chapter a study of the photochemical

behaviour of the 2-naphthalenols lOa and 11a in various solvents as well as in the presence of oxygen, is described_

hv 1 fast

Me

H

Me Me E-1a

E-2R,4R (+ 2S,4S)

Me

Me Et H Z-1b

Z-2S,4R {+ 2R,4S}12

28 4R,9S (+ 4S,9R)

hv

slow

2b 4R,9R (+ 4S,9S)

3

SCHEME 2.1

30 ASYMMETRIC SYNTHESIS OF CHIRAL NAf'HTHALFrNOLS

2.2 Synthesis

2.2.1 Synthesis of the optically· active precursor

Optically pure (2S,4R)- and (2R,4R)-1,2,3,4-tetrahydro-4-ethyl-l,1,4-trimcthyl-(3Z)­

ethylidene-2-naphthalenols lOa and lla were synthesized in seven steps from 2-phenyl­

butanc nitrile. In the following we describe each of the steps as summarized in scheme

2.2.

In this preparative procedure the optically pure (R)-2-methyl-2"phenyl-butanoic acid

4 has been employed as a key starting material for the synthesis of the naphthalenol

derivatives. The racemic acid has been prepared on large scale by the methylation of 2-

phenyl-butane nitrile in anhydrous THF with lithium diisopropylamide as the appropriate

base. Subsequent alkylation with iodomcthane resulted after standard work up in the crude

2-methyl-phenyl"butane nitrile in 95% yield. Hydrolysis of the crude nitrile with

potassium hydroxide, in diethyleneglycol at reflux. temperature, yielded after crystalliza­

tion from pentane the analytically pure (±) acid in 86% yield. 1S ,19 Resolution of the

racemic acid has been achieved by fractional crystallitation of the quinine salts from 70

% aqueous ethanol. The ratio alcohol vs. water has been optimized. Upon addition of too

much water, i.e. until almost turbidity,£'O cooling at 4 °C results in separation of an oil

and, hence, no resolution of the quinine salts. The formation of enantiomerically pure 2-

methyl-2"phenyl-butanoic acid 4 (97.5 g) by the resolution has been verified by the

determination of the optical rotation: [(>"]25D "" -29.20 (c 4.9, benzene; o.p. > 99%).20,21 The absolute configuration of (-) 4 is R.n Cram and coworkers established

the configuration of acid 4 by correlating it with the configuration of 2-phenyl-butane.22

They converted the activated isomers of add 4 without configurational modification in

lICveral steps tu 2-phenyl-butane, which in tum has been related, indirectly to that of (+)

tartaric acid whose enantiomer, at the time was determined by X"ray diffraction, directly.

The corresjXInding (R)-2-methyl-2-phenyl·butanoyl chloride 5 is easily obtained by

reaction of thionyl chloride with butanoic acid 4.

2.2.2 SYlllhesis of thf optically pure letmalky/ naphthalenones

An essential step in the reaction route is the synthesis of the optical pure tetraalkyl­

tetralone 6. Racemic crowded tetraalkyl-3,4-dihydro-2(1lJ)-naphthalenones are usually

synthesized via Friedel-Crafts alkylation using tetrahydro-2,2,5,5-tetraalkyJ furanone,

©XC=N 1. UNOPrb ~O 1. quinille ~ SOCG ~ .. •

2. Mer 2. t;+ . 0 H Et

3. KOHl H+ Me Et Me -"Et

t4 42R

Me Me Me Me

~O Se0:2 ©Q:0 1. (C~C= CHI SnCl4 ~ 0 ... .. 2_ BFsO(C:2Hib

- 0 - 0 Me Et

Me 'El Me Et 5 6 7

Me ~e Me Me 0 H

LiAIH4 ""'OH Me Me

Me Me ©Q:0 H o + Ph:JP=CH:H:3 Sa

. 0 Me Me H Me Me Me "-Et ©OeMe ~ 7 0, :::H Me

-__ 0

32 ASYMMETRIC SYNTHESIS 01' CHIRAL NAI'HTliALENOLS

which in turn is prepared via hydratation of (un)symmetrical acetylenic glycols. 23 Since

extensive rearrangements take place uSing this method, and the fact that separation of the

diastercomers has to be performed in one of the next steps it is highly desirable to make

use of another method. As a matter of fact in the first attempts we tried to prepare

diastereomers via the incorporation of an optically pure 2-methyl-butyl group close to the

stereocentre of the racemic (rae) analogue of naphthalenedione 7 via a Wittig reaction.

However, the Wittig reaction on (rac)-7 using 2'-methyl-butylidene triphenylphosphorane,

which in tum was prepared from the brominated chiral alcohol and triphenylphospinc,

failed because of significant sterie interactions of the bulky 2-methyl-butyl group with the

substrate. (It should be noted that Wittig reactions with ethylidene- and propylidene­

phosphoranes proceeded successfully.) Hence to obtain optically pure material we made

use of another method in which the synthesis is based upon acylation of the branched

alkene (iso-butene) by the optically pure 2-methyl-2"phenyl-butanoyl chloride 5, followed

by Friedel-Crafts cyclization of the resulting unsaturated ketones to the polyalkyl 3,4-

dihydro-2(1H)-naphthalenoncs.24 The conditions for acylation of isobutenc with butanoic

chloride 5 have been optimi7.ed; excess isobutene and 0.6 equiv. stannic chloride in

carbon disulfide were used, resulting in a mixture of ketones containing 45% 2-chloro-

2,5·dimethyl-5-phenyl-4-hl.1'tanone 6a, and 40% 2,5 dimcthyl-5-phenylcyclohepten-4-one

6b and 15 % of the naphthalcnone 6. All intermediates were initially isolated and cyclizcd

further with e:xcess of boron trifluoride etherate in dichloromethan.?4. Eventually

naphthalenone 6 was obtained in 45 % overall yield.

In order to avoid extensive rearrangements, the acylation has been carried out with a

minimal amount of stannic chloride in carbon disulphide. Larger amounts or longer

reaction times result in dimerization of the acykhloride via loss of chloride and subse­

quent deca.rbonylation. Hence complete cyclization is accomplished by using boron

trifluoride etherate. Oxidation with Se02 furnishes optically pure 1,4-dihydro-l-ethyl-

1,4,4-trirnethyl·2,3-naphthalenedionc 7 ([OI'f3o "" + 100°, c 2, methanol). A Wittig

reaction on 7, using ethylidene-phosphorane, gives rise to the OI',tJ'-unsaturated ketones Sa

and 9a in a ratio of 40:60. Small traces of E-iwmers were easily separated from the Z­

isomers by column chromatography and were formed in very low yields, due to the steric

interactions of the phosphorane with the bulky methyl and ethyl groups of the dione 7.

Some difficulties arose in the separation of the ethylidene naphthalenones 8a and 9a.

Repeated argcntation chromatography did not result in the separation of both isomers.

(Argentation chromatography has found limited application for enhancing the separation

of E- and Z-isomers of oleflns and is performed on silver nitrate coated silica gel by

evaporating a slurry of 10% AgN03 and silicagel in acetonitrile to dryness).25 The

2. 2. 3 Characterisation of the TUJphthale.nones 33

maximum differentiation among the isomers waS obtained by liquid solid chromatography (LSC). In spite of the disadvantages of this method (critical water content and no gradient

elution), HPLC on silica gel (5 ~m) using rather critical conditions (99.75% hexane and

0.25 % ethyl acetate saturated with 0.1 % water) resulted in the separation of Sa and 9a.

GLC showed both ketones to contain still 5 % of the other isomer.

2_2.3 Characterisation of the naphthalmones

In order to distinguish between the l-ethyl-l,4,4-trimethyl and the 4-ethyl-l,I,4-

trimethyl derivatives, two different methods have been used: first, by carrying out IH NMR Eu(fodh experiments; secondly, by correlating relative positions of the I3C NMR

resonances of the quaternary carbons in several tetraalkyl naphthalcnones, by changing

the substituents On the quaternary carbon from methyl to ethyl and the carbonyl to

ethylidene. Using the first method, the shift reagent coordinates with electronegative atoms in

the substrate and modifies the magnetic field experienced by the neighbouring protons_

Since the strength of this effect varies with the distance from the reagent, the chemical

shift of each proton is modified to a different amount. Relating the extent of the shift to

the concentrations of the shift reagent and being aware of the fact that the protons closest

to the donor atom (the carbonyl group) are shifted by a larger extent for a given amount

of reagent added, thc proposed structures as in scheme 2.2 is unambiguously confirmed

by these experiments.

The IH NMR Eu(fadh shift experiments, as shown in figure 2.1 for the a,{3"

unsaturated ketones Sa and 9a, indicate that for a given amount of reagent added, both

methyl groups on C1 in ketone 8a are shifted to a larger extent, compared to ketone 9ft in which only one methyl group on C1 exhibits such a large shift. A strong indication is the

difference in shift of the ethyl group: in case of ketone Sa, this group is slightly affected upon addition of the shift reagent, whereas in 9a the ethyl group is shifted even to a

larger extent than the two methyl groups at C4 •

Using the second method in determining the structure of both ketones, we compared

the l3C NMR resonances of the quaternary carbons of 1,4-dihydro-l,1,4,4-tetramethyl-

2,3-naphtha\enedione, 1 ,4-dihydro-l-ethyl-1 ,4,4-trimethyl-2( lH)-naphthalenedione and

3,4-dihydro-l, I ,4,Hetramethyl-(3Z)-ethylidene-2(1H)-naphthalenone17 to the resonances

of the optical pure ketones 8a and 9a (see figure 2.2). Upon replacement of one methyl

34 ASYMMm'RIC SYNTHESIS OF CHIRAL NAf'HTHALENOLS

Llp(Hz) Llr(Hz)

200 Mel a10 Mltl

Me.

Mel

8 1

100 100

40 80 12U 110 200 <II 110 nO 1110 200

10-3 eq Eu(fodhl C4 9a

FIGURE 2. I. Plot.~ of the induced chemical shift fl/!, versu~ the amount of shift reagent added for protons of ketones Sa and 98,

by an ethyl group in the dione derivatives, the Be NMR resonance of the

corresponding quaternary carbon is shifted downfield by 4.0 ppm, from 51.6 to 55.6

ppm. Going from dione to ethylidene group in the tetramethyl derivative its Be NMR

resonance undergoes an up field shift of 8.8 ppm, from 51.6 to 42.8 ppm. These data.

further support the assignment of the structures 8a and 9&, the isomer showing chemical

shifts of 49.2 and 45.8 ppm is proposed to be 8a and the isomer featuring chemical shifts

of 42 and 53.4 ppm represents ketone 9a.

©Q:Me }""e 0

. 0 Me ~t

7

Me tJle o

H

Sa

51.0 (.c.S, ~: Me Me

.. 55.6 (+4.0)

calcd. IOllnd

51.6 - 1.3 - 0.6 = 49.7 49.2

Me 51.6 + 4.0 - B.a = 46.S 45.8

51.6

51.6

Me Me H

Me

9a

50.3 (-1.3)

Me 42.S (..a.8)

caled. found

51.6 - S.B . 0.6 '" 42.2 42.0

51.6 + 4.(1 . 1.3", 54.3 53.4

FIGURE 2.2 Estimated BC chemical sh ifu. of the auatemarv carbons for the tetraalhl nanhthal~nnne rI~_r>v~tivp_~

36 ASYMMETRIC SYNTHESIS OP CHIRAL NAPHTHAJ...JaNOLS

2.2.4 Synthesis and characterisation of the rwphthale1wls

Upon LiAIH4 reduction of 'ketone 8a, the diastereomeric allylic alcohols lOa (2S,4R)

and lla (2R,4R) are formed in a ratio of 70:30. As might be expe.:;:ted, the presence of

the larger ethyl group induces hydride attack to occur more readily from the opposite

side_ Both diastereomers arc dis.tinguished by determination of the configuration on C2.

This has been reali7..ed initially by the assignment of the quaternary carbons C1 and C4

and adjacent substituents in a two-dimensional (2D) NMR INADEQUATE experiment

(see figure 2.3) and a heteroouclear lHJ 13C 2D chemical shift correlation NMR spec·

trum,26 followed by IH NMR Eu(fodh experiments to correlate the shift of the protons

closest to the hydroxyl group and/or to the shifting of the ethyl and methyl substituents on

C4 (sec figure 2.4).

10

ppm

70

CSC~CJ c.' C9 11.1 ___ 1.1

.:'.; "-.. -... -.. == __ .--_. - .. _-_'""_"'-_-. -.. _--- .. _-... -_~.:_-_ .. _-.-.-.---';'-=l>=i="" ... ---...... .

"f---"-r ,_... . ... : _=_ .. ·----==_c .. -----..,)._===;;:==

---- ._--_ ..... _-- -------_ .. ,.----_ .. -._-. .---'-.. "---r" ._-_., .. -.---,----. , '--" ... ---~ .... ' '-._._-_. ----,

120 90 60 30 -_ ........ --_ Dr m

rlGURo 2_3_ Two dimensional INADEQUATE spectrum of lOa showing 13C_IJC connec­

tivitie~. The hod7.0ntal F2-axis represents 13C chemical shifts or coupling~_ The FI-axis

r~'"PTesents 0.55 x I:lC chemical shifts. SW IISW = 3.6, 1" = 6 mS t 256 increments with a

spectral width of 17 kH7.; 32 ~cans per increment.

2.2.4 Sy1lJhesis and characterisation of the nilphthalcnols 37

From the results presented in these figures an unambiguous assignment of the can·

figurations for both diastereomers could be obtained. In the INADEQUATE-2D-COSY­

like symmetry experiment (INADSYM) for lOa the whole skeleton of the molecule could

be traced by determination of the 13C_ 13C connectivities, thereby starting from the clearly

assignable signal of the cthyl group (Cd at 9.8 ppm which is connected to the methylene

at 0.55 X 38 "" 20.9 ppm on the Pl-axis. In figure 2.3 the correlations appear as off­

diagonal cross peaks of the two adjacent carbon atoms as. in COSY. Coordinates of cross

peaks are given in the DC-chemical shifts and J3.CJ)C couplings on the F2 domain

(horizontal axis) and 0.55 x shifts and couplings (SWlISW2 -- 3.6) on the F I domain

(vertical axis) respectively. The horizontal lines are drawn for the purpose of clarity.

They indicate that we are dealing with the coinciding carbon atom.

From figure 2.3 it can be clearly seen that the horizontal line starting from the

(C II ,Cd crosspeak to the (C4,C ll) crosspeak lies exactly at 0.55 x 38.0 ppm on the

vertical axis. This chemical shift is exactly concurring with the methylene carbon Cll .

Thus C II is conneded to e\2 and a quaternary carbon at 42.8 ppm (and therefore denoted

as C4) This quatemary carbon has also a crosspeak at 0.55 x 31.7 ppm, which is

attributed to the adjacent methyl substituent Cn. The other quatemary carbon at 39.6

ppm is correlated (vertical line) to two methyl groups at approximately half shifts ("'" 0.55

X) of 26.0 and 30.0 ppm respectively and therefore it is denoted as C1. In combination

with the other crosspeaks for C1 (C2 and C5) these results of the INADEQUATE

spectrum strongly confirms the skeleton of the molecule.

With a 2D-13Ct 1H correlation eXperiment in combination with selective decoupling

experiments the 13C signals for lOa and 11a are correlated to the proton resonances in the

IH spectrum (sec experimental section for more details).

The next step in the determination of the configurations of lOa and lla, is to

examine the shift of proton resonances from the groups and ethyl groups at CI and C4

upon addition of Eu(fodh. Figure 2.4 clearly shows lhal in case of lOa the methyl signal

(Cd is shifted to much larger extent than that of 11a, where only one methyl (~ or Cs)

exhibits a very large shift. As a consequence, the hydroxyl group in the molecule has to

be attached to the same side as the ethyl group of alcohol lOa and to the opposite side in

11a. Hence, we assign the 2S,4R configuration to alcohol lOa and the 2R,4R configur­

ation to alcohol 11a.

38 ASYMMF-TRIC SYNTHESIS OF CHIRAL NAPHTHALENOLS

.io(ppm) Llo(ppm)

0-05 ICo C!2

0.85

C7~

.0.52 0.52 C710 Cg C1J C8(1

0.39 CIG 0.39

CIV1 CIS

0.28 U8 C,o C12

0.13

o 5 ID 16 20 a 30 !IIi 4D 46 5 10 15 20 25 30 35 oW ..s

10-3 eq Eu(fodhl eq lOa 10.3 eq Eu(fod»)/ eq lla

FIGURJ:;. 2.4 Plots of the induced proton chemical shift .:).6 of naphthalenols lOa and lta,

versus the amount of shift n:agent added.

2.3 Concluding Remarks

It has been demonstrated that the introduction of a stereocentre In a 2-phenyl

alkanoyl chloride for acylation of a branched alkene, followed by a Priedel Craft

cyclization is a significant improvement in the synthesis of tetrahydro-polyalkyl naph­

talenones_ This method allows the preparation of optically pure materials for the purpose

of asymmetric photochemistry and photooxygenation which is described in the next

chapter. The structures are unambiguously determined by m1R spectroscopy_ Especially

2.4 Experimental section 39

with use of the INADEQUATE spectrum the whole skeleton of the target molecules has been traced.

2,4 Experimental Section

Gem:ral procedures

All solvents and commercial reagents were reagent grade and dried with the appropriate drying agents. Argon was dried consecutively over concentrated sulphuric

acid and KOH pellets. Column chromatography was performed using Merck silica gel 60,

230-400 mesh as the stationary phase. Proton and carbon-13 NMR spectra were recorded

on a Broker AM200 or a Broker AM400 spectrometer uSing tetramethylsilane (fMS; 0

ppm) as internal standard. OC analyses were carried out On a Kipp Analytica 8200

instrument with FID detection (25m'" 0.25 mm 10, column type: WCOT fused silica,

stationary phase CP SIL-8 CB, film thickness 0.25 f.l.m). All HPLC separations were run on a system which consists of a LKB 2248 pump a 16 oj< 100 mm Lichrosorb 60 (:)t4m)

column, and a LKB 2510 UV detector (254 nm). Mass spectral data were obtained on a

Hewlett Packard 5970A system by electron ionization at 70 eV.

(-)(R)-2-Melhyl-2-phenyl-butanoic acid (4)

Racemic acid was prepared by thc methylation of 2·phenyl-butane nitrile (95%)

followed by hydrolysis of the 2-methyl-2-phenyl-butane nitrile18,19. To a solution of

diisopropylamine (72 g, 0.71 mol) in 1 L THF at 0 °C and 276 mL 2.5 M BuLi in hexane was added slowly at ·20 Ge, 100 g (0.69 mol) 2-phenyl·butane nitrile. After

stirring for 4 h, Mel (101 g, 0.71 mol) was added rapidly with external cooling (tempera­

ture was kept below 0 ~C). After stirring for another 2 h the reaction mixture was

allowcci to warm to room temperature. A minimum amount of water was added to

dissolve the precipitated salt. The resulting mixture was extracted with diethyl ether. The

organic phase was washed with water, dried, and evaporated to give 104 g of a residual

oil which appeared to be 95 % pure 2-methyl-2,phenyl"butane nitrile.

lH NMR (CDCI3) 5: 7.28-7.45 (5H, m, Ar), 1.95 (2H, m, 8 3), 1.68 (3H, s, Hs), 0.93

(3H, t, HJ.

Subsequently, the crude nitrile mixture was held at reflux with 80 g, (1,43 mOl) of

pOtassium hydroxide and 1 L of diethylene glycol for 24 h. The solution was cooled,

40 ASYMMETIUC SYNTHESIS OF CHIRAL NAPHTHALENOl,S

mixed with three volumes of water, and the resulting mixture was extracted with diethyl

ether. The aqueous phase was acidified with cold concentrated Het; the acid that

separated was extracted with diethyl ether. The ether layer was washed with water, dried

over MgS04 and evaporated to give 117 g (0.66 mol, yield 92%) almost white solid

material. This material was crystallized from pentane at 0 °C to give 104 g of racemic 2-

roethyl-2-phenyl-butanoic acid m.p. 56.0-57.5° (reported 56-58Y) yield 86%.

I H NMR (CDCI) 0: 11.72 (IH, s, HI), 7.20-7.40 (5H, m, Ar), 2.10 (1 H, m, H:\), 1.99

(lH, m, H)), 1.55 (3H, s, Hs), 0.85 (3H, t, A4).

!3C NMR (CDCl) 0: 182.85 (s, C1), 142.79 (s, Ar), 128.35 (2C, d, Ar), 126.84 (d,

Ar), 126.23 (2e, d, Ar), 50.33 (s, Cz), 31.56 (t, C), 26.25 (q, C5), 9.02 (q, C4).

To obtain optically pure material ,20,ZI quinine (832 g, 2.57 mol) and racemic acid

(685 g, 3.84 mol) wcre dissolved in 2500 mL ethanol and 1075 mL water. U]XIn cooling

(4 or:, slowly), the solution deposited the quinine salt (350 g, mp. lt8-121) which after

fivc r!':Crystallizations and conversion to the acid, gave 97.5 g. white pure (-) crystals

with mp 86.2·!1.7.0 °C and [afsO = -29.2° (c 4.9, benzene) and c.c. > 99%. Literature

[1X]Z5o = -29.5° (c 4.8, ben:r.ene).18

(2R)-2-Methy/-2-pheny/-butanoyl chloride (5)

A mixture of 97.5 g (0.55 mol) of 4 and 132 g (1.1l mol) of thionyl chloride was

stilTed for 20 h. The excess of thionyl chloride was evaporated under vacuum. Two

successive portions (25 mL) of anhydrous toluene were added and evaporated under

vacuum. The rcsidue was distilled to give 105 g (95% yield) of acid chloride as a

colourless liquid bp 122-124 °C.

IH NMR (CDCI) 0: 7.20-7.40 (5H, m, Ar), 2.20 (lB, m, H3), 1.99 (lB, m, H3), 1.63

(3H, s, Hs), 0.87 (3H, t, H4).

DC NMR (CDC1) lj: 178.85 (s, C t ), 140.42 (s, Ar), 128.88 (2e, d, At) I 128.21 (d,

Ar), 126.79 (2C, d, Ar), 60.70 (s, C2), 31.49 (t, C3), 22.13 (q, Cs), 8.74 (q, C 4).

(I R)-3.4-Dihydro-I-ethyl-I.4. 4-trimethyl-2(l H)-naphtha/enoM (6).

A three-necked jacketed reaction vessel equipped with a magnetic stirrer, ther­

mometer and a gas inlet tube was charged with butanoyl chloride 5 (33 g, 0.17 mol) and

freshly distilled carbon disuJphide as solvent,24 The mixture was cooled just below -10 °C

by means of a cIo!lCd circuit filled with methanol, and 26 g (0.10 mol) stannic chloride

was added. Subsequently the isobutcnc (excess) was condensed in a cold trap and slowly

passed into the stirred reaction mixture through the gas inlet tube. The progress of the

reaction was followed by taking aliquots periodically for IH NMR measurements. After

2.4 Experimental section 41

stirring for 24 h, when all starting material had disappeared, the reaction mixture was

poured into ice-water. The mixture was extracted with diethyl ether, and the organic layer

washed with 10% sodium hydroxide, then with water and dried with MgS04

monohydrate. Removal of the solvents gavc a residue which was eluted with

hexane/dichloromethane 2: 1 (v/v) over silica. The major fraction containing a mixture

(28.1 g) of three ketones was collected. Column chromatography (silica 60, hexanes/ethyl

acetate 95:5 (v/v» afforded a chloride 6a (45%), an alkene 6b (40%) and the

naphthalenone 6 (Rt=r0.18) respectively (15%).

(5R)-2-chloro-2,5 -dimethyl-5-phenyl-4-heptallone (00)

IH NMR (CDC!3) 5: 7.20-7.46 (5R, m, Ar), 2.75 (2H, AS Jab = 12.7 Hz, H3), 1.98

(2R, ro, Ri)' 1.66 (3R, s, HI or Hg), 1.61 (3H, s, Rs or HI)' 1.45 (3H, s, H9), 0.77 (3H, t, H7).

13C mAR (CDCI3) 208.76 (s, C,J, 142.02 (s, Ar), 128.76 (2C, d, Ar), 126.88 (d, Ar),

126.40 (2C, d, Ar), 67.95 (8, ~, 56.15 (s, C5), 51.47 (I, e3), 31.96 (t, e6), 31.86 (q,

CI Or Cg), 29.66 (q, C8 or e l ), 20.36 (q, e9), 8.48 (q, C7).

2,5-dimelhyl-5-phenyl-hex-l-en.e-4-on.e (6b)

tH NMR (CDCI) 0: 7.20-7.46 (5H, m, Ar), 4.82 (1 H, m, HI, cis or trans), 4.55 (lR,

m, HI, tran8 or cis), 2.90 (2H, A 1a == 12.7 Hz, H3), 2.0 (2R, m, H6), 1.62 (3R, s,

Hs), 1.48 (3R, 8, R9), 0.77 (3R, t, H7).

BC NMR (CDCI]) 209.48 (s, CJ, 142.21 (s, Ar), 139.36 (8, Cz), 128.36 (2C, d, Ar),

126.55 (d, Ar), 126.30 (2C, d, Ar), 114.13 (I, Ct), 55.97 (8, C5), 46.13 (t, C), 29.52

(t, C6), 22.22 (q, C g), 20.16 (q, C9), 8.38 (q, C7).

Compounds 6a and 6b were cyclized to the corresponding naphtha1enone 6 with

boron trifluoride ethyl etherate in dichloromethane.24 Subsequently, all ketones (28.1 g,

,., 0.127 mOl) were dissolvoo in 250 mL dichloromethane to which 135 mL of BFJ -

O(C2H5h was added. After refluxing for three days the mixture was poured into a

mixture of crushed ice-water, and made basic with ammonia. Usual work up and column

chromatography with hexane/ethyl acetate 95:5 (v/v) afforded 17.85 g (ovcrall yield 45%)

of pure 6. Upon addition Of tris[3-[(heptafluoropropyJ)-hydroxy-methylcnc]-d-cam.

phorato]europium(lII) [( + )-Eu(hfch, 9.4 mg, 0.0078 mmol] to a solution of 6 (20 mg,

0.078 mmol) in benzene-d(i, the enantiomeric excess was found to be > 99%, ([af3D ==

-24.2°, c 2, methanol). Upon addition of (+ )·Eu(hfc~ to a solution of racemic material,

the individual methyl signals of the enantiomers could be resolved easily. Elem. Anal.

42 ASYMMETRlC SYNTHESIS OF (,tHRAL NAPHTHALENOLS

Calc. %C 83.28, %B 9.32; Found %C 83.34, %H 9.10

IH NMR (Cl)Cl3) 0: 7.20-7.46 (4H, m, Ar), 2.63 (2H, A J ... = 12.9 Hz, H), 1.98 (2R,

m, H7), 1.42 (3H, s, H9), 1.34 (JH. ~, HIO or Hll ), 1.30 (3R, s, Hll or H IO). 0.67 (3H,

t, H8),

DC NMR (CDCl]) 0: 213.45 (s, C2), 144,35 (s, Cs or C6), 141.50 (s, C6 or C5), 126.90

(d, Ar), 126,81 (d, Ar), 126.41 (d, Ar), 124.61 (d, Ar), 52.49 (l, C~), 52.38 (s, C I),

37,99 (5, C~, 33.96 (t, C7), 31,22 (q, C lO or Ctt ), 30.56 (q, Ctt or CIO), 28.32 (q, C9),

9.87 (q, Cs).

(J R)-J ,4-dihydro-J-ethyl·J ,4, 4-trimethyl-2, 3-naphthalenedione (7)

3 ,4-Dihydro-l·ethyl-1 ,4,4-trimethyl"2(1 H)-naphthaJenone 6 (47.4 g, 0.22 mol) and

selenium dioxide (25.1 g, 0,23 mol) were dissolved in 500 mL of dioxane and the

mixture was heated under reflux for 30 h.24 The solution was filtered over hyflo and the

solvent evaporated. Crystallisation in the cold from pentane and rapid filtration gave a

yellow solid. However on storage at room temperature compound 7 turned out to be a

yellow oil, storage at 4 QC gave yellow crystals.

IH NMR (CDCI) 0: 7.27-7.38 (4H, m, Ar), 1.97 (IR, m, H7), 1.83 (IH, m, H7), 1.57

(3H, s, H 10 or H 11), 1.52 (3H, $, H9), 1.46 (3H, 5, Hll or HlO)' 0.74 (3H, t, Hg).

DC NMR (CDCl) 205.65 (s, C2 or C), 205.44 (5, C) Or CZ)' 141.12 (s, C,S or C6),

139.10 (s, C6 or Cs), 128.06 (d, Ar), 128.03 (d, Ar), 126.53 (d, Ar), 126.35 (d, Ar),

55.57 (5, C I), 50.99 (s, C4), 34.65 (I, C7), 28.03 (q, C9), 25.29 (q, C lO or CII), 23.21

(q, C ll or C to), 9.18 (q, Cs).

(4Rj-3, 4-Dihydro"4-ethyl-1,1, 4-trimethyl- (3Z)-ethyliden.e-2 (I H)-naphthalenQ7/.e (Sa) and

(I R)-3,4-dihydro·l "ethyl-1 ,4,4-trimethyl- (3Zj-ethylidene -2(1 Hj-naphlhalenom~ (9a)

n-Butyllithium (24 mL of a 2.5 M solution in hexane, 0.06 mol) was added dropwise

to a stirred suspension of 21.9 g (0.06 mol) (ethyl)triphenyl-phosphonium bromide in

100 mL of anhydrous diethyl ether, whereupon the deep red colour of the ethylidene­

phosphorane was produced. The mixture was then stirred for three h at room tempera­

ture. At the end of this period, 13,8 g (0.06 mol) of 7 was added dropwise, whereupon a

white precipitate formed. The mixture was then cooled and filtra.ted. The filtralc was

washed with water, the organic layer separated and dried over MgS04. Evaporation of

the ~olvents left a residue which was purified by column chromatography using

hexane/ethyl acetate 95:4 (v/v) as eluent Rt '" 0.21 and 0,18. Separation of both isomers

was carried out with HPLC on silica (5~m) using n-hexane/ethyl acetate/water

99,75:0.25:0.001 as eluent, Yielding 5.6 g of ketone 8a and 6.4 g of ketone 9a. The z·

2.4 Experimental section 43

isomers were separated from the corresponding E-isomers, yield present in less than 5%.

GLC showed the original product mixture to contain ketones 8a (mp 65-67.2 0c) and 9a

(mp 50.1-52.3 ~C) in a ratio of 40:60.

Sa IH NMR (CDCI}) 0: 7.21-7.33 (4H, m, Ar), 5.75 (lH, q, H9), L81 (3H, d, H IO), 1.53

(2H, m, H Il), 1.51 (3R, S, H7 or Hg), 1.48 (3H, 5, Ha or H7), 1.41 OR, s, H IJ), 0.71

(3R, t, Hd. l3C NMR (CDCI3) 209.65 (5, C2), 143.82 (s, C5 or C6), 143.06 (s, C6 or Cs), 142.81

(s, C3), 126.88 (d, Ar), 126.81 (d, Ar), 126.58 (d, Ar), 125.42 (d, Ar), 124.97 (d, ~),

49.18 (5, Cil, 45.81 (s, C1), 37.83 (t, ell)' 29.60 (q, Cn), 25.99 (q, C7 or Cal, 23.07

(q, Cg or C7), 14.96 (q, C9), 8.73 (q, Cu ).

GC/MS: 242.2 (Mi'), 227.1, 215.1, 214.2, 213.2 (base peak), 198.2, 17L3, 170.2,

155.2, 141.2,129.1,115.1,43.1.

9a 'H NMR (CDC!) 5: 7.21-7.40 (4H, m, Ar), 5,81 (IH, q, HLO), 2.15 (1H, m, H7), 1.85

(IR, m, Hi), 1.83 (3H, d, li ll), 1.43 {JR, s, H9), 1.41 (3R, s, HI2 or HlJ), 1.38 (3H,

s, Hn or Rd, 0.65 (3R, t, Hg).

13C NMR (CDC13) 209.65 (s, C:J, 146.01 (5, C5 or C6), 144.26 (s, C6 or Cs), 140.87

(s, C3), 126.78 (d, Ar), 126.74 (d, Ar). 126.70 (d, Ar), 124.45 (d, Ar), 124.27 (d, C/O),

53.44 (5, C1), 42.0 (s, C4), 34.02 (t, e7), 30.53 (q, C12 or Cll), 29.55 (q, Cn or Cd, 27.60 (q, C9), 14.97 (q, CIl)' 9.65 (q. Cg).

(2S,4R)- and (2R,4R)-J ,2,3,4-U!trahydro-4-ethyl-J .1,4-trimethyl-(3Z)-elhylidene-2-naph­

tOOlenol (lOailla)

To a stirred solution of I g (3,4 mmol) naphthalenone 8a in 25 mL anhydrous

diethyl ether was added 0.19 g (3.4 mmol) LiAIH4. After 30 min additional stirring,

respectively 0.5 mL of water, 0.3 mL of a 5N NaOH solution, and 5 mL of water were

added. Filtration. separation of the organic layer, and removal of the solvent afforded

0.94 g of a mixture of lOa (mp 52.0-53.6 QC) and lla (mp 69.9·71.6 0c) in a ratio of

70:30. Separation was accomplished by repeated column chromatography with hexane!

ethyl acetate 96:4 (v/v) as eluent.

(2S.4R) lOa

lR NMR (CDCl3) 0: 7.18-7.38 (4H, m, Ar), 5.72 (lH. q. ~), 4.52 (lH, s, H2), 2.05

44 ASYMMETRIC SyNtHESIS OP CHIRAL NAPflTHALE:NOLS

(lH, m, H II ), 1.86 (3R, d, H w), 1.81 (1 R, m, R lI ), 1.56 OR, s, R7 or Hg), 1.38 (3R,

s, Hu ), 1.07 (3R, s, Rg or H7), 0.48 (3R, t, H I2).

DC NMR (CDC I) b: 143.44 (,s, Cs), 142.92 (s, C6), 140.78 (s, C), 126.77 (d, Ar),

126.50 (d, Ar), 126.06 (d, Ar), 125.92 (d, Ar), 123.29 (d, C9), 73.93 (d, Cz), 42.75 (s,

C4), 39.60 (s, C 1), 37.95 (t, Cll)' 31.74 (q, Cn), 29.99 (q, C7 or Cs), 15.97 (q, Cg or

C7), 13.12 (q, C1Q), 9.80 (q, Cn>.

GUMS; 244.3 (M+), 226.2, 215.2, 197.2 (base peak), 182.15, 171.1, 167.15, 157.15,

156.15,155.1,141.05,128.1,115,1,91.15,43.1,32,0,29.1.

(2S,4R) 110.

lH NMR (CDCI) &: 7.12-7.38 (4H, m, Ar), 5.72 OH, q, .H9), 4.58 (lH, s, H1), 2.15

(IH, rn, Hll ), 1.82 (lH, rn, H II ), 1.81 (3R, d, RlQ), 1.55 (3R, s, H 13), 1,48 (3R, S, H7

Or Hg), 1,08 (3R, s, Hg or H7), 0.48 (3H, t, Hd.

l:lC NMR (CDCI) li: 142.60 (s, Cs), 141.89 (s, C6), 140.95 (s, C3), 126.50 (d, Ar),

125.88 (d, Ar), 126.06 (d, Ar), 125.92 (d, Af), 121.00 (d, C9), 74.44 (d, Cz), 43.51 (s,

C4), 39.60 (s, CI), 37.54 (q, Cd, 35.84 (t, ell)' 29.64 (q, C7 or Cs), 26.63 (q, Cs or

C7), 13.14 (q, C lO), 9,65 (q, Cu).

