Photochemical reactions of chloranil with cyclooctene, 1,5-cyclooctadiene, and cyclohexene revisited

FULL PAPER

DOI: 10.1002/ejoc.201001304

Photochemical Reactions of Chloranil with Cyclooctene, 1,5-Cyclooctadiene,

and Cyclohexene Revisited

Manfred Christl,*^[a]Max Braun,^[a]Oliver Deeg,^[a]and Stephan Wolff^[a][†]

Dedicated to Professor Siegfried Hünig on the occasion of his 90th birthday

Keywords: Cycloaddition / Dechlorination / Endiones / Photochemistry / Polycycles

Depending on the reaction conditions, excited chloranil

(³CA) and cis-cyclooctene (cis-CO) furnished three types of

products, which all contain carbocyclic four-membered rings.

The illumination of a solution in benzene by light bulbs gave

rise to the all-cis 1:1 cycloadduct (11a) in 76% yield. On its

part, this compound reacted with cis-CO under the action of

better sources for UV light with the formation of three iso-

cloadduct that resulted from ³CA and cis-CO in addition to

two symmetrical 1:2 cycloadducts. Also, excited 11a and cy-

clohexene (CH) furnished a bis(cyclobutane) derivative, a

significant portion of which was subject to a dechlorination

to give a cyclobutene derivative. cis,cis-1,5-Cyclooctadiene

(COD) and CH were transformed by ³CA to 1:1 and 1:2 cy-

cloadducts, analogous to cis- and trans-CO. In addition to

meric 1:2 cycloadducts. Two of them underwent a photo- this, ³CA suffered photoreduction in the presence of COD

chemical dechlorination leading to a cyclobutene derivative.

Albeit rather sluggishly, ³CA and trans-cyclooctene (trans-

CO) brought about an unsymmetrical 1:1 and a 1:2 cycload-

duct. By contrast, excited 11a reacted smoothly with trans-

CO and afforded a 72% yield of the unsymmetrical 1:2 cy-

and CH. The [2+2] cycloaddition onto a carbonyl group of

³CA with formation of a spirooxetane occurred to a small ex-

tent only with COD, but not at all with cis- and trans-CO and

CH.

dimethylindene,^[6]indene,^[5]diphenylvinylene carbonate, and

a number of stilbenes.^[7]If the subunit of the intermediate

originating from the alkene has the possibility to undergo

a rearrangement, spirotetrahydrofurans can emerge in ad-

dition to spirooxetanes as with benzvalene^[8]and norborna-

diene.^[9]With homobenzvalene only products resulting from

a rearrangement arise, namely two spirotetrahydrofurans

and a spirotetrahydropyran.^[9]If the alkene has allylic hy-

drogen atoms, the radical-ion pair may react to give tetra-

chlorohydroquinone (TCH) or the respective monocycloal-

kenyl ether of it, as in the cases of 1,3,5-cycloheptatriene,^[10]

1,3-cyclohexadiene,^[11]cyclohexene, indene, allyl ethyl

ether,^[5]2,3-dihydrofuran, and 3,4-dihydro-2H-pyran.^[12]

Introduction

Among the light-induced cycloadditions of quinones

with alkenes^[1,2]those of chloranil (CA) are of particular

importance because of the great variety of possible pro-

cesses. The rapid intersystem crossing ensures that all con-

versions of excited CA, leading to a chemical change of

CA, start from its triplet state (³CA).^[3]Depending on the

3

properties of the substrate, CA behaves as a diradical or

an electron acceptor. Based on their experimental results

and the application of the Weller equation,^[4]Xu et al.^[5]

3

concluded that CA takes an electron from alkenes with an

oxidation potential of Ͻ 2 V and reacts as a diradical with

alkenes having one of Ͼ 2 V (vs. SCE).

3

The attack of CA as a diradical at an alkene leads to

3

The result of an electron transfer from an alkene to CA

[2+2] cycloadducts of a dichloroethene moiety of CA, that

is, cyclobutanes. Examples of such alkenes are 2-methylpro-

pene, 1,3-butadiene, 2,3-dimethyl-1,3-butadiene,^[13]allyl

ethyl ether, methyl methacrylate, vinyl acetate, styrene, α-

chlorostyrene, α,p-dichlorostyrene,^[5]and 1,1-diaryleth-

enes.^[14]Cyclobutanes are also formed with cyclopentene

(CP), norbornene, and bicyclo[2.1.1]hex-2-ene, but signifi-

cant portions of these cycloalkenes undergo a cycloaddition

is a triplet radical-ion pair, which may collapse to give a

cycloadduct onto a carbonyl group of CA. As intermediates

in this process, diradicals as well as zwitterions can be con-

sidered, in which an oxygen atom and a carbonyl carbon

atom of CA, respectively, have formed a bond to the alkene.

Typical cycloadducts are spirooxetanes as in the case of 1,1-

3

with CA that eventually gives rise to tetrachlorobenzofu-

[a] Institut für Organische Chemie, Universität Würzburg,

Am Hubland, 97074 Würzburg, Germany

Fax: +49-931-3184606

E-mail: christl@chemie.uni-wuerzburg.de

[†] Deceased on May 29, 2009

ranone derivatives.^[15]Apparently, a deep-seated rearrange-

ment of the subunit stemming from CA proceeds in these

reactions. Scheme 1 summarises the compounds obtained

968

Eur. J. Org. Chem. 2011, 968–982

Photochemical Reactions of Chloranil

from CP, with which ³CA undergoes three parallel reac- to the substrate. The product ratio depends on the solvent

tions: the photoreduction to give TCH (8% yield), a [2+2] polarity in such cases.^[5,7,14,15]

cycloaddition, resulting in cyclobutane 1a or 1b (9%) and

When dividing the alkenes into two kinds of different

eventually also in 2 (8%), and a [4+2] cycloaddition. The behaviour towards ³CA, that is, electron transfer to ³CA

3

latter leads to a diradical (Scheme 1), a C–C bond of which and addition of CA as a diradical, the oxidation potential

breaks with formation of an α-chloro-β-oxo ketene. This of 2 V (vs. SCE) is not a strict borderline. There are excep-

compound rearranges to give the tetrachlorobenzofuranone tions to this rule, for example (E)-β-bromostyrene and α,β-

3, which was not isolated but exposed to methanol. The unsaturated carbonyl compounds, which have oxidation po-

methyl ester 3a (15%) and the methyl pseudoesters 3b,c tentials of 2.3 and about 2.4 V, respectively, and yet exclu-

(13%) were then obtained.^[15]By using norbornene as a sively produce spirooxetanes.^[5,7]Steric hindrance of the ad-

substrate, the pseudoacyl chloride analogous to 3 was iso- dition of these alkenes onto a dichloroethene moiety of ³CA

lated in 67% yield.^[15]

in favour of that onto a carbonyl group has been invoked

to explain the outcome.^[5,7]

The present work describes the addition of CP onto ³CA

under conditions different from those in the previous ex-

periment^[15]and deals with the reactions of cis- and trans-

cyclooctene (cis- and trans-CO), cis,cis-1,5-cyclooctadiene

(COD), and cyclohexene (CH). Those of cis-CO, COD and

CH have already been investigated; the structures of the

products claimed are collected in Figure 1.

Figure 1. Structures of the products described in the litera-

ture^[5,16,17]to be formed from excited chloranil (³CA) with cis-cyclo-

octene (cis-CO), cis,cis-1,5-cyclooctadiene (COD), and cyclohexene

(CH).

Bryce-Smith and Gilbert^[16]reported the formation of the

spirooxetane 4 and the 1:2 cycloadduct 5, having two cyclo-

butane subunits, on irradiation of CA in benzene in the

presence of cis-CO. They also noted that “1,5-cyclooc-

tadiene shows closely parallel behaviour to that of cyclooc-

tene”. Kim et al.^[17]described the production of the spiro-

oxetanes 6 and 7 with COD and CH, respectively, both dis-

solved in dichloromethane together with CA. Further, they

claimed the conversion of CA into the oxepin 8 during the

Scheme 1. Compounds formed in the photochemical reaction of

chloranil (CA) with cyclopentene (CP) in the ratio of 1:2 and the

pathway to products 3a–c according to ref.^[15]

Alkenes with an oxidation potential close to the value of photoreaction of CA with CH as well as with 1,3-cyclo-

2 V (vs. SCE) may give products that result from an electron hexadiene and 1,3,5-cycloheptatriene. Without reference to

transfer to ³CA and from the addition of ³CA as a diradical the Korean work,^[17]Xu et al.^[5]published their results on

Eur. J. Org. Chem. 2011, 968–982

www.eurjoc.org

969

M. Christl, M. Braun, O. Deeg, S. Wolff

FULL PAPER

the irradiation of CA in the presence of CH, for which they all, the corresponding 1:1 cycloadduct 1 was observed in

utilised a filter solution that prevented the entrance of light addition to the 1:2 cycloadduct 2 in the case of CP

of λ Ͻ 400 nm into the reaction vessel. They obtained the (Scheme 1). However, Xu et al.^[5]did not mention a respec-

1:2 cycloadduct 9 and TCH from the photolysis in benzene tive compound derived from CH, and Bryce-Smith and Gil-

solution, whereas the reaction in acetonitrile brought about bert^[16]explicitely excluded such a compound derived from

only TCH and the monocyclohexenyl ether 10 of TCH. cis-CO.

When dichloromethane served as solvent, all of the com-

Results and Discussion

1. Reaction of Chloranil (CA) with Cyclopentene (CP)

under Conditions Different from Those of the Previous

Experiment

pounds 9, 10, and TCH were found.

As to the oxepin 8, we believe that the authors^[17]are in

error, since neither Xu et al.^[5]gave any indication nor did

we find a clue for 8 (see below). Rather, TCH was mistaken

for 8, as the melting point of 240 °C is very close to that of

Illustrated in Scheme 1, the products of the first experi-

ment had ensued from CP and CA in a ratio of 2:1 in ben-

zene, irradiated by a Hanovia medium-pressure mercury

lamp (450 W) through a glass filter that was believed to

prevent the passage of light of λ Ͻ 400 nm. Now, we em-

ployed a Rayonet photochemical reactor RPR-100, radiat-

ing at λ = 350 nm, and used CP and CA in a ratio of

1.00:1.22. This changed the product spectrum (Scheme 2)

somewhat, as not only one of the 1:1 cycloadducts 1 was

obtained, but both of them resulted (1a and 1b). However,

it still remains open whether configuration 1a or 1b has to

be assigned to the previously described compound and

which configuration to the newly observed one. As expected

from the substoichiometric quantity of CP, just a very small

amount of the 1:2 cycloadduct 2 was formed. That only one

of the methyl pseudoesters 3b and 3c was observed and no

TCH may be an artefact of the chromatographic workup.

1

TCH (m.p. 236–237 °C^[18]), and the two signals in the H

NMR spectrum (δ = 8.90 and 3.75 ppm), assigned to OH

and CH of 8, most probably originate from OH of TCH

and residual water of [D₆]acetone used as solvent. The final

proof is provided by the other published spectroscopic data

(MS, IR, UV),^[17]which virtually exactly match our data of

TCH. Only in the IR spectrum did we find important bands

at 1411 and 1307 cm^–1in addition to those mentioned by

Kim et al.^[17]

The first motivation for the present study was to find out

whether or not CH and cis-CO give rise to tetrachloroben-

zofuranone derivatives analogous to 3 (Scheme 1). In the

case of CP, 3a–c, the consecutive products of 3 with meth-

anol, constituted about 50% of the amount of the products

isolated, and a spirooxetane was not observed. Norbornene

furnished the tetrachlorobenzofuranone derivative as major

product by far and a small amount of a cyclobutane analo-

gous to 1, when the substrates were irradiated in benzene.

The utilisation of a 1:1 mixture of benzene/methanol as sol-

vent led to a dramatic setback of the yields, but also to the

formation of a small amount of the spirooxetane. Dissolved

in benzene, bicyclo[2.1.1]hex-2-ene gave a cyclobutane and

a tetrachlorobenzofuranone derivative in addition to the

spirooxetane, which was the main product. In the series of

these three cycloalkenes, the tendency for spirooxetane for-

mation increases with decreasing oxidation potential: CP,

2.03 V; norbornene, 1.95 V; bicyclo[2.1.1]hex-2-ene, 1.88 V

(all vs. SCE).^[15]

The oxidation potentials of CH and cis-CO are 2.14 and

2.01 V (vs. SCE),^[19]respectively, and are thus significantly

higher and only marginally lower than that of CP. Hence,

the electron transfer from these cycloalkenes onto ³CA

should occur more difficult than and similarly difficult as

from CP, where merely TCH, isolated in a yield of 8%, can

be rated as an indication of an intermediate radical-ion pair.

If it is accepted that the formation of spirooxetanes also

proceeds via radical-ion pairs, the production of 7 and 4

(Figure 1) from CH and cis-CO, respectively, is hard to con-

ceive on the basis of the oxidation potentials of the latter.

Thus, the examination of whether or not CH and cis-CO

bring about spirooxetanes with ³CA in benzene was the sec-

ond motivation for our investigation.