GClMS: 244.3 (M+), 226.2, 215,2, 197.2 (base peak), 186.1, 182.2, 171.2, 157.15,

155.1,128.0,115.1,91.1,77.1,55.0,43.1,41.1,29.1,28.1.

References

1. 8. S. Mulliken, Phys. Rev. 41, 751 (1932).

2. (a) L. Salem, Science 191, 822, (1976). (b) W. G. Dauben, L. Salem and N. J. Turro, Acc. Chern. Res. 8, 781 (1969).

3. 1. Michl and V. Bonaeic-Kouteckj, In Electronic aspects of organic photochemistry, John Wiley & Sons, Inc., New York, 1990, chapter 4 and references therein.

4. (a) J. A. Marshall, Science 170, 137 (1970). (b) 1. A. Marshall, Acc. Chern. Res. 2t 33 (1973). (c) J. A. Marshall and H. FaubJ, J. Am. Chem. Soc. 92,948 (1970).

5. (a) P. 1. Kropp, Pure Appl. Chern. 24, 585 (1970). (b) P. 1. Kropp and H. J. Kraus, J. Am. Chern. Soc. 89, 5199 (1967).

6. P. Wan, M. 1. Davis and M.A. Teo, J. Org. Chern. 54, 1354 (1989).

References 45

7. F. G. West, P. V. Fisher and C. A. Willoughby, J. Org. Chern. 55, 5936 (1990).

8. F. W. A. M. Mie.fen, I-J. C. M. Baeten, H. A. Langetmans, L H. Koole, and H A. Claessens, Can. 1. Chern. 69, 1554 (1991).

9. (a) P. J. Kropp and H J. Kraus, J. Am. Chern. Soc. 91, 7466 (1969). (b) R. T. Arnold and J. F. Dowdall, J. Am. Chern. Soc. 70, 2590 (1948).

to. (a) H. R. Fransen and H. M. Buck, 1. Chern. Soc., Chern Commun. 786 (1982). (b) H R. Fransen, G. J. M Dormans, G. J. Bezemer and H. M. Buck, Reel. Trav. Chim. Pays-Bas 103, 115 (1984).

11. W. J. G. M. Peijnenburg and H M. Buck, Tetrahedron 44, 4821 (1988).

12. I should be noted that a racemic mixture of 2R,4R and 2S,4S naphthalenols is denoted as la; a racemic mixture of 2S,4R and 2R,4S as Ib; the optical pure analogues i.e. 2S,4R as lOa and 2R,4R as lla. The configurations and systematic names are corrected with respect to the original article according to the IUPAC convention. However in order to obtain as much consistency as possible, the numbers in the NMR data were assigned as in Scheme 2. I.

13. J. R. Scluiffer, K. S. Bhandari, R. E. Gayler and R. A. Wostrailowski, J. Am. Chern. Soc. 97, 2178 (1974).

14. A. M. Brouwer, L. Beumer, 1. Cornelisse and H. J. C. Jacobs, Reel. Trav. Chim. Pays-Bas 106, 613 (1987).

15. P. J. Wagner, S. Zhou, T. Hasegawa and D. L. Ward, 1. Am. Chem. Soc. 113, 9640 (1991).

16. S. L. Murov, In Handbook of photochemistry, Marcel Dekker, New York, 1973.

17. W 1. O. M. Peijnenburg and H. M. Buck~ Tetrahedron 44,4927 (1988).

18. D. J. Cram and J. Allinger, J. Am. Chem. Soc. 76,4516 (1954).

19. W. Kinnse and P. Feyen, Chcm. Rer. 108, 71 (1975).

20. D. J. Cram and J. D. Knight, I. Am. Chern. Soc. 74, 5835 (1952).

21. C. Riidwrdt and H. Trautwein, Chern. Ber. 98, 2478 (1965).

22. D. 1. Cram, K. R. Kopecky, F. Hauck and A. Langemann, J. Am. Chern. Soc. 81, 5754 (1959).

23. (a) 11. A. BrusOl!, F. W. Granl and E. Bobko, 1. Am. Chern. Soc. 80, 3633 (1958). (b) A. S. Medvedeva, L P. So.frol!ova, I. D. Ko.likhn/an and V. M. l'lasov, Fzv. Akad. Nauk. Ser. Khim. no. 5, 1175 (1975).

46 ASYMMETRlC SYNTI-lESIS of CHIRAL NAPHTHAL£NOLS

24. L R. C. Barclay, K. L. Adams, H. M. FOOle, E. C. Sanford and R. H. YOII-ng, Can. 1. Chern. 48, 2763 (1970).

25. (a) B. Vonach and G. Scht)mhurg, J. Chrornatogr. 149,417 (1978). (b) R. R Heath, -'- H. Tumlinson and R. E. Doolittle, J. Chromatogr. Sci., 15, 10 (1977). (c) O. K. Gulla and 1- Janak, J. Chromatogr. 68, 325 (1972).

26. A. H. /Jerom(~, In Modern NMR techniques for chemistry research., 2nd cd., Pergamon, Oxford, (t 987).

3 Abstract

Photochemistry and Photooxygenation of

(2S,4R)- and (2R,4R)-1;2,3,4-tetrahydro-

4-ethyl-1 ,1 ,4-trimethyl-(3Z)-ethylidene-

2-naphthaleno!sH

The photochemistry and photooxyge.nation of optically pure (2S,4R)- and (2R,4R)-1,2,3,4-

tetrahydro-4·ethyl-l, 1,4-trimethyl-(3Z)-ethylidene-2-naphthalenols lOa and lla are described.

Irradiation of both alcohols gives ri~e to E-Z isomerization under inert conditions, followed by

photodegradation reactions, leading to the formation ofindane 12 by the elimination of foonalde­

hyde. Irradiation of oxygenated solutions of lOa reveals a rapid photoaddition of oxygen to the

double bond, leading to the gtereo- and regioselective formation of hydroperoxide 15 and epoxy

alcohol 17a, which hydrolyses on standing to triol 18a. The observed stereochemistry is explained

by the steroosele<::tive coordination of the hydroxyl group at the chiral a1lylic ~ite with the. attack­

ing dectrophilic oxygen. In the case of l1a a mMe complicated reaction mixture has been ob­

tained, in which analogously the epoxy alcohol 17b and triol ISb were present. Semiempirical

MNDO, AMI and PM3 calculations reveal that in the preferential ground state conformation, the

hydroxyl group occupies an orthogonal position with respect to the exocyclic double bond.

Fluorescence measurements point out that the reactive state in the photochemistry of the.~e naph­

thalenol systems in inert conditions i~ the triplet state. In the photoreactions of optically pure

compounds lOa and lla, we have in fact been unable to identify products analogous to 2, claimed

to originate from the racemic precurSors 1 via a planar [I ,3]-OH shift from a singlet (in the

double bond) excited state. Clean photochemical reactions are observed only when loa and lla

are irradiated in oxygenated solutions, resulting in stereo- and regiospecific photoproducts.

'F,W.A.M. Miescn, I.LJ. van DoDgen and E.W. Meijer, pl>.rt of a paper, accepted for pu!)he.1ltlon ill Red. Trav. Chim. Pays-Bas.

47

48 PHOTOCHEMISTRY AND PHOTOOXYGi;NATION

3. 1 Introduction

Photochemical reactions are often controlled by the conformation of the reactants

the ground state. I ,2,) Especially when restrictions of rotation in intermediates do occur,

this conformational control leads to differentia.tion in photoreactivlty. For example io the

photochemistry of tetrahydro-l,4-naphthoquinone ring systems, the -y-hydrogen abstrac­

timi by the carbonyl oxygen is controlled by the twist conformation in the ground state. I

This gives rise to the formation of blradical intermediates with the same basic shape. In

these systems conformational isomerism during the excited state and biradical lifetime is

restricted by the bulky bridgehead groups and the biradical intermediate is frozen in the

conformation from which only one enone alcohol product is possible. I

As a matter of fact, the a.llylic photochemical rearrangement observed in the 2-

propenol unit of 8-hydroxygcrmacrene B (germacrol, 3), is considcred to be controlled by

the ground state. 4 X-ray analysis of the germacrene-silver adduct in combination with

MNDO calculations revealed that in the preferential ground-state conformation the

hydroxyl group is situated almost in the plane of the exocyc1ic double bond. As a re .... ult,

theoretical considerations based on the sudden polarization model,5 propose that an

efficient reaction could take place only when the molecule adopts this orientation.6 In this

hv 85%

5%

Me

hv Me{H) ---X ..... 15%

H{Me)

Z I (E) -1b 2R,4R (+ 2S,4$)10

SCHEMI> 3.1

2a 4R,9S (+ 4S,9R)

Me

Me -Et Me 2b

4R,9R (+ 4S,9S)

3.1 Introduction 49

model, the hydroxyl group is assumed to migrate in the plane of the three carbon atoms

of the propenol subunit. The stereospecific photochemical formation of product 2a depicted in scheme 3_1 for the racemic mixture of (2R,4R)- and (2S,4S)-1,2,3,4-

tetrahydro-4-ethyl-1,1 ,4-trimethyl-(3Z)-cthylidene-2-naphthaleno!s Z-la, was explained in

tenns of sudden polarization.7 For the diastcreomeric alcohols lb, that failed to give thc

allylic rearrangement, it was reported that according to MNDO calculations, these

isomers adopt an orthogonal orientation of the hydroxyl group with respect to the

propenol carbon atoms, being unfavourable for such a reaction. It has been postulated that

the steric intern.ctions dictated by the stereocentres leads to preferential twisting polariled

singlet excited diene mOiety and consequently to differentiation of reactivity_7

It should be noted, however, that direct excitation of the olefinic chromophore in

la.b into its S I state, must be considered unlikely due to the fact that simple alkenes

exhibit poor abSOlJltion in the spectral regions accessible to the photochemical equipment

employed. Actually, the alkyl benzene chromophore in la,b might be expected to be

primarily excited. Such an aromatic chromophore is known to be able to sensitize the

triplet state of alkyl ethylenes (Rr = 84 kcal/mol for benzene, 81 kcallmol for

tetrahydronaphthalene, and 79 kcal/mol for ethene),8 which would as a consequence

rather produce the triplet than the singlet excited state of the exocyclic double bond in

la.b_ In spite of the decreasing photoreactivity of b,b in more polar non-hydroxylic

solvents,9 an efficient reaction was found to take place in methanol for germacrol 3.

Moreover, upon UV-irradiation of la,b addition of methanol concomitant with elimina-

3

hv -ooe 4a

+

SCHEME 3_2

+

OH OH

4b

OH

4c

50 PHOTOCIIEMISTRY AND PHOTOOXYGENAY10N

tion of H20 was claimed to occur. The conflicting experimental results of 1a compared to

Ib and gennacrol 3 (vide infra, subsection 3.1.1) and increasing mechanistical interest in

stereochemistry combined with photochemistry of hydroxylic bichromophoric compounds,

has motivated us to reinvestigate the photochemistry of the 2-naphthalenol compounds.

Together with the fact that quenching with oxygen might lead to reactions exhibiting a

pronounced regio- and stereospecificity, and that the nature of the processes whereby

electronically excited singlet states interact with ground-state o)(ygen might give informa­

tion considering thc products or intermediates involved in the photochemistry of 2-naph­

thalenols, we present a study of the photochemistry of the opticaUy active starting

materials of la and lb (referred to as 11a and lOa respectivcly)IO in various solvents as

well as in the presence of oxygen.

3.1.1 Brief review on the photochemistry of allyUc a/(;ohois

A remarkable photochemical rearrangement has been claimed to occur in the 2·

propenol units of both 8-hydroxygermacrene »4 and 1 ,2 ,3,4-tetrahydro-4-ethy 1-1 ,1,4-

trimethyl-(3Z)-alkyl idenc-2-naphthalenol compounds.7 Irradiation of 8-hydrox ygermacrcne

(germacrol, 3) in methanol resulted in the formation of three major compounds, in which

the occurrence of the allylic rearrangement was demonstrated by NMR spectroscopy. In

the lH NM:R spectrum of 4a,b a single triplet signal was observed at 5.8 and 5.3 ppm

respectively representing an olefinic proton adjacent to a methylene group. In combination

with the resonances in the 13C NMR spectrum, these data are in line with the occurrence

of the allylic rearrangement for 3 as depicted in Scheme 3.2. In the photochemistry of

germacrol, the endocyclic diene moiety is a prerequisite for the allylic rearrangement,

since any small modification in position or removal of onC of the double bonds, did not

give the corresponding reaction.

The mechanism of this photochemical rearrangement has been studied by means of

semi-empirical MNDO calculations, followed by a limited (3 X 3) CI. 6 The calculations

involve the separation of charge upon excitation of the double bond. This phenomenon

known as 'Sudden polari:rntion" was first described by Salem & Coworkers.5 In the first

e)(cited state of electronically excited ethylene, both SI and T I undergo a twisting motion.

However, as in the S 1 state of ethylene the angle of twist approaches 900, a very weak

polarizing perturbation such as a nearby charge, causes a complete separation of charge

and results in zwitterionic states (Z[ and Z2). Either of these species is anticipated to

undergo ionic additions in competition of other polar reactions of S[ and Z. It should be

3.1.1 Photochemistry of allylic alco/z()ls 51

noted however that these primary reactions of these states will always be competitive with

deactivation pathways to the ground state. For the lowest excited singlet state of 2-

propenol the polarization leads to a negative charge on the ccntnll carbon, and a positive

charge On the terminal carbon of the double bond. The OH-functionality which bears a

small negative charge may now shift towards the positively charged terminus in the plane

of the double bond. After radiationless transition from a second twisted conformation the

reaction proceeds on the ground-state potential surface, towards the shifted product In the

calculations for this "planar [1,3]-OH shift, the activation enthalpy is only 17.1 kcaJlmol

and therefore considered as the preferential pathway. In a suprafacial mechanism, con­

fonn the symmetry rules for pericyclic reactions, the calculated activation enthalpy was

about 56.8 kcaJ/moL /I

TS H

H~ H

I

?:; .. ... '(r' I

H ~ H . ", "'H

H H H

~~ H:(o - 0 .. H ....... H H ., ..... H

H -'H ''H H

FIOURE 3.1 The propose<! mechanism for the rearrangement in the 2-propen-I-01 unit. involv­

ing sudden pOlarization.

To find experimental support for the theoretical considerations for the 2-propenol

systems, a model compound was developed, in which the stereochemical outcome would

give an indication of the preferred reaction pathway. Upon UV-irradiation of a mcemic

mixture of (2R,4R)- and (2S,4S)-1 ,2,3,4-tetrahydro-4-ethyl-1, I ,4-trimethyl-(3Z)­

ethylidene-2-naphthalenol, la, not only a fast E-Z isomerization has been found to take

place yielding a photostationary state for E:Z of approximately 1: 1, but also a slow

stereospecific fonnation in low yields, of photoproduct la is reported.7 Based on the

theoretical data, it was suggested that the photoreactivity was controlled by the preferen-

52 PHOTOCHEMISTRY AND PHOTOOXYGI>NATION

tial ground-state conformation, since the other stereoisomers Ib (a racemic mixture of

2S,4R and 2R,4S) did not give 2. It waS postulated that the sterle interactions dictated by

the stereocenters lead to a differentiation of reactivity in a twisting polarized singlet

alkene mOiety. According to the theoretical considerations on 2-propenol, and germacrol,

a planar migration of the zwitterionic alkene moiety accounted well for the observed,

stereospecificity in the photore.activity. As a result of the proposed planar mechanism, the

absence of photoproducts like 2 for Ib was explained by the fact that in the calculated

preferential ground-state conformation of lb, the hydroxyl group waS located in an

orthogonal position with respect to the plane of the exocyclic double bond. A suprafacial

mechanism was rejected on the bas.is of the observed stereQChemistry. In a suprafacial

m~hanism the stereoselectivity would be. reversed due to allylic strain.7

In order to find experimental support for the occurrence of s.udden polarization, tJV­

irradiations for the tetramethyl derivatives were conducted in more polar solvents. 9 The

decreasing reactivity in the more polar solvent, indicated indeed that solvatatjon inhibited

the progress of the reaction. Also other derivatives, in which the alkyl substituent on the

exocyclic double bond was replaced by a phenyl group, have been studied. However this

caused a radical change in the photochemical behaviour. 9

3. ].2 Interaction of oxygen with ollJins

The interaction of ground-state oxygen eBg) with electronically excited singlet and

triplet states has been a subject of considerable interest in many fields of research.!!

Molecular oxygen, plays an important role in many photochemical processes. Because of

its unique properties, conSidering reactivity, its low lying excited state and high energy

content molecular oxygen in either physical or chemical processes may be accompanied

with the formation of an even more reactive intermediate: singlet oxygen, denoted as

Ole fig). The mechanism of the interaction of electronically excited singlet (1M·) and

triplet eM) states with oxygen are summarized in the equations below. 12 Formation of

02(1.:1g) via the singlet manifold is dependent on the SI-TI energy gap of 22 kcall mol , the

0,0 excitation energy of 02(lfig).

1 M* + OzeEg) .....

1M" + 02eEg) -+

3M· + 0ze A$)

3M· + 02eEg)

Because this splitting is often insufficient and lifetimes of S I states are short, OZ(! ag)

formation is often restricted to the triplet manifold, where only the singlet encounter

3J2 Interaction of oxygen with Qle./ins 53

complex, leads to formation of O2<' .ig). This mechanism operates in the photochemical generation of singlet oxygen via photoscnsitation of dyes; this feature is discussed in more

detail in chapter 5.)

3M • + 02eEg) P:

3M* + 02eEg) P

3M'" + 02eE8) P

l[M"'OV+ - 1M + 02e.l.&)

3[M·"02]* .... 1M + 02eEg)

5[M···OzJ~

The interaction of singlet oxygen with olefins may be physical and/or chemical in

character. As a consequence of the low excitation energy of Oze.6.g), physical quenching

via triplet-energy transfer of singlet oxygen to olefins is restricted only to a smaller

number of highly conjugated molecules_ In this physical quenching, deactivation of singlet

oxygen to the ground state is catalyzed only when triplet energies lie below that of the

excited state of oxygen. Hence an alternative, but obviously slower, pathway, via a

chemical reaction is often more readily to proceed. ll ,13

Depending on the fact that monoolefins contain allylic hydrogens, monoolefins react

with singlet oxygen, to give either dioxetanes 6 or allylic hydmpefOxides 7 (scheme 3.3)_1l The overwhelming literature on the mechanism in recent years of both types of

reaction indicates that the controvers.ial discussions on this subject go on and that the

mechanism is still not exactly known. Initially, reactions were considered to proceed

through a concerted pathway, nowadays a consensus for a stepwise mechanism has

developed.14 In the stepwise model these reactions, the ene reaction in particular, pro­

ceeds through an intermediate. Several charge transfer intermediates,15 as well as biradi­

cals16 or even exciplexes17 have been put forward to explain the results for the substrates

studied. It is in the secondary process that intennolecular reactions take place yielding the

appropriate allylic hydroperoxide or dioxetane. In the concerted mechanism the

ch:uacteristic bond shifts take place through cyclic tmnsition state Sa. In the case of a

lwitterionic (5c) or biradica1 intermediate (5e,f), formation of the C-O bond (scheme 3.3)

is the first event.

In the other intermediate, the perepoxide (3d), the olefin forms two bonds with the

oxygen, and has been considered for a long time as the most important transition state, 13

accounting for many experimental results. However recent data of rate and activation

parameters, together with the results of Schuster et al. 17•18 lead to the suggestion that a

reversible exciplex (5b) is formed. As a result, the reactions of singlet oxygen are now

considered to proceed via the mechanism as depicted in scheme 3.3 for quencher Q.

However, the products may be still the intermediates SC-C. As a matter of fact both

54 PI-IOTOCHEMISTRY AND PHOTOOXYGENATION

perepoxide 5d and reversible exciplex 5b are assumed to have similar geometries.)4

5e

SCHEME 3.3

Q + 02eAg) :;::::! [Q"'02eAg)] ~ products .

.£. ISC

Q + 02eL~) ...... [Q .. 'OzeLg)]

5f 7

Since we are interested in the regiosclectivy and stereochemistry of allylic photooxy­

gcnation of the double bonds for the 2-naphthalenols, we briefly mention here the aspects

of the regioselective addition of singlet oxygen to alkenes. Three pronounced effeds of

regioselcctivily have been observed: the cis or syn effect,19 the gem effect, 20 and the

large group effecl. Z1 ,22 In the, by now, well-documented ci~-effect there is a strong

preference for hydrogen abstraction on the higher substituted side of trisubstituted olefins.

In the gem effect electron-withdrawing groups in either vinylic or allylic relationship, or

e.g OI',j3'-unsaturated ketones, increase the tendency for geminal hydrogen abstraction. This

effect has been recognized also for non bonding large groups, where regioselective double

bond formation occurS at the substituent with the (larger) group at the same carbon of the

olefin. Although these factors were rationalized in terms of rotational barriers,20b it was

pointed out later, that barriers to rotation do not dictate regiosclectivity, but regioselective

cne product distribution depends on the free energy difference of the isomeric transition

states.:22 It is also generally accepted that methyl and methylene hydrogens arc reactive,

and that isopropyl C-H and certain conformationally inaccessible hydrogens are nOt. 23

A unique coordinating cis effect of the hydroxyl group dictating regiosclectivity and

diastereoseIectivity was recently reported for flexible chiraI allylic alcohol derivatives by

3.1.3 Formation of epoxides in pht)tooxygenation 55

Adam el al.24 ,25 The ene reactions of the chiral alcohols as e.g. for 8 proceeds in high

regio- and stereoselectivity (see scheme 3.4).

The results are explained by the energy differences of the preferred conformations of

the allylic alcohols in their transition states (cf. Sa). Provided coordination between the

incoming enophile 102 and the allylic hydroxyl functionality of the chiral alcohol is

efficient, allylic sterle interactions dictate the reaction pathway as shown in scheme 3.4.

This explanation is supported by the fact that suppression of hydrogen bonding, by polar

solvents or functionaIization leads to decreased diastereoselectivity. For the acetyl deriva­

tives the diastereoseIection compared to 8 was even inverse, in this case the classical cis

effect operates. It was reported also that the C-C-C-O dihedral angle of the allylic

alcohol in the transition state of this reaction was in the order of c.a. 90-130°.24.25

t OH

- HaC~ 1-00"" CHa

HsC "'=::::,

C~ 8 8a 9

SCHEME 3.4

3.1.3 Fonnation of epoxides in photooxygenation

In cases of adamantalideneadamantane, 7,7-bisnorbomilidene and 2-norbornene it is

observed that dioxetane fonnation is accompanied by epoxidation. 26•27 Initially. cpoxide

formation has been considered to proceed through a perepoxide intennediate. Since no

appreciable oxidation of the solvent was observed, removal of the distal oxygen atom by

such a solvent or other reagent, indicated that the epoxide could not taken as a proof for

the perepoxide intennediate. However mOre recent investigations indicate that in epoxidc

fonnation a superoldde radical anion is involved. Epoxide formation strongly depends on

the sensitizer that is used and to a lesser extent on the solvent. Solvents in which the

lifetime of singlet oxygen is longer, ought to favour mOre epoxide. Depending on the

nature and the amount of the sensitizer strained olefins in dye sensitized oxygenation give

56 PHOTOCHIlMISTRY AND I'HOTOOXYGENATION

either dioxctanes or epoxides in high yields. Epoxidation is suppressed completely by

adding a radical scavenger and greater amounts of sensitizer (leading to the dimcr of Rose

Bengal) results in increased amounts of epoxidc_ Also in the DCA (dicyanoanthracenc)

sensitized photooxygcnation of organic compounds epoxides are formed. 28,29 In these

DCA sensitized photooxygenation reactions two competing mechanisms ("Type I" or

"Type W) are possible. While the type I mechanism follows an electron transfer path.

way, yielding variable amounts of epoxide, the type II mechanism, predominandy yields

typical singlet oxygen derived products like dioxetanes and hydroperoxides_ For TS (trans

stilbene), both radicaJ-cation (TS"+) and radical anion nCA·- have been observed by

ESR. The subsequent reactions of 302 with the radical ions of DCA and Rose Bengal lead

via formation of a superoxidc radical anion (which may be converted into hydrogen

peroxide) to benzaldehyde and variable amounts of TS-oxide_ As a matter of fact in the

initial exciplex with triplet oxygen, electron rich oleflns may act as an electron donor to

produce an olefinic radical cation and a superoxide radical anion. The subsequent oxida"

tion of the olefin with the supcroxide radical anion may then result in the formation of

epoxides (radical autoxidation)_

These results indicate, that quenching with oxygen may lead to epoxidalion of thc

olefin via a superoxide radical anion °2" -. It should be noted that the electron transfcr

mechanism operates es~;ally in polar solvents, due to solvatation of the ionic inter­

mediates, although for Rose Bengal, epoxides are formed in apolar solvents as well, while

in non-polar solvents generally the singlet oxygen mechanism operates.

3.2 Photochemistry

3.2.1 Re.mlts photochemistry oJ 2-naphtha!efl()ls

In order to reconsider the mechanism of the photochemical rearrangements for the

racemic naphthalenoi compounds la and lb, we reinvestigated the photochemical

behaviour of these compounds, by using the optically pure enantiomers lOa and lla as

starting materials, Irradiations have been conducted in inert medium as well as in media

potentially capable of entering into a (photo)rcaction with the substrates_ Inert conditions

are approached by purging the solution continuously with a stream of dry argon, and

using n-hexane (spectroscopic grade) as solvent Photolysis experiments have been carried

out in (air saturated) protic solvents (2 % H20 in CH)CN or MeOH) to study the possible

3.2.1 Results photochemistry of 2-nIlphtha!ello!s 51

photoaddition of oxygen (see subsection 3.3), water or methanol to the double bond and the subsequent behaviour. Studies. have been performed by photolyzing 4*10-3 M sol­

utions of the substrates in 500 mL of solvent. Kinetics of formation and disappearance of materials have been monitored by GC analysis.

6 B ----... Ilmin)

1= 5mi n

FIGURE 3.2. Stack plot of GC-chrornatograms of aliquots taken from the irradiation miJl;ture

of lOa (Z) after photolysis times of 1, .s and 10 minutes and two hours, showing the Z/E

isomerization.

In inert medium, UV -irradiation of alcohol lOa results primarily in a fast E-Z

isomerization around the double bond. Prolonged irradiation leads to formation of (Z)­

and (E)-I-etbyl-l,3,3-trimethyl-2-ethylidene-indanes 128 and 12b besides in low yields

(Z)- and (E)-ketones 13a and 13b (scheme 3,5).

As can be dearly seen from the chromatograms (GC) in figure 3.2, the 1:} photo­

stationary state for Z- and E-isomerS of both alcohols is. reached in about 10 minutes.

Evaluation of the data at longer photolysis times (table 3.2, experimental section) shows that prolonged irradiation leads to the almost complete photodecomposition of both E- and Z-isomers, affording a complicated reaction mixture. Despite repeated column chromato­graphy, and HPLC, only mixtures of photoproducts could be isolated in low yields, together with an unidentified polar polymer with an elemental compOsition of ap-

58 PHOrc>CHEMISTR,Y AND PHUfOOXYGENATION

prox.imately -C4H60- as one of thc main products. The combination of IH NMR and

GC/MS analysis of the major fractions has resulted in the characterization of E- and 1.­

ketones 13a,b and the U- and Z- indanes 12a and 12b. It should be noted that alcohol

lla shows a similar photochemical behaviour as lOa. No indication of the allylic rear­

rangement in the 2-propenoJ unit, as suggested in scheme 3.1, for the 2-naphthalenol

enantiomer I la has been obtained. Again IH NMR spectra and GC/MS analys.is indicated

a fast E-Z isomerization as the primary and major photoprocess, and also prolonged

irradiation ultimately led to the previously mentioned degradation products and the in­

danes by loss of formaldehyde.

~~> ~MO Me Et H

10a (or 11 a) Z.2S,4R (or Z-2R,4R)

:,11 ~> ~H

Me Et Me

10b (or 11b) E-2S,4R (01 E-2R.4R)

hv

slow

SCHEME 3.5

Me ~e

~• Me(H) o -~ H(Me)

Me Et

12a (+12.b)

Me Me

*MO(Hl Me Et H(Me)

13a (+13b)

Since photoaddition of oxygen and subsequent photochemistry turned out to be selec­

tive we have studied the photochemical behaviour of both alcohols lOa and Ila in the

protie solvents methanol and water and in inert conditions as well. However, irradiation

of lOa and lla in methanol does not result in photoaddition to the exocyclic double bond

of either the E- or Z-isomer, but again (after longer photolysis times, 8 h..) an unidenti-

3.2.2 Fluorescence studies 59

tied polymer has been found as the main substance. The photochemical behaviour observed in n-hexane does not appear in methanol. Photoaddition of water to the ex­

ocyclic double bond has been purs.ued, by dissolving both starting materials in CH]CN as oo-solvent. In 100% CHleN, only B-Z isomerization has been observed, secondary

photochemical processes. are very slow. Upon addition of 2% (10 mL, 0,5 mol) water to the photostationary mixture (1: 1) again no photoaddition to the exocyclic double bond has

been observed.

3.2.2 Fluorescence studies

In order to expand our insight in the mechanistic details of the photochemical behavi· our of the 2-naphthalenol derivatives we have examined whether reactions occur from the

singlet (SI) Or triplet (f 1) excited state. To decide whether S I or T 1 is the reactive state,

we: measured the UV -absorption and fluorescence of naphthalenol lOa and reference

alcohol 14, in which the exocyciic double bond is absent (ketone Ii is reduced to the cor­responding alcohol 14 viz. figure 3.3b). As can be seen from figure 3.3a the UV-absorp-

0.19

0,12

230 250 0170

... - .... - ... - .... _-

290 310 330 ... '-.(nm)

3.4 E '5

-

~OH ~·'H

MEi"Me 14

335 355 '-. (n Ill!

FIGURE 3.3A. UV-absorption sp~tta of lOa (dotted line,~) and refere~ce compound 14 (solid

lines) in n-hexane with >mu = 264 nin. FIGURE 3.3B. Normali~ed fluorescence spectra of

both compounds.

60 PHo·rOCHEMJSTRY AND PHOTOOXYGENATION

tion spectra of lOa and 14 are identical in shape in the spectral region from 230 to 300

nm. Both spectra have their maximum intensity at "m<tt = 264 nm. The normalized

fluorescence spectra in figure 3:3b show that (also at 77K) both compounds emit light of

the same wavelength with almost equal quantum yield. In fact the fluore5cence of lOa is

even slightly more intense than that of 14, sec figure 3.3b. Normalization was performed

by dividing the fluorescence intensities by the intensities in the absorption spectra at A,nax = 264 nm.

3.2.3 Conji)rmationai studies

The conformational properties of lOa,b and lla,b have been investigated on the

basis of three semiempirical MO calculations: MNDO,30 AMI]I and PM3.32 In these

compounds the exocyclic double bond, can adopt two possible orientations, one in which

the hydrox.yl group is situated in the plane of the exocyclic doubJe bond and one in which

the hydrox.yl group is oriented in an orthogonal position with respect to this. plane. In the

starting geometries the dihedral angles are set to 0° for the planar naphthalenol

conformers and 9(1' for the orthogonal conformers. Em:rgy has then been minimi7ru by

means of MM calculations with respect to all internal coordinates.

TABLI;'. 3.1 Heats of formation (fl.H r ) and r~lative populations (at ° ('C) of all stable conformers of

Z- and E-isomf.)fS of (2S,4R)-IOa,b and (2R,4R)-lIa,b. =

Heats of fotmation Relative population Naphthalenol aUf (kcal mol· l ) (%)

conformer MNDO AMl PM3 MNDO AMI PM3

lOa Z-2S,4R (orthog) -4.8 -35.5 -41.5 100 99.7 100

Z-2S,4R (planar) -0.2 -32.3 -36.5 0,0 0.3 0.0

Iia Z-2R,4R (orthog) -4.4 -36.1 -42.2 99.7 99.2 100

Z-2R,4R (planar) -1.2 -33.5 -36.3 0.3 0.8 0.0

lOb E-2S,4R (orthog) -1.8 -33.3 41.7 99.8 96.5 99.5

E-2S,4R (planar) +1.7 -31.5 -38.1 0.2 3.5 0.5

lib E-2R,4R (orthog) -l,2 -33.6 -42.1 95.8 96,5 100

E-2R,4R (planar) +0.5 -31.8 -37,7 4.2 3.5 0.0

3.2.4 Disc/1.Jsion photochemistry of 2-TUlphthaleno/s 61

The geometries of 2S,4R and 2R,4R are then optimized further using the MNDO, AMI

and PM3 Hamiltonians of the MOPAC program. lO•n The relative populations of all

stable confonners of lOatb and lla,b have bccn calculated from the Boltzmann equation

based on heats of fonnation at 00 only.

Table 3.1 shows that all methods of calculation for all conformers lead to the con­

clusion, that for diastereomers lOath and lla,b the preferential confonnation is the ortho­

gonal orientation and the absolute dihedral angle C(9)-C(3)-C(2)-O(2) ranges from 96 to

110". Although local minima for the planar geometries, where the absolute dihedral angle

C(9)-C(3)-C(2)-O(2) ranges from 6 to 12", are found, the planar conformation is un­

favourable in view of the very low populations calculated by MNDO, AMi and PM3.

3.2.4 Discussion phOtochemistry of 2-naphthaJenols

When using inert conditions, both optically pure compounds lOa and Ha, gave a

complex. mixture of photoproducts upon long inadiation times. Only E- and Z­

isomerization and photodcgradation reactions were observed, resulting in the formation of

(E)- and (Z)-I-cthyl-i ,3,3-trimethyl-2-ethylidene-indanes l2at b. Since the weakest bond

Me Me

12c

SCHEME 3.6

could well be the C(1)-C(2) (or C(3)-C(4») bond, formation of the indanes originates

probably from the benzyl"allyl-biradica1 12c, as the initial intennediate (scheme 3.6)

which by loss of CH20 and recombination of the radicals sites can yield the indane. As

an alterna.tive, the ketones 13a and 13b (M+ = 242) are formed in low yield by loss of

hydrogen. When using inert conditions, we did not obtain any evidence for the formation

of photoproducts found by Peijnenburg et. al 7

Reconsidering the mechanistic details of the photochemical behaviour of the 2-

62 PHOTOCHEMISTRY AND PHOTOOXYGENATION

naphthalenol derivatives, an important aspect was whether the reaction took place via the

planar or orthogonal ("twisted") state or via both species_33 Excited alkenes in both the

singlet and triplet states are able to twist efficiently. If a photoreaction will take place via

the singlet state of the exocyclic double bond in lOa, this state should eithcr be populated

by direct excitation or by singlet-singlet energy transfer from the singlet cxcited state of

the aromatic moiety _ Although both processes are a priori quite improbable, the results

for the absorption spectra show that the exocyclic double bond exhibits poor absorption in

the spectral regions accessible to our photochemical equipment. As a matter of fact alkyl

benzenes promote triplet formation of double bonds as a result of high intersy~tem cross"

ing yields and energy transfer. Hence, a singlet reaction could only be possible if singlct­

singlet energy transfer occurs. If energy is transferred from the S[ of the alkylbenzene

towards S] of ethylene the fluorescence intensitity for lOa has to be lower than for the

reference compound 14. However, no quenching of fluorescence (the fluorescence of lOa

is even slightly more intense) in lOa as compared to 14 is observed. Thus, the fluor­

escence measurements show that singlet energy transfer to the double bond is very unlike­

ly, even at lower temperatures (77K). Apparently, the reactive state in the photochemistry

of these naphthalenol systems is the triplcl state. Hence, a large increase in the dipole

moment as the twist angle of the double bond approaches 900 (referred to as sudden

polarir.ation) which is possible for olefin singlet states a~ the reactive intermediates, is not

plausible, as is further indicated by the absence of addition reactions by hydroxyJic

solvents_ In the photoaddition of water and methanol to 3-nitro-styrenei'4 evidence is

presented that an efficient addition occurs only from a planar polarized triplet state_

Therefore the absence of photoaddition of the triplet states of hoth naphthalenol com­

pounds and the fast B-Z isomerization, might be an experimental indication that the

intermediate is a twisted biradical exhibiting efficient E-Z isomerization as the major

deactivation pathway to the ground state34_

Another aspect concerning the photochemical behaviour was the investigation of the

conformation of the ground state geometries for the starting materials_ Semiempirical

quantumchemical calculations (MNDO, AMI and PM3) for the naphthalenol starting

materials show that not only in the 2S,4R7 but also in the 2R,4R isomer an orthogonal

orientation of the hydroxyl group is preferred over a planar one (i_c. the hydroxyl-group

in the plane of the exocyclic doubJe bond). So if conformational control plays an impor­

tant role (the conformation in the excited state might be different as a result of the twis­

ting double bond) the hydroxyl group in neither of the diastereoisomers occupies an ideal

position to react in a planar fashion with the double bond, as proposed earlier?_

In the photoreaction of optically pure compounds 10 and II we have in fact been

3.3.1 Results photoorygenalion o/2-naphthalenols 63

unable to identify products analogous to 2, claimed to be formed from the racemic pre­

cursors 1 via a. planar [l,3]yOH shift from a singlet double bond excited state, Although it

is still difficult to find out was has gone wrong in the earlier work and by what cause, we

examined the spectral properties of compounds 1 a and Sa 7 on the Onc hand and lOa and

11a on the other hand. Apart from their optical purity these compounds should be two by

two identicaL Since the first difference occurs in the spectral data of t 7a 7) and 8a, of

chapter 2, and the fact that we had to separate the isomeric optically pure naphthalenones

8a and 9a by HPLC, we suppose that the chemical purity of the compounds studied in the earlier work is doubtful. It should be noted that the wavelengths of maximum absorp­

tions of 1a and Sa (240 and 245 nm, in ref. 7, respectively) are different from what one

would expect for a tetrahydronaphthalene derivative (about 270 nm), whereas for com­

pounds loa and 17a the agreement is much better.

3.3 Photooxygenation

3.3.1 Results photooxygenation of 2-1UlphthalenoL~

UV-lrradiation of lOa in oxygen saturated hexane for 15 minutes at tcmpern.tures

below 0 °c, results in the formation of two products 16 and 17a, in 34 % and 60 % yield

at 50% conversion, respectively. We assume that 16 originates from the primarily formed

photoproduct 15 (see subsection 3_3.2), lrradiations have been stopped at maximum con­

centrations of the photoproducts at a ratio of Z, to E-isomers of approximately 7:3,

Longer photolysis times enhance the complexity of the reaction mixtures_ Isolation of the

photoproducts has been accomplished by column chromatography using hexane/ethyl

acetate (96/4) as eluent. Structures of photoproducts have been determined on the basis of

two-dimensional tHine correlation spectra and available literature data on the

photooxidation of al\ylic alcohols. The proton spectrum of the first eluted product 16

shows vinyIic reSonances at 6.44, 5.23 and 5.51 ppm in which the proton a.t 6,44 ppm is

coupled to the protons at 5_23 and 5.51 ppm eJ -- 10_67 and 16.82 Hz). Each of the

vinylic resonances represents One single hydrogen. The DC mfR spectrum of 17a shows

among others three characteristic resonances at 65.63 (s) and 58_85 (d) (-C-O~xirane)

and 77JJ4 ppm (d, -CH-O, alcohol). For a mOre detailed structure elucidation we refer to

the discussion section since 2D·NMR IH/13C correlation experiments do not give absolute

64 PHOTOCHEMISTRY AND PflOTOOXYGENAT!ON

proof on the connectlvities of oxygcn.35 Photoproduct 17a, obtained in relatively high

yield, is readily hydrolysed on standing in CDCl3 in the presence of light to afford com·

pound 18a. In the 13C NMR spectmm now three -C-O- resonances are visible at 67.82

(8), 62.54 (d) and 62.41 (d) ppm respectively, and the methyl on C9 of the exocydic

oxygenated double bond has shifted down field (7 ppm). 2D-NMR lH/13C Correlation

experiments revealed the structure of triol 18a (see scheme 3.7 and experimental section

for further details).

Me Me

~H o ··"'OH .".,. ------.. ""'OH "- :;;;:;:=.

Me "Et 302

16 10a 2S,3R,4R

hv

Me Me

'~'-...

SCHEME 3.7

Irradiation of l1a in the presence of oxygen results in a more complicated reaction

mixture. Repeated l1ash column chromatography still affords mixtures of photoproducts.

However, characteristic IH NM:R reSOnances in combination with the results obtained for

naphthaleno! lOa, indicate that an epoxy alcohol and a triol arc present in the collected

fractions (i.e. 03.33 (q) and 03.62 (s) for the epoxy alcohol 17b and 0 4.33 (q) and 3.28

(5) for the triol ISb). No ene reaction as observed for lOa has been found. Instead

characteristic signals in the lH NMR spectra of other fractions showed a singlet at 5.82

ppm and a quartet at 4.47 ppm.

3.3.2 Discussion photooxygenation of 2-naphthalenols 6S

3.3.2 Discussion photooxygentltion of 2-naphthalenols

Upon UV-irradiation of compound lOa in the presence of oxygen, we observed a fast

reaction leading to the regio- and stereospecific formation of two products._ Product 16,

the alcohol derivative of the allylic hydropcroxide 15, is without doubt the result of allylic

oxidation of the C=C double bond in a stereo- and regiosdective ene reaction, the

hydroperoxide group in 15 being easily converted to its alcohol, on C3, under the ex­perimental circumstances used, by performing column chromatography on silica geL The ene reaction of olefins is already known for three decadcs and is considered to proceed

with singlet oxygen as the reactive species_ 14 It should be noted that in apolar solvents

(hexane) the ene reaction is strongly favoured OvCr dioxetane formation at the experimen­

tal temperatures. Decomposition of the dioxetanes would lead to the corresponding

ketones in the reaction mixtures, however, no such ketones were found. As a result the

occurrence of the ene reactioJl implicated the fonnation of singlet oxygen (02(1..0.8) in the

reaction mixture, generated either by triplet Or singlet quenching of the naphthalenone

derivative, concomitant with thc deactivation to the ground state Or intersystem crossing, respectively. 12

The lH NMR spectrum of 17a indicates that the double bond has been photooxy­genated, since in combination with resonances. of 3.52 (q) and 3.40 (d) ppm, vinylic

resonances are absent. However, On the basis of spectral data in literature35 and our own spectral data we can establish the molecular structure of 17a as being the epoxy alcohol depicted in scheme 3.7. The structure of photoproduct 17a concurs with the results

reported by Adam et al. for the phQtooxygenation of olefins in thc presence of titanium

(IV) catalysts.35 Especially, the anomalous low·field 13C shift for the (doublet) hydroxy

carbon at 78 ppm of (R* ,S)-o:-(t, t-dimethylethyl)-2-methyl-oxiranemethanol is in ex­

cellent agreement with our own data (77 ppm). In combination with the other data and the

subsequent hydrolysis to trial 18a in which the relief of rigidity is evidenced by the

upficJd shifting of the methyl group at e9, product l7a is hence identified as the epoxy

alcohol shown in scheme 3.7.

Formation of an epoxide is not a straightforward reaction since on singlet

photooxygenation hydroperoxides and dioxetanes are generally formed as the main prod­

ucts. Only in certain circumstances epoxidation may occur (vide supra). Although we are

aware of the fact that epoxidation might be the result of triplet oxidation of the olefin in a

thennal process, high tripleHtate olefins are quenched efficiently by oxygen. Hence,

intersystem crossing from triplet to the ground state is probably the most significant

pathway _ Since we are dealing with a strained electron-rich olefin in a non-planar triplet

66 PHOTOCHEMISTRY ANO PHarOOXYGENATION

state, charge transfer intermediates of oxygen or even a 1,4-biradicaIJ6 are readily

formed_ As a matter of fact in the initial exciplex with triplet oxygen, electron rich

olefins, such as the ene moiety of lOa may act as an electron donor to produce an olefinic

radical cation and a superoxide radical anion.29 A subseqUlJ'Jlt reaction of thc olefin lOa

with the superoxide radical anion would then account for the high yield of epoxy alcohol

l7a together with the typical singlet oxygen derived product like the allylic

hydroperox ide_

Concerning the stereochemistry of the photooxygenations only two photoproducts

with a pronounced stereo- and regioselcctivity were found (> 95%). The determination

of the stereochemistry for these products is very cumbersome. Configurational assignment

is mainly based on factors controlling the regio and stereochemistry of the cnt reaction

and formation of cpoxides combined with our own chemical experiments on the crowded

tetraalkyl naphthalenols. As might be expected from steric shielding by the ethyl and

methyl groups, the attack of oxygen is blocked from this side and attack occurs exclusive­

ly at the side of the resident hydroxyl group. Since isopropyl erH and certain conforma­

tionally inaccessible hydrogens are not reactive, this explains why in the intermediate

e)(ciplex hydrogen abstraction occurs only from C-HlO and not from C-H2- The latter

occupies an inaccessible (quasi)equatorial position. 23 The observed high regio- and

diastereoselectivitics can be explained only by the stcreoelectronic properties of the

hydroxyl functionality, since sterlc control alone cannot account for the observed

diastereoselectivities, for as well 16, 17a and 17b_ Coordination of the hydroxyl group at

the allylic stereocenter with the incoming electrophilic singlet oxygen generates a

perepoxide-like transition state leading to only one isomer for these rigid allylic alcohols.

For compOund 17a, the Observed stereoselectivity is somewhat puzzling, since to

some extent, although relatively slow, ZIE isomerization took place, indicating that

oxygen quenches the triplet state effectively _ Singlet quenching would lead to enhanced

intersystem crossing and consequently faster Z/E isomerization. One explanation for our

observations might be that the stereosclective photochemical epoxidation is controlled by

the cis effect of the hydroxylic stereocentre on Cz, leading to a preferential attack of

oxygen on the less hindered hydroxylic side of the starting materiaL It is reasonable to

assume that in both photochemical pathways one initial, common transition state, an

exciplcx with perepoxide geometry, is involved in which steric interaction and allylic

strain is minimi7,e(\.J7 Attack only at the hydroxyJic side implies that the classical cis

effect operates for the E-isomer and thus reactivity is less. Although at present, absolute

proof was not obtained, we tentatively assign the reaction products to have the stereoche­

mistry depicted in scheme 3.6.

3.4 Concluding remarks 67

Upon stereo selective hydrolysis of the epoxy alcohol, by decomposition in the

presence of light and traces of water in CDCll , triol 188 is readily fonned. Nucleophilic

attack on the less shielded side of C9 gives then only one stereoisomer in which the

proton chemical shifts for C2 are almost identical to that in 16 suggesting that the hydro­

xyl groups are oriented in the same direction.

For the diastereomer lla the regio- and stereochemical selectivities are apparently

lower. IH NMR spectra do indicate formation of epoxy alcohol 17b and triol ISb as the main products, in which only the configurations now are opposite to that in 17a or 18a. Surprisingly for lla no ene reaction was found. Instead, several other products, presum­

ably by competing ene reactions with inversed regiospecificity, were fonned, which could

be explained by the fact that in lla the higher activation energy, due to larger sterlc

interactions of the ethyl group with the allylic moiety, could diminish regiQselective reac­tivity38 and lead to other competing reactions. As a matter of fact we anticipated that

photooxydation of the exocyclic double bond would lead to the fonnation of

photoproducts similar to 2a in scheme 3.1. Although no proof for such a product was

obtained, the quartet appearing at 4.47 ppm in combination with the singlet at 5.82 ppm,

indicates that addition of oxygen and an allylic migmtion of the double bond in an ene like reaction pathway does occur.

3.4 Concluding Remarks

It has been demonstrated that clean photochemical reactions are obscrved only when

optically pure (2S,4R)- and (2R,4R)-1 ,2,3,4-tetrahydr0-4-ethyl-1, 1,4-trimethyl-(3Z)­

ethylidene-2-naphthalenols lOa and 11a are irradiated in oxygenated solutions. While in

inert conditions no stereoselective reactions were observed, quenching with oxygen of the

excited states of lOa and 118 results in a remarkable high stereoselective photooxidation

of the double bond. This selectivity has been explained by the strong coordinating proper­

ties of the hydrox.yl group. The formation of an epoxide, as well as the photochemical

behaviour in inert conditions indicates that we are dealing with triplet biradica1s. This is

confirmed by several experimental results: the OCcurrence of a rapid E-Z isomeri~tion, a

stereospecific ene reaction indicating singlet oxygen fonnation and the fact that in­

tramolecular singlet-singlet energy transfer is precluded from the fluorescence

measurements.

68 PHOTOCHEMISTRY AND PHOTOOxYGENATION

In the photoreactions of optic.aJly pure compounds lOa and lla, we have in fact been

unable to identify products analogous to 2, claimed to be formed from the racemic pre­

cursors 1 via a planar [1 ,3]-OH shift from a singlet (in the double bond) excited state.

Semiempirical MNDO, AMI and PM3 calculations reveal that in the preferential ground­

state conformation, the hydroxyl group occupies an orthogonal position with respect to the

exocydic double bond.

3.5 Experimental Section

3.5.1 UV and jlMresceflce measurements

Fluorescence spectra were recorded on a Spex Fluorolog n emission spcctrometer##.

The excitation wavelength was 255 nm, All samples for fluorescence measurements were

degas~ by purging with argon for at least 15 minutes and had an absorbance (tem) of

0.10 ± 0.04. Electronic absorbtion spectra were recorded on a Hcwlcu- Packard 8451 A

diode array spectrophotometer. 2 Samples for low temperature (77K) measurements were

degassed by 4 freere-pump-thaw cycles,

(J R, 2S)-1 ,2 ,3 ,4-tetrahydro-1-elhyi-l ,4 ,4-trimethyl-2-naphthalenol (14)

To a stirred solution of I g (4.6 mmol) naphthalenonc 6 in 25 mL anhydrous diethyl

ether was added 0.26 g (4.6 mmol) LiAIH4 . After 30 min additional stilTing, respectively

0.5 mL of water, 0.3 mL of a 5N NaOH solution, and 5 mL of water was added. Filtra­

tion, separation of the organic layer, removal of the solvent, and separation accomplished

by column chromatography with hexane/ethyl acetate 96:4 (v/v) as eluent, afforded 0.3 g

pure (lR,2S)-naphthalenol 14.

lH NMR (CDCI) S: 7.20-7.33 (4H, m, Ar), 3.92 (lH, dd, H2), 2.05 (lH, dd, H), 1.73

(JH, dd, H), 1.54 (2H, m, H7), 1.38 (6H, s, HIO and H ll ), 1.33 (3H, 5, H9), 0.83 (3R,

t, 1-18)'

13C NMR (CDCI) 0: 144.05 (s, Cs or CD)' 141.59 (5, C6 Or C5), 127.36 (d, Ar), 126.79

(d, Ar), 125.97 (d, Ar), 125.31 (d, Ar), 73.13 (d, C2), 42.44 (s, C1), 42.41 (t, C3)

35.16 (s, C4), 34.03 (t, C7), 32,57 (q, C IO or C ll ), 28.29 (q, C Il or C IO)' 23.43 (q,

C9), 9.65 (q, C8),

## f~iH~i~~ ~,f lh~ lab,)tatoty for organic chemistry Qt th~ Unive"~ity of Am8terdam.

3.5 Experimental :5ection 69

3.5.2 Irradiation procedure

All photochemical experiments (as shown in table 3.2) On lOa and lla were per­

formed in 500 mL solvent of spectroscopic grade, containing 3 mmol of olefin. Ir­

radiations were performed uSing a 5OO-W medium pressure Hg lamp (Hanau TQ78)

through quartz. Cooling of the lamp was accomplished by means of a closed circuit filled with methanoL In this way the temperature in the reaction vessel was maintained below 0

QC. When using inert conditions, before and during irradia.tion, the solvent was purged by

a. stream of dry argon, to remove traces of oxygen. AU photoreactions were monitored by

means of gas chromatography. After irradiation, the solvents were removed on a rotatory

evaporator. The reaction mixtures were separated by means of HPLC (hexane/ethyl

acetate/water 99.75:0.25:0.001) or repeated column chromatography, using hexane ethyl

acetate (96:4) as eluent.

TABLE 3.2. Irradiations of (Z)-ethylidene naphthalenois in inert4 and oJlygenatcdb solutions.

yield (%)e

NaphthaJenol Amount Time

(mg) (min) (Z) e (E) C 12 13 ,,!d

lOa or l1aa 500 15 50 50

60 40 40 5 c 5

480 <5 <5 15 <5 >70

(Z)C (E)C 16 17

lOa b 400 15 35 15 17 30

lla b 400 15 30 10 23

·JrT~i;'lI.iQIlS in inert eonditions wete eo,M'\ucte.\ ll) '" SOO mL n-heXll1le containinl\, ~ 2 mmol of naphthalenoL

bJrradiatioru; were conducted in the P"'S<oIl~" of a continuous stream of dti~(1 oJ'.yg"n in .. 400 mL lI-heXll1le.

cYields were deterrniuo04 by gi\l>-chromatographic analy~is of aJi'lIlO!Jl remQved periodically from till) irradilltioll

oo.ixtul'e. 4 ~ompo5ed and polymerized materials. ~ All percentages hased On isolated yitld~ unleSB indicated

othetwiu.

70 PHOTOCHii:MISTRY ANO PHOTOOX,\,OENATJON

3.5.3 Characterisation of phmopmducts I 0

(2S,4R)-I. Z, 3.4-J'etrahydm-4-ethyl-l.l. 4-trimethyl-(3E)-ethylidene-2-nilphlhalenol (10b)

IH NMR (CDCl]) a: 7.22-7.42 (4H, m, Ar), 5.85 (IH, q, H9), 3.83 (IH, d, Hz), 2.25

(tH, m, H ll ), 2.03 (lH, m, H tt ), 1.86 (3H, d, H10>, L54 (3H, S, H7 or Rg), 1.39 (3H,

s, HlJ), 1.20 (3R, s, Hg or H7)' 0.84 (3H, t, Hd· lJC NMR (CDCl]) a; 143.30 (5, C5), 141.96 (s, C6), 138.72 (s, C 3), 126.82 (d, Ar),

126.10 (d, Ar), 125.97 (d, Ar), 125.88 (d, Ar), 123.36 (d, C9), 73.98 (d, C2.), 43.80 (s,

C4), 39.64 (s, Ct), 34.04 (t, C ll or Cg or C7), 32.84 (q, Cg or C7 or ell), 29.14 (q,

Cn), 25.56 (q, Cs or C7), 15.49 (q, CIO), 10.44 (q, Cd· GC/MS: 229.3 (M+ -eH), 226,4 (M+ -H20), 215.5,197.2,174.1,173.1,159.1,145.2

(base peak), 128.2, 117.1,91.1,77.1,43.2,39.1.

(2R. 4R)-J, 2, 3, 4-Tetrahydro-4-ethyl-], I, 4-1rimethy!- (3E)-ethyliden.e-2-nilphthalenol (11 h)

IH NMR (CDC13) a: 7.18·7.37 (4H, m, Ar), 5.74 (1H, q, H9)' 3.88 (lH, d, H2.), 2.35

(1R, nI, H l1), 2.15 (lH, nI, H ll ), 1.91 (3H, d, H IO), 1.62 (3H, s, HlJ), 1.32 (3H, s, Hg

or H7), 1.24 (3H, S, H7 or Hg), 0.60 (3H, t, H d· 13C NMR (COCl]) 0: 143.12 (s, CS)' 142.34 (s, CJ, 140.94 (s, C3), 126.48 (d, Ar),

126.09 (d, Ar), 125.53 (d, Ar), 125.53 (d, Ar), 121.03 (d, C9), 74.46 (d, Cz), 44.19 (s,

C~, 39.35 (s, Cl), 35.84 (I, C) 1 or Cn), 32.58 (q, Cn or ell)' 29.64 (q, C7 or Cg),

26.6..1 (q, C7 or Ca), 15.05 (q, Cl~' lOAl (q, Cd. GC/MS: 244.1 (M+), 229.2, 226.2, 216.1, 215,3, 197.2, 173.2, 159.3, 145.2 (base

peak), 128.1, 117.1,77.2,43.1.

E- and Z-J·Ethyl-l,3.3-trimethyl-2-ethylidelle"indanes (12a112b)

lH NMR (CDC I) 05: 7.41-7.15 (4H, m, Ar), 5.52 and 5.42 (lH, q), 1.90 and 1.85 (3R,

d), 2.05, 1.77 and 1.67 (lH, m), 1.48 and 1.43 (3R, s), 1.33 and 1.30 (3H, s), 1.29 and

1.27 (3R, s), 0.52 and 0.50 (3H, t).

GCrMS: 214.3 (M+), 199.2, 186.2, 185.2 (base peak), 170.15, 157.2, 155.15, 143.0,

128.1,115.1,91.1,77.1,41.1 32.0,29.1,28.0.

GCfMS: 214.20 (M"'), 199.2, 186.1, 185.1 (base peak), 170,15, 155.15, 143.1, 128.1,

91.2, n 1,41.1,32.0,29.1,28.0.

3.5 Experimelllal section 71

(4R) -3,4-Di hydro-4-ethyl" J , 1, 4-trimethyl- (3 -ZI E)-ethyJidene-2 (1 H)-flaphthalenones

(13a/13b)

GC/MS: 242.2 (M+), 227.1, 215.1, 214.2, 213.2 (base peak), 198.2, 171.3, 170.2,

155.2, 141.2, 129.1, 115.1, 43.1.

GC/MS; 242.3 (M+), 227.2, 215.2, 214.2, 213.2 (base peak), 198.2, 197.2, 182.5,

171.2, 170.2, 155.1, 141.2, 128.1, 115.1 43.1, 41.1.

(2S,3R.4R)-1 ,2.3.4 Tetrahydro-4-ethyl-l,1 ,4-trimethyl-3· (etheny/)-2 ,J-naphthalenedioi (16)

IH NMR (CDC!3) lj: 7.14-7.37 (4l-:\, m, Ar), 6.44 (lH dd, H9, 3J = 10.67 and 16.82

Hz) 5.51 (IH, dd, H IO, trans' 2J "= 1.9 and 3) = 16.82 Hz), 5.23 (JR, dd, H iO• cis' 2J ""

1.9 Hz and 3J = 10.67 Hz), 2.70 (IH, s, H;), 2.10 (lH, ill, H Il ), 2.02 (IH, m, H tt ),

1.52 (3H, S, Hu), L34 (3H, $, Hs or H7) , 1.32 (3R, S, H7 or Hs), 0.82 (3H, t, Cd. BC NMR (CDC!3) .5: 139.04 (s, Cj or C6), 138.02 (s, C6 or Cs), 135.11 (d, C9), 127.52

(d, Ar), 127.48 (d, Ar), 127.26 (d, Ar), 126.88 (d, Ar), 112.12 (t, ClO), 69.03 (d, C2),

65.63 (s, C3), 43.23 (s, CJ 39.2 (s, Ct), 32.78 (t, ell), 32.54 (q, Cn), 28.03 (q, C7 Or

Cg), 26.83 (q, Cg or C7), 9.98 (q, Cd.

(2S,3R,4R,1 'S)-! ,2,;,4-Tetrahydro-4-ethyl-I.I.4-trimethyl-2-naphthalenoI-spiro-J,; '­

methyl-oxirane (I7a)

IH NMR (CDC!3) li: 7.18-7.35 (4H, m, Ar), 3.52 (lH, d, HiI, 3.40 (IH, q, ~), 1.95

(lH, ill, Hu), 1.82 (IH, m, H l1), 1.62 (3H, d, H1iJ, 1.43 (3H, s, H 13), 1.38 (3R, s, Hs

or H7), 1.28 (3R, s, H7 or Hg), 0.63 (3R, t, Cd.

13c NMR (CDC1) 5: 143.82 (s, Cs or C6), !41.22 (s, C6 or Cs), 127.03 (d, Ar), 126.60

(d, Ar), 126.25 (d, Ar), 125.00 (d, Ar), 77.04 (d, C2) 65.63 (s, C3), 58.85 (d, C9),

43.23 (s, C4) 39.2 (s, C1), 31.24 (t, Cll), 31.06 (q, Cn), 27.89 (q, C7 or Cs), 25.04 (q,

Ca or C7 ), 15.17 (q, CIO)' 11.08 (q, Cd·

(2S.3S,4R.1 'R)-I,2,3.4-Tetrahydro-l-ethyl-3-(l'·hydroxy-ethyl)-1.4.4-ttimelhyl-2,3-

naphthalenediol (1/Ja)

lH NMR (CDCl) a: 7.18-7.30 (4H, m, Ar), 4.4 (IH, q, )-19), 3.05 (lH, s, HiI, 2.18

(2H, m, Hll), 1.55 (3H, 5, Hu ), 1.34 (3H, d, HlO), 1.32 (3H, s, Hs or H7), 1.28 (3H,

s, H7 or Hg), 0.87 (3H, t, Hd· DC NMR (CDC!3) 0: 140.71 (s, Ar), 139.10 (s, Ar), 126.61 (d, Ar), 126.43 (d, Ar),

126.31 (d, Ar), 125.97 (d, Ar), 67.82 (s, C3), 62.54 (d, C9), 62.41 (d, C;), 41.42 (s,

CI), 34.55 (s, C4), 31.13 (q, Cg), 31.02 (t, Cd, 29.59 (q, Cn ), 27.50 (q, C7), 22.51

(q, ClO), 10,04 (q, Cd.

72 PHOTOCHUMISTRY AND PHOTOOXYGENATION

(2R ,3S,4R,l 'R)-l,2,3, 4.Tetrahydro-4-ethyl-l, 1 ,4-trimethyl-2-naphthalenol-spiro-3,3 '­

methyl-oxjral1e (17b)

JH NMR (CDCI]) 0; 7.20-7.40 (4H, rn, Ar), 3.62 (lH, s, Hz), 3.33 (lH, q, H9), 1.82

(2H, rn, HlI ), 1.52 (3H, d, H 10)' 1.43 (3H, s, H 13), t.41 (3R, s, Hs or H7), 1.35 (3H,

s, R7 or Hg), 0.55 (3B, t, R l2.).

DC NMR (CDCI3) 0: 142.41 (s, Ar), 140.57 (s, Ar), 127.31 (d, Ar), 126.44 (d, Ar),

126.28 (d, Ar), 126.02 (d, Ar), 77.1 (d, C~) 64.31 (s, C]), 61.14 (ct, C9), 42.21 (s, C I),

40.01 (s, C4), 33.24 (q, Cg), 30.04 (t, Cn), 28.32 (q, CD), 26.41 (q, C7), l5.60 (q,

C10), 10,20 (q, C til,

(2R,3R,4K,] 'S)-l ,2,3,4-Tetrahydro-] -ethyl-3- (1 '-hydroxyethyl)-l, 1 ,4-trimethyl .. 2,3-naph­

thilLenediol (l8b)

IH NMR (CDCI]) 0: 7.18-7.34 (4B, rn, Ar), 4.33 (lH, q, ~), 3.l8 (lH, s, H2), 1.80

(2H, rn, HI])' 1.56 (3H, s, B13), 1.43 (3H, d, RIO)' 1.41 (3H, s, H7 or Hg), 1.38 (3R,

s, Hg or H7), 0.58 (3B, t, Hd.

References

1. 1. R. Scheffer, K. S. Bhtl1uklrj, R, E. Gayler and R. A. Wostradowski, J. Am. Chern. Soc. 97,2178 (1974).

2. A, M. Brouwer, L. Bezemer, J. Comelisse a.nd H. J. C. Jacobs, Reel. Trav. Chim. Pays-Bas 106, 613 (1987).

3. P. 1. Wagner, 8. Zhou, T. Hasegawa and D. L. Ward, J. Am. Chern. Soc. 113, 9640 (1991).

4. (a) H. R. Fransen and 11. M, Buck, J. Chem. Soc., Chern. Cornrnun. 786 (l982). (b) H. R. Fransen, G. J. M. fJonnans, G. J. Bezemer and 11. M. Buck, Reel. Trnv. Chim. Pays-Bas 103, 115 (1984).

5. (a) L, Salem, Science 191, 822, (1976), (b) w. G. Dauben, L. Salem and N. 1.Thr­ro, Acc. Chern. Res. 8, 781 (1969). (b) 1. Michl and V. Bon.alic-Kouteclcj, 111 Electronic aspects of organic photochemistry, John Wiley & Sons, Inc., New York) 1990, chapter 4 and references therein.

6. (a) G. 1. M. Donnans, H. R. Fransen and H. M. Buck, J. Am. Chern. Soc. lO6, 1213 (1984). (b) G. J. M, Donnans, W. J. G. M. Peijnenburg and II. M. Buck, J. Mol. StruCL (Theochern) 119,367, (1985). (c) G. J. M. Dormans, O. C. Groenen" boom and 11. M. Buck, J. Chern. Phys. 86, 4895 (1987).

References 73

7. W. J. G. M. Peij7l£nburg and H. M. Suck, Tetrahedron 44, 4821 (1988)_

8. S. L. Murov, In Handbook of photochemistry, Marcel Dekker, 1973.

9. W. J. G. M. Peijnenburg and H. M. Huck, Tetrahedron 44, 4927 (1988)_

10. It should be noted that a racemic mixture of 2R,4R and 2S,4S naphthalenQls is denoted as 18; 8 racemic mixture of 2S,4R and 2R,4S as. Ib; thc optically pure analogues i.e. 2S,4R as lOa and 2R,4R as 118. The configurations and systematic names are corrected with respect to the original article according to the lUPAC convention. However in order to obtain as much consistency as possible, the num­bers in the NMR data were assigned as. in Scheme 3, 1 and 3.6 names are corrected with respect to the original article.

11. (a) A. A. Frimer, Ed. In Singlet oxygen, eRe Press: Boca Raton, FL, (1985)_ (b) H. H. Wassennan and R_ W, Murray, Eds. In Singlet oxygen, Academic Press: New York, (1979). (c) A. A. Connan and M. A. 1. Rodgers, Chern. Soc. Rev. 10, 205, (1981)-

12. (a) J. Saltiel and B. W. Atwater, Adv. Photochem. 14, 1 (1988). (b) A_ A_ Gonnan, 1. Hamblett and M. A. J. Rodgers, J. Am. Chern. Soc. 106,4679 (1984)_

13_ D_ R_ Keams, Chern. Rev. 71, 395 (1979)_

14. M. OrjaMpou/os, I. Smonou and C. S. Foote, J_ Am. Chern. Soc. 112, 3f!:1J7 (1990)_

15. (a) C. W. Jefford and C. G. Rimbault, J. Am. Chern. Soc. 100, 6437 (1978). (b) C. W. JejJord, J, Boukouvalas, S. Kohmoto and G, Bernardinelli, Tetrahedron 41, 2081, (1985).

16. L. B. Harding and W. A. Goddard Ill, J. Am. Chem_ Soc. 102, 439 (1980).

17. (a) A. A. Gorman, I. R. Gould and I. liamblett, 1. Am_ Chern, Soc. 104, 7098 (1982). (b) A. A. Gorman, I. Hamblett, C. Lamben, B. Spencer and M. C. Standen, J. Am. Chern_ Soc_ 110, 6583 (1988).

18. 1. R. Hurst, S- L. Wilson and G. B. Schuster_ Tetrahedron 41, 2191 (1985).

19. (8) M. OrjaMpoulos, M. B. Grdina and L. M. Stephenson, 1. Am_ Chem_ Soc. 101, 275 (1979). (b) K. H Schulte-Elte and V. RaUlensrrauch, J_ Am. Chern. Soc. 102, 1738 (1980)_

20. (a) H. E. Ensley. R. V. C Carr, R. S. Manin and T. E. Pierce, J. Am. Chem_ Soc_ 102; 2836 (1980). (b) E. L. Clennnn, X Chen and J. J. Koola, L Am_ Chern. Soc. 112, 5193 (1990).

21. M. Orjanopoulos, M. Stratakis and Y_ E/(~mes, Tetrahedron Lett. 6903 (1989),

74 PHOTOCHEMISTRY AND PHOTOOXYGENATION

22. M. Oljanopou/os, M. Stratakis and Y. Elemes, J. Am. Chem. Soc. tt2, 6417 (990).

23. R. Matush and G. Schmidi, Relv. Chim. Acta 27,51 (1989).

24. W. AdiIm and B. Nestler, 1. Am. Chern. Soc. 114, 6549 (1992).

25. W. Adam and B. Nestler, J. Am. Chern. Soc. 115, 5041 (1993).

26. C. W. JejJord and A. W. Boshung, Relv. Chirn. Acta 60, 2673 (1977).

27. C. W. Jejford, Chem. Soc. Rev. 22, 59 (1993).

28. S. Konuma, S. Aihara, Y. Kuriyama, H. Misawa, R. Akaba, H. Sakuragi and K. Tokumaru, Chern. Cett. 1897 (1991).

29. (a) R. C. Kannner and C. S. Foote, L Am. Chern. Soc. 114, 678 and 682 (1992). (b) J. Eriksen and C. S. Foote, J. Am. Chem. Soc. 102, 6083 (1980).

30. (a) M. J. S. Dewar and W. Thiel, J. Am. Chern. Soc. 99,489 (1976). (b) H. Halim, N. Heinrich, W. Koch, j. Smidt and G. Frenkling, J. Compo Chern. 7, 93 (1986).

31. (a) M. J. S. Dewar, E. G. Zoebish, E. P. Healy and J. J. r. Stewart, J. Am. Chern. Soc. 107, 3902 (1985). (b) M. J. S. Dewar and 1 . .l. P. Stewart, QCPE Bull. 6, 24 (1986); QCPE Program No. 506 (version 2.10).

32. (a) j. j. P. Stewart, J. Compo Chern. 10, 209, (1989). (b) J. J. P. Stewart, J. Compo Chern. 10, 221 (1989).

33. N. j. Thrro, In Modem molecular photochemistry. The Benjamin/Cummings publis-hing Co., Menlo Park, Cal. 1978 p. 419 and 476.