Scheme 2. Products of the photochemical reaction of chloranil

(CA) with cyclopentene (CP) in a ratio of 1.22:1.00.

2. 1:1 Cycloadduct of cis-Cyclooctene (cis-CO) onto

Chloranil (CA)

Finally, there was the question for 1:1 cycloadducts of

CH and cis-CO onto CA containing a cyclobutane moiety,

which do have to be considered as intermediate compounds

The first experiment (Entry 1 of Table 1) was conducted

en route to the 1:2 cycloadducts 9 and 5, respectively. After as the reactions of CP, norbornene, and bicyclo[2.1.1]hex-

970

www.eurjoc.org

Eur. J. Org. Chem. 2011, 968–982

Photochemical Reactions of Chloranil

Table 1. Conditions and products of a series of photochemical reactions of CA with cis-CO.

Entry cis-CO

CA

Solvent

Light source

Duration of illumination Yield of 11a Yield of 5 Yield of 12

[mmol] [mmol]

[h]

[%]

[a]

1

2

3

4

5

6

7

8.14

16.2

81.2

32.6

72.6

461

4.07

8.13

40.7

3.25

16.3

C₆H₆

Hanovia 450 W,

glass filter, λ Ͼ 400 nm

Rayonet RPR-100,

λ = 350 nm

Rayonet RPR-100,

λ = 350 nm

light bulbs,

total power 500 W

light bulbs,

5

13

32

[b]

[a]

2.5

3

39

2

C₆H₆

14

18

[b]

[a]

C₆H₆

60

76

[c]

[a]

CH₃CN

CH₂Cl₂

C₆H₆

3.5

36

total power 1000 W

light bulbs,

total power 1000 W

medium-pressure Hg lamp,

150 W, Pyrex immersion well

[d]

46

13.5

33^[e]

[a] No 12 observed. [b] No 5 observed. [c] The products 5c and 11a were formed in a ratio of 1.0:1.3. [d] No 11a observed. [e] Yield of

pure 5c. The combined amount of 5a, 5b, and 5c in the crude product corresponds to a yield of 79%.

3

2-ene with CA.^[15]Hence, the crude product of the irradia- than benzene, it has to be concluded that 11a is very effec-

tion was treated with methanol, but no compounds con- tively excited by the weak UV radiation generated by light

taining a methoxy group were discovered. The chromato- bulbs and is highly prone to cycloaddition in that state.

graphic workup provided three fractions, the first two of

Finally, we employed conditions that approached those

which consisted of three 1:2 cycloadducts of the consti- described by Bryce-Smith and Gilbert^[16]as close as pos-

tution 5 (see Section 3). Owing to the chemical shifts and sible to us (Entry 7 of Table 1). Instead of a Hanovia S-500

the number of the signals, the ¹³C NMR spectrum of the medium-pressure mercury lamp, we utilised a 150 W lamp

third fraction immediately disclosed constitution and sym- of the same kind. The amounts of CA, cis-CO, and benzene

metry of the 1:1 cycloadduct 11a, which was obtained as a given yield a suspension of CA, which is why we stirred the

slightly contaminated colourless oil.

mixture. The irradiation was then performed at 10 °C for

Whereas cis-CO and CA were used in a ratio of 2:1 in 13.5 h, whereas it had been conducted at 30 °C over 64 h on

the above experiment, we chose a ratio of 1:1 of the sub- the fivefold scale previously. We exclusively observed three

strates, again dissolved in benzene, for the second experi- products 5 and isolated 5c (see Section 3) in 33% yield. On

ment (Entry 2 of Table 1), which was carried out in a the basis of the melting point, this compound had been

Rayonet photochemical reactor, radiating at λ = 350 nm. obtained in 42% yield earlier.^[16]There was no indication

The 1:1 cycloadduct of the first experiment was formed as for both 11a and the spirooxetane 4. The isolation of the

the only product of a significant amount and purified by latter in 51% yield had been claimed in 1964.^[16]The fact

chromatography to give crystals in 39% yield. A single-crys- that we did not find a 1:1 cycloadduct could have had its

tal X-ray diffraction analysis corroborated the constitution cause in the different lamp employed and the complete con-

and established the all-cis configuration as illustrated by version of CA in our case as compared to only 30% pre-

formula 11a.^[20]After a 2:1 mixture of cis-CO/CA in ben- viously.^[16]

zene had been irradiated in the Rayonet reactor (Entry 3 of

As to the spirooxetane 4, we suspect that Bryce-Smith

Table 1), only a small quantity of 11a was observed as well and Gilbert^[16]mistook 11a for 4. In 1964, ¹³C NMR spec-

as some 5 and, being the main result, two new compounds, troscopy was not available, and the authors had to rely on

which were identified to be the consecutive products 12 of less direct evidence. However, even this evidence is at vari-

5 (see Section 3). The best yield of 11a (76%) was achieved ance with the properties of 11a. Strong IR absorptions at

by illumination of a 2:1 mixture of cis-CO/CA just with 1675 and 965 cm^–1were reported as well as a UV band at

light bulbs (Entry 4 of Table 1).

262 nm (logε = 3.81).^[16]Compound 11a shows strong IR

The change of the solvent from benzene to acetonitrile absorptions at 1715 and 1697 cm^–1, whereas the one closest

and dichloromethane (Entries 5 and 6 of Table 1) furnished to the value of 965 is at 945 cm^–1, but only of medium inten-

different results on irradiation with light bulbs, as 5 sity. The maximum of the UV spectrum lies at 282 nm with

emerged in addition to 11a and exclusively, respectively. logε = 4.08. In addition, the catalytic hydrogenation of the

Since the latter solvents are more transparent for UV light alleged 4 was carried out and believed to result in a

Eur. J. Org. Chem. 2011, 968–982

www.eurjoc.org

971

M. Christl, M. Braun, O. Deeg, S. Wolff

FULL PAPER

monohydric phenol with a melting point of 190 °C.^[16]We whereas that of 12 increased. Constructed from data of the

hydrogenated 11a and obtained the hydroquinone derivative irradiation over 2.5 h, the turnover diagram is in line with

13 (Scheme 3), the structure of which was easily deduced the conversions of CA into 11a, of 11a into 5, and of 5 into

from its ¹³C NMR spectrum and which has a melting point 12. In these processes, the second and third step require

of 208 °C. Barltrop und Hesp^[13]had arrived at similar hy- direct excitation of the respective substrate by light, as CA

droquinones on hydrogenation of the 1:1 cycloadducts of is not present anymore and could thus not act as a sensitiser

1,3-butadiene and 2,3-dimethyl-1,3-butadiene.

after 1 h.

In view of the longest-wavelenght absorption of 11a (λ =

282 nm), it is surprising that 11a was excited and underwent

cycloaddition with cis-CO even under irradiation with light

of λ Ͼ 400 nm (Entry 1 of Table 1). The UV spectrum of

11a supports an ε value of 10 Lmol^–1cm^–1or less at 400 nm

as compared to 12023 Lmol^–1cm^–1at 282 nm. Therefore,

either the excitation of 11a must have been very efficient or

the glass filter utilised in the experiment did not completely

inhibit the passage of light with λ Ͻ 400 nm. We tend to

believe the latter, since the illumination of a benzene solu-

tion of the substrates with light bulbs brought about 11a

exclusively (Entry 4 of Table 1). Under these conditions, the

excitation of 11a by 400-nm light would have been possible

but apparently did not occur. However, if solvents more

transparent for UV light than benzene were taken, the for-

mation of 5 proceeded smoothly (Entries 5 and 6 of

Table 1), even promoted by the weak UV irradiation of the

light bulbs. The exclusive production of 11a on irradiation

in the Rayonet reactor (λ = 350 nm, Entry 2 of Table 1) was

due to the ratio CA/cis-CO = 1:1. Obviously, cis-CO was

Scheme 3. Hydrogenation of 11a and its reduction with sodium

borohydride.

Interestingly, the reduction of 11a with sodium boro-

hydride did not lead to 13 but to a product that contained

2 equiv. of hydrogen chloride more than 13, and for which

we propose structure 14a or 14b (Scheme 3). This result

may be taken as indication against the hydrogenation of the

carbonyl groups en route to 13, as it would give rise to an

isomer 14 as well. Rather, the hydrogenolysis of the C–Cl

bonds^[21]at the four-membered ring could have occurred,

with the resultant product tautomerising to 13.

3

very efficiently used up by CA, and the excitation of 11a

was much less successful than that of CA under these condi-

tions, which is why after the consumption of CA no cis-CO

was left for the reaction with 11a.

Previously, the question of isomerism possible for the

constitution 5 remained unaddressed.^[16]In addition to

some 11a, our first experiment (Entry 1 of Table 1) fur-

nished a mixture of three diastereomers 5. One of them had

already been isolated by Bryce-Smith and Gilbert^[16]and

was now again prepared as a pure compound in the experi-

ments of Entries 1, 6, and 7 (Table 1). Owing to only six

signals in the ¹³C NMR spectrum, three of them stemming

3. Cycloadducts of cis- (cis-CO) and trans-Cyclooctene

(trans-CO) as well as Cyclohexene (CH) onto the 1:1

Cycloadduct 11a

In order to confirm the assumption that the route to the from the methylene groups, the molecule must have two

1:2 cycloadducts 5 of cis-CO onto CA proceeds via the 1:1 symmetry elements. Since the shape of half the molecule is

cycloadduct 11a, we monitored the course of the reaction determined by the C_ssymmetry of its precursor, i.e. 11a,

by HPLC. The ratio of the substrates, the solvent and the the molecular point group must be either C_2hor C_2vand,

irradiation conditions were close to those of Entry 3 of hence, the structure 5c or 5c^Ј(Figure 2). En route to 5cЈ,

Table 1; only the scale was reduced to one quarter. Re- cis-CO would have to approach the excited 11a from the

corded before the lamps of the reactor were turned on, the endo side, which is sterically strongly hindered by the cyclo-

first chromatogram showed already a small signal of 11a in octane subunit, and thus we prefer 5c over 5cЈ, although

addition to that of CA. Apparently, the daylight had started there is no reliable experimental evidence for this assump-

the photocycloaddition. After 30 min of irradiation, the tion. This uncertainty also pertains to the configuration of

concentration of CA had substantially decreased and that the other isomers 5 (Figure 2) as well as 9 (Scheme 8), 12

of 11a was at its maximum. No further compounds in sig- (Scheme 5), 15 and 16 (Scheme 6), and 18 (Scheme 7).

nificant quantities were discernible at that time. After 1 h,

The identity of the two further diastereomers of 5 was

CA had been consumed completely, the concentration of clarified by an independent experiment, in which 11a was

11a had decreased somewhat, and two new products were illuminated in the presence of trans-cyclooctene (trans-CO)

present, which were characterised to be 5 and 12 on the to give an unsymmetrical diastereomer 5 in 72% yield

basis of the retention times by injection of authentic sam- (Scheme 4). This compound turned out to be one of the

ples. On further irradiation, the decrease of the concentra- diastereomers of 5 observed in the experiments of Entries 1

tion of 11a continued, and that of 5 diminished as well, and 3 (Table 1) by comparison of the ¹³C NMR spectra.

972

www.eurjoc.org

Eur. J. Org. Chem. 2011, 968–982

Photochemical Reactions of Chloranil

not be isolated. Its configuration was deduced from the ¹³

C

NMR spectrum of the three-component mixture by sub-

traction of the signals of 5a or 5aЈ and 5c or 5cЈ. Eight lines

were the result, in particular one of carbonyl, two of CCl

and two of CH groups. This supported a structure with a

symmetry plane, if the symmetry plane of the precursor

(11a) is accounted for. The missing of three signals originat-

ing from methylene groups was readily explained by super-

position of absorptions of 5a or 5aЈ and 5c or 5cЈ, which

cause 15 signals in the narrow region from δ = 23.4 to

29.4 ppm. Based on the steric hindrance of the endo ad-

dition of cis-CO onto excited 11a, already advanced with

regard to the formation of 5a or 5aЈ and 5c or 5cЈ, we prefer

structure 5b over the alternative 5bЈ (Figure 2). By the way,

the configuration of 5b corresponds to that of 2, the only

1:2 cycloadduct of CP onto CA (Scheme 1) known so

far.^[15]

A fourth diastereomer with the constitution 5 was ob-

tained when a mixture of CA and trans-CO in benzene was

irradiated. Its ¹³C NMR spectrum contained 11 signals, and

hence the C_iand C_ssymmetric structures 5d and 5dЈ (Fig-

ure 2), respectively, are possible. We prefer 5d, since en route

to this compound the excited precursor 11b (Scheme 4)

would have bound trans-CO at the exo side, whereas 5dЈ

would have resulted from an endo addition, which should

be impeded by steric overcrowding.

Figure 2. Possible configurations of 1:2 cycloadducts of excited

chloranil (³CA) onto cis- (cis-CO) and trans-cyclooctene (trans-

CO). It is discussed in the text why we favour 5a–d over 5aЈ–dЈ.

From the expected 22 signals, 21 were resolved, and the

double intensity of the line at δ = 24.7 ppm indicated the

coincidence of two signals. By taking into account the C_s

symmetry of 11a, again two possibilities are in accord with

the data, namely the structures 5a and 5aЈ of Figure 2.