34. P. Wan, M. J. Davis and M. A. Teo, J. Org. Chem. 54, 1354 (1989).

35. W. Adam, J. Am. Chem. Soc. 111, 203 (1989).

36. A. A. Gorman and M. A. J. Rodgers, J. Am. Chcm. Soc. 111,5557 (1989).

37. R. W. Hojfm(mn, Chern. Rev. 89, 1841 (1989).

38. M. OrjoTWpoulos, M. Statakis and Y. Elemes, 1. Am. Chern. Soc. 113,3180 (1991).

4

Abstract

Intramolecular Hydrogen-Transfer and Cyclo­

Addition Photochemistry of Cyclic I ,3-Dienes#

With use of one- and two-dimensional NMR spectroscopy and deuterium labelling. the

photochemistry of 9-endo-hydtOxy-9-exo-vinyl-bicyclo(4.2.1Inonadiene (1) and the 9-exo­

(I, l-dimethylvinyl)- (2) and 9- an-ethyl- (3) analogues has been studied. Irradiation of J, 2 and 3

in n-hexane gave novel 8-membered ring systems 4-6 by a light induced rearrangement ptOcess. in

which the hydroxyl proton is transferred on One side of the molecule towards OnC of the termini of

the endocyclic diene. This rearrangeml:nt process thus involves a formal hydrogen transfer, during

which either H+ Or H- may be transferred to a reactive diene intermediate. Replacement of thc

hydroxyl proton by deuterium in 1-3, and 2H NMR of the corre.~ponding photoptOduCtS confirmed

that the hydrogen translocation occurs intramolecularly. Prolonged irradiation of 4 and 5 re.~ult~ in

the formation of pyran products 10 and 11 by an intnunolecular photocycloaddition of the triplet

excited state of the ct,j'j-unsaturated ketone to 1,3-cyclooctadiene, via a stabilized bisallylic biradical

intermediate. Conformational studies of the structurally more rigid system 10 which is derived from

4, revealed that the hydroxyl proton was transferred on the endo side of the molecule.

'Baud Oil the foUowi!l.S p«per, Fmnciscus W.A.M. Micsen, HatlB C.M. Bae[eT;l, li<trm A. Langermans, Henk A. Ci~e=w; and Leo H. Kooie, C31\. J. Ch~ffi. 69, ISS4 (1991).

75

78 INTRAMOLECULAR J-IYDROGEN"TIlANSPF.R

We have studied the photochemistry of these compounds partly on the basis of deuterium

labelling experiments. The structural and conformational properties of the isolated

photoproducts were investigated with one" and two-dimensional NMR techniques, low

temperature I3C NMR, and semiempiricaJ AM t quantumchemicaJ calculations.

4.2 Results and Discussion

4.2. J symhesis, purification and identification

Compounds 1··3 were prepared in two steps from bicyc10(4.2.1]nona"2,4,7-trien- 9-one

(scheme 4.4).13 The bicyclic ketone is prepared readily and in large quantity from dilithium

cyclooctatetraenide and dimethyl-carbamoyl chloride in diethyl ether and subsequent

hydrolysis with sulphuric acid. The ketone is a fairly stable compound when stored in the

dark and freezer. The next step comprises selective hydrogenation of the ercg bond in the

presence of the other double bonds using Ni2B as a catalYSt. 14- 16 The NizB catalyst is

prepared in situ from two common reagents nickel(lI)acetate and sodium boronhydride and

is completely non-pyrophoric. The product bicyclo[4.2.1]nona-2,4-dien-9-one waS Obtained

in pure form after silica gel column chromatography using dichloromethane as eluent.

o (i) - (iii) -1 R = -CH=CH:!

2 R = -CH=CH(CH3~

3 R = -CH2CH3

The next step in the synthesis of 1·3 is a Grignard reaction with vinyl-,

2,2-dimethylvinyl- or ethylmagnesium bromide, respectively. It should be noted that this

reaction affords the 9-endo-hydroxy epimers ei<c!usively.17,ts The n-molecular orbital of the

carbonyl group interacts directly with the butadiene moiety via homoconjugation.

4.2.2 Photochemistry 79

When weak electrophiles are involved, homoaromatic charge delocalization polarizes the

attacking reagent and starts to engage in bonding. So when this carbonylgroup is increasingly

polarized by organometallic or hydride reagents, homoconjugation between these two

molecules is observably enhanced. To the extent that this conformation is relatively inflexible

to those reagents that attack, the bridgehead carbon is forced to anti approach because of

substantial steric shielding on the syn surface by the diene bridgc.

Compounds 1-3 were purified by silica gel column chromatography; their identity and

purity were assessed by one- and two- dimensionallH NMR (COSY), l:lC NMR {including

lNEPTP),19 two-dimensional IH/lJC correlation maps,19 and mass spectrometry. As an

example, strucrure elucidation of the.ore compounds has been carried out as follows. We start

with the assignment of the tertiary proton on CI . This carbon is easily identified from the

two-dimensional INEPTP spectrum in which the multiplet. structure is retained and the

desired multiplicity selection is chosen in such a way that CH, CIi] are positive and CH2

resonances are negative. The spectrum shows a positive doublet at relatively low field for

the aliphatic region. When the chemical shift of the C1 carbon is known, its adjacent proton

HI is also known from the IH/lJC correlation spectrum. Subsequently its coupling

neighbours, the protons of C2 and Cs can be assessed from the two-dimensional COSY

spectrum. In this way the protons on C2 and Cg are known, followed by the assignment on

CJ and C7 etc. Assignment of the geminal protons each, has heen done by determining the

vicinal coupling constants of the protons (vide infra and experimental section). All

photoproducts were isolated via preparative HPLC separation of the crude reaction mixtures.

Molecular structures could be unequivocally established by the above-mentioned methods.

4.2.2 Photochemistry

UV·irradiation of compounds 1-3 in n-hexane led to the formation of the novel

8-membered ring systems 4-t'i, resulting from a new intramolecular formal hydrogen transfer

reaction (scheme 4.5). Simultaneously, the expected products of a disrotatory e1cctrocyclic

ring closure reaction 7-9 were formed. Prolonged irradiation of photoproducts 4 and 5 led

to the strained pyron structures 10 and 11 (scheme 4.6).

In studying the photochemistry of 1-3, we monitored the kinetics of formation and

disappearance of materials at different photolysis time intervals (GC analysis). The results

for the irradiation of I arc presented in figure 4.1. Our data clearly show that 1 is initially

converted to 4 and 7; compound 10 is formed from 4 in a secondary photoreaction.

R R 11 I 10

7 5

1 R= H

2 R"' CH3

OH

3

R

hv

hv

R 11 I 10

B 3

7 4

6 S .

4 R" H

5 R = CH3

INTRAMOLECUI.AR HYDROGEN-TRANSFER.

R R 11 I 10

e

7

6 5

6

SCHEME 4.5

hv --..

7

SCHF-MI:l4.6

+

+

R

5

10 R;;H

11 R" CH3

3

R R 111 10

7 3

4

9

4.2.2 Photochemistry

100r------

% 90

eo

70

60

50

---€>-~-~ . --~r-...._.

B_1) - rl' ~ 1 _ 7 ""'----CL_,_. -_ ~I-_-_-..l-[:-:-~.- _ .

, - • ~1-, , fI- - - , ~ • ~

~-....---.,,----,----• _"-- __ .. ---;t,-

2 3 4

_--c-

5 6 7 8 ---- h

9 10

81

FIGURE 4.1. Plot for conversion (%) of cumpound 1 to photoproduct~ 4, 7 and 10 as function

of time (h).

Analogously, irradiation of compound 2 initially produces 5 and 8, and compound 5

is converted to 11 upon prolonged irradiation. The experiments with compound 3 merely

showed formation of 6 and 9, i.e., compound 6 does not show the secondary photoreaction.

Verification of the stereochemistry of these reactions could be obtained with compound

la (scheme 4.7). Transfer of deuterium toward C2 was evident from the If!- and I3C NMR

spectra of 48. The 2H NMR spectrum of 4a, recorded at 77 MHz, showed a singlet signal

at li 2.3 ppm, exactly at the position of One of the H2 protons in compound 4 (compare

figures 4.2a and 4.2b). This demonstrates that the transfer of deuterium occurs at One face

of the molecule exclusively.

The light-induced conversions 1 .... 4, 2 ..... 5 and 3 .... 6 may be formulated as. proceeding

via an initial photoprotonation process followed by a collapse of the resulting carbocation

intermediate. (c.f. The photo fragmentation of homoallylic alcohols):~,20,21 Cyclohexenyl.

alcohol afforded on sensitized UV-irradiation cyclohexane and acetaldehyde in high yield.

The photofragmentation was considered to proceed via intramolecular protonation, since mOre

rigid cyclohexene analogues did not undergo photoprotonation and did not show

INTRAMOLECULAR HYDROGEN-TRANSFER

photo fragmentation .

The cis - tranS isomerisation of common 1,3-dienes around a double bond is a very

common photochemical process and is helieved to proceed by twisting. Upon singlet

excitation one bond can isomeri7.e, whereas triplet excitation may lead to isomeri~ation of

both bonds. In contrast to twisted ethylenes the 7.witterionic strUCtures that contribute to the

lowest excited states do not mix equally because the allyl moiety and methylene moiety have

different electroncgativilies so that the resulting states are polar, which would make

photoprotonation plausible. As a result, such a photoprotonation process, concomitant with

the bre.'lking of one of the bridgehead bonds, would be in line with the involvement of an

olefinic intermediate exhibiting enhanced reactivity. 4,7

(a)

7

2' 2 T 6 8'

--.,....----r-.. ""' ..•. r'-'-~'-'T ·~~-T'· . I"-"~-'--' .,--r------r---.10 2.0 I.~,

(ppm)

(b)

••.• ---,---' ........ '" I' •• ,._-.,....-...,-_ •.. '! ·--~·--'T· "','-""---'" .1.0 1.0 I~

(ppm)

f"!GURE 4.2 (a) The IH NMR (400 MHz) a.nd FroURE 4.2 (b) The '"H NMR (77 MHz) spectrum

of 4.a in CDCI3. The assignrnent~ of protons were obtained from two-dimensional homonuclear

NMR spcctr<l (COSY).

4.2.2 Photochemistry 83

The formation of the photoproducts 7-9 involves a ring closure of the endocyclic dienes

in 1-3. As can be Seen in figure 4.1, the photoprotonation reaction and the ring closure occur

with approximately equal probability during the ilTIl.diation of 1. Similar results were

obtained for 2 and 3_ The photochemical ring closure of 1-3 follows a disrotatory route, as predicted by the Woodward & Hoffmann rules. 22 The endo orientation of the protons H2 and

Hs was established on the basis of the NMR s.pin-spin coupling constant JH(I)-H(2)' and comparison with literature data of cyclobutane formation for other bicyclo-dienes. 23

The products 7-9 of the direct photolysis of 1-3 are probably derived from the lowest excited singlet excited state because intersystem crossing appears to be unimportant in similar

cycloheptadiene systems in the cyclobutene formation. The e1ectrocyclic closure of the

excited singlet state to the cyclobutenes 7 - 9, which is not observed for twisted 3,5-

cycloheptadienones, suggests that a planar diene moiety is a prerequisite for singlet concerted, symmetry-allowed disrotatory ring closure. However when the diene moiety is twisted, cyclobutene formation can be observed only on sensitized irr.idiation via the lowe.<;t

triplet excited state_ In such cases the reaction is explained by the thermal con rotatory ringclosure of the cis-tIans geometry_ This reflects again the importance of ground-state

control in photochemical reactions_ If these reactions are compared again with the

decarboxylation process in the cycloheptadienone systems, photoproducts 4 - 6 may also be

formed as a result of singlet excited state process, since decarboxylation proceeds only from

the excited singlet state. It should be noted that for the triene derivatives of 4 - 6 the

bridgehead substituents are also lost upon UV-irradiation, which seems similar with the

decarboxylation process. Although on direct photolysis of the diene systems other photoproducts are obtained it is. not plausible that this alters the singlet course to triplet.

FIGUI\Ja 4.3 Schematic representation for the formal hydrogen-transfer reaction.

However, at present the exact nature of this photoreaction cannot be defined and transfer

of the hydroxylic proton in the form of H· or even W cannot be excluded, Even the cyclic

reorgani7.ation of ele.::trons (as shown in figure 4_3) involved in the overall reaction may be

84 INTRAMOLECUtAR I!YDROGEN-TRANSI'ER

formulat.ed as an 8-electron analogue of an oxy-retro-cne reaction.24 This reaction which is

the oxygen analogue of the retro-ene reaction is thought to proceed via a cyclic transition

state.

00

hv hv .. ..

1a 4a 10a

SCHEME 4.7

We could not establish e-xo or endo orientation of deuterium in 4a, since the

conformational flexibility of the 8-membered ring preclude.o: the application of Karplus-type

equations to translate the vicinal proton-proton I-coupling constants into proton-proton

dihedral angles. 25 However, for the stnlCturally more rigid system lOa, which is derived

from 4a (scheme 4.7), it could be unequivocally established that deuterium is located on the

eJUJo-faec of lOa and hence also on the enda-face of 4a (vide infra). Additional evidence for

the presence of deuterium on C2 was obtained from a comparison of the 13C NMR spectra

of 4 and 4a. In 4a, the Cz signal is split into three lines of equal intensity and I = 19.4 Hz

due to I-band J-coupling with deuterium (I = 1). Interestingly, a small signal is also visible

at i5 1.6 ppm in the zH NMR spectrum of 4a. The small peak is found exactly at the position

of onc of the Hg protons in compound 4 (compare figures 4.2a and 4.2b). The explanation

for the appearance of the small peak in figure 4.2b may lie in the occurrence of a thermal

supraladaJ 11 ,51-H sigmatropic shiH, occurring during the thermal isomcrization of

cix,lrans-con to l:is,cis-COD, according to scheme 4.8. 26,27 Evidently, this U.5JH·shift

would place deuterium in an isochronous position in the 2H NMR spectrum. when compared

with the endo-located H~ in compound 4.

Of course, the occurrence of this [1,5]-H shift is not detectable from the IH and DC

NMR spectra of compound 4. It may be noted that the UV irradiation of 1 and 2 does not

induct a sigmatropic [1.3]-OH shift in the head group. In fact. we anticipated that this

reaction could occur, in analogy with the previous work of our group on the photochemistry

of 8-hydroxygermacrene B. llu ,2S Although this reaction was also claimed to occur in racemic

tetra alkyl-2-naphthalenol derivatives,28-30 the results presented in chapter 3, led to the

4.2.2 Photochemistry 85

negative conclusion with respect to the actual feasibility and occurrence of a planar [1 ,3]"OH

shift for these compounds.

[1,5] H shift .. ....

SCHEMli 4.8

The 8-membered ring systems 4 and 5 are converted into the strained pyrans 10 and 11

in a secondary photoreaction. To the best of Our knowledge, this reaction represents a novel

photocycloaddition, which is reminiscent of the Patemo-Buchi reaction of ketones with

alkenes, leading to the formation of oxetanes,31,32 These reactions generally proceed from

an attack of an (nT *) state of the carbonyl compound on an unsaturated substrate. In general

photocycloadditions of (n1/) states are expected to proceed via biradical intermediates. 33,34

These intermediates are produced either directly or indirectly from S I (n'lf *) or T t(n ... ·) states

in the oxetane formation.

lntersystem crossing in most carbonyl compounds is an efficient and facile process,

whcn the lowest energy singlet excited state is of the (n'll'") type, especially when the crossing

can lead to a ('If'lf.) triplet. Obviously if the rate of ISC is faster than that of diffusion in

solution bimolecular reactions of the excited singlet state are precluded. In comparison with

the aromatic carbonyl compounds the rate of ISC for saturated (aliphatic) ketones is relatively

slow. Hence, these aliphatic ketones may have the opportunity to react in the excited singlet

state. The cycloaddition reaction may be deactivated or not OCCur at all by radiative decay

(fiuoresct..'1Icc) from its singlet excited state. Efficient quenching of this fluorescence, in

diaIkyl-ketones, however, may result again in the photocycloaddition product. As a result it

is clear that photocycloaddition from the singlet excited state occurs only when the

deactivation to the ground state is slower. In contrast if quenching of the aliphatic ketone

fluorescence is left unaffected the photocycloaddition reaction is inefficient. However both

the singlet and triplet excited states can be deactivated mostly by energy transfer from the

excited state of the reactant molecule to the ground state of the quencher. An important issue

is the fact that in this type of reactions the unsaturated moieties, themselves, especially when

they are conjugated may act as a quencher. Irradiation of the carbonyl compounds in the

86 INTRAMOLECULAR HYDROGEN-TRANSFER

presence of dienes usually lead to isomerization and dimerization (the latter especial in

concentrated solutions), but nO cycloaddition products. Thus. such dienes may efficiently

quench the carbonyl triplet. If the triplet energy of the dienes is below that of carbonyl

compounds, dienes quench phot.oaddition as well as other reactions from the triplet excited

state. Concomitantly, sensilation of the dienes by the (n1/" +) excited slatc leads to

isomerization of the olefin. If the triplet energy is below that of the diene triplet-triplet,

energy transfer will become inefficient and photocycloaddition may occur. To examine which

excited state is responsible for these photochemical reactions, one needs to know the energy

of the excited slates of the aJkanone and diene moiety which are involved.

TABLe 4.1 Appro)(imatc singlet and triplet energy levels.

Compound

0/ .(3-U nsaturate.d ketone!!.

Saturated ketone

cis ,Cis-Cyclooctadicneb

cis,trans-Cyclooctadieneb

"I)""" ,)btaill~d from N.J. Tuno. J~ bO"L" /.ItLaiMd from R.S.H. l-il). 31

51 [kcallmol]

74

84

>80

T] [kcal/mol]

70

78

70-73

<70

In our system the Gt,/3-unsaturated ketone would have a singlet (n7/') state of lower

energy than the singlet state of the conjugated cis,cis-COD system, while the triplet (n'lr~)

state is almost equal to the triplet of the cis,Gis-COD system. The triplet excitation energy

in the COD ~ystcms is considerably higher lhen the known values for common 1,3-dienes.

From calculations (see subsection 4.3 on conformational analysis) the diene moiety is far

from planar. Since the triplet energy of an isolated bond is considerably higher. an increase

in the triplet stale energy is expected as a diene unit looses conjugation by rotation around

the single bond into a configuration where double bonds are close to orthogonal.

For the saturated ketone, however. the photocycloaddition is absent, implicating that the

reaction, which is a secondary process. does not proceed via a mechanism in which initial

protonation of one of the double bonds of the dil.'11e system is followed by formation of the

cyclic structure. The reason for the absence of photocycloaddition may lie in the fact that the

triplet excited state of the saturated ketone is considerably higher than the triplet of the

4.2.2 P}wtocil£mistry 87

cis,cis-COD systcm. Consequcntly the triplet of the saturated ketone would be quenched by

energy transfer, many times faster than the (l,p-unsaturated ketones, because of the large

exothermicity of the process. For approximate energy levels see table 4.1.

Energy transfer to the cis,cis- or cis,trans-COD system results in a photostationary state

composition of cis,cis- and cis,trans-cyclooctadienes. 36,37 It should be noted that the c;S,trans

isomer, may isomerize to cis,cis either by a cis,trans isomerization about the trans double

bond or by a [1,5]-H shift for the thermal isomerization.Z6,:n

Since intersystem crossing (ISC) is very fast in (l,p-unsaturated ketones, and singlet

processes are normally too slow to compete, a mechanism involving a triplet excited state

is preferred over a singlet excited process. It has been proposed that the reaction of enones

generally proceeds via excitation of the eoone which may reversibly go to a triplet exciplex

of the olefin and the triplet Cnonc. The next step is the formation of the biradicals which may

either collapse to the strained pyrans or undergo radiationless decay to the ground-state. 1,3-

Dienes may also interact with the electronically excited singlet states of the aJkanones,

implicating that not only triplet quenching occurs but also the alkanone singlet may undergo

substantial quenching. A mechanism involving the formation of a charge-transfer, stabilized

singlet exciplex, leading also to a biradical is rather remote for the cis,cis-COD system. 38 ,)9

However, in the oxetanc formation of acetone and COD (ISC is slower in this case) the total

rate of tluorescence quenching of the ketone is signi!1cantly slower than the reaction rate.38

Therefore we assume the pyran formation to occur predominantly from the triplet state of

the I)( ,S- unsaturated ketone, and a chemical reaction can indeed compete with the triplet

energy transfer to the COD, because of the decrease in diene character of the 8-membered

ring. Based on an alkoxy model,40 the intramolecular pyran formation like intramolecular

cycloadditions is expected to obey a "rule of five" (i.e. the initial step being the formation

of the five membered ring),41 involving the stabilized biradicaloid intermediate (scheme 4.9).

SCHEME 4.9

88 INTRAMOLECULAR HYOROGEN"TRANSf£R

43 Conformational studies

The IH NMR spectrum of "the compounds 1-3 shows a long-range spin-spin coupling

between HID and the hydroxyl proton tJHCCOH '" 1.5 Hz). This reveals that the coupling

path is in all trans ("W") conformation, i.e., the OH group resides in the plane of C9 , CIO,

and C II· The C to-Cll double bond in 1 and 2 is therefore approximately orthogonal with

respect to the endocyclic diene system. The conformational properties of 1-3 were

investigated further 00 the basis of semiempirical MO calculations. Dreiding models of 1-3

provided initial starting conformations.. The geometries of 1-3 were optimi:zerl uSing the AM!

Hamiltonian of the AMPAC program42 ,4J The optimized conformations of 1- 3 showed trans

conformation of the coupling path H-O-C9-C IO-H IO in the head group (figure 4.4a).

1 10

FIGURE 4.4 (a) Optimized conformation of 1 ~tudied with AMl calculation method showing

tranS contormation of the I;oupling path in the hCild group (.1.Hf = -2.78 kcal/mol). fiGURE 4.4

(b) Optimi7.cd conformation of 10 from which proton-proton torsion angles for the Karplus

relations have been extracted (.1.Hf ;; -8.03 keal/mol).

Inspection of the AM1-optimizcd conformation of 1 confirms that the hydroxyl proton

is proximate to Cz (rOH.pt()tcI11-C(2) = 2.52 A). Essentially thc same conformation was found

in the AMl calculation on structures 2 and 3. (The AMI Ca.lculations on 1"3, 10 and n,

89

resulted in the following enthalpies of formation (AHf): .:1H f (1) = -2.78 kcalJmol, .:1H( (2)

= -17.80 kcal/mol, .:1Hr(3) "" -34.67 kcallmol, LlHd10) = ·8.03 kcalJmol, .o.Hdll) =

-23.47 kcaJ/mol.)

The confonnational properties of the 8-membered ring compounds 4-6 are more difficult

to characterize, since there is no obvious starting geometry for the AMI calculations_ Rather

it must be expected from the work of Anet and Yavari on cis,cis-l,3-cyclooctadiene and

cis,ci.f-l ,3-cyclooctadiene mono-epoxide 44-46, that compounds 4·6 are involved in a complex

conformational equilibrium. They reported two distinct (energy minimum) conformations,

the twist boat chair (THC) and the twist boat (TB) conformation for

cis ,cis-l ,3-cyclooctadiene_ Both TBe an TB have strongly twisted diene chromophores, the

calculated twist angles being 54 0 and 43 0 respectively and the length of one double bond

being greatly enlarged, while the other double bond is normaL These data reflects again the

decrease in dime character of the eight-membered ring. We obtained convincing evidence

for the flexible nature of compounds 4-6 from variable-temperature 13C-NMR measurements.

Upon lowering the sample temperature from 20 Lo - 90~C in C~CI2 we observed

decoaJescence phenomena for 4-6_ As a typical example, figure 4.5a and 4.5b show the

olefinic regions in the )lC NMR spectra of compound 6, as mcasured at 20°C and -90°C.

The observation of two sets of Be NMR peaks at -90°C clearly shows that at least two

distinct conformations co--exist at this temperature_ It should be noted that our DC NMR

spectral data on compounds 4-6 closely resemble the data on Anet and Yavari on

cis ,cis-l ,3-cyciooctadiene and cis ,cis-cyclooctadiene mono-cpoxide_ 44-46 The confonnational

properties of strained pyrans 10 and 11 were again studied with the AM 1 calculation method.

Slaning geometries were obtained from Dreiding molecular models. In the optimized

confonnation of 10 (figure 4.4b) we have specially focused on the proton - proton torsion

angles 4>. [H 1-C 1-Q-H21 = -2.r; "'2 (H)-C) ~"H:;rl "" 119.2°; oP3IHrC3-CrH2] = -23.6 0;

q,4 [H]-C3 -Cz-H2·] = 96.4". Thus, from a simple Karplus dependency of 3'HH on the

proton-proton torsion angle,:zJ one expects that 3 'H(().H(2,end,,) and 3 JH(;:S)-H(2.~lJdo) are

substailtially larger than 3JH(1}_H(2',u<l) and 3JH(.!)"H(2"~xo) -

Interestingly these tentativc conclusions are in line with the experimentallH NMR data

on compound 10, which show that the downfidd Hz proton (2_24 ppm) has relatively large

coupling constants with its vicinal neighbours Hl and H;:s (i.e 10.1 and 6_2 Hz, respectively,

(see Experimental Section), while the upfield Hz, proton (1.63 ppm) shows much smaller

3JHl( couplings with HI and H) (i.e. < 1 Hz in both cases). Therefore, we tentatively assign

the down field H2 proton as H2,e'/ldo ' and the upfield Hz, proton as H2"ao • Having this

assignment, it is worthwhile to reexamine the lH NMR spectrum of the C2-deuterated

compound lOa. Now it seems likely that it is H2,~ndo and not H;L~Q that is replaced by

90 INT~AMOLECULAR HYDROGEN-TRANSFi>&

deuterium by lOa. Thus the photochemical formal proton transfer reaction 1 - 4, proceeded

on the endo-face of the molecule exclusively. We anticipate that this holds also for the

analogous photochemical reactions 2 - 5, and 3 ..... 6.

6 -4 5

, .--..,----•• --••• c ," ,.".""----r------.- . , .'.".,-._-,---" III 1.10 m 1M I~ In

(ppm)

""-'" -~- , " -~--"""T' .. , .. '/'-'~"--"T-.'

130 126 m m

FIGURE 4.5 (a) Olefinic region of the 1JC NMR spectrum of 6 at 200c (100 MHz, solvent

CD2CI2). The assignments are ba.~ed on a 13C JH correlation experiment. FIGURF. 4.5 (h)

Decoalesccnce in the olefinic regjon of LlC NMR spectrum of 6 at -90 9 C (100 MHz, soivl;rtt

CD2CIz). Clearly, twO distinct molecular \:onformations are present in a ratio of approximately

I : 2 at this temperature. These observations dosely re.~emble the data of Anet et al. on

cis,cis"cyc1ooctadienc" and cis,cis-cydooctadiene monoepoxide). 44-46

4.5 Experimental section 91

4.4 Concluding Remarks

UV irradiation of compounds 1-3 leads to a new photochemical protonation of the olefin, in

which the hydroxyl proton is transferred on the endo-side of the molecule toward one of the

tenninals of the cndocyclic diene. However the reaction can also be considered as an oxy"

retro-ene type reaction. Thc 8-membered ring structures 4 and 5 are converted relatively

slowly to the strained pyrans 10 and 11 in a secondary photoreaction. This reaction is not

observed for the ethyl counterpart 6. Most likely, the reaction of 4 and 5 is a photo­

cycloaddition, resembling the Paterno-Buchi reaction. We assume that the reaction develops

from the T 1 excited state of the C/,,B-unsaturatcd ketone, leading to the strained pyran

products, which may arise from a bisallylic-radical intermediate.

4.5 Experimental Section

4.5.1 General procedures

All solvents and commercial reagents were reagent grade and dried with the appropriate

drying agents. Argon was dried over Drierite pellets. Column chromatography was

perlonned using Silica-60 as the stationary phase. Proton, deuterium, and carbon-l3 NMR

spectra were recorded on a Bruker AM400 or a Bruker AM600# spectrometer in CDCI]

solution. Chemical shifts are reported in ppm, relative to tetramethylsilane as an intcmal

standard. Coupling constants (J ) are given in Hz. GC analyses were performed on a Kipp

Analytita 8200 instrument with PID detection (25 m '" 0.22 mm !D, Column type: WCQT

Fused silica, stationary phase CP WAX 51). All HPLC separations were run on a system

which consists of a Waters 6000A pump, a 4 '" 100 mm Lichrosorb 60 (5 fLm) column, and

a Philips Unicam PU4020 UV detector (254 nm). Mass spectral data were obtainw on a

Hewlett Packard 5970A system by electron ionization at 70 eV.

Bicyclof4. 2.1 jnt)no.-2, 4-trie.n.-9-071,(.

This compound was readily preparw according the procedure of Antowiak et al.13 To a

vigorously stirred solution of diethyl ether (250 mL) in a 1 L-threeneck.ed flask equipped

#600 MHz lH NMR spectra were tecotdedOI\!he Jl.r\ll(~r AM600 Nl\.n~. spectrometer of the Dutch National hlNMR facility at Nijmegetl, The Neth"d.u)<h

92 INTRAMOLECULA~ HYDROGEN-TRANSFER

with a mechanical stirrer, and condenser, small pieces of lithium 1.6 g were added. Purging

the reaction mixture continuously with a stream of dry argon. The mixture was cooled to -70

°e and the cyclooctatetraene (lOA g) was added at once. After stirring for 4 h, removing

the dry ice-acetone bath, and stirring fo, another 10 h (ovcmight), sufficient dry diethyl ether

was added to dissolve the precipitated dilithium cyclooctatetraenidc.

Dimethyl-carbamoyl chloride (11.9 g, 0.11 mol) in dry diethyl ether (100 mL) was

added dropwise to the solution of the dilithium compound. The mixture was stirred for 2 h

and warmed to 15~C. Dilute sulphuric acid (3 N, 100 mL) was added to the stirred reaction

at lO°e. the mixture was separated and the aqueous layer was extracted with diethy1 ether.

The diethyl ether was washed with water saturated sodium bicarbonate solution, and water,

dried over magnesium sulphate and concentrated (50 mL) under reduced pressure. Distillation

of the liquid residue yielded bicyclo[4.2.I]nona-2,4,7-trien-9-one (yield 7.0 g, bp 46-4TC

(0.3 mm Hg») as a colourless liquid.

Bicyclo/4.2.1Jnona-2,4-dien-9-one This compound was prepared via selective hydrogenation of the CrCg bond in

bicyclof4.2.1]nona-2,4,7-triene-9-one (scheme 4.4). This reaction is not straightforward, as

is evident from the work of Brown,l4 GillissenHi and Schuster. 17 We used Ni2B as

hydrogenation catalyst, which was prepared as follows: Ni(Ach.(H20)4 (230 mg, 0.92

mmol) was dissolved under an argon atmosphere with magnetical stirring in 20 mL of ethanol

in a 500 mL erlenmeyer flask. A solution of 33 mg NaBH4 in 20 mL of ethanol was

transferred dropwise into the reaction vessel, yidding a black suspension of Ni2B.

Subsequently, bicyclo[4.2.I]nona-2,4,7.triene-9-one (1 g, 7.4 mmol) was added, and

hydrogen gas was allowed into the flask (atmospheric pressure). Stirring was continued until

170 mL (6.9 mmol) hydrogl-'TI gas was absorbed. Ethanol was evaporated, and the residue

was dissolved in 100 mL of dry diethyl ether. Standard work-up (addition of 50 mL saturated

aqueous NalIeO), filtration, and TL1JdIted washing of the aqueous phase with dicthyl ether)

yielded a mixture of the desired diene, and the starting compound (ratio'" 5 : 1), which was

chromatographed on a sHica gel column using dichloTOmethane as eluent (~diene = 0.43;

Rf triene = 0.49). This afforded 460 mg of the desired product (yield 45%).

IH NMR (CDCI]) 0: 5.58-5.96 (48, m, H2.5), 2.69-2.87 (2H, m, H),6)' 2.10·2.46 (4H, m,

H7,g)·

Be NMR (CDCI]) Ii: 216.44 (s, C9), 129.55 (d, Cz,s), 126.12 (d, C3,J, 50.35 (d, C1,6)'

34.55 (t, C7,g).

EI-MS: 134 (m+), 106,91,78 (base peak), 51,39,27.

4_5 Experimental section 93

9-endo-Hydroxy-9-exo--viTlyl-bicyclQ[4. 2. IJnona·2. 4-diene (1)

Magnesium turnings (1.27 g, 52.7 mmol) and 15 mL of dry THF were transferred into

a 250 mL reaction flask equipped with a reflux cooler (connected with a methanol cryostat

(-600C), a dropping funnel, and a mechanical stirrer. During the experiment, dry argon was

slowly passed through the reaction vesseL Vinyl bromide (0.5 g, 4.7 mmol), and a smaIl

iodine crystal were added. After start·up of the reaction, 15 mL of dry THF was added

immediately, followed by the main portion of vinyl bromide (6.25 g, 58.3 mmol) in 10 mL

of dry TIJF, over a period of 1 h. After completion of the addition, the mixture was stirred

for 1 h, and cooled to OQC with an ice bath. Bicyclo[4.2.1]nona-2,4-dienc-9-one (1.68 g,

12.5 mmol), dissolved in 15 mL of dry THF, was added dropwise over a period of 1 h. The

reaction was allowed to proceed for 6 h at OQC, and for 14 h at ambient temperature. Then,

the reaction was tenninated through addition of 35 mL of 3 N HCl at 0 QC. THF was

removed and the residue was mixed with 100 ml of diethyl ether. The organic layer was

washed with NaHC0:l, dried over MgS04 , and concentrated. Column chromatography uSing

dichloromethane as eluent afforded 5.19 g of pure 1 Rf 0.35, yield 91 %.

IH NMR (CDCl3) &: 6.18-6.34 (lB, dd, HlO), 5_75-6.01 (4H, m, H1-S), 5.38-5.49 (lB, dd,

Hll tranJ, 5_12-5_22 (IH, dd, HI! ei$)' 3.07 (lB, d, OH), 2.48-2.58 (2H, m, HI,!i)'

2.02-2.08 (4H, m, H7 s)'

ue NMR (CDC13) 0: '142.81 (d, ClO), 136.13 (d, C2-S), 126.14 (d, C3,4), 113.88 (t, ClI)'

77.49 (s, C9), 49_44 (t, C1•6), 38.17 (t, C7,s)'

EI-MS: 147, 115, 107,95,91, 79 (base peak), 65, 55, 39, 27_

9-e"do..flydroxy-9-exo- (2 '-melhylprapenyl)-bicydo[4. 2. ljnona" 2.4-diene (2)

The Grignard reagent was prepared from I-bromo-2-methytpropene (2.35 g, 11.4

mmol), and magnesium turnings (0.37 g, 15.2 mmol) in 30 mL of dry THF. The reaction

was allowed to proceed for 6 hat 40 Q C. Bicyclo[4.2.I]nona-2,4-diene·9·one (1 g, 7.45

mmol) in 10 mL of dry THF was slowly added in 30 min_ The reaction mixture was stirred

for 48 h, poured in crushed ice, and 100 mL I N NH4Cl solutio!] was added. Subsequently,

the pH of the solution was lowered to 5, via addition of an 1 N HCl solution. Usual work-up

procedure and repeated column chromatography, using hexane-dichloromethane (30 : 10) as eluent, afforded pure 3 (Rf 0.20, yield 0.44 g, 31 %)_

IH NMR (CDCll ) 0; 5.70-6_11 (4H, m, H2_S), 5.45-5.55 (JR, m, HIO), 2.87-2.88 (lH, d,

OH), 2_55-2.72 (2H, m, H I,6), 1.95-2.04 (4H, m, H,.s), 1.91-1.92 (3H, d, CHJ ). 1.77-1.78

(3H, d, CH3).