Again we favour the result of the exo addition of the olefin

(trans-CO) onto the excited 11a, i.e. 5a, since the endo ad-

dition, from which 5aЈ would emerge, should severely suffer

from steric encumbrance by the cyclooctane subunit of 11a.

3

It is interesting to note that the reactions of CA with

cis- and trans-CO did not lead to a common product.

Somewhat differently, the additions of excited 11a onto cis-

and trans-CO both yielded 5c or 5cЈ, which was the sole

product emerging from trans-CO, but only one of three iso-

mers in the case of cis-CO.

On irradiation of a 2:1 mixture of cis-CO/CA in benzene

in the Rayonet reactor (λ = 350 nm), two diastereomers

with the constitution 12 were formed as major products

(Entry 3 of Table 1). When a 1:1 mixture of 11a/cis-CO was

treated in the same way, only the isomers 12 were obtained.

Their origin from the 1:2 cycloadducts 5 was obvious from

the monitoring of the reaction course by HPLC (see above).

This dechlorination was corroborated by the irradiation of

a pure sample of 5c or 5cЈ, dissolved in C₆D₆in an NMR

tube. Although a number of byproducts were generated, the

conversion of 5c or 5cЈ into only one of the diastereomers

12 was proved beyond doubt. Provided the structure 5c is

correct, the configuration of the dechlorination product

must be 12a (Scheme 5). Akin to 12a, its diastereomer is

also characterised by a plane of symmetry, which is why it

cannot be a consecutive product of 5a or 5aЈ, but had to

arise from 5b or 5bЈ. We assign structure 12b or the corre-

sponding alternative configuration to it (Scheme 5).

The formation of the cyclohexenediones 12 occurred

only on irradiation in the Rayonet reactor (λ = 350 nm),

but did not with the other light sources. Probably, in the

case of the latter, the intensity of the UV light necessary to

excite the saturated ketones 5 is too low. The dechlorination

of 5 should start with the breaking of a C–Cl bond, which

Scheme 4. Products of the photochemical reactions of chloranil

(CA) and the 1:1 cycloadduct 11a of CA onto cis-cyclooctene (cis-

CO) with trans-cyclooctene (trans-CO).

Observed in the experiments of Entries 1 and 3 (Table 1) is a β cleavage within an excited ketone. Such reactions are

and having the constitution 5, the third diastereomer could well known for α-functionalised ketones, among them α-

Eur. J. Org. Chem. 2011, 968–982

www.eurjoc.org

973

M. Christl, M. Braun, O. Deeg, S. Wolff

FULL PAPER

With respect to the dechlorination of 15, there are two

possibilities, one at the side of the cyclohexane and the

other at the side of the cyclooctane moiety. Giving rise to

16, the latter mode was proved by NMR experiments. At

first, the signals of cyclobutene protons were identified by

means of an HMBC spectrum due to their coupling with

the olefinic carbon atoms. And secondly, it was shown by

an H,H-COSY spectrum that the CH groups of the cyclo-

butene subunit are parts of an eight-membered rather than

a six-membered ring.

4. Reaction of Chloranil (CA) with cis,cis-1,5-

Cyclooctadiene (COD) and an Attempt To Use 1,3,5,7-

Cyclooctatetraene (COT) as Substrate

Scheme 5. Photodechlorination of the 1:2 cycloadducts 5b and 5c.

For the reaction with cis,cis-1,5-cyclooctadiene (COD),

CA was dissolved in benzene and excited with light bulbs.

Thus, the conditions were similar to those of the cycload-

dition onto cis-CO as given in Entry 4 of Table 1. A large

excess of COD was used (ratio of COD/CA = 20:1) in the

first experiment, which gave rise to an oxetane 6 (7% yield),

a 1:2 cycloadduct 18 (22% yield), the monocyclooctadienyl

ethers 19 (24% yield) of tetrachlorohydroquinone (TCH)

and TCH (6% yield) (Scheme 7). Signals of a 1:1 cycload-

duct 17 could not be discerned in the NMR spectra of the

halo ketones.^[22,23]In addition, photochemical β elimi-

nations of two chlorine or two fluorine atoms with forma-

tion of a C–C double bond were discovered some time

ago.^[24,25]

Using 11a as an easily available and excitable substrate,

we not only performed the addition of cis- and trans-CO

but also that of cyclohexene (CH). Because of the Rayonet

reactor (λ = 350 nm) being the light source, the [2+2] cy-

cloadduct and its dechlorination product were obtained in

yields of 33 and 45%, respectively. We assign them the

structures 15 and 16 (Scheme 6). As 20 signals in the ¹³C

NMR spectrum indicate, the [2+2] cycloadduct does not

possess any symmetry. On the basis of the C_ssymmetry of

11a, CH was attached antarafacial. Thus, the newly formed

bicyclo[4.2.0]octane system has a trans ring junction (see

Section 5 for the strain energy difference of cis- and trans-

bicyclo[4.2.0]octane). We propose structure 15, because the

assemblage of the central unit with two cyclobutane moie-

ties located trans at the cyclohexanedione subunit should

suffer from much less steric congestion than the alternative

with mutually cis-orientated cyclobutanes.

Scheme 6. Products of the photochemical reaction of 11a with cy-

clohexene (CH).

Scheme 7. Products of the photochemical reactions of chloranil

(CA) with cis,cis-1,5-cyclooctadiene (COD).

974

www.eurjoc.org

Eur. J. Org. Chem. 2011, 968–982

Photochemical Reactions of Chloranil

crude product. In order to obtain information whether or the cyclohexanedione core for steric reasons (see Figure 2).

not 17 is an intermediate en route to 18, we performed the Even then, two configurations have to be taken into ac-

second experiment with a ratio of COD/CA = 1:1. Indeed, count for 9, which are illustrated in Scheme 8.

two diastereomers with the constitution 17 (Scheme 7) were

obtained in addition to the two isomers 19 and a small

amount of 6. The spectra of the crude product provided no

evidence for 18. Although 6 and 19 could be removed from

the mixture by chromatography, we failed to separate the

diastereomers 17 and even to prepare an analytically pure

sample of their mixture. However, as judged on the basis of

their ¹³C NMR chemical shifts, the configurations of the

diastereomers 17 (Scheme 7) are likely to be analogous to

those of 11a and 11b.

Beyond the fact that 18 is unsymmetric, we have no evi-

dence as to its precise configuration. However, we prefer

the trans orientation of the two cyclobutane entities at the

six-membered ring (Scheme 7) according to the reasoning

advanced for the case of the compounds 5 (Figure 2). Also,

we made no attempt to establish the junction of the four-

and the eight-membered ring of 6 to be cis or trans. Further,

the routine NMR spectroscopic data do not allow the as-

signment of the structure 19a or 19b to the major one of

the two isomers.

In view of the fair yield of 18, we were encouraged to

attempt the cycloaddition of 1,3,5,7-cyclooctatetraene

(COT) onto CA, all the more so as this substrate has no

Scheme 8. Products of the photochemical reactions of choranil

allylic hydrogen atoms and hence cannot furnish products

like 19. Unfortunately, we did not observe any reaction at

all. The reason for this may be the colour of COT, which

(CA) with cyclohexene (CH).

Isolated in 10% yield, the third product was character-

does have the absorption maximum at 280 nm, but the band ised to be the 1:1 cycloadduct 20a by the ¹³C NMR spec-

extends into the visible region, which is why COT is yel- trum due to its similarity to the spectra of 11a and 11b. The

low.^[26]Thus, it may prevent the excitation of CA by com- number of lines indicates an unsymmetrical structure, which

pletely consuming the relevant light.

should be caused by the trans junction of the four-mem-

bered and the saturated six-membered ring. Another origin

for the missing symmetry could be the trans junction of the

cyclobutane and the cyclohexenedione subunit. However,

this possibility seems unlikely because of the rigidity of the

5. Reaction of Chloranil (CA) with Cyclohexene (CH)

Three products were isolated after a benzene solution of unsaturated six-membered ring, which would result in a sig-

a 2:1 mixture of CH/CA had been irradiated with a Hano- nificantly higher strain energy than that accepted by the

via medium-pressure mercury lamp, which was surrounded trans annulation of the cyclohexane subunit.

by a glass filter supposed to inhibit the passage of light of

In a second experiment, a 1:1 mixture of CH/CA, dis-

λ Ͻ 400 nm. According to the melting point and the ele- solved in benzene, was irradiated in a Rayonet photochemi-

mental analyses, two of the products were identified as the cal reactor with light of λ = 350 nm. Two of the three prod-

1:2 cycloadduct 9 (30% yield) and the monocyclohexenyl ucts formed were the same as in the first experiment,

ether 10 (20% yield) of TCH, which had been obtained by namely 10 and 20a. In addition, a symmetrical diastereomer

Xu et al.^[5]previously. These authors did not address the of 20a was obtained, for which the two structures 20b are

configuration of 9. Based on the ¹³C NMR spectrum, we possible (Scheme 8). The 1:2 cycloadduct 9 was not ob-

ascribe 9 an unsymmetrical structure. The ranges of the ¹³

C

served. Apparently, all CH was consumed by excited CA,

NMR chemical shifts of the CH₂, the CH, and the CCl and hence it was no longer available for the addition onto

groups of 9 are well in accord with the presence of a cis- the excited enediones 20.

and a trans-bicyclo[4.2.0]octane system, since the values of

In none of the two experiments, was any evidence found

the respective parent hydrocarbons exhibit substantial dif- for the spirooxetane 7. Thus, we believe that the report of

ferences.^[27]As estimated by quantum chemical calculations, Kim et al.^[17]is in error with regard to 7. On treatment of

the strain energy of trans-bicyclo[4.2.0]octane exceeds that the crude product of the first experiment with methanol, no

of the cis isomer by only 4.6 kcalmol^–1.^[28]Certainly, this compounds carrying a methoxy group were generated. This

quantity is not large enough to restrict the system to the cis finding excludes a photochemical process that would give

annulation. As in the case of 1:2 cycloadducts 5, we assume rise to a tetrachlorobenzofuranone derivative analogous to

that the cyclobutane entities should be orientated trans at compound 3 in the reaction of CP (Scheme 1).

Eur. J. Org. Chem. 2011, 968–982

www.eurjoc.org

975

M. Christl, M. Braun, O. Deeg, S. Wolff

FULL PAPER

6. Comparison of the Cycloalkenes as to Their Reactivity

towards Excited Chloranil (³CA) and the Stereochemical

Course of Their Cycloadditions

since a spin inversion has to take place. Thus, diradicals are

necessary intermediates. The unsymmetrical products 17a

and 20a, which arose with COD and CH, respectively, in

addition to the symmetrical ones 17b and 20b, show that

In contrast to that of cyclopentene (CP), the reactions of the intermediate diradicals exist in two conformations, one

cyclohexene (CH) and cis-cyclooctene (cis-CO) with excited of which undergoes the ring closure to cause a cis and the

CA do not produce tetrachlorobenzofuranone derivatives. other the trans annulation. Only one diradical conformation

Such a transformation seems to require the angle strain of each seems to be effective with cis- and trans-CO, namely

CP and its derivatives. No spirooxetanes emerged from CH that which closely approaches the shape of the starting cy-

and cis-CO. In this respect, these cycloalkenes behaved like cloalkene and hence ensures a suprafacial cycloaddition.

CP, which is in accord with the rule of Xu et al.^[5]that an

The mechanism of the reactions that the 1:1 cycloadducts

3

electron transfer from the alkene onto CA initiates spiro- of CA are subject to with another molecule of cycloalkene

oxetane formation, but occurs only with alkenes having an should be very similar to the pathway leading to them.

oxidation potential of Ͻ 2 V vs. SCE. On the basis of the Nevertheless, the stereochemical course of some of these

values of 2.01 (cis-CO),^[19]2.03 (CP),^[29]and 2.14 V processes deviates from that giving rise to the 1:1 cycload-

(CH),^[19]the previous reports about the production of the ducts. Thus, 11a was formed from CA and cis-CO as the

3

spirooxetanes 4^[16]and 7^[17]from cis-CO and CH, respec- sole isomer and reacted with cis-CO to give the three 1:2

tively, seemed unlikely and have indeed to be revised. How- cycloadducts 5a–c or 5aЈ–cЈ, two of which (5b,c or 5bЈ,cЈ)

ever, the outcome of the photochemical reaction of CA with exhibit a cis and the third (5a or 5aЈ) a trans configuration.

cis,cis-1,5-cyclooctadiene (COD) shows that the oxidation In contrast, the additions of trans-CO and CH onto 11a

potential is not a reliable criterion for or against spiro- proceeded highly stereoselectively and yielded 5a or 5aЈ and

oxetane formation. The value of 2.07 V vs. SCE^[19]is larger 15 indicating a suprafacial and an antarafacial cycload-

than those of cis-CO and CP, and yet the spirooxetane 6 dition, respectively. The suprafacial course was exclusively

was isolated in 7% yield in addition to the cyclobutane de- observed also on addition of trans-CO onto 11b, albeit the

rivatives 17 and 18, tetrachlorohydroquinone (TCH), and low yield of 5d or 5dЈ limits the significance of this result.