13C NMR (CDCI3) 0: 138.21 (s, ClI), 136_37 (d, C2-S), 130.59 (d, C IO), 125.76 (d, C3,,v, 16.51 (s, C9), 48.58 (d, C1,6)' 37.92 (t, C7•S)' 27.83 and 20.60 (q, CH3)-

94 INiR.AMOLECULAR HYOltoGEN-TRANSI'ER

EI-MS: 190 (m+), 175, 147, 105, 83 (base peak), 55, 41, 39, 29, 27.

9-endo-Hydroxy-9-exo·ethyl-hicydo[4. 2. J lnona-2, 4-diene (3)

The Grignard reagent was prepared from ethyl bromide (1.21 g, 1 L 1 mmol) and

magnesium turnings (0.27 g, 11.1 mmol) in 30 mL of dry diethyl ether.

Bicyclo[4.2.1]nona-2,4-diene·9-one (1.36 g, 10.1 mmol), dissolved in 15 mL of dry diethyl

ether, was added dropwise in 30 min at OQC. After completion of the reaction (30 min at

oDe and 90 min at ambient temperature), the suspension was poured into 30 g of crushed ice.

Hydrochloric acid (25 mL of a 1 N solution) was slowly added. Standard work-up and

column chromatography with dichloromethane as eluent yielded 0.92 g (54%) of pure 2 (Rf

0.33).

'H NMR (CDCl) 0; 5.80-6.05 (4H, m, H2_5), 2.84 (lH, 5, OH), 2.40-2.55 (2H, m, H 1,6)'

1.95-2.05 (4H, m, H7,s)' 1.69-1.86 (2H, q, H10), 0.92-1.11 (3H, t, Btl).

l:)C NMR CDC1) 0: 136.45 (d, Cz_s), 125.71 (d, eJ.~' 77.44 (s, C9), 48.17 (d, C1,6), 37.83

(t, C lO), 32.05 (t, C7,a) 9.34 (q, C ll).

EX-MS: 164 (m+), 135, 117, 107,91,79,65,57 (base peak), 41, 39, 29.

4.5.2 Irradiation procedure

TA8l...E 4.2 Irradiation of bicyclonolJadienes~

Olefio Amount Time Yield (%)h

(mg) (h) 1-3 4--6 7-9 lO,lt

1 460 1.5 22 41 30 4

10 21 30 30 16

2 440 1.5 23 18 56

8 15 73 6

3 460 1.5 16 34 50 o

10 34 65 o

"Itradiations w<:rt> "ollductcd as del/CTibeJ ill this section u~ing $00 mL of n-bclUIlle> oontaining J mmol of olefin. b

Yields werc de-tennhled by gas-chr()rt]tlto8,rapbic an.a.Iysis of aliquot!l removed petiodic.ally from the itt:adi.won

mixture:.

4.5 Experimental sectiOn 95

All photochemical experiments (as shown in table 4.2) on 1-3 were performed in 11-

hexane (concentrations ... 6 mM), which was dried on sodium prior to use. Irradiations

were performed using a 500 Watt medium pressure Hg lamp (Banau TQ78) through quartz.

Cooling of the lamp, and the reaction mixture was accomplished by means of a closed circuit

filled with methanol. In this way, the temperature in the reaction vessel was maintained

around 0 ~C. Beforc and during irradiation, the reaction mixture was purged by a stream of

dry argon, in order to remove oxygen. All photo-reactions were monitored by means of gas

chromatography. After irradiation, the solvent was removed on a rotatory evaporator. The

reaction mixture was separated by HPLC using dichloromethane/hexane (10 : 90 or 15 ; 85)

as eluent.

4.5.3 CharacterisaJioll 0/ phOtoproducls

The nomenclature of compounds is according to the IUPAC convention. However, in

order to obtain as much consistency as possible, the numbers in the NMR data were assigned

as in scheme 4.4. 4.5 and 4.6, and may differ from the IUPAC convention.

6-(1 '-Qxo-propenyl)-1,3-cis,cis-cycloocradiene (4)

lH NMR (CDC13) b: 6.46 (IH, dd, H1o), 6.26 (IH. dd, H II , traIlJ, 5.95-5.84 (2H, m, H4,s).

3J J{(4)_H{3) = 11.3 Hz, 3JH(4)_H(S) = 4.28 Hz, 3J J{(5)-H(6) = 11.3 Hz), 5_76 (IR, ddt HIl ,

ciJ, 5.65 (2H, m, Hl ,6), 3JH(3)-l-I(2) "" 7.12 Hz. 3JH(3).(2') = 2.5 Hz, 3JH(6)_H(7) ::::; 7.04 Hz,

3JH(6)_H(7') '" 2.50 Hz), 2.93 (lB, m, HI> 3JU(1).H(2) ::::; 9.13 Hz, 3)H(I)_H(2') "'" 2_27 Hz,

3JH(1)_H(S) = 4_36 Hz, 3 JJ{(l)-fJ(8') = 8.86 Hz), 2.48 (IH, m, H2')' 2.48 (lH, m, H7 or H.,,),

2.30 (lH, m, ~), 2JH(2).H(2') ;;;;;; 15.29 Hz), 2.12 (tH, m, HT or H.,), 1.85 (lH, m, Hg or

E g,), 1.63 (IH, m, Hs' or Hg).

Be NMR (CDC1) 0: 204.50 (s. C9), 136.11 (d, ClO), 132.06 (d, Cl or C6), 130.17 (d, C6

Or C3), 129.03 (t, C 11), 128_64 (d, C4 or Cs), 127.15 (d, C j or C4), 45.48 (d, C1), 30.60

(t, Ci). 27.67 (t, C7), 26.75 (t, Cg).

EI-MS: 162 (m+), 147, 145, 107,91.79 (base peak), 55, 41, 39, 27_

(2-DeUlero)-6(1 ' -oxo-propellyl)-l ,3-cis ,cis-cyclooCIadiene (40.)

lH NMR (CDCI3) 6: Identical to 4 except for the H2 signal at "" 2.30 which is lacking.

2H NMR (CnCIJ) {j; 2.32 (lD, S, Dv, 1.64 (lD. s, nl).

#Signal due to (1 ,SJ·H shift.

96 INTRAMOLECULAR ItYDROOSN-TRANSPER

DC NMR (CDCl3) 0: Identical to 4 except for the Cz signal: "" 30_66 (m, ez).

EI-MS: 163 (mo +),162 (mll+)' 148, 147, 107,91,79 (base peak), 55, 39, 27.

6·3 '-methyl-(1 '-oxo-bUl-2 '-enyl) " 1 ,3"ds,cis-cyclooctadiene (5)

IH NMR (CDC!) 0: 6_12 (1H, m, Hu0, 5_93-5_80 (2H, m, H4,j), 5-64 (2H, m, H(),6)'

2.58 (lH, m, Hi)' 2.45 (lH, m, H2,), 2.45 (lH, m, H7 or HT ), 2_24 (lH, ro, Hz), 2.14

(3H, d, Hll or H 13), 2.07 (tH, m, HT or H7), 1.90 (3H, d, Hu or Hd, 1.83 (lH, m, Hs

or Hs')' 1.57 (lH, m, Hs' or Hg).

DC NMR (CDC!) 0: 205.30 (5, C9), 156.77 (s, C ll), 132.18 (d, C lO), 130.64, 128_35 (d,

CJ,6), 126.99,124.15 (d, C4.S), 48_73 (d, Cd, 30.66 (t, Ci), 28.86, 21.85 (q, el2,H)' 27.80

(t, C7), 26.59 (t, Ca>.

EI-MS: 190 (m+), 175, 147,83 (base peak), 55, 39_

6-(1 '-oxo-propyl)-l ,3-cis ,cis-cyclooctadiene (6)

1H NMR (CDC!) a: 5.94-5.80 (2H, m, H4S)' 5.60 (2A, m, H) 6)' 2.63 (lH, m, HI'

3 J H{I)-H(2) "" 3.87 Hz), 2,49 (2H, q, HlO), 2.42 (tH, m, H2,) 2.42 (lR, m, B7 or H7')' 2_27

(lH, ro, Hz), 2.08 (tH, m, H7· or H7), 1.85 (lH, m, Hs or Hs'), 1.60 (IH, m, Hs- or Ha>,

1.06 (3R, t, H ll).

DC NMR (CDCI3) {i: 215_77 (s, C9), 132.08 (d, C3 or C6), 130.2.1 (d, C6 or C3), 128.62

(d, C4 or Cs), 127.01 (d, C j or C4), 47.91 (d, C.), 35_21 (t, C lO), 30.56 (t, Cz), 27.87 (t,

C7), 26.69 (t, Cal, 9.06 (q, C ll ).

EI-MS: 165, 164 (m+), 146, 135, 107,91,79 (base peak), 57, 41,39,29.

9-endo-Hydroxy· 9-exo-vinyl-lricycJo/4. 2. 1. (j.5 ]nona-3-ene (7)

·H NMR (CDCl]) 0: 6.23 (tH, dd, HlO), 6.23 (2H, m, H),4J, 5.40 (lH, dd, H ll , tr<mJ, 5.15

(IH, dd, Htl • ~i~)' 3.33 (2H, ro, H2_5), 1.95 (2H, m, H I ,6)' 1.55 (2H, m, H7,8 or H7"s,),

1.42 (2H, m, H7',8' or H)"S)'

DC NMR (CDCI3) 0: 142.42 (d, ClO), 140.70 (d, C3•4), 115.66 (t, ell)' 93.61 (s, C9),

50.72 (d, CZ-S)' 47_84 (d, C.,6)' 24.63 (t, C7,g).

EI-MS: 161, 147, 134, 107,96,91,79 (base peak), 55, 39, 27.

9-endo-Hydrt)Xy-9-exo-(2 '-mefhyl-propeny/J-tricyclo[4.2.1_ cY.5jno1Ul-3-ene (8)

IH NMR (CDC1]) 0.: 6.27 (2H, m, H3,4)' 5.50 (lH, dd, H IO), 3.29 (2B, dd, H2_S)' 2.03

(2H, m, HI•fi), 1.85 (3H, d, Hl2 or H13), 1.85 (21-1, m, H7,g or H7',8')' 1.74 (3H, d, H13

or Hd, 1.45 (2H, m, H7·,s' or H7.S>' He NMR (CDCl:3) 0: 140.72 (d, ~.4)' 129.65 (d, CIO)' 50.72 (d, C2_j ), 47.70 (d, C.,6)'

4.5 Experimenlal section

27.74 (q, Cn or Cn), 24.39 (t, C7,s), 20_63 (q, C 13 or Cd. EI-MS: 190 (m+), 175, 147, 109,91,83 (base peak), 55, 39-

9-endo-Hydroxy~9--exo-proJlYl-tricyclo{4. 2_1_ rj.5 jnotul-3-ene (9)

97

lH NMR (CDC!3) 5: 6.28 (2H, m, HJ,4), 3_28 (2H, dd, Hz_s), 1.85 (2H, m, H I,6), 1.69

(2H, q, Hlo), 1.51 (2H, m, Hi,a or H7"g')' 1.37 (2H, m, HT,s' or H7,s), 1.01 (3H, t, HIl)­

DC NMR (CDC1) 0.: 141.06 (d, C3,,j, 95.43 (s, C9), 50.60 (d, C2•S)' 4552 (d, C I ,6)'

30.80 (t, CJo), 24,73 (t, C7,~, 9.47 (q, CIl).

EJ·MS: 164 (m+), 163, 149, 135, 107,91,79 (base peak), 57,41,39,29.

9-0xa-8-viny/-tricyclo[5. 2, J, rI· 8 jdec-2-ene (10)

IH NMR (CDC13) 5: 6.13 (lH, dd, H IO), 6.13 (lH, dd, H4 , :3 JH(4)-H(S) = 4,8 Hz, ] JH(4)-lt(3)

= 9,6 Hz, )JH(4)-H(6) '" 1.0 Hz), 5.53 (lH, dd, Hs, J J H(5)-H(6) "'- 3.5 Hz), 5.27 (lH, dd,

Hu , IranJ, 5_12 (lH, dd, Hu , eiJ, 4.44 (lH, m, H], 3Jf:l(3)_H(2) ::::: 6_2 Hz, 3 i HO).H(2') = < 0.1 Hz), 2.70 (lH, m, H6), 2.34 (1R, m, HI' 3Jl-I(1)_H(2) = 10.1 Hz, 3JHm-H(2') = < 0,1

Hz), 2.24 (lH, m, H2 Z)H(2)-H(2') "" 1 1. 7 Hz), 2.10 (IH, m, H7 or Hg or H7, or Hd, 1.63

(lH, m, H2,) 1.63 (1H, m, Hg or H7 or H7, or Hs'), 1.80 (1H, m, tiT or HS or H7 or Hg,),

l.51 (IH, m, Hs, or Hs or H7' or H7).

Be NMR (CnCiJ) 0: 140.80 (d, C IO), 113_92

(t, Cll)' 134.39 (d, CJ, 129_50 (d, C5), 93.51 (s, C9), 73.95 (d, c), 45,92 (d, CJ, 45.63

(t, C\), 44,86 (t, Cy, 32.03 (t, C7 or Cg), 29.42 (t, Cs or C7).

El-MS: 163, 162 (m+), 147, 134, 133, 120, 107,91,79 (base peak), 55, 41,39,27.

(lO-deUlero)-9-tJX.a-8-vinyl-lricyclo[5. 2.1.fj-8Jdec-2~ne (lOa) IH NMR (CDCl]) 8: Identical to 10 except for the RIO signal at = 2.24 which is lacking_

2H NMR (CDC~) 0: 2.21 (lD, s, D0. Be NMR (CDC 1) tJ: 141.03 (d, ClO), 134.20 (d, CoV' 129.59 (d, C5), 113.89 (t, ell)'

93_53 (s, C9) 73.88 (d, Cl ), 46.08 (d, C6), 45.78 (t, CI) 44_61 (m, Cz), 32.02, 29.52 (t,

C7,~-EJ-MS: 163 (mD +),162 (mH +), 148, 147, 129, 108, 107 (base peak). 92, 91,79,55,39.

9-oxa-8~(2 '·methyl-propen·l '-yl)-lricyclo{5.2.1. (j.8Jdec-2-ene (11)

Could not be isolated as pure 11. U's existence was verified by GC analyses (6.5% after

8 h).

98 INTRAMOLlicuLAR HYDROGEN-TRANSFER

References

1. 1. A. Marshall, Acc. Chern. Res. 2, 33 (1973).

2. P. J. Kropp and H. J. Kraus, J. Am. Chern. Soc. 89, 5199 (1967).

3. P. J. Kropp, Pure Appl. Chem. 24, 585 (1970).

4. 1. A. Marshall and H. Fauble, j, Am. Chern. SOC. 92,948 (1970).

5. P. J. Kropp, J. Am. Chern. Soc. 91, 5783 (1969).

6. R. R. Sauers, W. Shinsky and M. M. Mason, Tetrahedron Lett. 4763 (1967).

7. P. J. Kropp, E. J. Reardon Jr., Z. L. F. Gaibel, K. F. Williard and J. H. Hattaway Jr., 1. Am. Chern. Soc. 95, 7059 (1973).

8. A. J. Merer and R. S. Mulliken, Chem. Rev. 69,639 (1969).

9. P. Borrell and F. C. James, Trans. Faraday Soc. 62, 2452 (1966).

to, P. Borrell and P. Cashmore, Trans. Faraday Soc. 65, 2412 (1969).

11. Ca) H. R. Fransen and H. M. Buck, J. Chern. Soc., Chern. Comrnun. 786 (1982). (b) w. J. G, M. Peijnenburg and H. M. Buck, Tetrahedron 44,4821 and 4927 (1988).

12. F. G. West. P. V. Fisher and C. A. Willoughby, J. Org. Chern. 55, 5936 (1990).

13. T. A. Antkowiak, D. C. Sanders, G. B. Trimitsis, J. B. Press, and H. Shechter, J. Am. Chern. Soc. 94, 5366 (1972).

14. C. A. Brown, 1. Chern. Soc., Chern. Commun. 600 (1969).

15. H. M. 1. Gillissen, P. Schipper, P. J. J. M. van 001 and fl. M. Buck, J. Org. Chern. 45, 319 (1980).

16. H. M. 1. Gillissen, PhD Thesis, Eindhoven University of Technology (1982).

17. D. Schuster and C. W. Kim, 1. Org. Chern. 40, 505 (1975).

18. A. Diaz, 1. Fulcher, R. Cetinn, M. Rubio and R. ReymOSQ, 1. Org. Chem. 40, 2459 (1975).

19. A. E. Derome, In Modem NMR techniques for chemistry research. 2nd Ed. Pergamon, Oxford. 1987.

20. P. J. Kropp and H. J. Krauss, J. Am. Chern. Soc. 9l, 7466 (1969).

99

21. R. T. A17I()ld and 1. F. Dowdall, 1. Am. Chem. Soc. 70, 2590 (1948).

22. R. B. Woodward and R. Hoffmann, In The conservation of orbital symmetry, Academic Press, New York. 1970.

23. D. I. Schuster and C. W. Kim, J. Am. Chern. Soc. 96, 7437 (1974).

24. J. March, ln Advanced organic chemistry. 3rd ed. John Wiley & Sons, New York. 1985. p.711, and references therein.

25. C. A. G. Haasnoot, F. A. A. M. de Leeuw and C. Altona, Tetrahedron 36, 2783 (1980). To calculate 31HH, the general Karplus equation has been extended with p:mtmeters which correct for the influence of electronegative substituents on 3 lllli'

3JHH = 13.22 x coslq:, ~O.99 x cosq:, + ELlXi{O.87-2.46 x COS2«(i<P + 19.91.aXi I)} q:, is the proton-proton torsion angle. LlXi is the difference in electronenegativity between the substituent and hydrogen according to the electronegativity scale of Huggins (0 for hydrogen and 0.4 for carbon), and ti a substituent orientation parameter (i.e. ± 1 ).

26. S. L. Murov, L. Yu and L. P. Giering, J. Am. Chern. Soc. 95, 4329 (1973).

27. l. J. Bloomfield and J. S. McConaghy, Tetrahedron Lett. 3723 (1969).

28. G. J. M. Dormans, H R. Fransen and H. M. Buck, J. Am. Chem. Soc. 106, 1213 (1984).

29. G.1. M. Dmmtms, G. C. Groenenboom and H. M. Buck, 1. Chern. Phys. 86, 4895 (1987).

30. G. l. M. DOrmanJ, W. J. G. M. Peijnenburg and H. M. Buck, J. MoL Struct (Throchem) 119, 367 (1985).

31. I. March, In Advanced organic chemistry 3rd ed. John Wiley & Sons, New York. 1985. p. 868.

32. D. R. Arnold, Adv. Photochem. 6, 301 (1968).

33. W. Herndon, Tetrahedron Lett. 125 (1971).

34. N. 1. Turro, Pure AppL Chern. 27, 697 (1972).

35. N. J. Turro, In Modem molecular photochemistry, Menlo Park, California: The Benjamin/Cummings Publishing Co. inc. 292 (1978).

36. W. 1. Nebe and G. 1. Fonken, J. Am. Chern. Soc. 91, 1249 (1969).

37. R. S. H. Liu, LAm. Chem. Soc. 89, 112 (1967).

38. K. Shima, Y. Sakai and H. Sakurai. Bull. Chern. Soc. Jap. 44,215 (1917).

100 INTRAMOJ,.ECULAK HYDROGEN-TRANSFER

39- R. R. Hautala and N. J. Thrro, J_ Arn_ Chern. Soc. 93, 5595 (1971).

40_ J. M. Sunur and M, P_ Bertrand, Bull. Chern. Soc_ Chim. France Sb, 1861 (1973).

41. R. Srinivasan and K_ -'_ Carlough, J. Am. Chern_ Soc_ 89, 4933 (1967)_

42. M. J. S. Dewar, E. G_ Zoebisch, E. P. Healy and J. J. P. Stewart LAm. Chem. Soc. 107, 3902 (1985).

43_ M J_ S. Dewar and J. J. P_ Stewart, QCPE Bulletin_ 6, 24 (1986); QCPE Program NO. 506 (version 2.10).

44. F, A_ L Anet and I. Yavari, Tetrahedron LeU. 43, 1567 (1975)_

45. F. A. I., AnN and J. Yavari, 1. Am_ Chern. Soc. 100,7814 (1978).

46_ N L. AlJinKer, j, P_ Viskocll, U. Burkert and Y_ Yuh, Tetrahedron 32, 33 (1976),

5 Abstract

Excited State Chirality. #

Synthesis and CPL of Optically Pure 3-( lmr J­(IS ~6R)-Bicyclo[ 4.4.0]decane-3 ,8-dione

Low-temperature photoo){ ygenation of (I S,6R)-3-(E,Z.methoxymethylene)-blcyclo[4.4.01de­

can-8-one, 19, prepared in 10 steps from 2,7-dimethoxy.naphthaJene, yields a mixture of four

isomeric l,2-diOJlctanes Sa-d with e.e. > 98 %. Upon th~rmal decomposition, the.~e 1,2-.dioxe­

tanes are all precursors for 3-(IO?r'")-(IS,6R)-bicyclo[4.4.01decane-3,8-dione (1\ an optically

active diketone in its locally excited 1m\"* state. The optical activity of this molecule is evidenced

by the non-vanishing cifcular polarization in the chemiluminescence (CPL) of Sa"'tl. The CPL of

Sa-d with a A.nax of 420 run and a g-factor of emission (g~) of -(1.53 ± 0.35) x 1(}·3 is

comparable to that of the regular f1uOn~sccnce of optically active (IS,6R,8S}-8-hydroxy­

bicyclo[4.4.0Jdecan.3-one 15. This similarity implies that the rate constant for intramolecular lmr+

energy trllIlSfer in 1+ is ~ 10+9 S·l and, hence, no racemization in the eJlcited state is observed.

The chiroptical results show that localization of excitation energy at th~ (;arbonyl at the g-position

of 1'" can be obtained by the synthetic route applied, yielding enantiomerically pure 3-(ln1r*)­(lS,6R)-bicyclo[4.4.0]decanc-3,8-dione (1\ an optically active molecule whose chirality is solely

due to the presence of an excited state.

#F. W. A. M. MicM'n, A, P. p. Wollersheim, S. C. l. Meskers~. H. p, ], M. Oek.kets* aitd KW. Meijer, accepted for publication in I. Am. Chcm. Soe.

101

102 EXCITED STATi:l CHIRALITY

5. 1 Introduction

Molecules in which chirality is solely due to small differences between the discrimi­

nating groups have always fascinated stcrcochemists. I Striking examples arC chiral

tctraalkyl methanes2 and fullerenes,3 and molecules which are chiral due to isotopic

substitution_4 One of the smallest deviations from achirality is found in systems in which

the chirality is solely due to the presence of localized electronic excitation energy. 5,6 This

can arise in centrosymmctric or meso compounds with two remote chromophores of

opposite chirality, as given in figure 1 for 1, 1*, and related structures 2 - 4. In the

ground state these diketones (R,S) are achiral, while in the excited state two enantiomcric

forms (R·,S and R,S*) are present, provided the excitation energy is localized.

1

o-Ctlr 6.a &;10 H H H

2 3 4

f1Gl)RI;;: 5, I Chirality due to the excited state, The diketones 1-4 arc achiral in the ground

state but chiral (1,) in their excited state, due to remote chromophores of oppOsite chirality_

'rhe excited state can be formed by the irradiation of R,S with circularly polarized

light. The difference in concentration of R-,S and S,R" is then governed by the dis­

symmetry factor in absorption (ga-::::; ..iO.D. I 0.0.); in fact thc resultant enantiomeric

excess (e.e.) equals ga I 2_ The optical activity of the excited state species is evidenced by

a non-vanishing degree of circular polarization of luminescence (CPL). Such excited-state

chirality has experimentally been demonstrated in 1,7-diketone 2*.'5 In the case of 1,5-

diketone 3~ a zero CPL was found which was ascribed to a fast racemization in the

5.1 Introduction t03

singlet excited state (lifetime Tf)' due to In'Jl"· energy transfer. The circular anisotropy, as

measured in these experiments, equals g(J.g~ 12 in the absence of energy transfer;

otherwise a lower anisotropy will be observed.S' Such energy transfer reduces the e.e.,

and thus the CPL signal by a factor of (2 .kET• T f + 1)01.

Since in photoseIection experiments the e.e. is governed by ga (which, for the n~·

bands of ketones will give an e.e. of at the most - 10%), a major improvement of

selectivity, has been foreseen by using chemiexcitation to produce R· ,S, potentially with

an e.e. of tOO%.4d,7 According to this approach, singlet excited state 2,4-adamantanedio­

ne 4* has been prepared by the thermolysis of the corresponding optically active 1,2-

dioxetanes. 7 These 1,2--dioxetanes are known to dissociate into two carbonyl fragments

accompanied with chemiluminescence. Upon thermal decomposition, one of the carbonyls

is produced in its (singlet or triplet) excited state (chemlexcitation). The radiative

transition from the singlet excited state to the ground state following chemiexcitation

accounts then for the chemiluminescence. Despite the fact that the excitation is directed

chemically to one of the carbonyls only, a vanishingly small circular polarization in

chemiluminescence of the 1,2-dioxetanes has been detected. A complete loss of optical

activity occurred within the lifetime of 2 ns of 4~ j due to the close proximity of both

carbonyIs in 4, wer-gy tr.msfcr is extremely fastJ

In this chapter we describe the synthesis and the CPL, of optically pure (1n'l"·)_

(lS,6R)-bicyclo[4.4.0]decane-3,8-diolle (1*), S a diketone where both carbonyls are more

remote. The synthetic route is based on our previous work for the chemiexdtation of 4.7,9-

Via thennal decomposition of the optically active 1 ,2-dio~etanes Sa-d, enantiomerically

pure 1- is obtained together with ground state methyl formate (scheme 5.1).

- + en> HCOC~

H

1*

SCHEME 5.1

104 EXCITED STATE CHIRALITY

5.1.1 Circular pOlarization in luminescence

Stereochemical and electronic features of chirnl molecular systems in their ground

state are normally reflected by optical rotation, Optically Rotary Dis.pers.ion (ORD) or

Circular Dichroism (CD).IO Structural features jn chiml luminescent molecular specjes,

however, arC investigated by the emission analogue of CD, mostly referred to as

circularly polarized lumincscence (CPL).11,U (See chapter 1, for a more detailed

description)

The important factors reflecting chirality in ground and emitting states, represented

by CD and CPL, are given by the dissymmetry factors in absorbtion (gd) and emission

(&...,) , respectively. Chiral molecules absorb and cmit left- and right-handed circularly

polarized light to different extent. For such molecules there is circular anisotropy in the

absorption and emission, define<! as:

where € and I are average extinction coefficient and emission intensity, C = (cL +€RJ 1 2

and 1 = (IL +Iill 2, respectively, and therefore -2 $ g $, 2. In general the g-factors

reported arc well below unity, typically < 10-2 , for many compounds like aromatics,

which possess fully aHowed 'Of""""''lI". transitions of large oscillator strength (f ~ 0.5 and €

= to-~ M-I cm"I). By contrast, some carbonyl compounds possessing a fully forbidden

electric dipole n-r· (but magnetic dipole allowed) tmnsition of vcry small f « 10-3) or

€ « I(X) M-1 em-I) show much higher g-factors up to 0.25-0.30 for local C2v symmetry

of the carbonyl Chromophorc. The transition borrows electric dipole intensity from the

perturbations due to the rest of the molecule, resulting in large dissymmetry ralios.

Optical activity in n ........ • transition of a series of ketones has been jnvestigated by

means of experimental CD and CPL by Dekkers and GlosS.1 3 Their study led to the

conclusion that optical activity in these transitions can be understood with the theory of

Moffit and MOSCQwitz, provided proper allowance is made for a legitimate role of the

excited state. The geometry of a ketone in its In ... ~ state differs in two distinct ways from

the geometry in the ground state: the length of the C="O bond has increased, and the

carbonyl is bent out of planc. In the fluorescent transition the electric dipole strength is

increased and is therefore responsible for a 5-10 fold decrease of the dissymmetry factor

(ge) in emission.

Concomitant with the thermal fragmentation of 1,2-dioxetanes to the corresponding

ketones, a bright blue chemiluminescence takes place due to ketone fluorescence only.

5.1.2 Ch€miexcitation via 1.2-dioxetanes 105

(emission maximum at 420 nm). Phosphorescence is effectively quenched by the presence of oxygen under the experimental circumstances used. 9,14 An intriguing phenomenon of

optically active I ,2-dioxetanes~ is that they yield optically active ketones concomitant with

a blue chemiluminescence that is to some extent circularly polarized. Optical activity in

chemiluminescence was firstly demonstrated by Numan, who developed an optically active l,2-dioxctane 6. Upon fragmentation of 6 One of the emitting ketones is chiraL 15

Detennination of the optical activity was accomplished by measuring the circular

polarization of chemiluminescence. 16 The noo-zero value for the circular anisotropy (gl-J

together with the additional results of Meijer for the 1,2-dioxetane 7 prepared from the

enol ether derivative 8, demonstrated indeed that gtwrz is equal to g~ of the corresponding ketone"" (5.1 ± 0.1) X 10-:); no methyl formate emission is observed .. In 6 the degree

of CPL (g11Otl = (3 ± 1) X 10-3 is determined by the distribution of the excitation energy

over the chiral species and adamantanone Assuming that the ratio of emission quantum

yields for both ketones remains the same (tPf for 7 = 0.87 X tPf of adamantanone) at the different temperatures of thermolysis, the distribution of Singlet-excitation energy has

been calculated to be 1:1.2 for adamantanone versus 7. Notc that ge for l,2-dioxetanes is

sometimes denoted as glum'

o~ ~~ CH'O~~ CH

30 to to

6 7 8

FIOUR£ 5.2 Optically active l,2-dioxetanes and enol ether for detennination CPL and

distribution of singlet excitation energy.

5.1.2 Chemiexciuuion via therowlysis of 1.2-dioxetanes

The first and probably most famous examples of chemiluminescence are found in

biological systems, and are mostly to be referred to as bioluminescence. 17,1!! This form of

luminescence is generally catalyzed by an enzyme, which increases the cificiency of

emission of light tremendously. For instance, the quantum yield of the American firefly

(photinus PyraIis) is close to 100%, whereas chemiluminescence of most organic oorn-

106 EXCITED SIA1'£ CHIRALITY

pounds have typical quantum yields of 0.1 to 10%, only.

In the first scientific experiments on bioluminescence, at the molecular level, Robert

Boyle (1627-1691) discovered that oxygen for the emission of light of some fungi was

essential. Through the years, well before they were actually prepared and isolated, 1,2·

dioxetanes were proposed as the key intermediates in numerous chemi- and

bioluminescent reactions. In the firefly reaction (as in all bioluminescent organisms), the

dioxetane intermediate is formed by the oxidative process (addition of oxygen) to thc

lucifcrasc substrate. This dioxetanone has been proposed to dissociate to produce CO2 and

the excited·state product, the decarboxy luciferin. Allhough, from symmetry rules this

excited state was predicted to be a triplet state, which would be non-fluorescent, experi­

ments revealed that to the contrary high quantum yields were observed. This dilemma was

rationalized by Koo and Schuster who proposed that the dioxetane undergoes a Chemical­

ly Initiated Electron Exchange Luminescence (CIEEL).19 In the case of the firefly

luciferin dioxetanone (scheme 5.2), an intramolecular CIEEL from the easily oxidi7.ed

polycyclic heterocycle portion of the luciferin to the high energy dioxetane moiety,

followed by rapid decarboxylation, generates a charge transfer resonance structure of the

product in its singlet state and could therefore be held responsible for the high quantum

yields observed.Also, high singlet quantum yields and instabilities of dioxetanes were

observed, when they were substituted with easily oxidisable or dcprotonated phenolic­

substituents. 20

CIEEL ..

lucif9lin-dioxetanone

SCHEME 5.2

In 1969 Kopecky and Mumford prepared thc first synthetic l,2-dioxetane and demon­

strated that upon thermolysis, 1,2-dioxetanes undergo decompOsition to produce two

carbonyl fragments, accompanied with the emission of light.21 These readily accessible

chemiluminescent substances which closely resemble the bioluminescence of luciferin are

in tact suitable for generating electronically excited carbonyl products, and can therefore

be considered as latent excited states of carbonyls exhibiting nQrmal fluorescence or

phosphorescence. 17 After their synthesis, these dioxetanes have become the centnll

5.1.2 Chemie.xcitation via 1.2-dioxetanes 107

molecules of investigations in the field of chemiluminescence, and are together with

hydroperoxides intennediates in the autoxidation of polymers, unsaturated lipids, and food

products_9 Especially, in the presence of light, these systems may undergo photochemical

oxidation in whiCh singlet oxygen may play a considerable role (vide infra). Simple

isolable 1,2-dioxetanes afford predominantly triplet excited states upon spontaneous (uncatalyzed) decomposition in solution. Actually simple alkyl substituted dioxctancs (e.g

tetrnmethyl-l.2-dioxetane) can yield about 50 triplets in proportion to every singlet excited state.

The thermal decomposition of dioxetanes, the kinetics being first order and usually

unimolecular, involves the breaking of two bonds. Two distinct mechanisms of dioxetane cleavage have been proposed; both dealing with the crucial question: at which stage does electronic excitation occur? It has been suggested that these bonds may cleave either in a

concerted or in a stepwise excitation mechanism. In the concened mechanism22•23 the

four membered ring twists fully synchronously before the bonds are broken. In theoretical considerations orbital and state correlations of ground state dioxetane and excited state of

the dicarbonyl product are discussed. The experimental singlet/triplet ratios are explained

by spin-orbit coupling. The non-concerted mode, proposed by Richardson and O'Neal,24i:

proceeds by a rate detennining oxygen-oxygen bond cleavage, leading to initial fonnation

of a singlet 1,4-biradical. The carbon--carbon bond is then broken in a second step to give

one singlet-excited biradical and one ground·state carbonyl. Triplet biradicals, are formed by internal conversion from the singlet biradical, which may subsequently result in triplet­excited Slate carbonyls_ 23

In the past, evidence has been presented for both type of mechanisms_ 25 Especially

the effect of substituents on the rate of thermolysis has been interpreted in terms of

biwlical intermediates, rather than transition states in the concerted pathway. For example the activation parameters for 3,3-dimethyl-, 3-methyl-3-phenyl- and 3,3-phenyl-

1,2-dioxetane were essentially identical and therefore favour a biradical mechanism. If a concerted mechanism were operative, One would expect increased thermal decomposition

in this series, since the. fonnal substitution of a methyl for a phenyl would result in

conjugation of the phenyl with the developing carbonyl. In absence of substituent effects

most values for activation energies are about 26 kca1/mol, which is approximately the ring

strain energy of the 1,2-dioxetane.

However, on the basis of excitation yields the strict biradical mechanism does not account for the: problem of preferential triplet-state production, whereas the activation

parameters do not hold for the strict concerted mechanism. More concordance was

achieved by the qualitative arguments of Turro and Devaquet.26 The selective production

108 Excf'rnO STATE CHIRALITY

of triplet state carbonyls was reasonably rationalized by the "merged" mechanism in

which the oxygen-oxygen bond is initially lengthened in an activated complex leading to a

biradicaL25,27 Qualitative symmetry considerations for dioxctane thermolysis pointed out

that crossing over points on potential surfaces play an essential role in the mechanism of

excitation. Another hypothesis concerning the fragmentation or stability of 1,2-dioxetanes

has been suggested by Hummelen. 28 Depending on solvent viscosity, the (concerted like)

twisting of the four-membered ring is more or less inhibited by the surrounding solvent

molecules, which implies a higher activation energy, for the molecules with larger

molecular weight.