3

its monocyclooctadienyl ethers 19. Thus, the claim of Kim

Since the addition of COD and CH onto CA gave rise

to two 1:1 cycloadducts each (17a,b and 20a,b, respectively),

et al.^[17]was correct with regard to the observation of 6.

In the case of trans-cyclooctene (trans-CO), the oxidation it is surprising that only one 1:2 cycloadduct each (18 and

potential is not a good lead either, since the value of about 9, respectively) was obtained. A tentative explanation is that

1.75 V vs. SCE, estimated from the first vertical ionisation the cycloalkenes added suprafacially onto the trans isomers

potential (8.69 V^{[30] [31]}

would support the expectation of an 17a and 20a and antarafacially onto the cis isomers 17b and

)

electron transfer onto ³CA and thus the generation of a 20b. Beyond the finding that trans-CO reacted exclusively

spirooxetane, which we did not observe. Surprisingly, the suprafacially, further trends cannot be discerned, which is

reaction of trans-CO did not proceed as smoothly as that why predictions are difficult as to the stereochemical course

of cis-CO, which is testified by our failure to isolate pure of new reactions of the present types.

11b. It seems that the strain energy of trans-CO does not

promote the addition onto ³CA, although it exceeds that of

cis-CO by 9.3 kcalmol^–1.^[32]

Conclusions

The only cycloalkene to produce a 1:1 cycloadduct with

³CA in good yield, namely 11a, was cis-CO. The consecu-

The merit of this work consists in the elucidation of

tive reaction of 11a with cis-CO could easily be avoided by courses and products of the photochemical reactions of CA

the choice of light bulbs for the irradiation and benzene as with several cycloalkenes. The best condition for the prepa-

3

solvent. With CP, CH, and COD, CA gave rise to two 1:1 ration of 1:1 cycloadducts of the cyclobutane type is the

cycloadducts each, the yields of which were diminished, irradiation of the substrates in a ratio of 1:1 in benzene as

however, by the photoreduction of CA, furnishing TCH or solvent. Particularly beneficial is the employment of light

its monocycloalkenyl ethers. These products resulted from bulbs for the illumination, since they require no special

3

the abstraction of an allylic hydrogen atom by CA and a equipment, emit only little UV radiation and thus restrict

second abstraction by the radical generated from ³CA or by to a minimum the excitation of the 1:1 cycloadducts, which

the collapse of this radical with the allyl radical that origi- gives rise to 1:2 cycloadducts if the cycloalkene is present

nated from the cycloalkene. Presumably, an allylic C–H in excess. The good yield of 11a, obtained from CA and cis-

bond prefers this reaction type the better the closer parallel CO, shows that benzene as solvent is able to completely

it is orientated relative to the p orbitals that establish the suppress the excitation of 11a. However, except for cis- and

π bond, since then the transition state can optimally take trans-CO, the photoreduction of CA occurs even under

advantage from the resonance of the incipient allyl radical. these conditions with formation of TCH and/or its monocy-

The halfchair conformation of CH may best meet this pre- cloalk-2-en-1-yl ether, if the cycloalkene possesses allylic hy-

requisite, whereas cis- and trans-CO cannot fulfil it all.

drogen atoms.

Particularly interesting is the stereochemical course of

the cycloadditions onto CA, which proceed in two steps, 11a is well suited for the preparation of bis(cyclobutanes)

Owing to its good accessibility and its photoreactivity,

3

976

www.eurjoc.org

Eur. J. Org. Chem. 2011, 968–982

Photochemical Reactions of Chloranil

Irradiation of Chloranil (CA) in the Presence of Cyclopentene (CP):

At variance with this reaction carried out previously,^[15]now the

ratio of CA/CP was 1.22:1.00, and the light source was a Rayonet

photochemical reactor RPR-100 (λ = 350 nm). A solution of CA

(1.00 g, 4.07 mmol) and CP (227 mg, 3.33 mmol) in benzene

(250 mL) was irradiated at 20 °C for 2.5 h. The solvent was then

quickly evaporated in vacuo. Anhydrous methanol (150 mL) was

immediately added to the remaining buff oil, and the mixture was

refluxed for 16 h. After evaporation of the methanol in vacuo, the

residue was subjected to flash chromatography (SiO₂; LP/EA, 25:1)

such as 5a or 5aЈ and 15. Presumably, numerous alkenes, in

addition to trans-CO and CH, readily react with excited

11a and give rise to 1:2 cycloadducts of CA. On the other

hand, a series of 1:1 cycloadducts was synthesised from

noncyclic alkenes and ³CA previously.^[5,13]Akin to 11a,

these compounds should be able to undergo a photochemi-

cal [2+2] cycloaddition with various alkenes with formation

of 1:2 cycloadducts of CA. Such reactions are easily per-

formed, and thus the resulting bis(cyclobutanes) could gain

synthetic importance, because the elimination of the four at –30 °C. Three fractions were collected, all of which were some-

what contaminated by CA and other impurities. The first fraction

(227 mg, 21%) mainly contained the previously isolated 3b,5,6,7a-

tetrachloro-2,3,3a,3b,7a,7b-hexahydro-1H-cyclopenta[3,4]cyclo-

buta[1,2]benzene-4,7-dione^[15](1a or 1b) and a small quantity of

the 1:2 cycloadduct 2^[15](ratio of 1a or 1b/2 = 15:1). The second

fraction (193 mg, 18%) mainly consisted of the previously isolated

1a or 1b and its isomer 1b or 1a in a ratio of 1.0:1.5. The third

fraction (309 mg, 30%) was mainly a 1.0:4.4 mixture of the methyl

pseudoester 3b or 3c that was obtained as the minor isomer pre-

viously and the methyl ester 3a.^[15]Newly observed (3bα,7aα)-

chlorine atoms should be possible. The products would be

quinones having two cyclobutene subunits annulated. The

first half of such a transformation is documented by the

photoeliminations 5b Ǟ 12b, 5c Ǟ 12a, and 15 Ǟ 16.

Barltrop and Hesp^[13]reduced the adduct 21 of ³CA onto

2,3-dimethyl-1,3-butadiene with zinc in methanol/water and

isolated mainly the hydroquinone 22 (Scheme 9). If these

conditions can be successfully applied to 1:2 cycloadducts

of CA such as 5, 15, and analogous compounds, which

should be accessible as proposed above, a route existed for 3b,5,6,7a-tetrachloro-2,3,3a,3b,7a,7b-hexahydro-1H-cyclopenta[3,4]-

cyclobuta[1,2]benzene-4,7-dione (1b or 1a): ¹H NMR (250 MHz,

the preparation of hydroquinones containing no chlorine

atoms and having two cyclobutene subunits annulated. The

synthesis of the target quinones mentioned above would

CDCl₃): δ = 0.90 (m, 1 H), 1.40–2.30 (m, 5 H), 3.21 (m, 2 H) ppm.

¹³C NMR (63 MHz, CDCl₃): δ = 26.6 (C-2), 29.7 (C-1, C-3), 46.3

(C-3a, C-7b), 71.7 (C-3b, C-7a), 145.0 (C-5, C-6), 182.4 (C-4, C-7)

then just be a matter of an oxidation of the hydroquinones.

ppm.

Irradiation of Chloranil (CA) in the Presence of cis-Cyclooctene (cis-

CO): Entry 1 of Table 1. A solution of CA (1.00 g, 4.07 mmol) and

cis-CO (897 mg, 8.14 mmol) in benzene (150 mL) was irradiated at

10 °C by a medium-pressure mercury lamp (Hanovia, 450 W) by

utilising a Pyrex immersion well containing a glass filter, which

surrounded the lamp and was supposed to prevent the passage of

light of λ Ͻ 400 nm, for 5 h. The solvent was then evaporated in

vacuo. Anhydrous methanol (150 mL) was added to the residue and

Scheme 9. Dechlorination of the 1:1 photocycloadduct 21 accord-

the mixture refluxed for 16 h. After evaporation of the methanol

in vacuo, the residue was subjected to flash chromatography (SiO₂;

ing to Barltrop and Hesp.^[13]

pentane/EA, 35:1) with three fractions collected. The first one was

a yellow oil (430 mg, 23%), consisting of a 4:2:1 mixture of the

three 1:2 cycloadducts (6aα,6bβ,7aα,7bα,13aα,13bα,14aβ,14bβ)-

(5a)

or

(6aα,6bα,7aα,7bα,13aα,13bα,14aα,14bβ)-

(5aЈ),

Experimental Section

(6aα,6bβ,7aα,7bα,13aα,13bα,14aβ,14bα)- (5b) or (6aα,6bβ,

7aβ,7bβ,13aβ,13bβ,14aβ,14bα)- (5bЈ), and (6aα,6bα,7aβ,

7bβ,13aβ,13bβ,14aα,14bα)- (5c) or (6aα,6bα,7aα,7bα,13aα,

13bα,14aα,14bα)-6b,7a,13b,14a-tetrachloroeicosahydrobenzo-

[1ЈЈ,2ЈЈ:3,4;4ЈЈ,5ЈЈ:3Ј,4Ј]dicyclobuta[1,2:1Ј,2Ј]dicyclooctene-7,14-

dione (5cЈ). The second fraction was also a yellow oil (170 mg, 9%)

and turned out to be pure 5c or 5cЈ, which crystallised on treatment

with EA. According to the m.p., this compound had been obtained

previously.^[16]The third fraction (colourless oil, 190 mg, 13%) was

shown to be almost pure (4aα,4bα,10aα,10bα)-2,3,4a,10b-tetra-

chloro-4a,4b,5,6,7,8,9,10,10a,10b-decahydrobenzo[3,4]cyclobuta-

[1,2]cyclooctene-1,4-dione (11a).

General Details: NMR: Bruker AC 200, AC 250, Avance 400, and

DMX 600 spectrometers; chemical shifts are given in ppm relative

to Me₄Si (δ = 0.00 ppm) by using signals of the solvent (CDCl₃) as

internal reference (δ = 7.26 and 77.0 ppm; ¹H NMR signal of

CHCl₃and ¹³C NMR signal of CDCl₃, respectively). The assign-

ments of the ¹³C NMR signals are based on DEPT and C,H COSY

spectra. IR: Perkin–Elmer 1420 ratio-recording infrared spectro-

photometer, Perkin–Elmer 1605 FT-IR. UV: Hitachi U-3200. MS:

Varian MAT CH 7, Finnigan MAT 8200, and MAT 90. Melting

points: Kofler hot stage apparatus from C. Reichert, Optische

Werke AG, Vienna, Austria. Elemental analyses: Carlo Erba Stru-

mentatione Elemental Analyser Mod. 1106 and LECO Elemental

Analyser CHNS 932. Frequently used solvents: LP = light petro-

leum ether (b.p. 40–65 °C), EA = ethyl acetate.

1

Compound 5b or 5bЈ: H NMR (250 MHz, CDCl₃): δ = 1.10–2.00

(m, 24 H, CH₂), 2.96–3.22 (m, 4 H, CH) ppm. ¹³C NMR (63 MHz,

CDCl₃): δ = 23.4, 24.7, 25.2 (CH₂, 3 further signals of CH₂groups

are superimposed by signals of 5a or 5aЈ and 5c or 5cЈ), 47.8, 52.7

(CH), 70.6, 78.8 (CCl), 194.2 (CO) ppm. The MS of the 4:2:1 mix-

ture of 5a or 5aЈ, 5b or 5bЈ, and 5c or 5cЈ is very similar to that of

pure 5c or 5cЈ (see below).

General Conditions for the Photochemical Reactions: Thoroughly

dried solvents were used. A stream of nitrogen was passed through

the solution of the substrates for 15 min prior to irradiation. The

progress of the reaction, i.e. the consumption of chloranil (CA),

was monitored by TLC (SiO₂; LP/EA, 15:1) or by drop analysis

with a saturated solution of hexamethylbenzene in benzene, which

gives rise to a red charge-transfer complex with CA.

Compound 5c or 5cЈ: M.p. 234–237 °C (ref.^[16]235 °C). ¹H NMR

(250 MHz, CDCl₃): δ = 1.19–1.76 (m, 24 H, CH₂), 3.13 (m, 4 H,

Eur. J. Org. Chem. 2011, 968–982

www.eurjoc.org

977

M. Christl, M. Braun, O. Deeg, S. Wolff

FULL PAPER

1

CH) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 24.0, 25.0, 29.1 (CH₂), Compound 12b: M.p. 136–140 °C. H NMR (250 MHz, CDCl₃): δ

53.5 (CH), 73.6 (CCl), 194.3 (CO) ppm. IR (KBr): ν = 2920 (s, = 1.15–2.08 (m, 22 H, CH₂), 2.13 (apparent br. d, line distance

˜

br.), 2855 (m), 1716 (s) (ref.^[16]1720), 1463 (m), 1445 (m), 1159 (m), 14 Hz, 2 H, CH₂), 2.66 (m, 2 H, CH), 3.29 (apparent d, line dis-

1139 (m), 1056 (s), 879 (m), 843 (m), 787 (m), 672 (m) cm^–1. MS

tance 11 Hz, 2 H, CH) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 24.6,

(EI, 70 eV): m/z (%) = 466 (0.1) [M]⁺, 435, 433, 431, 429 (0.7, 6, 24.7, 25.2, 25.8, 28.8, 29.8 (CH₂), 46.5, 47.5 (CH), 76.9 (CCl), 157.2

16, 16) [M – Cl]⁺, 395 (14), 394 (15), 393 (18), 110 (87), 109 (60),

(C=C), 186.4 (CO) ppm. IR (KBr): ν = 2924 (s), 2852 (m), 1686

˜

95 (34), 91 (14), 82 (100), 81 (51), 79 (14), 77 (11), 69 (12), 68 (23), (s), 1464 (w), 1447 (w), 1321 (w), 1290 (w) cm^–1. MS (CI, NH₃,

67 (77), 55 (32), 54 (25), 41 (40). C₂₂H₂₈Cl₄O₂(466.3): calcd. C

56.67, H 6.05; found C 56.44, H 6.13.