The present status of dioxetanc thermolysis is now rationalized by a sort of compro­

mise where there may be a spectnlm of biradical-stepwise to concerted reactions involv­

ing a more Or less asymmetric activating complex or exciplex to explain both the

dioxetane stabilities and excitation yield. 29

Since 1,2"dioxetanes serve as an interesting tool for generating excited st.ates, the

concomitant chemiluminescence on thermal activation is well suited for various applica­

tions in biochemistry and clinical chemistry: chemilumincscent labels are frequently used

for analytical purposes, since they allow detection at very low conccntrations.28.30.31

5.1.3 Dye-sensitized photooxygenarion

The most convenient route towards 1,2-dioxetanes is the 1,2-cycloaddition of singlet

excited-state oxygen (02e .:l~), usually referred to as 102) to electron rich oldins. 9,32

Singlet oxygt:n is the first excited state of oxygen and plays an important role in the

degradation of all organic materials and living organisms. The ground state of molecular

oxygen involves assignment of the two highest energy electrons to degenerate molecular

orbitals and according to Hund's role, is a triplet state (lEg·). Occupation of the samC

orbitals gives In addition two singlet states, which are in fact the lowest excited states of

02. The energy diffcrence with the triplet ground state is 22.5 keal/mol for the first

excited state and 37.5 kcal/mol for the second excited state. The multiplicities of ground

state oxygen and the excited states are responsible for the unique behaviour of oxygen in

the presence of electronically excited molecules. Hence, ground state molecular oxygen is

a general and efficient quencher of the Sl and Tl of organic molecuJes. B Quenching is

often accompanied with the generation of 1°2; ~(1Ea) when fonned in solution, is

expected, due to its short life time. to decay rapidly to t02. Singlet oxygen can therefore

be produced photochemicalIy34 (400W sodium or UV 500W filtered by kapton or

5.1.3 Dye-senstitized photow:ygenation 109

chromatelcuprisulfare) with thc aid of a sensitizer. In this process the sensitizer is

primarily excited to its singlet excited state by irradiation. After intersystem crossing

(Isq, the triplet-state sensitizer, can transfer its energy to ground-state molecular

oxygen, generating singlet oxygen, which can react or return to its ground state. In this

triplet sensitized generation by aromatic molecules the Quenching of the triplet sensitizer

is assumed to proceed via an encounter complex in which an electron exchange mechanu

ism is proposed. Only one of the nine encounter pair spin states can lead to quenching by

aromatic electronic excitation transfer in a process which is overall multiplicity allowed. 33

For energetic considerations sensitizers with triplet state energies between 22 and 38

kcalImol can provide only 02(16.g), whereas sensiti~rs with triplet energies in the range

of 38 to 45 kcalJmol can generate both 02(IAg) and 02eEg). Usually dyes are used as a

sensitizer, for example Rose Bengal (Er = 40 kca1/mol), methylene blue <Er "" 34

kcalImol). The disadvantage of these sensitizers is that they are difficult to remove

completely; especially in the presence of labile products. This problem can be overcome

by making use of a heterogeneous sensitizer; the polymer"bounded Rose Bengal.

Singlet oxygen may react with mono-olefins in two different ways. It can fonn the

diox.etanes or allylic peroxides in an ene reaction. The ene reaction has already been

discussed in chapter 3. In this paragraph we focus on 1,2.-dioxetancs, but both reactions

are very closely related. Although the mechanism is still unknown, both reactions have

received considerable mechanistic attention, and in all cases a charge brulsfcr character

plays an important rolc. Proposed mechanisms for these conversions include, like for the

ene reaction, concerted and stepwise sequences proceeding through a biradical, a

zwitterion, or a perepoxide.35 The results of theoretical calculations are contradictory.J6

GYB-CI ab initio calculations rule out the perepoxide intermediate, which has been

claimed as the common first intermediate in favour of a 1,4-biradical, for a long time.37

The proposed charge transfer intermediates of the 102 and ethylene reaction, placed in the

descending order of enthalpy of formation, are a zwitterion, a perepoxidc and a 1,4-

biradical, respectively. However this view has been challenged by Hotokka, Roos and

Siegbahn on the basis of CASSCF calculations.38 On the other hand ST0-3G and unre­

stricted MINDO/3 (UM3) calculations exclude both the perepoxide and biradical

mechanism and favour the concerted procesS.39 Some of the same intermediates have

also been suggested for the 2 + 2 addition of oxygen to electron rich olefins to prodll(:e

dioxetanes. It is not certain whether these reactions proceed through similar brulsition

states, but the pathway to 2 + 2 addition seems to be more polar than the ene route, since

the fonner reaction is favoured by polar solvents in cases where the two reactions

compete. Evidence for dipolar intennediates, by trapping with solvent, has however come

110 EXCITED STATE CHIRALrrY

mainly from substrates in which the cation is stabilized by oxygen or aromatics_ It is

known that enol ethers react with singlet oxygen rapidly,40 to yield isolable 1,2-

dioxetanes. Despite, the electron donating properties of the methoxy group that enhance

the reactivity of the double bond towards electrophllic singlet oxygen, enol ethers beadng

allylic hydrogen atoms also give hydroperoxides as the main byproduct .. via the cne reac­

tion. However, based on previous reports for model systems, product distributions of

dioxetanes vs. hydropemxides for photooxygenation of enol ethers respond to solvent

polarity and reaction temperature. Asveld and Kellogg have investigated thc 102 reactions

(using tetraphenylporphyrin and Rose Bengal as sensitizers) of enol ethers 21-23 as

function of the solvent polarity and reaction temperature. 41

As an example from Asveld's work, photooxygenation in ben7.ene at rOOm tempera­

ture of methoxymethylene-cyC\ohexanc gives nearly exclusively the allylic hydropcroxide.

Due to the conformational flexibility in the cyclohexane ring, the axial hydrogens are

readily available to undergo hydrogen abstraction. If the flexibility is more confined, e.g,

by an tett-butyl group in 23, photooxygeoation at room temperature in benzene results in

dioxetane formation. When benzene is replaced by dichloromethane a considcmble portion

of dioxetane (20%) is formed. Lowering the temperature to -80a C increased the yicld of

1,2-dioxetanes further to 60 % (and 40% hydroperoxides),41

5.2 Synthesis of 1,2-dioxetanes

The synthesis of the optically active dioxetanes was accomplished by the stereoselec­

tive and stereospecific enzyme-catalyzed reduction of the achiral trfJlIS­

bicyclo[4.4.0]dccane-3,8-dione l. 42 This reduction results in the optically active

(lS,6R,8S)-8-hydroxy-bicyc1o[4.4.0]decan-3-onc 15 in an enantiomeric excess of 98%.

Protection of the hydroxyl group and trdIlsformation of the carbonyl into the enol ether

results into a stereo- and regioselcctive incorporation of the electron rich double bond,

essential for preparing the optically pure diketone, in which chirality is engendered solely

by one of the caroonyls in its excited state. Deprotection and subsequent oxidation results

in the precursor for the synthesis the optically active 1,2-dioxetanes, The optically active

1,2-dioxetanes 5 were obtained by photooxygenation of the E- and Z- isomers of 19 with

02 and methylene blue in dichloromethane at low temperature (vide infra: scheme 5.5).

In the following subsections we describe each of the steps as summarized in scheme 5.3.

5_2 Sy7llheSis of 1.2-dioxeulIIes III

JQrVrOC~ ~OCH' (a) ... I I I (b) ..

CH30 CH30 9 10

;X)0 0..(X)° (c) + ...

11 12 m) (d) H O~

0$0 (e) .. .. 13

H 14

H

ili~': mo (f) (g) ... .. o ~

H H

1 15

ili,~AC (h) -liJ:C (i) .. H , ,..

o =- A H 16 OCH3 17

-liJ~: U) .. H : H

A OC~ 18

SClifME 5.3 (a) Na I NH) (b) H+ (c) TsOH, glycol (d) Li I NH3 (e) H" (f) HLADH I

NADH (g) Acp (h) Ph3PfCH20CH31Br I BuLi (i) LiOH G) PCC.

112 EXCITED STATE CI-IlRALITY

5.2. J Synthesis of optically active (J S,6R,8S)-8-hydmxy-bicyclo/4. 4. O/decan .. J..one 1.542 ,43

The route to 15 started from 2,6-dimethoxy-naphthalene 9 as shown in scheme 5.3.

Sodium or lithium reduction in liquid ammonia and THF/EtOH and crystallisation from

acetone resulted in pure 2,6-dimethoxy-2,4,6,8-tetrahydro-naphthalene 10.44 This bis-enol

ether was then hydrolysed with dilute hydrochloric acid giving both 11 and {3,'Y.isomer 12

in a ratio of 9 : I. Since the direct lithium reduction in liquid ammonia of 11 and isomer

separations in the next steps turned out to be difficult for the large scale preparations in

particular, it was ne<:essary to ketalizc the mixture with 1,2-ethancdiol and p-TsOH in

benzene. Replacement of benzene by toluene results again in isomerization to the {j,,,{­

isomer, due to increasing reflux temperature. Separation of the isomers after ketalization

is still difficult; isolated yields of pure material are relatively low due to the rearrange­

ment of the o-,{j-unsatumted ketone. Rapid column chromatography on small scale yields

the pure a,{3-unsaturated ketone 13. Lithium reduction in liquid ammonia followed by

deketalization, yielded the desi red tran s-diketone 1. A lternati vel y, this diketone was also

prepared without purification of the apparently unstable a,{3-unsaturated derivatives. The

cl"Ude mixture of unsaturated ketaJs has been reduced again with Li/NH3 , and then

separated with column chromatogmphy with hexane/ethyl acetate 60;40 as eluent.

Deketali7.ation and column chromatography yielded 1 as a white powder. This diketone

was then crystallized from ethyl acetate. Slow crystallisation from THF resulted in

crystals of the space group P2l/n (see subsection 5.5.1).

Horse liver alcohol dehydrogenase (HLADH), a commercially available en:r.yme

catalyses the stereospecific oxidation and reductions of alcohol and ketone substratcs. 42

This versatile enzyme is nicotineamide coenzyme dependent and is able to reduce, with

ethanol as the coupled substrate for recycling, the symmetrical decalindione 1 with

concurrent regio- and stereospecificity, to give good yield (60%) of enantiomerically purc

(lS,6R,8S)-8-hydroxy-bicyclo[4.4.0}decan-3-one (hydroxy ketone) 15. The rate of

reduction is somewhat low (8 days). The absolute configuration on C3 is S as was

evidenced by Jones et at. by converting the hydroxy ketone to the lmown optically active

trans decanols in a Wolff-Kishner reduction. 42 Diastereomeric (d.e.) and eoantiomeric

excess (e.e.) determinations were carried out on the racemic and the optically active

decalin methyl ethers. Baseline separations of the OCA3 peaks in IH NMR spectra of

these compounds enabled the d.e. (> 98 %) and c.c. (> 98 %) of the optically active

hydroxy ketone reported to have [ct}25p == +24.SQ (c= 0.7, ethanol). To compare the

e.e. and d.e. for our hydroxy ketone 15, we determined the optical rotation under

identical conditions. Prom an analytical sample, crystallized from ethyl acetate, the [af5D

tumed out to be +24.5~, indeed.

5.2.3 Pht)too:cygenation of model enol ethers 113

5.2.2 Symhesis of (1 S,6RJ-3-(E,Z-methoxymethylene)-bicyclo[4. 4. OjdecaJ1--S-one 19

Protection of the hydroxyl group of the hydroxy ketone 15 with an acetyl group by

acetic anhydride in dry pyridine gave quantitatively 16. It should be noted that in the t3C

NMR spectrum of 16 resonances of other isomers (after isolation) were present for < 2%, only. A Wittig reaction on 1(i USing 1.1 equivalent methoxymethylene phosphorane

in TIIF gave the E- as well as the Z-isomer of (lS,6R)-3-(methoxymethylene)·

bicycIo[4.4.0]decanc-8-acetate 17 in a ratio of 1: 1 in 55 % yield. Removal of the protec­

tive acetyl group using LiOR in ethanol proceeded almost quantitatively to give 18 in

95% yield. Oxidation of the hydroxyl group of the (lS,6R,8S)-3-(E,Z·methoxy­

methylene)-bicyc1o[4.4.0]decan·g·oI18 was carried out with a small excess of pyridinium

chlorochromate (PCC), added in portions.4S Since long reaction times and large amounts

of PeC, leads to fonnation of diketone 1, the reaction was monitored by TLC and

stopped at approximately 2 h. Standard work up furnished the optically pure (lS,6R)-3-

(E,Z-methoxymethylene)-bicyclo[4.4.0]decan·g·one (enol ether) 19 as the precursor for

photooxygenation ([t:rJ25D ;;;;;; -54.5 9, C 0.213, ethanol) towards the 1,2·dio;cetanes Sa-d.

(yield 54%).

5.2.3 PhotooxygerltItion of model enol elhers

To prepare optically active 1,2-dioxetanes via dye-sensitized photooxygenation we

have to enhance the ratio dioxetanes versus hydroperoxides as much as possible. There­

fore, we checked the findings of Asveld and Kellogg41 for polystyrene bound Rose

Bengal at different temperatures for model compounds 22 and 23 in dichloromethane (see

table 5.1 and scheme 5.4). For optimization of the synthesis for the optically active

dioxetane, we studied the formation of methoxymethylene-cyclohexane-l,2-dioxetane 2S

and methoxymethylene-adamantane-l,2-dioxetane 24 dissolved in n~heptane.4()(, Methoxy­

methylene·adamantane 20, is a very suitable model compound for preparing a pure 1,2-

dioxetane, due to the fact that formation of double bonds at the bridge head carbons is

extremely difficult (Bredt rule): no allylic hydroperoxides are formed upon photooxygena­

tion. All enol ethers were prepared via the standard methods, uSing methoxy(methylene)­

triphenylphosphonium chloride and a suitable base (BuLl or NaB in DMSO).46,47

Photooxygenation of 22 fOt 3 h at 0 9C, filtration of the sensitizer and concentration

furnished a ratio of 22:78 dioxetane versus hydroperoxide, whereas photooxygenation (3

h) at -60 "C yielded a ratio of 40:60. The ratio dioxetanes vs. hydroperoxides were

detennined from integrations in the If{ NMR spectrum (CDCI3)·

114 EXCITED STATE CHIRALITY

Qr(~ H - ........ ~CCH' 20 24

0=<~H3 ____ m~H,· exCCH, H

21 25 ~ H;, H - +

22 26

~~~ -+<F~H,.~~ 23 27

SCHEMEl 5.4

The results in the photooxygenations of the model enol ethers shown in table 5.1

contlrm that increasing the polarity of the solvent (subsection 5.1.3) and lowering the

temperatures favours 1,2-dioxetane formation at the expense of ene-reaction products.

Howcver, photooxygcnations proceed more efficiently, when methylene blue is used;

irrddiation takes about one hour only. As a result, photooxygenation has been carried out

in diehloromethane at -80 ~C with methylene blue as sensitizer.

Determining the ratio dioxetane versus hydroperoxidc is one thing but isolating the

1,2-dioxctanes is another problem- Whereas Asveld and Kellogg performed their studies

on the crude reaction mixtures, we wen: able to remOve the a1lylic hydroperoxides from

all the 1,2-dioxetanes 24-27 by medium pressure column chromatography on silica-gel at

-30 °C and dichloromethane elution. The collected fractions were analyzed from

dioxetane content by spraying an aliquot of the solution on a hot plate in the dark, upon

which a bright blue light fmm the chemiluminescence of the 1,2-dioxetanc is apparent_

5.2.3 Photooxygenation of model enol ethers 115

TABLE 5.1 Ratio of a1lylic hydroperoxide vs, 1,2-dioxetane formation in the photooxygenation of

enol ethers 20-23 in dichloromethane.

Enol ether

20

21

22

23

temperature [QC]

-60

-80

0

-45

-60

-60

1,2-dioxetane hydroperoxide

[%] [%]

100 0

60 40

22 78

38 62

40 60

43 57

The wlvent of the fractions that produced the most light intensity was removed under reduced pressure (below 0 QC). Analysis by NMR spectroscopy showed dio)(etane

solutions of approximately 95% purity_ The characteristic resonance of the four­

membered ring in 1 H NMR Spectra for all the 1,2-dioxetanes is siruated at 5.1 ppm,

whereas allylic peroxides show resonances at 5.0 and 5.8 ppm. In BC NMR specba of

the 1,2-dioxetanes the characteristic chemical shifts arc located at 90 (8) and llO (d) ppm,

respectively. This is shown for 1,2-dioxetane 25 in figure 5.3.

Il I .,.,. ,......"

1~Q no I

40

(ppm)

FlOURE 5.3 Uc NMR spectrum of methoxymcthylene-cyclohexane-l ,2..dioxetane

2S in CDCI3 _ Note the characteristic carbon resonances at 90 and 110 ppm of the

fuur-membered ring_

I

70

116 Excrmo STAT!'; CHIRALITY

5.2.4 Synthesis of optically active J .2-dioxewncs 5a-d

The optically active l,2-dioxetanes 5a-d were prepared from the optically active (a

mixturc of E and Z isomers, I: 1) enol ether 19 ({(X]25n "" -54.5", c 0.213, EtOH),

without separation of the isomers, according the general procedure in dichloromethane at

·80 °c. The enantiomeric excess (e.c.) is assumed to be equal to the c.c. of the hydroxy

ketone 15. Column chromatography On silica gel at -30 °C of the reaction mixture and

19

E and Z

30% do..~O +

H>Z:H3T ~ 5a

H 0

30% p,--~ +

CH~X-- T ~

40%

3 H Sc

H~H 0 o d +

H oc~ H

28a,b

SCHEME 5.5

H., ~o CHp~

0.......0 H

5b

H 0

c~o" r---1---I H~

0--0 H

5d

H~H 0 o d H OCH H

;'\

28c,d

subsequent dichloromethane/ethyl acet.ate (60:40) elution, afforded a :solution consisting of

a mixture of four 1,2-dioxetanes and four allylic peroxides as byproduCls 28a-d in a ratio

of 6:4. Purification of the 1,2·dioxetanes (by removal of more polar peroxides) was

performed by concentration (,.. 3 X) of the crude dioxetane fractions and subsequent

5.3.1 Decomposition oItm: 1,2-di()JCetanes 117

column chromatography (~30 0c) by using diehloromethane!acetone (96:4) elution,

respectively. As a result, analysis by I H NMR spectroscopy showed the 1, 2·dioxetanes to

be better than 95 % purity (vide infra).

5.3 Characterisation of the 1,2-dioxetanes

Characterisation of the 1,2-dioxetane solution was performed by lH NM:R and 13C

NMR spectroscopy, Circular Dichroism (CD) and Circular Polarization of Luminescence

(CPL), the latter is discussed in morC dctail in subsection 504. Integration of the resonances of the proton of the dioxetane ring at 5.4 ppm in the 1 H

NMR spectrum in toluene-dg reveals that the ratio of the mixture of four isomeric 1,2-

dioxetanes Sa-d is about 1: 1 : 1: L In combination with all characteristic resonances in the

13C NMR spectrum of the dioxetane ring Le. two carbons at ,.. 109.5 and four at ...

89.4 ppm, these data confirm the expected structures in schem~ 5.5.

___ JIL~ ... _ .... _-L-______ ._' ...... __ '

5.5 r '''"-,. J 5~O j

(0 ~~-----'-I··'·

(ppm)

FIGURE 5.4 Characteristic resonances in the lH NMR SpectruM {or dioxetanes Sa-d.

5.3.1 Decomposition o/the 1,2-dioxetanes

The stability of 1,2 dioxetanes is expressed by the rate of decomposition at a certain

tempoature. The decomposition of the 1,Z-dioxetanes has been analyzed by monitoring

the concentration of the diQ;tetanes by IH NMR spectroscopy, by direct chemilumines-

118 EXCITED STATE Ctl.lRALITY

cence decay (CPL) measurements, as well as by CD measurements. All dioxetanes

decompose cleanly with chemiluminescence into the corresponding ketones and methyl

formate as depicted in scheme 5.1. for compound 5a-d.

When the four isomers of the 1,2-dio:xetanes 5a-d were heated, considerable differ­

ences in stability were found. The results for the lH NMR analysis in toluene-dg at 60ue are depicted in figure 5.5. It can be seen that two isomers (solid circles) decompose

remarkably faster than the other two isomers (open circles). The rate constants for the

thermal decomposition of the individual isomeric dioxetanes obey first order kinetics with

estimated k's of 0.00096 and 0.00065 S-l at 60 °C, respectively.

2 .

---~ ~-:.-:.---------..I!:~=,"--*

--r... 0

~~~ .--~-:-~----.. ------

---.. ----------~~-- --.----. ----. ..,.---------.

--------.-------"----Ll_

___ • -0' - --'2-_0._3

~---- ..... __ "tl-Tr-ir_r:r __ -----... -____ ... _ i---. • ---~-_-___ __...I~

-.<~~~~= ----------=~~- --~---

---... ----... _-

0

>. .... (/)

C (I) -1 -C

s:;:

-2

<r-_ -----'!.

----... ___ -----_-.11 --------~---- ir-- __

-3 --.-----

-4 1,.""._--'--.... .. ___ ._.J

o <120 640 1260 1680 2100 2520

time [s]

FIGURE 5.5 lH NMR analysis of the decomposition at 60 °C in tQluene-ds of the four

isomeric l,2-dioxetancs, by monitoring the intensities of the characteristic protons of the four

membered ring. 'lbe upper Jines represent the methoxy protons, whereas the lower lines

represent the olefinic protons. It should he noted that two proton chemical shifts do coincide

(line in the middle).

5.3.1 Decomposition of the 1,2-dioxetanes 119

CD measurements taken on a dilute sample of pure dioxetane in bis(2-methoxyethyl)

ether (diglyme) at fixed wavelength (295 nm) at 80 ~C confirm that the decomposition for

two isomers is faster and than for the other two. A non-first order exponential decay of

the elipticity (AO.D.) is found versus time. Based on the results for the thennolysis at 60

"C, frrst-order decay for the two individual pairs of l,2-dioxetanes is assumed. From

integrntions of the characteristic resonances in the 1 H NMR spectrum we determine the

molar fraction for each of the two pairs of 1,2-dioxetane: 0.54 and 0.46, respectively. The simulation of the of the normalized CD decay is then given by the equation: CDt -

CD"" / COIO-CDo. = 0.54 X e-k1 I + 0.46 X e-1c2 t ; in which kl = 0.0072 and k2 =

0.0023 s·l, respectively.

time Is] o 500 1000 Or----~--

-50

·100 o d -150

<I .200

·250

-300 (a)

1500 2000

; I \~~c~-=.-=(b=)

o 5 10

time [minI

15 20

FIClURE 5.6 (a) The experiment.al thermolytic decay at 80 ·C of the 1,2-dioxetanes Sa-d in

diglyme by monitoring the optical activity by circular dichroism. It should be noted that the

curve fit is not depicted, since both the curvefit and experimental decay do coincide. FIGURE

5.6 (b) The decay of chemiluminescence intensity of the 1 ,2-dioxetanes Sa-d in diglyme.

The normalized experimental decay is simulated with two exponential functions with a

correlation coefficient of > 0.999. CDIO - CD", represents the total CO which is

equivalent to the total dioxetane concentmtion at t ;;;;;; O. Simulation of the CPL intensities

upon thermolysis at 80 °C in diglyme reveal similar rate constants 0.016 and 0.0030 S-I,

respectively. Combining the results of the NMR analysis at 60 °C and those from CD and

chemiluminescence at 80 ~C, estimated values for the activation energy can be calculated

by: Ea = R x (In (~,80 I ~,(0» / (liT so· 1/T60 ). When we neglect the solvent effect on thermolysis of the dioxetanes (R = 1.989 cal/mol/K and T in K), Ea,1 "" 23.5 kca1Imol

and EIl,2 -14.8 kcal/mol for the CD measurements and EIl,1 = 29.1 kcal/mol and Ea,2

= 17.7 kcal/mo1 for the chemiluminescence.

120 ExertED STATE CHIRALITY

5.4 Circular polarization in chemiluminescence

Thermolysis of the optically active dioxetaoes Sa-d is accompanied with

chemiluminesccnt emission, the spectrum of which (figure 5_7) closely resembles the

fluorescence of 1 and 15. As expected, no phosphorescence is observed since 3n'l"+

species formed in the decomposition reaction are effectively quenched undcr the experi­

mental conditions used. Upon thermolysis at 80 °C in bis(2-dimethoxy)-ethyl ether

(diglyme) and collecting the emission at 430 om, we observe48 for the dioxetane 5a-d a

dissymmetry factor of chemiluminescence g~ "" -1.53 (± 0.35) x 10-3 • For a test

sample of mt:thoxymcthylene-adamantane-I,2-dioxetane we found for high emission

quantum yields g(! "" 0 (± 1) x 10-3.'

ge = - 1. 5 x 1 0-3

1\ "\~

-'-IJ "\

. \", I ~~

/ -'\...t--~_V\ . o ., (-. c·- c-·/ I. u •• • ••• -_.-:'::"-::'-'::---'--",

300 350 400 450 500 550 600

wavelength [nm]

FICltJJt£ 5.7 Fluorescence spectrum of the dioxetanes 5a-d showing normal ketone emission.

To determine the excited-state chirality, we measured the absolute circular

anisotropies 8u and ge for an analytical sample of optically pure hydroxy ketone 15. The

factor g" is determined at wavelengths of the maxima in the CD and UV spectra. In the

UV measurements the absorbance, O.D. = log Io f I =€CZ, is automatically recorded,

whereas in thc CD spectra the ellipticity, 8, is the measured quantity. To determine the

'Luminescence e;o.p~ri!l"lellt8 \Wre carried QuI .. t the Leiden Univer~ity by df_ H. P. J. M. Dekket~ and df&_ S_ C. ). Met;!C.llt$.

5.4 Circular polarization in chemiluminescence 121

ga"factor, the ellipticity has to be converted to the same unit in the absorption measure­

ments. Since e -32.90 CZ~€, 8a = LlO.D. / 32.90 D.D.; g~ or glum is determined simultaneously from the CPL I fluorescence measurements. CD, UV spectra are shown in

fIgure 5.8. For 15 we observed a positive Cotton effect in the 265·315 nm range. The

dissymmetry factor in absorption measured for 15 is Sa = 9.6 X 10-2. The diss.ymmetry

factor of fluorescence ge "" -0.85 (± 0.16) X 10-3• The results are given in table 5.2.

0.14 ,--

0.12 0 d 0.1

" (b) c 0.08 1"11

C 0.06 d ..(j 0.04

0.02 (a~ __ ~

/ ---"'----0 --=:-:---,... 215 265 315 365

Wavelength [nml

FIGURE 5.8 AO.D. (a) and O.D. (b) of the hydrOl(Y katonc 15 as function of the wavelength

innm.

TABLE 5.2 Dissymmetry factors of IS in absorption and emission at different wavelengths.

absorbance ga emiSSion ge wavelength

x 102-wavelength

x loJ [nm] [nm]

285 8.2 420 -0.86 ± 1.1

295 9.6 430 -0.85 ± 0.16

300 10.0 440 -LSO ± 0.5

To verify that the g-factor is determined by optically active excited states, only, we

recorded CD spectra of diluted and original 1,2·dioxetane solutions in diglyme before and

after complete decomposition. The CD spectrum of a diluted sample of (lS,6R)-5a~d in

diglyme shows a single negative Cotton effect at A.nax = 295 nm, as givw in figure 5.9.

122 ExCrtED STATE CHUl.ALITY

After thermolysis a.t 80 ~C only a trace amount of optically active impurity gives rise to a

small residual CD of 5a-d. In the origina.l 1,2-dioxetane solution (i.e. identical to the

concentration in the CPL measurements), no CD signal at 430 nm, due to a CI) artifact,

was observed_

Waveleogth [nm]

230 280 330 380 430

o ---__ --(b)----.-.->/

o o o .... c:i c:i -10 <I

-15

(a)

!

I !

480 530

FlOORS 5.9 CD spectra of a diluted 1,2-dioxetane solution (a) before and (b) after

thermol ysis.

5.5 Structural aspects of centrosymmetric and meso diketones

5.5. J Crystal slructure#

The molecular structure of (tS,6R)-bicyclo[4.4.0)decane-3,8-dione 1 is shown in

figure 5.10. The fractional coordinates, along with equivalent isotropic temperature

parameters are summarized in table 5_3_ The bond distances and bond angles between

non-hydrogen atoms are giv(.'T1 in table 5.4. A Crystal with dimensions (mm) 3 x 2 X I

have been obtained from a solution of 1 in tetrahydrofuran, by :s1ow evaporation.

IIX_ffly aJISIysis w~ p"c(otmed by A. Sch~.mte!l, and L KrooD, :8ijvoet Center for Biomolecular Reswr~h, P.e.4u.alaan II, 3584 CH Utncht, the NethcrhlDd$; to be published.

5.5.1 Crystal Structure 123

FIGURE 5.10 X-ray structure of (IR,6S)-bicyclo[4.4.0[decane.3,8-dione 1.

TABLE 5.3 Fractional coordinateS of Non-Hydrogen Atoms, and Equivalent rsotropic Thennal

Parameters Ucq (A2).

X Y z Ue-q

0(2) 0,3235(2) 0.1775(1) 0.3769(1) 0.0664(4)

C(1) 0,3588(3) 0,1524(1) 0.0832(2) 0.0476(4)

C(2) 0.4446(3) 0,1357(1) 0,2684(2) 0.0460(4)

C(3) 0.6915(3) 0,0657(2) 0.3121 (2) 0.0521(5)

C(4) 0.7010(3) -0.0540(1) 0.2069(2) 0.0500(4)

C(5) 0.3686(2) 0.0308(1) -0.0177(2) 0.0395(4)

H(5) 0.239(3) .0.028(2) 0.018(2) 0.061(2)

H(II) 0.481(3) 0.210(2) 0.038(2) 0.061(2)

H(12) 0.184(3) 0.190(2) 0,064(2) 0.061(2)

H(31) 0.829(3) 0.120(2) 0.282(2) 0.061(2)

H(32) 0.7189(3) 0.049(2) 0.436(2) 0.061(2)

H(41) 0.575(3) -0.117(2) 0.245(2) 0.061(2)

H(42) 0.877(3) -0.091(2) 0,234(2) 0.061(2)

124 ExcrI'l>O STATE CHIRAJ..,ITY

The X -ray analysis of 1, which crystallizes in the monoclinic spaccgroup P2 j /n (no. 14),

contlrms the transoid molecular stmcture. Symmetry code (a) -x, "y, -z. Crystal data: a

= 5.2490(1), b = 10.6801(3) , c = 7.9484(2) A, {3 = 96.90t(2t; Z :;:; 2. The

intramolecular distance (R<lb) between the carbonyl is determined at 7.508 A for oxygen

atoms and S.248 A for the carbons (symmetry operation: l-x,-y,-z).

TABLE 5.4 Bond distance..<: (A) and bond angl~ n.

BOnd Length Bond Bond angle

O(2)-C(2) 1.216(2) C(2)-C{l )-C(5) 112.9(1)

C(1)-C(2) 1.497(2) O(2)-C(2)-C(1) 122.2(1)

C(1)-C(5) 1530(2) O(2).C(2)-C(3) 122.0{l)

C(2)-C(3) 1.500(2) C( I )-C(2)-C(3) 115.9(1)

C(3)-C(4) 1.532(2) C(2)-C(3)-C(4) 111.9(1)

C(4)-C(5)a 1.525(2) C(3)-C(4 )-C(5)a 1125(1)

C(5).C95)a 1.524(2) C(1)-C(5)-C(4)a llLl(l)

C(1)-C(5)-C(4)a 110.9(1)

C(4)a-C(5)·C(5)a 110.9(1)

TAlIl.!': 55 PM) and MM calculations for diketones 1-4.

diketone method LlH f carbon-carbon oxygen-O)[Ylen kcal/mol distance A distance

1 PM3 -102.1 5.210 7.449

MM 5.238 7.341

2 PM3 -99.3 6.964 9.385

MM 6.948 9.332

3 PM3 -79.6 4.474 6.&93

MM 4.433 6.886

4 PM3 -90.1 2.456 4.183

MM 2.498 4.352

5.6.1 Discussion on photooxygenation 125

5.5.2 Modelling diketones with J'emil!mpiricai PM3 calculations

To establish the intnunolecular distances (Rab , vide infra) between the carbon atoms

of the two carbonyl groups in meso-diketones 1-4, we optimized the geometries obtained

from MM (conjugate gradient and Charmm forcefieId) with semiempirical

quantumchemical PM3 calculations. 49 The results are given in table 5.5.

5.6 Discussion

5.6.1 Discussion on photoo;rygenation

Photooxygenation of enol ethers carried out in dichoromethane at -80 9C with

methylene blue as sensitizer has proven to be the best method to prepare pure 1,2-

dioxetanes Sa-d. The ratio of approximately 1: 1: 1: 1 indicates that no directing effect for a

preferential singlet-oxygen attack at the enol ether 19 is opera.tive. Dichloromethane is the

most suitable solvent for photooxygenation, it is relatively polar, inert and easily evaJXlr­

ated at low temperatures. The l,2-dioxetanes are stable only when kept in solution below

-20 ac. Evaporntion at low temperatures is possible only when using volatile low boiling

solvents. It should also be noted that concentration to dryness leads to rapid decomposi­

tion of the 1,2-dioxetane. A polar low boiling solvent like methanol results in addition of

methanol, trapping probably a zwitterionic Or pcrepoxide intermediate and methanol is

therefore unsuitable as solvent for preparing pure dioxetanes.

In the preparation of the bicyclo[4.4.0]decane 1,2-dioxetanes some difficulties al'05e

in the isolation and storage of these compOunds. Because of the increased polarity of the

compounds, column chromatography of the reaction mixture according the standard

procedure failed and a more polar elution is necessary. Hence, to obtain pure 1,2-

dioxetane the crude dioxetane solution was concentrated prior to dichloromethanelacetone

(96:4) elution in a second trial. Although the polar solvents were more difficult to remove

at ambient temperatures, we were able to remove these solvents by addition of the

appropriate high boiling solvents toluene-d,. diglyme or heptane under rrouced pressure

by using a vacuum pump.