150 eV): m/z (%) = 433, 431, 429 (10, 68, 100) [M + N₂H₇]⁺, 416,

414, 412 (10, 48, 72) [M + NH₄]⁺, 359 (18), 342 (13). C₂₂H₂₈Cl₂O₂

(395.4): calcd. C 66.83, H 7.14; found C 66.73, H 7.09.

For the data of compounds 11a and 5a or 5aЈ, see the next experi-

Entry 4 of Table 1: A stirred suspension of CA (10.0 g, 40.7 mmol)

in a mixture of benzene (200 mL) and cis-CO (8.95 g, 81.2 mmol)

was illuminated at 10 °C by several light bulbs (Osram, total power

500 W) for 60 h. Then the mixture was concentrated in vacuo and

the solid residue recrystallised from LP/EA to give pure 11a (11.0 g,

76%) as light yellow crystals.

ment and another one further below.

Entry 2 of Table 1: A solution of CA (2.00 g, 8.13 mmol) and cis-

CO (897 mg, 8.14 mmol) in benzene (250 mL) was irradiated at

20 °C in a Rayonet photochemical reactor RPR-100 (λ = 350 nm)

for 2.5 h. After evaporation of the benzene in vacuo, the remaining

buff oil was purified by flash chromatography (SiO₂; LP/EA, 35:1).

The resulting light yellow crystals of 11a (1.43 g) were washed with

pentane and recrystallised from EA to give colourless crystals

Entry 5 of Table 1: A solution of CA (800 mg, 3.25 mmol) and cis-

CO (3.59 g, 32.6 mmol) in acetonitrile (20 mL) was illuminated at

10 °C by several light bulbs (Osram, total power 1000 W) for 3.5 h.

Then the mixture was concentrated in vacuo and the residue ana-

lysed by ¹H NMR spectroscopy. In addition to 5c or 5cЈ and 11a in

a ratio of 1.0:1.3, further products were not present in a significant

quantity.

1

(1.13 g, 39%), m.p. 108–109 °C. H NMR (250 MHz, CDCl₃): δ =

1.20–1.50 (m, 8 H, CH₂), 1.60–1.85 (m, 4 H, CH₂), 3.21 (m, 2 H,

CH) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 23.9, 24.7, 28.9 (CH₂),

52.5 (CH), 71.4 (saturated CCl), 146.4 (olefinic CCl), 180.7 (CO)

ppm. IR (KBr): ν = 2940 (s), 2915 (s), 2911 (s, br.), 2870 (m), 2850

˜

(m), 1715 (s), 1697 (s), 1561 (s), 1461 (m), 1266 (m), 1198 (s), 1173

(m), 1121 (m), 1066 (m), 1022 (m), 945 (m), 772 (m), 702 (s) cm^–1.

UV (CH₂Cl₂): λ_max= 282 (logε = 4.08) nm. MS (EI, 70 eV): m/z

(%) = 325, 323, 321, 319 (2, 21, 63, 64) [M – Cl]⁺, 227 (21), 225

(21), 110 (25), 109 (54), 108 (20), 95 (29), 93 (20), 82 (63), 81 (46),

69 (20), 68 (23), 67 (100), 55 (43), 54 (34), 41 (56). C₁₄H₁₄Cl₄O₂

(356.1): calcd. C 47.22, H 3.96; found C 47.12, H 3.90.

Entry 6 of Table 1: A stirred suspension of CA (4.00 g, 16.3 mmol)

in a mixture of dichloromethane (100 mL) and cis-CO (8.00 g,

72.6 mmol) was illuminated at 10 °C by several light bulbs (Osram,

total power 1000 W) for 36 h. About 90% of the volatiles were then

evaporated in vacuo, and the oily residue was dissolved in a small

amount of LP. Storage of this solution at –35 °C afforded pure 5c

or 5cЈ (3.47 g, 46%) as colourless crystals.

Entry 3 of Table 1: A solution of CA (2.00 g, 8.13 mmol) and cis-

CO (1.79 g, 16.2 mmol) in benzene (300 mL) was irradiated at

10 °C in a Rayonet photochemical reactor RPR-100 (λ = 350 nm)

for 3 h. The resulting brown solution was filtered through silica gel

(5 g) and then concentrated in vacuo. The residue was subjected

to flash chromatography (SiO₂; LP/EA, 20:1) with four fractions

collected, which were characterised by their ¹³C NMR spectra.

First fraction (456 mg, 12%): a mixture of the 1:2 cycloadducts 5a

or 5aЈ, 5b or 5bЈ (minor component), and 5c or 5cЈ (major compo-

nent). Second fraction (155 mg, 4%): a 1:1 mixture of 5c or 5cЈ and

the 1:1 cycloadduct 11a. Third fraction (260 mg, 8%, of

Entry 7 of Table 1: A stirred suspension of CA (4.00 g, 16.3 mmol)

in a mixture of benzene (30 mL) and cis-CO (60 mL, 461 mmol)

was irradiated at 10 °C by a medium-pressure mercury lamp

(Hanau, 150 W) by utilising a Pyrex immersion well for 13.5 h. The

volatiles were then evaporated in vacuo, and the residue was ana-

lysed by ¹³C NMR spectroscopy. In addition to the signals of cis-

CO and the 1:2 cycloadducts 5a or 5aЈ, 5b or 5bЈ, and 5c or 5cЈ

(ratio of 2:5:5) those of unidentified components were present, but

the search for absorptions of the spirooxetane 4 was of no avail,

and even for 11a no evidence was found. The crude product was

extracted with hot LP until it had turned into a colourless solid,

which was identified to be 5c or 5cЈ (1.69 g, 22%). The combined

extracts were concentrated to dryness in vacuo, and the solid resi-

due was recrystallised from LP/EA to furnish a second crop of 5c

or 5cЈ (846 mg, 11%). A yield of 42% was reported previously.^[16]

yellow

crystals

after

crystallisation

from

LP/EA):

(6aα,7aβ,7bβ,13aβ,13bβ,14bα)- or (6aα,7aα,7bα,13aα,13bα,14bα)-

7a,13b-dichloro-1,2,3,4,5,6,6a,7a,7b,8,9,10,11,12,13,13a,13b,14b-

octadecahydrobenzo[1ЈЈ,2ЈЈ:3,4;4ЈЈ,5ЈЈ:3Ј,4Ј]dicyclobuta[1,2:1Ј,2Ј]-

dicyclooctene-7,14-dione (12a). Forth fraction (306 mg, 10%, of

yellow crystals after crystallisation from LP/EA): (6aα,7aβ,

7bα,13aα,13bβ,14bα)- or (6aα,7aα,7bβ,13aβ,13bα,14bα)-7a,13b-

dichloro-1,2,3,4,5,6,6a,7a,7b,8,9,10,11,12,13,13a,13b,14b-octa-

decahydrobenzo[1ЈЈ,2ЈЈ:3,4;4ЈЈ,5ЈЈ:3Ј,4Ј]dicyclobuta[1,2:1Ј,2Ј]di-

cyclooctene-7,14-dione (12b).

Hydrogenation of the 1:1 Cycloadduct 11a: In a solution of 11a

(2.10 g, 5.90 mmol) in methanol (100 mL), Pd/C (100 mg) was sus-

pended, and the mixture was stirred at room temperature in a hy-

drogenation apparatus under hydrogen (5 bar) for 5 d. The catalyst

was then removed by filtration through Celite, and the solvent was

evaporated in vacuo. The residue was recrystallised from glacial

1

Compound 12a: M.p. 136–140 °C. H NMR (250 MHz, CDCl₃): δ acetic acid to give pure (4bα,10aα)-2,3-dichloro-4b,5,6,7,8,9,10,10a-

= 1.00–2.10 (m, 22 H, CH₂), 2.12 (apparent br. d, line distance

14 Hz, 2 H, CH₂), 3.08 (m, 2 H, CH), 3.28 (apparent d, line dis-

octahydro-1,4-dihydroxybenzo[3,4]cyclobuta[1,2]cyclooctene (13;

1.47 g, 87%) as colourless needles, m.p. 208 °C. ¹H NMR

tance 11 Hz, 2 H, CH) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 24.2, (250 MHz, CDCl₃): δ = 1.35–1.80 (m, 10 H), 2.15 (apparent br. d,

24.6, 24.8, 25.7, 29.0, 29.8 (CH₂), 47.3, 52.4 (CH), 75.0 (CCl), 157.4 line distance 12 Hz, 2 H), 3.43 (apparent d, line distance 10 Hz, 2

(C=C), 185.0 (CO) ppm. IR (KBr): ν = 2943 (m), 2919 (s), 2850

H, CH), 5.12 (s, 2 H, OH) ppm. ¹³C NMR (63 MHz, CDCl₃): δ =

˜

(m), 1695 (s), 1680 (s), 1627 (m), 1466 (m), 1442 (m), 1290 (w), 25.9, 26.0, 30.1 (CH₂), 46.3 (CH), 116.9 (CCl), 131.4 (C-4a, C-10b),

1100 (w), 810 (w), 694 (w) cm^–1. MS (CI, NH₃, 150 eV): m/z (%) 141.3 (COH) ppm. IR (KBr): ν = 3300 (s, br.), 2920 (s), 2850 (m),

˜

= 433, 431, 429 (10, 55, 85) [M + N₂H₇]⁺, 416, 414, 412 (17, 76, 1420 (s), 1340 (m), 1320 (m), 1260 (w), 1240 (w), 1220 (w), 1180

100) [M + NH₄]⁺, 359 (13), 342 (15). C₂₂H₂₈Cl₂O₂(395.4): calcd. (w), 875 (m), 835 (w), 805 (w) cm^–1. MS (EI, 70 eV): m/z (%) =

C 66.83, H 7.14; found C 66.58, H 7.15.

290, 288, 286 (10, 63, 100) [M]⁺, 245 (25), 243 (30), 232 (26), 231

978

www.eurjoc.org

Eur. J. Org. Chem. 2011, 968–982

Photochemical Reactions of Chloranil

(44), 230 (38), 229 (46), 217 (21), 207 (23), 206 (22), 205 (23), 204

(m) cm^–1. MS (EI, 70 eV): m/z (%) = 435, 433, 431, 429 (0.4, 5, 16,

(26), 195 (26), 194 (20), 193 (30), 181 (22). C₁₄H₁₆Cl₂O₂(287.2): 18) [M – Cl]⁺, 396 (16), 395 (19), 394 (24), 393 (20), 321 (11), 320

calcd. C 58.55, H 5.62; found C 58.25, H 5.70.

(11), 299 (14), 110 (34), 109 (96), 95 (40), 93 (12), 91 (22), 82 (47),

81 (44), 79 (23), 77 (17), 70 (16), 69 (15), 68 (15), 67 (100), 61 (11),

55 (38), 54 (19), 45 (15), 43 (91), 42 (21), 41 (59), 39 (13).

C₂₂H₂₈Cl₄O₂(466.3): calcd. C 56.67, H 6.05; found C 56.39, H

5.97.