126 EXCITED STAT!'. CHIRALITY

5.6.2 Discussion on the themJ(}/ysis of 1,2-dioxetanes

Reconsidering the stabilities of the 1,2.dioxetanes 5a-d, the CD spectra revealed that

the decompositition did not obey first order kinetics. To estimate the first order rate

constants of the individual isomers kinetic studies at 60 °C using lH NMR spectroscopy

arc pcrfonncd. Since after decomposition still a residual negative Cotton effect was

visible (CDoo) due to trace amounts of optically active impurities (probably a trace of

{:nc-products) we correcLed and normalized the measured CD effect for the decay corre­

sponding to 1,2-dioxetanes, only. The results indicate that two dioxctancs are consider­

ably more stable than the other two. The difference in stability, observed, originates from

the different energy-content of the four·membered ring for the equatorial or axial oriented

oxygens. So An explanation of the difference in stability must lie in the proximity of

methoxymethylene group and the trans oriented cyclohexane ring, when oxygen is axial

or equatorial; no cOnsiderable differences in steric repulsion arC expected for the E or Z

isomers. 10 5b,d the conformation with oxygen-2 axial, is locked, since both rings adopt

a chair conformation and the large methoxymethylcne is situated equatorial. Due to

greater sterk interactions with oxygen-2 in an axial than in an equatorial position, these

isomers are more stablc to cleavage. Unlike the conformation of 5b,d, that of

stereoisomers Sa,e (with oxygcn-2 equatorial) is expe>:;ted to be more complicated and not

limited to one conformation and the energetically more unfavourable axially-oriented

methoxymcthylene group may lead to a boat conformation. As a consequence dioxetaru.:s

5a.c will have a lower activation energy, which may give rise to a non-first order

exponential decay of the luminescence intensity as well as for CD. The values for the

enugy of activation, although incorrect, are in good agreement with the available data for

activation parameters tor 1,2-dioxetanes. Especially the energy of activation for Sb,d is in

line with the value tor the similarly frozen conformation of cisA-tert-butyl-methylene­

cyclohexane-l,2-dioxetane (24.1 kcallmol). However, it should be noted that in polar

solvents (diglyme) decomposition is faster than in apolar (toluene-ds) solvents.

5.6.3 Discussion on circular polarization in chemiluminesceru::e

Thennolysis of the optically active dioxetanes 5a-d furnishes initially the excited

diketone in its locally excited In'lf- state. The optical activity of this molecule is evidenced

by the non·vanishing circular polarization in the chemiluminescence. Tn the absence of

racemization, the circular anisotropy in emission of the hydroxy ketone 15, g~, should be

5. 7 Concluding Remtlrks 127

- - - - - . SImilar to gl! of the chemllummescent dilretone 1, since both carbonyl groups are

contained in more or less identically chiraIly twisted cyclohexane rings. CPt measure­

ments at 80 °C, and collecting at 430 nm, (bandwidth 30 nm) using a solution of 1,2-

dioxetanes in bis(2-dimethoxy ethyl) ether (see experimental section for further details),

revealed a g~ factor of - 1.53 (± 0_35) X to-3 for the dioxetanes_ Thc correspondence of

this value and the dissymmetry factor in the (light-excited) fluorescence of hydroxy

ketone 15 (ge "" -0.85 (± 0_16) X 10-3, at the same wavelength) implies that race­

mization due to energy transfer is relatively small: kET :5 109 S-I, calculated with Tf = 2

ns. S 1 This result for 1,6-diketone 1 is in line with the available data on the energy

transfer for ketones 2 - 4. The distance between both carbonyl carbons in 1, ~b' as

found from X-ray analysis amounts to R~b = 5.25 A (figure 5. 10). For the 1, 7-diketone 2

energy transfer is relatively slow (kET :s;; 107 s-l) for an estimated Rab ,., 6.95 A (PM3

caleulation49), while for l,5-diketone 3 and l,3-diketone 4 for estimated Rab's of 4.47 A and 2.50 A, respectively, the energy transfer is fast (kET > 1010 s-I).

The CD spectrum of a diluted sample of (l R,6S)-5a-d in diglyme shows a singlc

negative Cotton effect at ~ "" 295 nm, concurring with the octant rule of Moffitt and

Moscowitz.S2 The absence of a CD-signal at 430 nm of the original 1,2-dioxetane

solution (i.e. identical to the samples used in the CPI~ measurements), rules out the

possibility of chiral absorption, due to e.g. n-cr dioxetane transitions of linear polarhed

luminescence of the dioxetanes. The chiroptica1 results indicate that the value of glum of

1 \s detennined only by optical activity due to the presence of a locally excited state in a

high enantiomeric excess, equal to the> 98 % e.e. and d.e. of the. hydroxy ketone 15_

5.7 Concluding Remarks

In conclusion, the results presented in this chapter show the successful synthesis and

detection of CPL of enantiomerically pure (e.e. > 98%) In~·-excited 1*, whose chirality

is solely due to the presence of localized excitation energy. The locaJiution of excitation

energy on the carbonyl at the 3-position of 1, has been accomplished by enantioselective

chemiexcitation of enol ethers to 1,2-dioxetanes 5a-d, which by decomposition are all

isomeric precursors of 3-(ln ... ")-1 *. The optical activity was established from the CPL in

the chemiluminescence of the 1,2-dioxetanes. No racemization, due to relatively slow

intramolecular energy transfer (kET ~ 10 + 9 S - 1) with respect to the singlet excited-state

lifetime (2 ns) was observed. Furthermore, this concept of excited state chirality of

128 EXertED Sl'AH; CHlilALllY

ground-state meSO molecules can be used to observe second harmonic generation (SHG)

in centrosymmetric molecules and assemblies by using a circularly polarized beam. For

instance, in centrosymmetric crystals of a racemic mixture SHG has already been observed. 53

5.8 E~perimental Section

5.8.1 General procedures

Melting points were determined on an Uoicam TAMS heating apparatu~. Infrared

(FT.IR) spectra were recorded on a Perkin Elmer 1605 Fr-IR spectrophotometer. CD

spectra were recorded On a Jaseo spectropolarimeter J600. UV-spectra were recorded on a

Perkin Elmer UV!VIS spectrophotometer Lamba 36. Elemental analy:;;es wcre performed

On a Perkin Elmer 240 apparatus. Optical rotations were measured on a Optical Activity

Ltd AAlO polarimeter. Proton, deuterium and carbon-13 NMR spectra were recorded on

a Bmker AM400 spectrometer using tetramethylsilane (TMS; 0 ppm) as intemaJ standard.

GC analyses were carried out 00 a Kjpp Analytica 8200 instrument with FID detection

(25 m '" 0.25 mm ID, column type: WCOT Fused silica, stationary phase CP SJI~-5 ca, film thickness 0.25 fGm). Chemiluminescence spectra were recorded on a Spex Fluorolog

II emission spectrometer. The circular polarization of the luminescence (CPL) experi­

ments were performed at the Leiden University on a custom built spectrometer, operating

in a photon counting modc. 48 For the measurement of the circular polari7.ation of the

hydroxy ketone 15 a 900 W Xenon arc lamp and a monochromator (Minimate, Spex) was

used as the excitation source. The excitation wavelength was 290 nm; the bandwidth 30

nm. During the chemiluminescence measurements, the temperature was maintained at 80

QC by a thermostated bath.

Column Chromatography was performed using Merck silica gel 60, 230-400 mesh as

the stationary phasc. AnalyticaJ thin layer chromatography was conducted on precooled

TLC plates, silica gel 60 F-254 layer thickness 0.25 m, using UV (254 nm) and Iodine

detection. Preparations of dioxetanes were carried out in a 700 mL reaction vessel.

AU solvents and commercial reagents were reagent grade. Djethyl ether was dried

over CaClz and stored over sodium wire. Dichloromethane was distilled from CaHz,

Pyridine was distilled from KOH pellets. DMSO was distilled under reduced pressure

5_8 Experimental section 129

from CaH2• Oxygen was dried over several traps consisting concentrated sulphuric acid

and two filled with KOH pellets. Tetrahydrofuran (THF) was distilled under dry argon

from sodium in the presence of benzophenone_

2, 6-dimethoxy-l ,4,5, 8-tetrahydronaphthalene (10)

In all reaction flask equipped with a condenser (filled with dry ice in 2-propanol),

thennometer and a gas inlet tube, was collected about 600 mL liquid ammonia at a

temperature of -70 to -40 °C by means of a cooling bath (filled with dy ice in acetone).

During the organometal reductions dry argon was continuously passed through the

reaction vessel. Subsequently, a solution of 2.6-dimethoxy naphthalene (9) (10.0 g, 53.1

romol) , in anhydrous tetrahydrofurane (125 mL) and absolute ethanol (125 mL), was

added slowly to the liquid ammonia. This mixture was now cooled further to -78 °c and

the condenser was replaced by a mechanical stirrer. In a period of 1_5 h about 13 g of

freshly cut sodium (0.57 mol) or 10_15 g lithium (1.4 mol) was added to the reaction

mixture_ A bluish cast to the solution developed which disappeared as the sodium was

consumed. The solution was stirred for a further 30 min., the flask was withdrawn from

the cooling bath, and the ammonia was allowed to evaporate overnight. The residue was neutralized with solid ammonium chloride, and addition 300 mL water produced a pale

yellow solution. Standard workup by extraction with diethyl ether, washing the combined

ether extracts with water (2 x 30 mL), drying over MgS04 and concentration afforded

95% crude material which was purified by crystallisation from acetone in 80% yield. mp

95-96°C_

lH NM:R (CDCy ) 0: 4.65 (2H, t, HJ and H4), 3.55 (6B, s, OMe), 2.73 (4H, m, HI and

H5), 2_65 (4H, m, H4 and Hs).

llC NMR (CDC~) 0: 152.88 (s), 122.75 (s), 90.15 (d), 53.85 (q), 32_57 (t), 31.06 (t).

Anal. caIro. for C12H 160 2: C 74.97, H 8-39; found C 74.92, H 8.29.

IR (KBr) em-1: 2810 (-O-CIi:J, C-H stretch (C-C-OMe), 1210 (=C~O"C~, c-o stretch)_

bicyclo[4. 4. Ojdec-] (2)-ene-3, 8-dione (II) and bicyclo[4. 4. Ojdec-J (6)-ene-3, 8-diOTll! (12)

A mixture of 8-5 g of 10 (44.2 mmol) in anhydrous TBF (40 mL) and a 1 M

aqueous Hel (100 mL) and THF (40 mL) was heated under reflux for I h in a nitrogen

atmoSphere and then cooled to room temperature and extracted with diehloromethane

(4 X, 50 rnL). The extract was dried on MgS04 and evaporated to give 6.9 g crude

product 11 containing 10% of .B,risomer bicyclo[4.4.0]decane-l(2)-3,8-dione 12.

Column chromatography on silica gel (eluent: ethyl acetate/dichloromethane, (10:90»

130 Excrn:o STATE CIIIRALITY

afforded pure 11 (yield 4.42 g, 61 %). An analytically pure sample was obtained by

crystallisation from ethyl acetate mp 92 QC.

11

lH NMR (CDCI) 0: 6.05 (IH, s, Hi), 2.1-2.9 (12H, m), 1.7-1.9 (IH, m).

I3C NMR (CDCI) 0: 208,67 (5), 198.48 (s), 161.84 (d), 126.Ql (s), 45.04 (d), 38.17 (t),

36.46 (t), 35.78 (t), 30.90 (t), 29.67 (t).

Anal. calcd. for 11 C lOH I20 2: C 73.14, H 7.37; found C: 73.35, H 7.21 %.

YR. (KBr): 3049 (C=C-H, C-H stretch), 1710 (C-O stretch, saturated), 1654 (C""O

stretch, OI,S-unsaturated, C=C stretch) cm· l .

12 lH NMR (CDCI) 0; 2.85 (4H, s, H2 and H7)' 2.45-2.55 (4H, t, H4 and H9), 2.30-2.40

(4H, t, H5 and RIO)' .

I3C NMR (CDCl) b: 209.27 (s), 126.38 (5), 43.47 (t), 38.07 (t), 29.66 (t).

8, 8·ethylenedioxy-bicyclo/4. 4. O/dec-l (2)-ene-J-one (13)

A mixture of 90% II and 10% 12 (4 g, 24 mmol), dissolved in anhydrous benzene

(200 mL) containing absolute 1,2-ethanediol (1.35 mL, 24 rumol) and p.TsOH (125 mg)

was. refluxed in a Dean-Stark apparatus for 30 min. After all water had been removed the

mixture was poured into saturated aqueous NaHC03 (300 mL) and extracted with diethyl

ether (4 x, 50 mL). The organic layer was dried with MgS04 and rotoevaporatated and

the yellow-brownish residual oil (4.6 g) obtained chromatographed on silica gel (chloro­

form/hexane/methanol (90:19:1». DC aod 18 NMR spectra revealed that the crude

reaction mixture contained 80% of 13 and 20% of 8,8.ethylencdiox~-bicyClor4.4.0]dec-

1 (6)-ene·3-one.

lB NMR (CDCI) 0: 5.9 (lH, s, H2), 4.05 (4H, s, OCH2), 1.4·2.8 (IIH, m).

13e NMR (CDCIJ) 0: 199.45 (s, C), 164.16 (d, C2), 124.75 (s, C1), 107.71 (s, Cs),

64.41 (t, OCH 2), 64.37 (t, OCH:J, 41.36, 36.32,35.00,34.04,32.13, 29.00.

8,8-nhylenedioxy-bicyclo/4. 4. OJdec-J (6)-ene-3-one

IH NMR (CDC1;l) b: 4.0 (4H, 5, OCH2), 1.6·2.9 (12H, m).

I3C NMR (CDC!]) li: 210.11 (s, C), 124.85 (s, Cl> C6), 107.95 (s, Cg), 64.31 (t,

OCl:-(2), 64.19 (t, QCH2), 43.53, 39.72, 38.31, 30.78, 30,55, 29.23.

(1 R,6S)-B,B-elhyJenedioxy-bicyclo[4. 4. OJdecane-3-one (Up The monoketal 13 (4.0 g, 19.2 mmo1) or a mixture of monoketals (5 g, 24.0 mmo1)

in anhydrous TAF (150 mL) was added rapidly to a mechanically stirred solution of

5.S Experime1'llal section 131

lithium (2.0 g, 0.29 mol) in liquid ammonia (collected in a similar way as for 10), kept

at -78 "C. After stirring for 30 min the blue reaction mixture solid NH4CI (45 g) was added. The flask was withdrawn from the dewar, and ammonia was allowed to evaporate.

The residue was taken up in water (400 mL) and extracted with dichloromethane (4 x, 50 mL). Rotoevaporation of the dried (MgS04) extract gave (lR,6S)-8,8-ethylenedioxy.

bicyc1o[4.4.0]decane-3-one 14 yield 2.4 g (75%). If the mixture of k.etaIs was used,

column chromatography on silica using hexane/ethylacetate 60:40 gave pure 14. Rf =

0.25 for 14 and 0.15 for the ~,1'.isomer.

I» N'MR (CDCI3).5: 3.95 (4H, s, OCH2), 1.2-2.4 (14H, m).

BC NMR (CDC1]) 0: 211.16 (s, C]), 108.80 (s, Cg), 64.31 (t, OCH2), 64.27 (t, OCHz),

47.61 (t, Cz), 41.84 (d, C6 or Cr), 40.94 (d, C r or C6), 38.58 (I), 33.92 (t), 32.87 (t),

30.91 (t).

(1 S,6R)-bicydo{4.4. Ojdecane-3.S-dione (1)

The ketal 14 (4.0 g. 19.2 mmol) in THF (300 mL) and 2 M aqueous HCl (300 mL),

was heated at 50 "c for 1 h. The cooled reaction mixture was the extracted with

dichlommethane (4 X, 75 mL) and the dichloromethanc washed with saturated aqueous

NaHCO], dried (KzC0J,) and roloevaporated. The white solid obtained was (2.8 g, 89%)

recrystallized from elhyl acetate or THF to give the trans-bicycIo[4.4.0]decane-3,8-dione

1. mp 137.5 - 138.1 .. C.

rH NMR (CDCIJ) 5: 2.45·2.55 (2H, m, Hz; H7), 2.30-2.45 (4H, m, H4; H9), 2.08-2.20

(2H, dd, H2; H7), 2.00-2.08 (2H, m, HI; H6), 1.75-1.90 (2H, m, HI; H6), 1.50-1.6:5

(2B, m, H~; HrO>. BC NMR (CDC~) 0: 209.86 (s, C3; Cs), 47.08 (t, Cl ; C7), 41.44 (d, C1; C6), 40.27 (t,

C4 ; ~), 33.06 (t, Cs; C1O>. Anal. calcd. for ClOH 140 2: C 72.26, H 8.49%; found C 72.44, H 8.46%.

IR (KBr): 1710 (C-O stretch, saturated)

(1 S,6R,SS)-S-hydroxy-hicyclo{4. 4. OJdecan-3·one (15)

A 0,1 M phosphate buffer (400 mL, pH 6.5) prepared from 0.1 M NaH2P04 and

0.1 M Na2HP04 solutions of millipore water, was sterilized for about an 30 min.

Diketone 1 (1 g, 6.02 mmol), NAD+ (700 mg), ethanol (6 mL) and HLADH (140 units)

were added to the buffer and the mixture kept in the dark at room temperature for 8

days. After 8-days maximum reaction time, the mixture was worked up by saturating with

NaCl, extracting with CHCI] (4X 50 mL) and rotoevaporntion of the dried (MgS0,J

CHC1] solution. The crude product mixlure was chromatographed on silica gel

132 EXCITeD STAY£ CHIRALITY

(dichloromethane/ethylacetate, 60:40 elution) to give the pure hydroxy-ketone product 15

(680 mg, yield 67%) a~ a white solid (unreacted diketone was recovered); mp (after

crystallisation from ethyl acetate) 77-79 °C; [af5D = +24.5 (c 0.7, ethanol).

lH NMR (CDCI3) 0; 4.20 (lH, m, Hg), 1.20-2.40 (14H, m).

])e NMR (CDO) 0: 211.55 (5, C3), 66.13 (d, Cg), 48.25 (I, C2), 42.90 (d, C t ), 41.61

(t, C4), 38.80 (t, C7), 34.80 (d, C6), 33.31 (t, CS), 32.00 (I, C IO), 27.55 (t, C9).

Anal. caled. for C IOH160 Z: C 71.39, R 8.59% found: C 71.49, H 8.82%.

IR (KBr): 3505 (-OH, broad), 1699 (C=O stretch, saturated) .

(1 S, 6R,8S)-8-acetyl-bicyclo/4. 4. OJdecan-3-on.e (16)

To a solution of 15 (1.5 g, 8.92 mOl) in dry pyridine (18 mI., 0.22 mol) was added

2.5 mL acetic anhydridc (26.8 mmol). After stirring for about 15 h at room temperature

the reaction mixture was poured into water (100 mL) and exlracted with hexane/ether

(l: 1). (4 x, 30 mL). The combined organic layers were washed again with water (2 x, 10

mL), dried (MgS04), and coevaporated with toluene and methanol to give 16 as a white

oiL (1.71 g, yieJd91%).

lH NMR (CDCI) 0; 5.1 (lR, br. HS)' 2.2-2.4 (3R, m), 2.0 (3R, s, CH;l), 1.80-1.95

(3H, m), 1.6-1.75 (lH, m), 1.25-1.50 (5H, m), 1.10-1.25 (IH, m).

Be NMR (CDCl) 0.: 210.55 (s, CJ), 170.08 (s, OCO), 69.13 (d, Cs), 47.79 (t, Ci),

42.13,41.16,35.66,35.23 (1),32.78 (t), 28.75 (t), 27.90 (t), 2Ll2 (q, CH3).

IR (neat); cm-1 1731 (-CO-O- stretch, saturated), 1715 (C=O stretch, saturated), 1367

(O=C-CH) deformation), 1240 (O-C=O stretch).

(l S, 6R,BS)- (3EIZ)-methoxymethylene-hil.yclo[4. 4. Ojdecane-8-acetate (17)

To a stirred suspension of methoxy(methyl)triphenylphosphonium chloride (4.24 g,

12.4 mmol) in 35 mL anhydrous THf' in a nitrogen atmosphere, was added 3.8 mL 2.5

M n-butyllithium (9.52 mmol), whereupon the deep red colour of the elhyIidene

phosphorane was produced. After stirring for about an hour, a solution of (IS,6R,8S)-S·

acetyl-bicyclo[4.4.0]decane-3-one 16 (2.0 g, 9.52 mmol) in anhydrous THF was added

rapidly to the dark red solulion. After stirring for another 2 h at room temperature 10 mL

of a saturated ammonium chloride solution was added. The reaction mixture was diluted

with dicthyl elher (250 mL) and washed with water (3 x, 20 mL). Extracting with ether,

drying (MgS04) and rotoevaporation of the organic layer afforded a dark red residue.

Column chromatography on silica gel (hexane / ethyl acetate, 92:8 elution) gavc 1.25 g of

(1 S ,6R, 8S)-3-methox ymethylene"bicyclo[ 4.4. 0]decane-8-acetate (17) as a colourless oil

(yield 55%).

5.8 Experimental sectiOn 133

tH NMR (CDCI3) 0: 5.8 (2H, s, HU)' 5.1 (2H, br, Hs), 3.55 (6H, d, OCR3), 2.75-2.85

(lH, dd, Rz), 2.65-2.75 (1H, m, H4), 2_05 (6R, s, CR), 0.9-2.1 (26H, m)_

l3C NMR (CDCI3) (\: 170.41 (s, O-C=O), 139.Ql (d, Cit), 138_89 (d, ell)' 117.24 (5,

C:3), 117.17 (s, e 3), 69.99 (d, Cg), 59.11 (q, OCR), 43.82, 42.53, 37.02, 36-98, 36.92,

36.83,34.84,33.49,31.91,29_94,29.71,29.64,27.90, 27.74, 24.86, 21.32.

lR (neat); cm-t 2839 (-O-CH), C-H stretch), 1735 (-CO-O- stretch, saturated), 1687

(C=C strW;h), 1364 (O",C-CH) deformation), 1245 (O-e=o stretch).

(lS, 6R. 8S)-3·metJwxymethylene-bicyclo/4. 4. Ojdecan-8-Ql (18)

To a solution of (IS,6R,8S)-3-methoxymethylene-bicyclo[4.4.0]decane-8-acetate 17

(1.5 g, 6.29 mmol) in 30 mL ethanol was added 15 mL of a aqueous solution of lithium

hydroxide (1 g, 24 mmol). The mixture was stirred for 16 h and the diluted with 250 mL

dichloromethane_ The organic layer washed with water dried (MgS04)' rotoevaporated to

give 18 as a white oil (1.17 g, yield 95%)_ The reaction was monitored by TLC (hexane /

ethyl acetate; 80:20 elution), Rf 0.15; [a]2SD = ·28_8 O(c 0_052, EtOH).

IH NMR (CDC1]) 0: 5_8 (2H, s, Hu ), 3.55 (6H, d, OCH3), 2.85-2.9 OH, dd, H,J,

2.75-2_85 (IH, m, H,J, 1.0-2.4 (24B, m).

13C NMR (CDCI) 8: 138.86 (d, ell), 138_76 (d, Cll)' 117.75 (s, C3), 117_68 (s, c), 66.61 (d, Cs), 66.57 (d, Cg), 59.19 (q, OCH3) , 44.26, 42.95, 39.89, 39-83, 37.16,

36_31,36.27,35.05,33.72,32.61,32_52,32_05,30.08, 27.30, 27.15,25_00_

IR (neat) em,l 3360 (-OH, broad), 2839 (-O-CH), C-H stretch), 1686 (C=C stretch).

(I S,6R)-3-Methoxymethylene-bicyclo{4. 4. OjdecQn-8-one (19)

To a stirred solution of 18 (L3 g, 6.62 mmol) in 40 mL dry diehloromethane in a

nitrogen atmosphere, was added in small portions 1.4 g (5.3 mmol) pyridinium

ehlorochromate (pCC). The reaction was monitored with TLe (eluent: ethyl

acetate/hexane, 20:80) Rf 0_25 and after lh another 0.35 g (1.4 mmol) PCC was added.

After stirring for another hour the contents was poured into 100 mL of dry diethyl ether_

The dark brownish residue was washed with dry diethyl ether (2 x 30 mL) and the ether

layer decanted. The collected ether layers were filtrated over MgS04 and florisiI (mesh

600) and the rotoevaporated. The crude product obtained was purified by chromatography

on silicagcl (hexane/ethyl acetate 80:20), to give 19 as a colourless oil (700 mg, yield

54%), [(~f50 "" -54.5 (c 0.213, EtOH). This product should be stored in the freezer

since it may readily hydrolyse to aldehyde derivatives.

IH NMR (CDCl) 0: 5.8 (2H, br, HlI ), 3.55 (6H, d, OCH;l) , 2.85-2_95 (IH, dd, Hz),

2.75-2.85 (lH, m, H,J, 1.0-2.4 (26H, m).

134 EXCITED ST A. TE CHJR.AlITY

13C NMR (CDCl l ) 0: 211.31 (Cs), 139.59 (C ll ), 139.49 (C ll), 116.25 (C), 116.13

(C), 59.30 (q, OCH3), 48.24, 48.16, 43.20, 42.88, 41.61, 41.24, 41.18, 36.19,35.35,

34.03,33.23,33.16,31.05,29.15,24.08.

lR (neat) em-I: 2837 (-O-CH), C-H stretch), 1713 (C",O stretch, saturate<l), 1687 (C=C

stretch).

5.8.2 General procedure for the synth{~sis of J ,2-dioxetanes derived from e.nol ethers

The photooxygenations of the enol ethers were performed at low temperatures. Amounts

of enol ether varying between 0.3 g and 1 g were dissolved in 450 mL dichloromethane

that contained a catalytical amount of methylene blue trihydrate (10-15 mg, Janssen) or

Rose Bengal B bound to polystyrene (0.5-1 g, Fluka).S4 The irradiations were carried out

with a Hanau TQ78 500W medium pressure mercury lamp, through quam"., using a sheet

of brownish polyimide (Kapton 500H, Du Pont de Nemours) wrapped around the lamp

compartment, as short wavelength filter (cut off 550 nm).28 Cooling of the lamp and

reaction vessel was accomplished by means of a closed circuit filled with methanol of a

cryostat. The temperature of -80 ~C is reached by extra external cooling baths of ethanol

and liquid nitrogen. The mechanically stirred enol ether dichloromethane solution, was

irradiated for I h (3 h, for Rose Bengal B bound to polystyrene) at -80 ·C (nnless stated

otherwise), while oxygen was bubbled through. It should be noted that irradiation in

absence of oxygen, leads to rapid decay of the 1,2-dioxetane.

When the irradiation was stopped, the cold (-50 to -30 °c reaction mixture was

purified hy flash column chromatography on silica gel, cooling the short 1,..'Olumn at -30

~C). For recording IH and DC NMR spectra, an aliquot of the stock solution was

concentrated at 0 QC. During evaporation cold CDClj was added, in such a way that the

1,2-dioxctane is always present in solution. For chemiluminescence spectra the same

procedure was used with the appropriate high boiling solvents, (hept.1.ne, diglyme or

tulucne-dg, all spectrophotometric grade). The 1,2-dioxetane was stored as solution in

dichloromethane below ·20 Q C.

Methoxymethylene-tldamantane (20)7 ,55

To a stirred suspension of methoxymethyl-triphenyl-phosphonium chloride (23.6 g,

68.8 mmol) in dry diethyl ether (125 mL) in a nitrogen atmosphere, was added a solution

of 2.5 M butyHithium (28 mL, 70 mmol) in hexane. After stirring for I h, a solution of

adamantanone (10 g, 67 mmol) in dry dicthyl ether (85 mL) was added dropwise to the

5.8 Experimenlal section 135

rcd-dark solution. The reaction mixture waS then stirred for another 15 h at room

temperature. The excess of ylide was removed with a few drops of water and 50 g of

water and 8 g of ZnCl2 was added to give an insoluble complex with the phosphor

compounds. Thc neatly clear ether layer was decanted and thc residue washed with ether.

The combined ether layers were washed three times with water, dried (MgSOJ, and COn­

centrated. The residue was purified by column chromatography on silica (ethyl acetate)

hexane elution, 1:99), yielding pure 20 as a colourless oil which solidified on storing.

Yield 6.9 g (62%).

IH NMR (CDCl:),) 0; 5.6 (IH, s, H-C=C), 3.5 (3R, s, OCR;}), 2.95 (lH, br), 2.25 (lB,

br) en 2.0-1.6 (12R, m).

Be NMR (CDCI3) 0: 134.79 (d), 126.72 (s), 59.29 (q), 39.83 (t), 38.39 (t), 37.33 (0, 33.99 (d), 29.05 (d), 28.74 (d) .

Methoxymethylene-adanumtane-l.2.dioxetan.e (24)7

Compound 20 (1 g, 7.94 mmol) and methylene blue 10 mg were dissolved in

dichloromethane and photooxygenated for 1 h at -60 °C. Column chromatography on

silica gel according the general procedure afforded pure dioxetane in solution. Small

impurities in the reaction mixture as a result from decomposition of the dye were

removed by trating the solution with activated carbon, and column chromatography at

low temperature.

lH NMR (CDCI:),) 0: 5.4 (lH, s, H-C-OO), 3.45 (3R, s, OMe), 2.7 (IH, br), 2.55 (lH,

br), 2.2-1.4 (12H, m).

BC NMR (CDCI;) 0: 110.41 (d), 93.03 (s), 55.93 (q), 35.75, 35.14, 34.70, 33.31,

31.61,31.20,30.34,26,14,25.80.

Methoxymethylene-cyclolu!xane (20l7

To sodium hydride (4.34 g, 99.5 mmol, 60%, washed with n-pentane) was added

under a nitrogen atmosphere 50 mL of dry DMSO. After stirring for 2 h at rOOm

temperature the solution was cooled in an ice/water bath and methoxymethyI-triphwyl­

phosphonium chloride was added, After 15 min stirring at room temperature freshly

distilled cyclohexanone (6 g, 61.2 mmol) was added dropwisc. The solution was stirred 3

h at room temperature, 2 h at 50 °c and for another 10 h. The mixture was then

fractionally distilled at 30 mm Hg. The fraction at bp 70-80 ac was collected in a flask

cooled at -20°C, washed with potasSium carbonate and extracted five times with n­

pentane. The combined pentane extraCts were dried (MgSO,), concentrated, and frac­

tionally distilled at 37 mm Hg. The fractions with boiling points ranging from bp 68-72

136 EXCITED STAT!! CIIIRALITY

were collected. The fraction of bp 71-72 °C being pure methoxymethylene-cyclohexane

(yield 2.6 g, 33%).·

lH NMR (eDCl]) 0: 5.63 (lH, s), 3.40 (3H, s), 2.08 (2H, m), 1.83 (2H, m), 1.32-1.44

(6H, br)_

lJe NMR (CDCI) 5: 138_72 (d), t 18.31 (d), 59- 17 (q), 30.49 (t), 28.33 (t), 26.98 (1),

26.83 (t), 25.38 (t).

Melhoxymethylene-cyclohexane-l,2-dioxetane (25)

Compound 21 (I g, 7.94 mmol) and methylene bluc 10 mg were dissolved in

dichloromethane and photooxygenated for Ih at -80 QC. Column chromatography on silica

gel acwrding the general procedure afforded. pure dioxetane in solution.

lH NMR (CDCI]) 0: 5_63 (lH, s), 3.40 (3H, s), 2.08 (2H, m), 1.83 (2H, m), 1.32-1.44

(6H, br),

DC NMR (CDC!) 8: 110.27 (d), 90-04 (d), 56_05 (q), 34.51 (t), 29.91 (t), 24.47 (t),

21.91 (t), 21.91 (t).

Melhoxymethylene-cyclododecane (22)

2,5 M butyllithium (23 mL, 57.5 mmol) was added to a suspension of methoxy­

methyl-triphenyl-phosphonium chloride (18.8 g, 55 mmol) in dry ether (130 mL)_

according the procedure for 20. After stirring for 15 min the solution was cooled to 0 0C,

and the cyclododecanone (10 g, 54.9 mmol) in dry diethyl ether (70 mt) was added_

Standard work up and chromatography On silica gel ethyl acetate/hexane elution 2;98).

afforded 22 as a colourless oil, yield 5.1 g (44%).

lH NMR (CDC1)) 8: 5.75 (l H, s, H-C =q, 3.5 (3R, s, OCR), 2.1 (2H, t), 1.9 (2H, t),

1.5-1.4 (4H, m), 1,4-1.2 (14H, m)_

13C NMR (CDCI) 0: 143_50 (d), 116.90 (s), 59.20 (q), 28_95, 25.45, 24.61, 24.45,

24.23,24_07,23_86,23.81,23.41,23.13, 22.88_

Metlwxymelhylene-cyclododecan.e- I ,2-dioxetane (26)

Photooxygenation for 3 hours of 22 (1 g) in the presence of rose bengal bound to

polystyrene in 450 mL dichloromethane, was carried out at 0, -45 and -60 °C respec­

tively. Column chromatography and treating with charcoal gave pure 1,2 dioxetane.

lR NMR (CDCl]) 5: 5,4 (IH, s, H-C-OO), 3.45 (3H, s, OMe) , 2.5-1.2 (22H, m).

13C NMR (CDCI]) 0: 110_12 (d), 93.35 (s), 55.91 (q) and resonances between 32 and 18

ppm_

5.8 Experimental section 137

Methorymethylene-4-tert-butylcyc!ohexane (23)

To a solution of methoxymetbyl-uiphenyl-phosphonium chloride (51.3 g, 150 mmol)

in dry diethyl ether (150 mL) was added 2.5 M butyllithium (33 mL, 82.5 mmol) in a

nitrogen atmosphere. After stirring for 30 min the reaction mixture was cooled to -30 °c and the 4-t-butylcyc1ohexanone (7.7 g, 50 mmol) in 75 mL dry ether added dropwise.

The reaction mixture was stirred for 1 h at -30 "C, 1 h at 0 °C, and one night at room temperature. Standard work up according to 20 (18 g ZnCI2 was added) yielded a yellow

oil (1 t g). Purification was accomplished by column chromatography on silica gel (ethyl

~tateJhexane 1:99) and distillation (5.8 g, bp 103°C at 12 mm Hg), yielding 4.73 g

(52%) pure 22.

IH NMR (CDCl3) 0: 5.6 (lH, br, H·C""C), 3.5 (3H, s, OMe) , 2.8 (tH, m), 2.1 (tH,

m), 2.0-1.7 (3R, m), 1.6-1.5 (IH, m), 1.2-0.8 (3H, m), 0.8 (9H, s, CH3).

IlC NMR (CDCI) 0: 138.41 (d), 118.22 (s), 59.22 (q), 48,45 (d), 32.46, 30.56, 29.03,

27.71,27.60, 25.39.

Methoxymethylen.e-4-teTl-burylcyclohexane-l,2-dioxetane (27)

Photooxygenation, according the general procedure, of methoxymethy1ene-4-ten­

butylcyclohexane 1.0 g (23) and 1.0 g rose bengal bound to polystyrene in 450 mL

dichloromethanc at -60 ac. furnished the pure 1,2-dioxetane.

IH NMR (CDCl3) 8: 5.4 (lB, 5, H-C-OO), 3.5 (3H, s, OMe), 3.0-0.8 (9H, m), 0,8 (9H,

s, CH3).

13C NMR (CDCI) li: 110.20 (d), 90.30 (s) and 8 resonances between 56 and 20 ppm.

l,2-dioxetane of 3-m,etho;cymethylene-bicyclo/4.4.0Jdecan-8-one (Sa-d)

Compound 19 (700 mg 3.5 mmol) and sensitizer methylene blue (10 mg) were

dissolved in 600 mL dichloromethane and photooxygenated for Ih at -80 "C. Column

chromatography at "30 DC using dichloromethanefcthyl acetate 60:40 elution resulted in

removal of the sensitizer, yielding a mixture of dioxetanes and allylic hydroperoxides in a

ratio of 60:40. Pure dioxetane was obtained by concentrating 700 mL crude dioxetane

solution to ,., 250 mL and subsequent column chromatography at -30 DC using

dichloromethanelacetone 96:4 as eluent, respectively. The relative low boiling solvents,

dichloromethaoe, acetone, ethyl acetate were removed in the presence of an appropriate

high boiling solvent: bis(2-methoxyethyl)-ether (diglyme) or toluenc-ds. CPL

measurements were carried out using a concentrated dioxetane solution (A solution of 550

mL 1,2-dioxetane was concentrated into 20 mL diglyme).

tH NMR (d-toluene) 8: 5.26, 5.22, 5.20, 5.18 (4 X lH, S, H II ), 3.28, 3.26 (2 x 3H, s,

138 EXCITED STATE CHIRALITY

OCHJ,) , 3.16 (6H, s, OCHJ ), 2.7-0.7 (envelope).