Reduction of the 1:1 Cycloadduct 11a with NaBH₄: A stirred solu-

tion of 11a (2.00 g, 5.62 mmol) in ethanol (15 mL) was treated with

small portions of NaBH₄(total amount 500 mg, 13.2 mmol) at

room temperature. Stirring was then continued for 1 h, after

which period the mixture was refluxed for 15 min, cooled to room

temperature and acidified with 2 m hydrochloric acid. Some insolu-

ble material was removed by filtration, and the filtrate was ex-

tracted with diethyl ether (5ϫ 20 mL). The combined extracts were

dried with MgSO₄and concentrated to dryness in vacuo. By

recrystallisation from glacial acetic acid, the residue furnished

colourles crystals of (1α,4α,4aα,4bα,10aα,10bα)- (14a) or (1α,4α,

4aβ,4bβ,10aβ,10bβ)-2,3,4a,10b-tetrachloro-1,4,4a,4b,

5,6,7,8,9,10,10a,10b-dodecahydro-1,4-dihydroxybenzo[3,4]cyclo-

buta[1,2]cyclooctene (14b) (1.72 g, 85%), which turned out to be

hygroscopic, m.p. 243–246 °C. ¹H NMR (250 MHz, CDCl₃): δ =

1.03–1.46 (m, 6 H), 1.53–1.69 (m, 4 H), 2.16 (br. dt, J = 14.1,

3.9 Hz, 2 H), 2.82 (br. d, J = 2.5 Hz, 2 H, OH), 3.01 (AAЈ part of

an AAЈXXЈ spectrum, 2 H, 4b-H, 10a-H), 4.44 (d, J = 2.5 Hz, 2

H, 1-H, 4-H) ppm; the assignment of the signal at δ = 2.82 ppm to

the OH groups was achieved by treatment of the sample with D₂O,

which made the br. d disappear and the d at δ = 4.44 ppm collapse

to a sharp s. ¹³C NMR (63 MHz, CDCl₃): δ = 21.1, 25.6, 30.1

(CH₂), 50.5 (C-4b, C-10a), 73.5 (C-1, C-4), 75.6 (C-4a, C-10b),

Irradiation of CA in the Presence of trans-CO. Formation of

(4aα,4bα,10aβ,10bα)-2,3,4a,10b-Tetrachloro-4a,4b,5,6,7,8,9,

10,10a,10b-decahydrobenzo[3,4]cyclobuta[1,2]cyclooctene-1,4-dione

(11b) and (6aα,6bβ,7aα,7bα,13aβ,13bα,14aβ,14bβ)- or (6aα,6bβ,

7aβ,7bβ,13aα,13bβ,14aβ,14bβ)-6b,7a,13b,14a-Tetrachloroeicosahy-

drobenzo[1ЈЈ,2ЈЈ:3,4;4ЈЈ,5ЈЈ:3Ј,4Ј]dicyclobuta[1,2:1Ј,2Ј]dicyclooctene-

7,14-dione (5d or 5dЈ): A solution of CA (2.00 g, 8.13 mmol) and

trans-CO (1.20 g, 10.8 mmol) in benzene (220 mL) was illuminated

at 10 °C by several light bulbs (Osram, total power 1000 W) for

4.5 h. After concentration of the mixture in vacuo, a ¹³C NMR

spectrum of the residue indicated the presence of 11b as major

component. Dissolution of the residue in the minimum amount of

acetone and storage of the solution at –35 °C gave rise to light

yellow crystals of 5d or 5dЈ (138 mg, 4% and 5% with reference to

CA and trans-CO, respectively). After concentration of the mother

liquor in vacuo, the residue was subjected to flash chromatography,

but the crystallisation of 11b could not be achieved, which is why

it is characterised by NMR spectra only.

Compound 11b: Oil (impure). ¹H NMR (250 MHz, CDCl₃): δ =

1.00–2.25 (m, 11 H), 2.37 (m, 1 H), 2.82–3.04 (m, 2 H) ppm. ¹³C

NMR (63 MHz, CDCl₃): δ = 26.2, 27.2 (double intensity), 27.8,

29.6, 29.7 (CH₂), 50.6, 53.9 (CH), 68.4, 77.2 (saturated CCl), 144.9,

145.6 (olefinic CCl), 180.0, 181.1 (CO) ppm.

Compound 5d or 5dЈ: M.p. 178–181 °C. ¹H NMR (250 MHz,

CDCl₃): δ = 1.05–1.40 (m, 10 H), 1.60–1.90 (m, 12 H), 2.15 (m, 2

H), 2.46 (m, 2 H), 2.84 (m, 2 H) ppm. ¹³C NMR (63 MHz, CDCl₃):

δ = 25.8, 27.0, 27.1, 27.5, 28.7, 29.2 (CH₂), 50.5, 55.1 (CH), 68.7,

128.5 (C-2, C-3) ppm. IR (KBr): ν = 3447 (s, br.), 3378 (s, br.),

˜

2952 (m), 2922 (m), 2848 (m), 1618 (w), 1468 (m), 1444 (m), 1418

(m), 1387 (m), 1348 (m), 1277 (w), 1144 (s), 1128 (m), 1114 (m),

1032 (m), 1007 (m), 943 (m), 825 (m), 652 (m) cm^–1. MS (CI, isobu-

tane, 120–150 eV): m/z (%) = 347, 345, 343, 341 (6, 25, 54, 42) [M –

OH]⁺, 311, 309, 307, 305 (4, 28, 93, 100) [M – OH – HCl]⁺, 289

(10), 287 (13), 271 (18), 269 (13), 253 (11), 225 (19), 223 (19), 213

(16), 211 (46), 209 (48), 127 (11), 110 (16), 109 (43).

C₁₄H₁₈Cl₄O₂·2H₂O (396.1): calcd. C 42.45, H 5.60; found C 42.63,

H 4.76.

80.3 (CCl), 191.7 (CO) ppm. IR (KBr): ν = 2933 (s), 2850 (m),

˜

1727 (s), 1450 (m), 1436 (m), 1177 (m), 1115 (m), 1077 (m), 1005

(m), 764 (m), 605 (m) cm^–1. MS (EI, 70 eV): m/z (%) = 466 (0.3)

[M]⁺, 435, 433, 431, 429 (0.6, 4, 11, 12) [M – Cl]⁺, 396 (18), 395

(17), 394 (27), 393 (15), 358 (14), 356 (11), 322 (10), 320 (11), 299

(16), 110 (19), 109 (100), 95 (43), 91 (14), 82 (24), 81 (27), 79 (14),

77 (11), 68 (11), 67 (83), 55 (39), 54 (11), 41 (34). C₂₂H₂₈Cl₄O₂

(466.3): calcd. C 56.67, H 6.05; found C 56.68, H 5.89.

Monitoring of the Photochemical Reaction Between CA and cis-CO

by HPLC: A solution of CA (500 mg, 2.03 mmol) and cis-CO

(450 mg, 4.08 mmol) in benzene (100 mL) was irradiated at 10 °C

in a Rayonet photochemical reactor RPR-100 (λ = 350 nm). Before

the reactor was turned on and then every 30 min up to 2.5 h, a

sample of the mixture was withdrawn and immediately analysed by

HPLC. Instrument: Bruker LC 21. Column: Phenomenex Prodigy

reversed-phase ODS-3 (20ϫ4.6 mm, 5 μm). Solvent: acetonitrile/

water (90:1). Detector: Bruker UVD 160, operating at 280 nm. De-

termined by injection of authentic samples, the retention times were

2.30 (CA), 6.06 (11a), 14.7 (12a, 12b), and 18.5 min (5c or 5cЈ).

Irradiation of 11a in the Presence of cis-CO. Formation of the De-

chlorination Products 12a and 12b: A solution of 11a (1.47 g,

4.13 mmol) and cis-CO (456 mg, 4.14 mmol) in benzene (250 mL)

was irradiated at 10 °C in a Rayonet photochemical reactor RPR-

100 (λ = 350 nm) for 2 h. The resulting brown solution was filtered

through silica gel (5 g). Then the solvent was evaporated in vacuo,

and the residue was subjected to flash chromatography (SiO₂; LP/

EA, 20:1), which furnished two yellow solids. The first fraction

consisted of 12a (131 mg, 8%) and the second of 12b (186 mg,

11%).

Irradiation of 11a in the Presence of trans-Cyclooctene (trans-CO).

Formation of 5a or 5aЈ: Pure trans-CO was prepared according to

ref.^[33]A solution of 11a (700 mg, 1.97 mmol) and trans-CO

(217 mg, 1.97 mmol) in dichloromethane (30 mL) was illuminated

at 10 °C by several light bulbs (Osram, total power 1000 W) for

48 h. The mixture was then concentrated in vacuo. A ¹³C NMR

spectrum of the crude product indicated the presence of almost

Photolysis of 5c or 5cЈ. Formation of 12a or the Corresponding Al-

ternative Isomer: A solution of 5c or 5cЈ (38 mg, 0.081 mmol) in

pure 5a or 5aЈ, which gave light yellow crystals (660 mg, 72%) on C₆D₆(0.7 mL) was irradiated at room temperature in a Rayonet

treatment with LP/EA, m.p. 164–166 °C. ¹H NMR (250 MHz, photochemical reactor RPR-100 (λ = 350 nm) for 75 min. There-

CDCl₃): δ = 1.05–2.05 (m, 24 H, CH₂), 2.80–2.98 (m, 2 H, CH), after, the ¹³C NMR spectrum documented a ratio of 5c or 5cЈ/

1

3.00–3.22 (m, 2 H, CH) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 12a or the corresponding alternative isomer = 1:2. The H NMR

23.4, 23.7, 24.7 (double intensity), 25.8, 26.5, 27.0, 27.5, 28.6, 29.14,

29.17, 29.4 (CH₂), 48.1, 52.88, 52.94, 55.5 (CH), 70.3, 72.1, 74.1,

spectrum was charcterised by a substantial line broadening. The

sample showed a smell of hydrogen chloride, and a wet test paper

81.5 (CCl), 191.0, 193.9 (CO) ppm. IR (KBr): ν = 2926 (s), 2858 indicated the presence of an acid. Continued irradiation did not

˜

(m), 1723 (s), 1459 (m), 1446 (m), 1158 (w), 1070 (m), 874 (w), 674

change the ratio of 5c or 5cЈ/12a or the corresponding alternative

Eur. J. Org. Chem. 2011, 968–982

www.eurjoc.org

979

M. Christl, M. Braun, O. Deeg, S. Wolff

FULL PAPER

isomer, but the H NMR spectrum displayed a progressive decom-

1

(20:1) 2 L, then (10:1) 1 L] to give four fractions, which were

colourless solids. The first one consisted of 18 [838 mg, 22%; in

addition to the configurations shown in Scheme 7, further possible

configurations are (6aα,6bα,7aα,7bα,13aβ,13bα,14aα,14bα) and

(6aα,6bβ,7aβ,7bβ,13aα,13bβ,14aβ,14bα)], the second of 6 (212 mg,

7%), the third of the two isomers 19 as a 2:1 mixture (689 mg,

24 %), and the forth of TCH (112 mg, 6 %). Because of the ex-

tremely low solubility of TCH in the eluant, the amount of TCH

actually formed may have been larger than the one isolated. The

first three fractions were recrystallised from LP/EA, whereby 6, 18,

and the major one of the isomers 19 were obtained pure.

position of the sample.

Irradiation of 11a in the Presence of Cyclohexene (CH). Formation

of (6aα,6bα,7aβ,7bβ,11aα,11bβ,12aα,12bα)- or (6aα,6bα,7aα,7b-

β,11aα,11bα,12aα,12bα)-6b,7a,11b,12a-Tetrachlorooctadecahydro-

cycloocta[3,4]cyclobuta[1,2-b]biphenylene-7,12-dione (15) and

(6aα,7aβ,7bβ,11aα,11bβ,12bα)- or (6aα,7aα,7bβ,11aα,11bα,12bα)-

7a,11b-Dichloro-1,2,3,4,5,6,6a,7a,7b,8,9,10,11,11a,11b,12b-hexa-

decahydrocycloocta[3,4]cyclobuta[1,2-b]biphenylene-7,12-dione (16):

A solution of 11a (2.00 g, 5.61 mmol) and CH (923 mg, 11.2 mmol)

in benzene (220 mL) was irradiated at 10 °C in a Rayonet photo-

chemical reactor RPR-100 (λ = 350 nm) for 4 h. The mixture was

then concentrated in vacuo, and the residue was subjected to flash

chromatography [SiO₂; LP/EA (20:1) 1 L, then (10:1) 0.5 L, finally

(5:1) 0.5 L]. In the order of elution, 15 (814 mg, 33%) and 16

(503 mg, 24%) were obtained. Recrystallisation from LP/EA fur-

nished a colourless and a yellow solid, respectively.

Compound 15: M.p. 269–271 °C. ¹H NMR (200 MHz, CDCl₃): δ =

1.15–2.00 (m, 20 H), 2.48 (ddd, J = 13.6, 9.6, 3.6 Hz, 1 H, 7b-H

or 11a-H), 2.70 (ddd, J = 13.6, 11.0, 3.3 Hz, 1 H, 11a-H or 7b-H),

3.05–3.21 (m, 2 H, 6a-H, 12b-H) ppm. ¹³C NMR (50 MHz,

CDCl₃): δ = 22.9, 24.5, 24.7, 24.8, 24.9, 25.1, 27.4 (double inten-

sity), 29.0, 29.2 (CH₂), 48.6, 52.5, 53.2, 55.2 (CH), 69.6, 74.2, 75.1,

1

Compound 6: M.p. 131–132 °C. H NMR (250 MHz, CDCl₃): δ =

1.39 (dddd, J = 12.7, 10.0, 5.6, 4.2 Hz, 1 H), 1.85 (tdd, J = 12.7,

6.5, 4.5 Hz, 1 H), 2.00–2.35 (m, 6 H), 3.60 (ddd, J = 12.5, 8.7,

5.5 Hz, 1 H, 1Ј-H), 5.14 (ddd, J = 10.0, 8.7, 5.5 Hz, 1 H, 8Ј-H),

5.55 (ddd, J = 10.5, 9.4, 7.1 Hz, 1 H), 5.71 (ddd, J = 10.7, 8.5,

7.1 Hz, 1 H) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 21.0, 23.2,

23.7, 32.0 (CH₂), 53.0 (C-1Ј), 84.8 (C-8Ј), 85.7 (C-1), 129.1, 130.2

(HC=CH), 129.9, 130.7 (C-3, C-5), 151.6, 152.4 (C-2, C-6), 169.8

(C-4) ppm. IR (KBr): ν = 3016 (w), 2938 (m), 2861 (w), 1686 (s),

˜

1607 (w), 1578 (s), 1459 (m), 1394 (w), 1278 (m), 1111 (s), 996 (w),

975 (m), 962 (m), 911 (w), 877 (w), 826 (w), 790 (w), 754 (m), 742

(m), 728 (s), 671 (w) cm^–1. MS (EI, 70 eV): m/z (%) = 358, 356,

354, 352 (0.1, 0.6, 1.0, 0.8) [M]⁺, 323, 321, 319, 317 (0.4, 2.3, 7.7,

7.8) [M – Cl]⁺, 248 (15), 246 (15), 245 (12), 108 (51), 107 (17), 93

(44), 91 (11), 87 (12), 81 (27), 80 (100), 79 (95), 78 (12), 77 (12), 67

(91), 66 (28), 55 (11), 54 (94), 53 (13), 41 (33), 39 (26). C₁₄H₁₂Cl₄O₂

(354.1): calcd. C 47.49, H 3.42; found C 47.50, H 3.28.