DC NMR (CDCI3) 5: 211.28, 21l.08 (s, 2 x Cg), 109.60, 109.44 (d, 2 x C 11 ), 89.54,

89.51, 89.43, 89.34 (s, 4x CJ), 56.04, 55.99, 55.69, 55.62 (q, 4x OCR)), 47.25,

47.98,46.91,46.74 (t, 4x C7), 41.16, 40.92,40.67,40.52,40.16,40.92,40.67,40.06,

39.37, 38.02, 37.11, 35.70, 35.38, 34.94, 33.79, 33.52, 32.64, 32.44, 32.28, 30.23,

29.2,28.04,27.7,26.27.

References

1. (a) K. Mislowand P. Bickart, hr. J. Chern. 15, 1 (1976177). (b) M. Farina and C Morandi. Tetrahedron 30, 1819 (1974). (c) L. Verbit. Progr. Phys. Org. Chern. 7, 51 (1970). (d) E. L Elid, 1. Am. Chern. Soc. 71, 3970 (1949). (e) E. L. Alexander and A. G. Pinkus. J. Am. Chem. Soc. 71, 1786 (1949).

2. (a) H. ll)nberg, G. L Hekken, J. P. M. Houbiers and H. W. Bosch, J. Am. Chern. Soc. 87, 2635 (1965). (b) H. ll)nberg and L A. Hulshoj; Tetrahedron 30, 1775 (1974). (c) W. Ten Hoeve and H. ",),nberg, J. Org. Chern. 45, 2754 (1980) (d) J. March, In Advanced organic chemistry, John Wiley & Sons: New York, 1984, 3rd ed., p. 86.

3. (a) R. Eul, l. Chuo, F. Diederich and R. L. Whetten, Nature, 353, 149 (1991). (b) F. !)jederich. R. L. "-'hetten, C. Thilgen, R. Eul, I. Chao and M. M. Alvarez, Science 254, 1768 (1991). (c) 1. M. Hawkins and A. Meyer, Science 260, 1918 (1993).

4. (a) G. Barth and C. Djerassi, Tetrahedron 37, 4123 (1981) and references therein. (1)) R. F. R. Dezentje and II. P. J. M. Dekkers, Chern. Phys. 18, 11 (1976). (c) w. C. M. C. Kokke and L. 1. Oosterhq{f, J. Am. Chern. Soc. 94, 7583 (1972). (d) E. W. Meijer and H. Wynberg, I. Am. Chem. Soc. 104, 1145 (1982). (e) D. Arigoni and E. L. Eliel, Top. Stereochem. 4, 127 (1969) and references therein. (t) J. W. Corrifonh, Tetrahedron 101, ISiS (1974). (g) J. Reter. J. Seibl, D. A rigoni, 1. W. Cornforth, G. Ryback, W. P. Zey/emaker and C. Veeger, Eur. L Biochem. 14, 232 (1970). (h) J. w. Cornforth, J. W. Redmond, H. Eggerer, W. Buckel, and C. GUlsclww, Eur. J. Biochem. 14, 1 (1970). (i) J. W. Cornfonh, Quart. Rev., Chern. Soc. 23, 125 (1969). (j) A. Colkt and G. Gottarelti, J. Am. Chern. Soc. 104, 7383 (1982). (k) A. Collet and G. Gottarelli. J. Am. Chern. Soc. 103, 5912 and 204 (1981). (i) J. M. Friedman, S. Freeman and J. R. Knowles, 1. Am. Chern. Soc. no, 1268 (1988) and references therein. (rn) C. E. Grimshaw, S. G. Sogo. S. D. Copley and 1. R. Knowles, J. Am. Chern. Soc. 106, 2699 (1984).

5. (aJ P. H. Schippers and H. P. 1. M. Dekkers, J. Am. Chern. Soc. lOS, 145 (1983).

References 139

6. P. H Schippers, PhD Thesis, Leiden University, 1982.

7. (a) E. W. Meijer and H. l+)'nberg, Angew, Chern. 100, 1004 (1988); Angew_ Chem. lot. Ed. Engl. 27, 975 (1988)_ (b) E. W. Meijer and H ~nberg, J. Am. Chem_ Soc, 104, 1145 (1982).

8. To obtain as milch consistency as possible, numbers in IUPAC nomenclature for all bicyclo compounds were assigned as in 1-, in which priority is given to the carbonyl in its lo-r+ state.

9. A. A. "'rimer, Ed_, In Singlet O2, CRC Press, Inc, Boca Raton, FL, 1985_

10. F_ Giardelli and p, Salvadori, Eds., In Fundamental aspects in recent developments in ORD and CD, London (1971).

1 L F_ S_ Richardson and J, P_ Riehl, Chern. Rev. 77, 773 (1977)_

12. 1. P. Riehl and E, S. Richardson, Chern. Rev, 86, 1 (1986).

13- H. P_ 1. M. Dekkas and L. E. Closs, 1. Am. Chern, Soc- 98, 2210 (1975).

14_ H_ Numan, PhD Thesis, University of Groningen, 1978,

15. H. Wynberg and H Nunum, J. Am. Chern. Soc. 99, 603, (1977).

16, H KYnberg, H. Nwnan and H. P_ 1. M. De/d,ers, J. Am. Chern. Soc. 99, 3870 (1977).

17. W. Adam and G. alenlo, Eds, In Chemical and biological generation of excited states., Academic Press, New York, (1982),

18. (a) J. W. Hastings and T. Wilson, Photochem, PhotobioL 23, 461 (1976). (b) T. Wihon, In Singlet O2, A, A. Frimer, Ed., CRe Press, Jnc, Boca. Raton, FL, Vol. il, chapter 2, 1985 and references therein.

19_ 1. -Yo KDo and G. B. Sc!utster, J. Am. Chern. Soc. 100, 4496 (1978).

20. K. A_ Zakliko., T. Kissel, A. L. Thayer, P_ A. Burns and P. Schaap, Photochem. PhotobioL 30, 35 (1979).

21. K_ R_ Kopecky and C. Mumford, Can. J_ Chem. 47, 709 (1969).

22_ N 1. Thrro and P. Lechtken, Pure Appl. Chern. 33, 363 (1973).

23. (a) F. McCaprQ, Pure AppL Chern. 24, 611 (1970); (b) D_ R. Keams, Chern. Rev. 71, 395 (1971).

140 ExcrrED ST ATE CHI~Al.ITY

24. (a) II. E. O'Neal and W. fl. Richardson, J. Am. Chern. Soc. 92, 6553 (1970). (b) H. E. O'Neal and W. li. Richardson, 1. Am. Chern, Soc. 93, 1828 (1971). (c) W. H. Richardson, F. C. Montgomery, M. B. Ydvington and lJ. E. O'Neal, J. Am. Chern. Soc. 96, 7525 (1974).

25. E. J. II. Becham and T. Wilson, J, Org. Chem. 45,5261 (1980).

26. N. J. Thrro and A. {)evaquet, J. Am. Chern. Soc. 97, 3859 (1975).

27. W, Adam and W. J. Baadu, 1. Am. Chern. Soc. 107, 410 (1985).

28. 1. C. Hummelen, PhD Thesis, University of Groningen, chapter 3, (1985).

29. w. H. Richardson, C. Batinica, J. Janota-Perret, T. Miller and D. She.n, 1. Org. Chem. 56, 6140, (1991).

30. M. DeLuca and W. D, McElroy, &:Is., In Bioluminescence and chemiluminescence. basic chemistry and analytical applications, Academic Press, New York, 1981, and references therein.

31. T. M. Luider, PhD Thesis, University of Groningen (1988).

32. (a) II. H. Wassermnn and R. W. Murray, Eds, In Singlet oxygen, Academic Press, New York, NY, (1979). (b) L. M. Stephenson and M. S. OrjGTU)pouim', Acc. Chern. Res. 13, 419 (1980).

33. (a) A. A Connan, I. Hambleft and M. A. J. Rodgers, J. Am. Chern. Soc. 111, 5557 (1989). (b) J. Saltiel and B. W. Atwater, Adv. Photochem, 14, 1 (1988). (c) A, A. Gorman and M. A. J. Rodgers, J. Am. Chern. Soc. 106,4679 (1984).

34. N. V. ShinkarenJw and V. B. Ale.~hmkii, Russ. Chern. Rev. 50,406 (1981),

35. (a) N M, Hasty and D. R. Kearns, 1. Am. Chem. Soc., 95, 3380 (1973). (b) C. S. Foote, S. Mazur, P. A. Burns, and D, Lerdal, 1. Am. Chern. Soc" 95, 586 (1973). (c) N, R. Easton. F. A. L. Anet, P. A. Burns and C. S. Foote, J. Am. Chem. Soc. 96, 3945, and 4339 (1974). (d) H. Takp.shita, T. Hatsui and O. Jinnai, Chern. Lett 1059, (1976).

36. K. Yamaguchi, In Singlet °2, A. A. Frimer, Ed., eRC Press, Inc, Boca Raton, FL, Vol. III, chapter 2, 1985.

37. L B. Harding and W. A. Goddard 111, J. Am. Chern. Soc" 102,439 (1980).

38. M. Hotokka, B. Roos and P. Siegbahn, 1. Am. Chern. Soc. 105,5263 (1983).

39. K. Yamaguchi, S. Yabu.~hita, T. Pu{~no and K N. Houk, J. Am. Chern. Soc. 102, 5043 (1981).

Rifi!.Tences 141

40. (a) P. D. Bartlett and A. P. Schaap, J. Am. Chern. Soc. 92, 3223 (1970). (b) E. W. Meijer and H. "ynberg, Tetrnhedron Lett. 3997 (1979). (c) E. W. Meijer and II. ~nberg, Tetrahedron Lett., 785 (1981).

41. E. w. H ASVf!ld and R. M. Kellogg, J. Arn. Chem. Soc. 102, 3644 (1980).

42. (a) D. R. Dodds and J. B. Jones, J. Am. Chern. Soc. t to, 577 (1988). (b) D. R. Dodds and J. B. Jones, 1. Chern. Soc., Chem. Commun. 1080 (1982).

43. J. B. Jones and D. R. Dodds, Can. J. Chern. 65, 2397 (1987) and references therein.

44. W. Kotlarek and M. Kacor, Bull. Acad. Polon. Sci. SeT. Sci. Chim. 9, 907 (1961).

45. a) E. J. Corey and J. W. Suggs, Tetrahedron Lett. 2647 (1975). b) E. J. Corey and G. Schmidt, Tetrahedron Lett. 399 (1979). c) G. Piancatelli, A. Scettri and M. D'Auria, Synthesis, 245, (1982).

46. G. Wittig, w. BlJll and K H. Kruck, Chern. Ber. 95,2514 (1962).

47. R. Greenwald, M. Chuylwwsky and E. J. Corey, J. Org. Chern. 28, 1128 (1963).

48. For a description of the instrument, see R. B. Rexwinkel, S. C. J. Me.skers and H. P. J. M. Dekkers, AppL Spcctrosc. 47, 731 (1993).

49. (a) J. J. P. Stewart, J. Compo Chern. 10, 209 (1989). (b) J. J. P. Stewart, 1. Compo Chern. 10, 221 (1989).

50. A. L. Baumstark, F. Niroomand and P. C. Vasquez, J. Am. Chern. Soc. 49, 4497 (1984).

51. N. J. Turro, In Modem molecular photochemistry; Benjamin/Cummings: California, 1978.

52. W. Moffit, R.B. Woodward, A. Moscowitz, W. Klyne and C. Djerassi, J. Am. Chern. Soc. 83, 4013 (1961).

53. (a) E. W. Meijer, E. E. Hovinga and G. L. J. A. Rikken, Phys. Rev. Lett. 65, 37 (1990). (b) E. W. Meijer and E. E. Havinga, Synth. Met. 57,4010 (993).

54. (a) J. J. Lambens and D. C. NeckeT$, J. Am. Chern. Soc. 105, 7465 (1983). (b) R. S. Hcurdlry, Tetrahedron Lett. 3183 (1985). (c) A. P. Schaap, Tetrahedron ldt. 1159 (1987).

55. A. H. Alberts, H. Wijnberg and J. Staling, J. Synth. Commun. 2, 79 (1972).

SumttUlry 143

Summary

This thesis describes a multidisciplinary study on the combination of

stereochemistry and mechanistic organic photochemistry of chiral bichromophoric

compounds. In particular the photooxygcnation and photochemical rearrangements of

these compounds are discussed. Furthermore, the synthesis, in high optical purity, of a

molecule whose chirality js solely due to the presence of a locaUzed excited state is

investigated; a phenomenon belonging prior to this research to the realm of fantasy.

Some of the bichromophoric compounds possess a 2-prop-l-enol unit with the

feasibility for a diastcreose1ective photochemical intramolecular rearrangement, conform

the observations for gennaerol and racemic mixtures of tctnl.a.lkyl-ethylidene-­

naphthalenols. According to quantum chemical calculations of Dormans, Fransen and

Buck on the mechanism of this photochemical rearrangement in 2-propenol, excitation of

2-propen-l-01 involves a 90(> twisted ene moiety, which is accompanied by charge

separation (sudden polarization). For 2-propenol this separation of charge is dictated by

the neighbouring electronegative OM functionality. As a result, the hydroxyl group then

shifts to the positively charged tenninus of the allylic fragment in the plane of the double

bond. This wplanar" photochemical [1,3]-OH shift involving sudden polarization was also

claimed to occur for racemic 2-naphthalenols. The results of chapter 2, 3 and 4, show

that the photochemical behaviour of 1,2,3,4-tetrahydro-4-ethyl-l, I ,4-uimethyl-(3Z)­

ethylldene-2(lH)-naphthaJenols, synthesized in their optical pure form, and endo hydroxy·

bieyclo[4.4.0]-nonadiene derivatives, is quite different with respect to the occurrence of

the sigmatropic photochemical shifts, and more common reactions have been found

instead.

Chirality solely due to the presence of a localized singlet e)(cited state was first

noted and discussed by Schippers and Dekke.rs for the irradiation of an achiral 1,7-

diketone with circularly polarized light The local excitation of such achiral ketones gives

rise to two enantiomeric fOnTIs, provided the excitation remains localized at the carbonyl

group. However, for a locally singlet-excited 1,3-diketone, synthesized by Meijer, via

thermolysis of an optically active l,2-dioxetane, racemization occurred within the excited

state. Due to the close proximity of the carbonyls, energy transfer was too fast to observe

optical activity of the excited state. The results in chapter 5 show, however, that one of

the smallest deviations from achirality, Le. chirality solely du~ to the presence of an

excited state, can be accomplished in high enantiomeric excess.

144 SUMMARY

Chapter 2 describes the synthesis and characterisation of optically pure (2S,4R)­

and (2R,4R)-1,2-3,4-tetrahydro-4-ethyl-l, 1 ,4.trimelhyJ.(3Z)·ethylidene-2(1 H)-naphthalen­

ols. A seven-step procedure has· been developed. The optically pure R-2-methyl-2-phenyl­

butanoic acid, obtained by resolution of the quinine salts of the racemic butanoic acid,

was employed as the key starting material. The acylation of isobutene with the optically

active 2-methyl-2-phenyJ·butanoyl chloride, followed by a Friedel-Craft cycJil..ation, an

oxidation and a Wittig reaction gave two isomeric 3,4-dihydro-ethyl-trimethyl-(3Z)

ethyIidene-2(lH)-naphthalenones. Separation of the isomers by HPLC and reduction gave

the optically pure target molecules for the purpose of photochemistry and

photooxygenation described in chapter 3. All compounds were fully characterized with

one- and two·dimensional NMR spectroscopy, including INADF.QUATE.

In chapter 3 the reinvestigation of the photochemistry of the optically pure 2-

naphthalenols is described. The results show that besides E·Z isomerization, indanes are

formed by the elimination of formaldehyde. Semiempirical MNDO, AMI, and PM3

calculations revealed that in the preferential ground-state conformation the hydroxyl group

(l(Xupies an orthogonal position with respect to the exocyclic double bond. Fluorescence

measurements confirm, that the reactive state is the triplet state. In fact we have been

unable to identify photoproducts via a planar [l,3]-OH shift as claimed for racemic 2-

naphthalenols. As a matter of fact, the results in this chapter lead to a negative conclusion

with respect to the feasibility and actual occurrence of the [1,3]-OH shift for the (2S,4R).

and (2R,4R)-1 ,2-3,4-tetrahydro-4-ethyl-l, 1 ,4-trimcthyl-(3Z)-ethylidene-2-naphthalenols.

Since photooxygenation of chiral alcohols attained increasing interest, we have

investigated the photochemistry of the optically active naphthalenols in the presence of

o~ygen. The results show that the double bond is rapidly ox.ygenated which leads to

stereoselectivc formation of an cpoxide and an ally lie hydroperoxide. The observed regio­

and stereospecificity is explained by the selective coordination of the hydroxyl group in

the ehiral 2.propen-l-ol unit with the attacking c1cctrophilic oxygen.

In Chapter 4 the photochemistry of 9-endo-hydroxy-9-"C~o-vinyl"bicyclo[4.2.1]

nona-2,4-diene analogues is described. On the basis of deuterium labelling experiments,

conformational analysis and one- and twodimensionaJ NMR spectroscopy, it was

established that a light-induced intramolecular formal hydrogen transfer occurred towards

one of the termini of the endocyclic diene. The hydrogen transfer takes place on the endo

side of the molecule and can also be considered as an oxy retro-ene reaction. Prolonged

irradiation revealed an intramolecular cycloaddition of the a,/J·unsaturated ketone in a

secondary photoprocess, in which strained pyrnns were fonned via a triplet bisa1lylic

biradical intermediate.

Summary 145

The results in Chapter 5 show that a compound, whose chirality is solely due to

the presence of localized electronic excitation energy, can been achieved_ The optically

active excited 1,6- diketone with an optical purity> 98% and a lifetime of ... 2 ns was

prepared by thennolysis of an optically pure 1,2-dioxetane. The synthetic route is mainly

based on the steJ"eO" and regiosclcctive enzymatic reduction with HLADHfNADH and

low-temperature photooxygenation of electron rich olefins_ The optical activity was

evidenced by a non-vanishing circular polarization in the chemiluminescence. In addition,

the work described in this chapter, shows, that circularly polarized chemiluminescence

generated from a 1,2-dioxetane precursor, provides some insight into intramolecular

energy transfer in diketones since this process would provide for racemization of the

chiral excited state_

146 SAMENVATI1NG

Samenvatting

In dit procfschrift wordt cen multidisciplinair onderzoek bcschrcvcn naar de

combinatie van stereochcmic en cxpcrimcntclc fotochemie van chiralc verbindingen die

twee Iiehtgevoelige gedeelten (chromoforen) bezitten. Met name de foto·oxidatie en het

mcchanisme van fotochemische omleggingen van deze verbindingen zijn bestudeerd_

Bovendien is de synthese, in een hoge optische zuiverhdd, van een verbinding onderzocht

die de chiraliteit sle<:hts dankt aan de aanwezigheid van een gelokaliseerde aangeslagen

toestand; ren fenomeen dat voor dit onder:1.Oek behoorde tot het rijk der fantasie_

Ecn gedeelte van dere "bichmmofore'" verbindingen bezit ren 2-propcn-l-ol

fragment, dat de mogelijkheid ber-it voor een diastereoselektieve fotochemische omlegging

overccnkomcnd met de waamcmingcn voor germacrol en raccmische mengsels van

tetraalkyl-ethylideen-2-naftalenolen. Volgens kwantumchemische berekeningen van

Dormans, Fransen en Buck naar het mechanismc van dezc fotochemische reactie in 2-

propen-I-ol, leidt de excitatie van 2-propen-I-ol tot een 90° gedraaid alkeen . .fragment,

waarbij een ladingsschelding optreedt (sudden poJari,,.ation). Deze ladingsscheiding in 2-

pmpen-lol is door de naburige elektmnegatieve OH-gnx..-p gcstuurd. De hydroxyl-groep

migrccrt dan in het vlak van de dubbele binding naar het positief geladen eind van het

ailylisch fragment. Gesteld werd dat deze "pJanaire" fotochemische [1,3]-OH shift, en het

optreden van "sudden polarization" ook voor racemische 2-naftalenolcn plaatsvindL De

resultaten van hoofdstukken 2, 3 en 4 toncn aan dat het fotochemisch gedrag van de

gesynthetiseerde optisch wivere l,2,3,4.tetrahydro-4-ethyl-l ,1 ,4-trimethyl-(3Z)-ethyli­

d«n-2( I H)-naftalcnolen, en cnda hydrox y-bicyclo[ 4.4. O]-nona-dieen derivaten nogal

verschilt met betrekking tot hd optre(icn van sigmalrope fotochemische shifts. Meer

gebruikelijke reacties bleken plaats tc vinden.

Chiraliteit slcchts veroorzaakt door de aanwezigheid van ren gelokaliseeroe

singulet aangeslagen toestand werd voor het eerst bemerkt Cn besproken door Schippers

en Dek:kers voor de bestraling van een achiraal 1,7 -dik.eton met circulair gepolariseerd

licht. De lokale exeitatie van dcrgelijk.e achiraJe ketonen geeft aanleiding tot twee

enantiomere vormen, mits de excilatie gelokaJiseerd blijft op de carbonyl groc-p_ Echter,

voor een lokaal aangeslagen singulet 1,3-diketon, gesynthetiseerd door Meijer via

thermolyse van cen optisch actief 1,2-dioxetaan, vond in de aangeslagcn tocstand

racemisatie plaats. Vanwege de dichte nabijheid van de carbonylen was energie

overdracht dermate snel, dat optische activitcit veroorzaakt door een aangeslagen

toestand, niet wcrd waargenomen_ De resultaten in hoofdstuk 5 laten echtcr :lien dat

chi rali teit, vcroorzaakt door een van de kleinste afwijk:ingen in symmetric, dat wil

Samenvalting 147

zeggen, chiraliteit slechts veroorzaakt door aanwe~igheid van een aangeslagen toestand, in

ern hoge optische zuiverheid verwezenlijkt kan worden.

Hoofdstuk 2 beschrijft de synthese en karakterisering van de optisch zuivere

(2S,4R) en (2R,4R)-I,2-3,4-tetrabydro·4-ethyl-l,l,4-trimethyl-(3Z)-ethylideen-2(lH)-naf­

talenolen. Ben altematieve synthese route bestaande Uil zeveo stappen is ontwikkeId.

Uitgegaan werd van het optisch zuivere R~2·methyl-2-fenyl-butaanzuur, dat werd

verkregen door resolutie van de kinine zouten van het racemisch butaanzuur. Acylering

van iso-buteen met het optisch actief 2-methyl-2-fenyl-butanoyl chloride, gevolgd door

een Friedel-Craft cyclisatie, een oxidatie en een Wittig reactie resultcerde in twee isomere

3 ,4-dihydro-ethyl-trimethyl-(3Z)-ethyJideen-2( IH)-naftalenonen. Scheiding van deze

isomeren met behulp van HPLC en reductie leidde tot de gewenste optisch zuivere

moleculen voor de bestudering van de fotochemie en foto-oxidatie, zoals beschreven in

hoofdstuk 3. AIle verbindingen zijn volledig gekarakteriseerd met een- en twee·

dimensionale NMR spectroscopie, waaronder INADEQUATE.

In hoofdstuk 3 is bestudering van de fotochemie van de optisch zuivere 2-

naftalenolen beschreven. De resultaten tonen aan dat naast B-Z isomerisatie, indanen

gevonnd worden door e1iminatie van formaldehyde. 8cmi-empirische MNDO, AMl, en

PM3 berekeningen geven san dat de OH"groep in de voorkeursconformatie in de

grondtoestand een onhogonale positie inntcmt ten opzichte van de exocyc1ische dubbclc

binding. Fluorescentiemetingen bevestigen dat de reactieve tocstand de triplet aangeslagen

toestand is. Identificatie van fotoproducten via een "planairew [l,3]-OH shift roals voor

racemische 2·naftalenolen kon niet gerea1iseerd worden. In [eite lcidcn de resultaten van

hoofdstuk 3 tot de conclusie dat de [1,3]-OH shift voor de (28,4R)- and (2R,4R)-1,2-3,4.

tetrahydro-4-ethyl-l, I ,4-trimcthyl-(3Z)-ethylideen-2-naftalenolen niet optreedt. Vanwege

een toenemende belangstelling voor foto-oxidatie van chirale alcoholen is de fotochemie

van de optisch ~uivere nafialenolen in aanwezigheid van zuurstof bestudeenL De

resultaten tonen san dat de dubbele binding snel geoxideerd wordt en leidt tot de

stereoselektieve vorming van een epOxide en een allylisch hydroperoxide. De

waargenomen regio· and stereospecificiteit is verklaard san de hand van de selektieve

co6rdinering van de OH-groep in het chirale 2-propen" 1-01 fragment met het elektrofieIe

aanvallende zuurstof-molecule.

In hoofdstuk 4 is de fotochemie van 9-endo-hydroxy-9-exo-vinyl-bicyclo[4_2.1]

nona-2,4-dieen analoga beschreven. Op basis van deuterium labelling experimenten,

confonnatic analyse en een- en tweedimensionale NMR spectroscopic kon worden

aigeleid, dat een door lichtge1nduceerde intramolekulaire formele waterstofoverdracht

plaatsvindt naar een van de eindstandige koolstofatomen van het endocyclische dittn. De

148 SAMF.NVATnNG

waterstofoverdracht vindt plaats op de en do positie van het molecule en lean ook

bcschouwd worden als cen oxy rctro-ccn rcactie_ Langdurige UV-bestraling leidt tot een

intramoleculaire cycloadditie van het a,/3-onverzadigd keton in een secundair fotoproces,

waarin via een bisallyIisch triplet biradicaal als intermediair, gespannen pyranen werden

gevormd_

De rcsultaten in hoofdstuk 5 tonen aan dat een verbinding, die de chiraliteit slechts

dankt aan aanwezigheid van gelokaliseerde excitatie-energie, verwezenlijkt lean worden_

Ret optisch actieve geexciteerde 1,6-diketon met ren optische zuivcrhcid > 98% en een

levensduur van ,.. 2 ns is bereid door thermolyse van een optisch 1:uiver 1,2-dioxetaan_

De syntheseroute is met name gebaseerd op de stereo- and regioselektieve enzymatischc

reductie met HLADR/NADH en foto-oxigenering van elcktronenrijke olefinen bij lage

tcmpcratuuL Dc optische activiteit is aangetoond door een blijvende circulaire polarisatie

in de chemiluminescentie. Het werk beschreven in dit hoofdstuk toont tevens aan dat

circulaire polarisatie in chemiluminescentie gegenereerd door ccn 1,2--dioxetaan, enig

inzicht lean verschaffen in de intramolcculaire energie-overdracht in diketonen, daar dit

proccs kan lciden tot racemisatie van de chirale aangeslagen toestand_

Curriculum Vitae 149

Curriculum Vitae

De auteur van dit proefSChrift werd geboren te Maastricht op 19 juni 1963. Na het

behalen van het a.theneum-p' diploma in 1982 aan de scholengemeenschap St-Ma.a.rtenscol­

lege te Maastricht werd in datzelfde jaar begonnen met de studie Chemische Technologie

aan de Hogere Technische School te Hccrlcn. In 1983 via de Wet Wederzijdse Doorstro­

ming zette hij de studie Scheikundige Technologie voort aan de Technische Univcrsiteit

Eindhoven. Ret afstudeerwerk werd verricht in de vakgroep Organische Chemie onder

begcleiding van dr. i:l:. M.RP. van Genderen en prof. dr. H.M. Buck. Het doctoraal

examen werd in oktober 1988 afgelegd. Vanaf november 1988 tot november 1993 was hij

rus onderzoeker in opleiding (oj.o) in dienst van de Nederlandse Organisatie voor

Wetenschappelijk onderwek en eerdergenoemde Universitcit binnen de vakgroep

Organische Chemie.

In dere periode were! het onderzoek, zoals beschreven is in dit proefschrift uitgevoerd_

Aanvankelijk stond het onderzoek onder leiding van prof. dr. RM. Buck, maar in ccn

later stadium onder leiding van prof. dr. E.W. Meijer.

150 nANl<WOOIU>

'Danksvoord

yt'tUJfj 'UJi[ i( teaerun fiartdijl( aanf:gn- die, op wdf:g wijze aan oot. JUeft 6yguirogen aan tie totsta.nakpming van ait proifsdirift. In fr.et hyMn-tier '{viI ~ 6eUk pronwtoretl prof ar. 'E. 'W. Meijer en prof. dr. J. W. 'llerfiowen noemen va.n-wege de pre.ttige wijze wa.arop zij li£t onaerzoet 1i£66en 6eg&itf. J-filtt steun en 6ereia.'UJi[figIieU om JUt orulerzoel( ,fa.t ten grondsfag [igt aan tiit praefsdirift op zicli te. nemen en voort te zetten, /ie6 ~ aer gewaanfeera. Ook. tir. ir. '1(Jne Jan.ssen 6en if( zeer erf<.§ntdiji( voor zijn 6eg&iding en nauwgt:Zette wija waarop Iiij ae teRJt van iit proejscFi.rift lieeft doorge.nomul. m van commentlUU /ieeft voorzitn. 'livens Den if( ved danl( versdiu[iigi aan ar. iT. Leo 1\9o[e VOOr zijn 6ijarage aIl7!- Iioofistui( 4 van. ait poefsclirift. Va.arnaast 'Wi£ if( tir_ Harrit 'IJetf<.§rs en ars. MaTed Me..sktrs uit Leitfen noemen Voor liun- onmis6are en zur p[ezi£rige e;rperimentdi! l1uaewerKjng en. TYlllI/geviJlgen hij tie interpretati£ van tie verR.regen resuftaten zoafs 6escfirwen i:n. liooftistul(S .

Verier wi!. if( tie afstuaeertiers ir. :Jfans 'Batten, ir. :J-{a:rm La.rrgmnans en ir­Ylngefino 'Wo{{aslieim 6edanf:gn voor at liet wert fiat zij fr.e66en verric/it. 'lirJens lieh if( ae sanumwerking 1Tt.ft al(e "Dewoners" van liet [a6 01flaniseli aftija a1s prettig ervare.n. 'v'oorts hw ik. ar. ir. MarrA van genaeren, dr. Jef ·VeKi-tfUln.S en tIr. Stet/en van 'Es

erf:gntdijl( VOM liun {urzame wetenscliapdij/(j aiscussi£s en eonstnu:tieve hijC£rape oon ait proefsdi.rift. 001( 'Wi1 if:... 1Tt.ft name mijn ~ __ {am£.rgenoten dr. ir. ofav 510gaarti en ar. ir. 9i"J£('1Jroeiers noem£tt vaM fum morefe steun en vritndsdiap.

Ook/Jen ik..Janf( verse/i.ufaiga aart ing. Joost van Vongen en ing. :J-fenl( CUussens '(.!O(Jf nun 'fiigli performance' 6i) tie peraparatieve HPLC sc.fieUingen en 5-i£nK.., 'Laing VOM

JUt vervaartiigen van tie vde i[[ustrati£s. ?t].et in at taatste pfoots wif ik... ar. rr'aeo Sdi.e.rer uit JUnsterdam noemen VOM zijn

mdtwerf.:.ing 6ij ae spectroscopiscfl.e m.etingen in Fioofdstuk.3-

Stelling en

1. Door het onlbreken van een experimentele verificatie van de door weinigen voor

mogclijk gchouden fotochemische [1,3]-OH shift via cen nict-Woodward-Hoffinann

mechanisme, verkeert de organische fotochemie weer in een meer rationele toestand.

Dit procfscluift hoofdstuk 3,

W, j, G, M, Pcijnenburg and H. M. Buck, Tetrahedron 1988, 44,4821.

2. Gczicn de aanbcvelingcn vOOr de acccntgcbiedcn omtrent het Toekomstig Chemisch

Onderzoek aan de Nederlandse universiteiten (TCO-rapport), is het wellicht beter het

acronym TCO te specificeren met "Toegepast Chemisch Onderzoek".

VNcr Onderwijs en On&rz.oek, 26januari 1994, 18.

3. De aanduiding "super" als vooNoegsel van cydofanen getuigt eerder van eeo gezonde

waardering voor de organische chemie dan van een systematische terminologie voor

deze verbindingen.

"Super" p~, R, GleilCr en D, KralZ, Ace, Chern, Res, 1993, 26, 31 J,

4. De door Womer en Miilhaupt uitgevoerde NMR interpretatie van poly(propyleen­

imine)-dendrimeren, is, zoals zij kunnen lezen in het daaropvolgend artikel, strijdig

met de toekenningen van de Brabander en Meijer.

C. W6mer en R. Miilhaupt, Angew, Chl;':rn, 1993,32, 1306.

E. M, M. de Brabandcr-van den Berg en E. W. Meijer, Angew. Chern. 1993, 32. 1308.

5, De maatregel waarin de veiHge drempd voor giftigc chemicalien uitsluitend bepaaJd

wordt door de maximaal toelaatbare concentratie, houdt ten onrechle geen rekening

met de ontoelaatbare verdunde emissie van dergelijke gevaarlijke stoffen in de

samenleving,

Profdr. N. Vermeulen in Elsevier, 13 november 1993, 102.

6. De ajwezighdd van het Thorpe-Ingold effect in de polymerisatie van 1,1,2,2-

tctramethyl- t ,2-disilacyc!opent-3"een is eerder te wijten aan het hoge penx'1ltage

trans-dubbele bindingen en sterische interacties in het polymeer, dan aan de

voorgestelde beperkte rQtatievrijheid van de Si-Si binding, tijdens de compiexatie aan

de katalysatoL

L. R. Sita cn S. R. tyon, 1. Am. Chcl1l. sO';. 1993, lIS, 10374

7. Het fcit dat de belangstclling van chemic; tegenwoordig mecr uitgaat naar grotere en

complexere (chirale) systemen, doet onrecht aan het wetenschappelijk belang van

onderzoek naar de klcinst rnogelijkc afwijking in asymmetric van (chirale) objecten.

Dit procfschrift, hoofdsruk 5.

Chiral Methyl Groups: Small is beautiful, Acc. Chel1l. Res. 1993,26. J 16.

8. De opvatting dat psoriasis een onschuldige huidafwijking is waaraan therapeutisch

wcinig te doen valt (en waarmee men zou moeten lercn leven). is onjuist.

9. Met het gebruik van technisch arbeidspotentieel via "stageplaatsen met behoud van

uitkcring", streeft men het doel voorbij om (tijdelijke) arbeidsplaatsen te creeren ten

tijdc van een economische recessie.

10.

F.W.A.M. Miesen

De beste vcrdediging voor de promovcndus (wit) en

zijn twee secondanten van deze stelling is nog altijd

de aanval op de hoog- en zeergeleerde opponentcn

spelend in het zwart met: Tetxe7t, ..

Eindhoven, 29 maart 1994