82.2 (CCl), 190.5, 194.8 (CO) ppm. IR (KBr): ν = 2935 (s), 2866

˜

(m), 1734 (s), 1724 (s), 1461 (m), 1448 (m), 1167 (m), 1133 (m),

1109 (m), 914 (m), 767 (m), 684 (m), 609 (m) cm^–1. MS (EI, 70 eV):

m/z (%) 442, 440, 438, 436 (0.2, 0.4, 0.7, 0.4) [M]⁺, 407, 405, 403,

401 (1, 7, 21, 22) [M – Cl]⁺, 368 (20), 367 (34), 366 (25), 365 (46),

109 (50), 95 (23), 91 (32), 82 (50), 81 (70), 79 (26), 77 (32), 67 (100),

55 (29), 41 (47). C₂₀H₂₄Cl₄O₂(438.2): calcd. C 54.82, H 5.52; found

C 54.61, H 5.60.

Compound 16: M.p. 270–272 °C. ¹H NMR (600 MHz, CDCl₃): δ =

1.21–1.37 (m, 3 H), 1.40–1.48 (m, 6 H), 1.55–1.68 (m, 3 H), 1.69–

1.76 (m, 3 H), 1.78–1.86 (m, 2 H), 1.97 (ddd, J = 13.1, 10.8, 3.6 Hz,

1 H, 7b-H or 11a-H), 2.05 (m, 1 H), 2.11 (dm, J = 15 Hz, 1 H, 1-

H or 6-H), 2.18 (dm, J = 15 Hz, 1 H, 6-H or 1-H), 2.57 (ddd, J =

13.1, 11.4, 3.2 Hz, 1 H, 11a-H or 7b-H), 3.25 (ddd, J = 12.0, 4.3,

1.9 Hz, 1 H, 6a-H or 12b-H), 3.31 (ddd, J = 12.1, 4.3, 2.0 Hz, 1 H,

12b-H or 6a-H) ppm. ¹³C NMR (151 MHz, CDCl₃): δ = 24.5, 24.8

(C-1, C-6), 25.1, 25.4 (C-9, C-10), 25.79, 25.82 (C-3, C-4), 27.3,

27.8 (C-8, C-11), 29.8, 29.9 (C-2, C-5), 47.2, 47.3 (C-6a, C-12b),

51.5, 51.8 (C-7b, C-11a), 72.4, 82.2 (C-7a, C-11b), 156.5, 157.3 (C-

6b, C-12a), 184.3, 185.3 (C-7, C-12) ppm; the assignments are

based on HMBC, C,H-COSY, and H,H-COSY spectra. IR (KBr):

Compound 18: M.p. 195–197 °C. ¹H NMR (250 MHz, CDCl₃): δ =

1.10–1.41 (m, 2 H), 1.68–1.83 (m, 2 H), 1.90–2.30 (m, 12 H), 2.72

(ddd, J = 11.8, 9.5, 6.1 Hz, 1 H), 2.80 (ddd, J = 11.6, 9.8, 5.9 Hz,

1 H), 2.92 (td, J = 11.8, 4.3 Hz, 1 H), 2.97 (td, J = 11.7, 4.8 Hz, 1

H), 5.54–5.71 (m, 4 H) ppm. ¹³C NMR (63 MHz, CDCl₃): δ =

22.5, 22.7, 23.2, 23.4, 27.9, 28.31, 28.33, 28.6 (CH₂), 45.2, 48.0,

52.9, 53.6 (saturated CH), 70.5, 73.1, 76.1, 78.8 (CCl), 129.8, 130.0

(double intensity), 130.2 (olefinic CH), 191.7, 192.8 (CO) ppm. IR

(KBr): ν = 3017 (m), 2941 (m), 2910 (m), 2852 (w), 1727 (s), 1465

˜

(m), 1439 (m), 1205 (m), 1186 (m), 1129 (m), 1088 (w), 1068 (m),

1010 (w), 875 (w), 756 (w), 734 (m), 671 (m) cm^–1. MS (EI, 70 eV):

m/z (%) = 464, 462, 460 (0.3, 0.7, 0.5) [M]⁺, 392 (20), 391 (21), 390

(32), 389 (23), 108 (49), 107 (99), 93 (32), 91 (42), 81 (20), 80 (43),

79 (100), 78 (21), 77 (37), 67 (53), 54 (28), 41 (37), 39 (26).

C₂₂H₂₄Cl₄O₂(462.2): calcd. C 57.17, H 5.23; found C 56.88, H

5.20.

ν = 2919 (s), 2850 (m), 1697 (s), 1686 (s), 1462 (m), 1321 (m), 1098

1

˜

Major Isomer 19: M.p. 113–115 °C. H NMR (250 MHz, CDCl₃):

(w), 993 (m), 808 (w), 688 (w), 590 (m) cm^–1. MS (CI, NH₃,

150 eV): m/z (%) = 405, 403, 401 (11, 65, 100) [M + N₂H₇]⁺, 388,

386, 384 (7, 39, 56) [M + NH₄]⁺. C₂₀H₂₄Cl₂O₂(367.3): calcd. C

65.40, H 6.59; found C 65.17, H 6.42.

δ = 2.19 (m, 1 H), 2.29–2.45 (m, 3 H), 2.64–2.86 (m, 2 H), 5.48–

5.71 (m, 4 H), 5.80 (ddd, J = 12.0, 5.5, 1.5 Hz, 1 H), 5.91 (s, 1 H,

OH) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 27.93, 27.95, 34.8

(CH₂), 82.4 (HCO), 119.0, 127.5 (CCl), 125.6, 129.4, 129.47, 129.54

(HC=CH), 145.7, 145.8 (aromatic CO) ppm. IR (KBr): ν = 3433

˜

Irradiation of CA in the Presence of cis,cis-1,5-Cyclooctadiene

(m, br.), 3020 (w), 2937 (w), 2876 (w), 2827 (w), 1436 (s), 1388 (s),

1320 (m), 1279 (m), 1203 (m), 1168 (w), 1033 (w), 986 (w), 967 (w),

939 (m), 902 (m), 880 (w), 816 (w), 791 (w), 717 (w), 668 (w) cm^–1.

MS (EI, 70 eV): m/z (%) = 254, 252, 250, 248, 246 (0.2, 2, 9, 17,

14) [C₆H₂Cl₄O₂]⁺, 107 (50), 106 (46), 105 (12), 91 (48), 79 (100), 78

(30), 77 (15), 41 (14). C₁₄H₁₂Cl₄O₂(354.1): calcd. C 47.49, H 3.42;

found C 47.30, H 3.53.

(COD)

Isolation of 2,3,5,6-Tetrachlorospiro[cyclohexa-2,5-diene-1,10Ј-

[9]oxabicyclo[6.2.0]dec[4]en]-4-one (6), 6b,7a,13b,14a-Tetrachloro-

1,2,5,6,6a,6b,7a,7b,8,9,12,13,13a,13b,14a,14b-hexadecahydrobenzo-

[1ЈЈ,2ЈЈ:3,4;4ЈЈ,5ЈЈ:3Ј,4Ј]dicyclobuta[1,2:1Ј,2Ј]dicyclooctene-7,14-di-

one (18), 2,3,5,6-Tetrachloro-4-(cycloocta-2Ј,6Ј-dien-1Ј-yloxy)phenol

(19a), 2,3,5,6-Tetrachloro-4-(cycloocta-2Ј,5Ј-dien-1Ј-yloxy)phenol

(19b), and TCH: A stirred suspension of CA (2.00 g, 8.13 mmol) in

a mixture of COD (20 mL, 163 mmol) and benzene (10 mL) was

illuminated at 10 °C by several light bulbs (Osram, total power

1000 W) for 30 h. Then the volatiles were removed in vacuo, and

the residue was subjected to flash chromatography [SiO₂; LP/EA

Minor Isomer 19: ¹³C NMR (63 MHz, CDCl₃): δ = 23.1, 28.0, 29.0,

29.4 (CH₂, one of the 4 signals is from an impurity), 83.1 (HCO),

118.9, 127.4 (CCl), 128.3, 128.7, 128.8, 129.4, 130.2 (HC=CH, one

of the 5 signals is from an impurity), 145.6, 146.1 (aromatic CO)

ppm.

980

www.eurjoc.org

Eur. J. Org. Chem. 2011, 968–982

Photochemical Reactions of Chloranil

Observation of 2,3,4a,10b-Tetrachloro-4a,4b,5,6,9,10,10a,10b-octa-

hydrobenzo[3,4]cyclobuta[1,2]cyclooctene-1,4-dione (17): A solution

of CA (2.00 g, 8.13 mmol) and COD (880 mg, 8.13 mmol) in ben-

zene (220 mL) was illuminated at 10 °C by several light bulbs (Os-

ram, total power 700 W) for 8 h. The solvent was then evaporated

in vacuo. A ¹³C NMR spectrum of the crude product indicated the

presence of significant amounts of both isomers 19 and a small

quantity of 6. Signals of further important components were as-

cribed to two isomers 17, whereas no evidence for 18 was found.

Although 6 and 19 were removed from the mixture by flash

chromatography (SiO₂; LP/EA, 20:1), an analytically pure sample

of 17 could not be obtained. Also, attempts to separate the isomers

17 failed. Number and intensities of the signals in the ¹³C NMR

spectrum indicate that a symmetrical compound 17 and an unsym-

metrical one were present in a ratio of 1:2. Red oil. ¹H NMR

(250 MHz, CDCl₃): δ = 1.0–3.4 (several m with the major one be-

tween δ = 1.9 and 2.3 ppm), 5.5–5.7 (m, 2 H) ppm. ¹³C NMR

(63 MHz, CDCl₃): δ = 22.4, 23.3, 24.4, 25.2, 27.8, 28.6 (CH₂), 47.8,

50.6, 52.1 (saturated CH), 69.4, 70.0, 76.1 (saturated CCl), 129.2,

130.0 (double intensity) (HC=CH), 145.0, 145.6 (olefinic CCl; 1

signal is missing, possibly due to too low intensity), 180.3, 180.5,

181.2 (CO) ppm.

Compound 10: M.p. 101–102 °C (ref.^[5]105–107 °C). ¹H NMR

(400 MHz, CDCl₃): δ = 1.62 (m, 1 H), 1.81 (m, 1 H), 1.92–2.08 (m,

3 H), 2.15 (m, 1 H), 4.72 (m, 1 H, HCO), 5.89 (s, 1 H, OH), 5.89

(ddt, J = 10.1, 3.7, 1.8 Hz, 1 H), 6.01 (dddd, J = 10.1, 4.0, 3.0,

1.0 Hz, 1 H) (HC=CH) ppm. ¹³C NMR (63 MHz, CDCl₃): δ =

18.7, 25.1, 28.8 (CH₂), 78.2 (HCO), 118.9, 127.7 (CCl), 125.8, 133.1

(HC=CH), 145.6, 146.0 (aromatic CO) ppm. IR (KBr): ν = 3440

˜

(m, br.), 3040 (w), 2950 (m), 2910 (w), 2880 (w), 2840 (w), 1691

(w), 1681 (w), 1439 (s), 1390 (s), 1358 (w), 1333 (w), 1322 (m), 1282

(m), 1205 (m), 1173 (w), 1162 (w), 1119 (w), 1003 (m), 941 (m),

922 (m), 911 (m), 889 (w), 881 (m), 736 (w), 719 (m) cm^–1. MS (EI,

70 eV): m/z (%) = 295, 293, 291 (0.8, 2.0, 2.1) [M – Cl]⁺, 250 (10),

248 (26), 246 (35), 244 (19), 211 (17), 209 (17), 89 (11), 87 (33), 81

(95), 80 (100), 79 (35), 53 (12), 41 (12).

1

Compound 20a: M.p. 118–119 °C. H NMR (250 MHz, CDCl₃): δ

= 1.20–1.45 (m, 3 H), 1.60–2.05 (m, 5 H), 2.12 (m, 1 H), 2.70 (m,

1 H) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 24.9, 25.2, 27.2, 27.7

(CH₂), 51.3, 52.6 (CH), 68.7, 78.4 (saturated CCl), 145.4, 145.8

(olefinic CCl), 179.8, 181.3 (CO) ppm. IR (KBr): ν = 2965 (m),

˜

2945 (m), 2880 (m), 2855 (m), 1701 (s), 1680 (m), 1559 (s), 1459

(w), 1449 (m), 1252 (m), 1222 (m), 1195 (m), 1150 (m), 1136 (m),

1120 (w), 1062 (m), 1050 (w), 1029 (m), 950 (m), 935 (w), 923 (w),

889 (w), 780 (w), 733 (m), 721 (m), 713 (m) cm^–1. MS (EI, 70 eV):

m/z (%) = 297, 295, 293, 291 (4, 32, 100, 100) [M – Cl]⁺, 246 (29),

244 (20), 211 (23), 209 (25), 87 (44), 81 (52), 67 (39), 55 (24), 43

(29), 41 (31). C₁₂H₁₀Cl₄O₂(328.0): calcd. C 43.97, H 3.07; found

C 45.38, H 3.18.

Irradiation of CA in the Presence of Cyclohexene (CH)

Isolation of 4b,5a,9b,10a-Tetrachlorohexadecahydrobenzo[3,4]cyclo-

buta[1,2-b]biphenylene-5,10-dione (9), 2,3,5,6-Tetrachloro-4-(cy-

clohex-2Ј-en-1Ј-yloxy)phenol (10), and (4aα,4bα,8aβ,8bα)-2,3,4a,8b-

Tetrachloro-4a,4b,5,6,7,8,8a,8b-octahydrobiphenylene-1,4-dione

(20a): A solution of CA (1.00 g, 4.07 mmol) and CH (669 mg,

8.14 mmol) in benzene (150 mL) was irradiated at 10 °C by a me-

dium-pressure mercury lamp (Hanovia, 450 W) by utilising a Pyrex

immersion well containing a glass filter, which surrounded the lamp

and was supposed to prevent the passage of light with λ Ͻ 400 nm,

for 12 h. The solvent was then evaporated in vacuo, and the residue

was subjected to flash chromatography (SiO₂; pentane/EA, 25:1).

In the order of elution, 9 [496 mg, 30%; ref.^[5]20%; in addition to

the configurations shown in Scheme 8, further possible configura-

tions are (4aα,4bα,5aα,5bα,9aβ,9bα,10aα,10bα) and (4aα,4b-

β,5aβ,5bβ,9aβ,9bα,10aβ,10bα)], 20a (140 mg, 10%, slightly con-

taminated by 9 and other components), and 10 (261 mg, 20%) were

obtained as oils. Dissolution of these oils in the minimum amount

of EA and storage of the solutions at –35 °C gave rise to colourless

crystals in all three cases.

Observation of (4aα,4bα,8aα,8bα)- or (4aα,4bβ,8aβ,8bα)-2,3,4a,8b-

Tetrachloro-4a,4b,5,6,7,8,8a,8b-octahydrobiphenylene-1,4-dione

(20b): A solution of CA (1.00 g, 4.07 mmol) and CH (334 mg,

4.07 mmol) in benzene (250 mL) was irradiated at 20 °C in a

Rayonet photochemical reactor RPR-100 (λ = 350 nm) for 4.5 h.

After evaporation of the benzene in vacuo, the remaining buff oil

was subjected to flash chromatography (SiO₂; LP/tert-butyl methyl

ether, 9:1). In the order of elution, 20b (125 mg, 9%), 20a (107 mg,

8%) and 10 (234 mg, 18%) were obtained as oils, which still con-

tained impurities to the extent of about 10%.

1

Compound 20b: H NMR (250 MHz, CDCl₃): δ = 1.18–1.41 (m, 4

H), 1.52–1.93 (m, 4 H), 3.24 (m, 2 H) ppm. ¹³C NMR (63 MHz.

CDCl₃): δ = 21.3, 22.3 (CH₂), 44.3 (CH), 69.4 (saturated CCl),

146.4 (olefinic CCl), 180.9 (CO) ppm.

[1] K. Maruyama, A. Osuka in The Chemistry of Quinonoid Com-

pounds (Eds.: S. Patai, Z. Rappoport), Wiley, New York, 1988,

vol. 2, part 1, chapter13.

[2] a) D. Creed in CRC Handbook of Organic Photochemistry and

Photobiology (Eds.: W. M. Horspool, P.-S. Song), CRC Press,

Boca Raton, 1995, pp. 737–747; b) A. Gilbert in CRC Hand-

book of Organic Photochemistry and Photobiology, 2nd ed.

(Eds.: W. M. Horspool, F. Lenci), CRC Press, Boca Raton,

2004, chapter 87.

[3] S. M. Hubig, T. M. Bockman, J. K. Kochi, J. Am. Chem. Soc.

1997, 119, 2926–2935.

[4] a) D. Rehm, A. Weller, Isr. J. Chem. 1970, 8, 259–271; b) A.

Weller, Z. Phys. Chem. (Wiesbaden) 1982, 133, 93–98.

[5] J.-H. Xu, Y.-L. Song, Z.-G. Zhang, L.-C. Wang, J.-W. Xu, Tet-

rahedron 1994, 50, 1199–1210.

[6] P. M. Rentzepis, D. W. Steyert, H. D. Roth, C. J. Abelt, J. Phys.

Chem. 1985, 89, 3955–3960.

[7] J.-H. Xu, L.-C. Wang, J.-W. Xu, B.-Z. Yan, H.-C. Yuan, J.

Chem. Soc. Perkin Trans. 1 1994, 571–577.

[8] M. Christl, M. Braun, Angew. Chem. 1989, 101, 636–638; An-

gew. Chem. Int. Ed. Engl. 1989, 28, 601–603.

[9] M. Braun, M. Christl, O. Deeg, M. Rudolph, E.-M. Peters, K.

Peters, Eur. J. Org. Chem. 1999, 2093–2101.

In a second experiment, the irradiation was carried out as above,

but, after the removal of the benzene, the residue was refluxed in

methanol overnight. Thereafter, the same products as above were

observed and no compounds containing a methoxy group.

In addition to elemental analyses, Xu et al.^[5]reported ¹H NMR

and IR spectroscopic as well as MS data for 9 and 10. Since these

data are not extensive and seem to be not very precise, we add our

data.

Compound 9: M.p. 204–206 °C (ref.^[5]212–214 °C). ¹H NMR

(250 MHz, CDCl₃): δ = 1.10–2.00 (m, 16 H), 2.61–2.81 (m, 2 H),

3.09–3.31 (m, 2 H) ppm. ¹³C NMR (63 MHz, CDCl₃): δ = 21.2,

21.6, 23.1, 23.6, 24.9, 25.1, 27.0, 27.6 (CH₂), 44.8, 46.0, 47.6, 55.3

(CH), 69.2, 71.9, 76.6, 82.6 (CCl), 190.6, 195.1 (CO) ppm. IR

(KBr): ν = 2950 (s), 2870 (m), 1730 (s), 1455 (m), 1192 (w), 1170

˜

(w), 1158 (w), 1137 (w), 1121 (m), 1099 (w), 1028 (w), 958 (w), 918

(w), 898 (m), 811 (w), 849 (w), 688 (m) cm^–1. MS (EI, 70 eV): m/z

(%) = 414, 412, 410, 408 (0.1, 0.6, 1.3, 0.7) [M]⁺, 379, 377, 375, 373

(1, 9, 27, 27) [M – Cl]⁺, 340 (16), 339 (33), 338 (25), 337 (44), 301

(19), 273 (15), 169 (15), 105 (17), 91 (17), 82 (36), 81 (100), 79 (28),

77 (32), 67 (68), 55 (19), 53 (16), 41 (39), 39 (17).

Eur. J. Org. Chem. 2011, 968–982

www.eurjoc.org

981

M. Christl, M. Braun, O. Deeg, S. Wolff

FULL PAPER

[10] G. O. Schenck, E. K. von Gustorf, B. Kim, G. v. Bünau, G.

Pfundt, Angew. Chem. 1962, 74, 510–511; Angew. Chem. Int.

Ed. Engl. 1962, 1, 516.

[11] G. Jones II, W. A. Heany, J. Phys. Chem. 1986, 90, 5410–5414.

[12] M. Braun, Dissertation, University of Würzburg, 1990.

[13] J. A. Barltrop, B. Hesp, J. Chem. Soc. C 1967, 1625–1635.

[14] J. Xue, J.-W. Xu, L. Yang, J.-H. Xu, J. Org. Chem. 2000, 65,

30–40.

[28]

[29]

[30]

J.-L. Mieusset, U. H. Brinker, J. Org. Chem. 2007, 72, 10211–

10219.

P. G. Gassman, R. Yamaguchi, G. F. Koser, J. Org. Chem. 1978,

43, 4392–4393.

C. Batich, O. Ermer, E. Heilbronner, J. R. Wiseman, Angew.

Chem. 1973, 85, 302–303; Angew. Chem. Int. Ed. Engl. 1973,

12, 312–314.

[31]

This estimation is based on the vertical ionisation potentials of

CP and CH (9.18, 9.12 eV: P. Bischof, E. Heilbronner, Helv.

Chim. Acta 1970, 53, 1677–1682), norbornene (8.97 eV: P. Bi-

schof, J. A. Hashmall, E. Heilbronner, V. Hornung, Helv. Chim.

Acta 1969, 52, 1745–1749), cis-CO (8.98 eV: C. Batich, P. Bi-

schof, E. Heilbronner, J. Electron Spectrosc. Relat. Phenom.

1972/73, 1, 333–353), and COD (9.06 eV: R. McDiarmid, J. P.

Doering, J. Chem. Phys. 1981, 75, 2687–2692) and the calcula-

tion of the corresponding oxidation potentials by means of the

empirical correlations (L. L. Miller, G. D. Nordblom, E. A.

Mayeda, J. Org. Chem. 1972, 37, 916–918; P. G. Gassman, R.

Yamaguchi, J. Am. Chem. Soc. 1979, 101, 1308–1310), as well

as the experimental oxidation potentials (see text for those of

CP, CH, cis-CO, and COD and ref.^[29]for that of norbornene).

[15] M. Braun, M. Christl, E.-M. Peters, K. Peters, J. Chem. Soc.

Perkin Trans. 1 1999, 2813–2820.

[16] D. Bryce-Smith, A. Gilbert, Tetrahedron Lett. 1964, 5, 3471–

3473.

[17] S. S. Kim, D. Y. Yoo, I. H. Cho, S. C. Shim, Bull. Korean Chem.

Soc. 1987, 8, 296–298.

[18] Handbook of Tables for Organic Compound Identification, 3rd

ed., CRC Press, Cleveland, 1967, p. 126.

[19] T. Shono, A. Ikeda, J. Hayashi, S. Hakozaki, J. Am. Chem. Soc.

1975, 97, 4261–4264.

[20] K. Peters, E.-M. Peters, M. Braun, S. Wolff, M. Christl, Z.

Kristallogr. – New Cryst. Struct. 2000, 215, 45–46.

[21] A. R. Pinder, Synthesis 1980, 425–452.

[22] K. Schaffner, Pure Appl. Chem. 1968, 16, 75–85.

[23] N. J. Turro, Modern Molecular Photochemistry, Benjamin,

Menlo Park, 1978, pp. 559–560.

[24] A. P. Avdeenko, A. L. Yusina, Zh. Org. Khim. 1995, 31, 458–

461; Russ. J. Org. Chem. (Engl. Transl.) 1995, 31, 418–421.

[25] W. M. Horspool, Photochemistry 1997, 28, 341–378.

[26] G. I. Fray, R. G. Saxton, The Chemistry of Cyclooctatetraene

and Its Derivatives, Cambridge University Press, Cambridge,

1978, p. 7.

[32]

[33]

S. W. Benson, Thermochemical Kinetics, 2nd ed., Wiley, New

York, 1976, p. 273.

a) Y. Inoue, N. Yamasaki, A. Tai, D. Akira, Y. Daino, T. Yam-

ada, T. Hakushi, J. Chem. Soc. Perkin Trans. 2 1990, 1389–

1394; b) Y. Inoue, H. Tsuneishi, T. Hakushi, A. Tai in Photo-

chemical Key Steps in Organic Synthesis – An Experimental

Course Book (Eds.: J. Mattay, A. Griesbeck), VCH, Weinheim,

1994, p. 207.

[27] P. Metzger, C. Cabestaing, E. Casadevall, A. Casadevall, Org.

Magn. Reson. 1982, 19, 144–147.

Received: September 17, 2010

Published Online: December 22, 2010

982

www.eurjoc.org

Eur. J. Org. Chem. 2011, 968–982

Article Doi

Photochemical reactions of chloranil with cyclooctene, 1,5-cyclooctadiene, and cyclohexene revisited

DOI: 10.1002/ejoc.201001304

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejoc.201001304

Full text of DOI:10.1002/ejoc.201001304

Products guided by the article

R&D Labs maybe for 1274903-90-0

Relevant to this article

Hot Product