Intramolecular anodic olefin coupling reactions and the use of electron-rich aryl rings

1578

J . Org. Chem. 1996, 61, 1578-1598

In tr a m olecu la r An od ic Olefin Cou p lin g Rea ction s a n d th e Use of

Electr on -Rich Ar yl Rin gs¹

Dallas G. New, Zerom Tesfai, and Kevin D. Moeller*

Department of Chemistry, Washington University, St. Louis, Missouri 63130

Received October 11, 1995^X

The utility of intramolecular anodic olefin coupling reactions involving electron-rich aromatic rings

for constructing fused, bicyclic ring skeletons has been examined. Reactions involving alkoxy-

substituted phenyl rings were found to benefit strongly from a 3-methoxy substituent on the phenyl

ring. Although overoxidation of the bicyclic product was observed in these reactions, this problem

could be minimized with the use of controlled potential electrolysis conditions when a monomethoxy

phenyl ring was used and avoided entirely with the use of a vinyl sulfide moiety as the initiator

when a more electron-rich phenyl ring was used. Reactions involving 4-alkoxy-substituted phenyl

rings as substrates did not lead to good yields of fused products. Furan rings were found to be

excellent coupling partners for the reactions and afforded products having fused, bicyclic furan

ring skeletons. Cyclizations involving furans were shown to be compatible with the formation of

both six- and seven-membered rings, the generation of a quaternary carbon, and the use of a variety

of electron-rich olefins as the other coupling partner. It appears that the furan can serve as either

the initiating group or the terminating group for the cyclizations. Finally, the reactions were shown

to be compatible with the use of a pyrrole ring as one of the participants.

The intramolecular anodic olefin coupling reaction can

provide a unique means for generating new carbon-

carbon bonds.²The potential of this method is high-

lighted by the general way in which electrochemistry can

be used to generate reactive radical cation intermediates

from enol ether functional groups under neutral condi-

tions. The net result is to reverse the polarity of the enol

ether so that it will undergo coupling reactions with

nucleophiles. To date, intramolecular anodic olefin cou-

pling reactions have proven compatible with the forma-

tion of five-, six-, and seven-membered rings, the gen-

eration of quaternary carbons, and the synthesis of

angularly fused tricyclic ring skeletons. Our interest in

the use of intramolecular anodic olefin coupling reactions

for constructing fused polycyclic ring skeletons led us to

consider the use of aromatic rings as participants in these

reactions (Figure 1). In accord with earlier observations,

coupling reactions of this nature were expected to lead

to an addition of the aromatic ring to the olefin with a

regiochemistry opposite to what would be anticipated for

the corresponding Friedel-Crafts type of alkylation.

F igu r e 1.

used to synthesize spirocyclic ring skeletons.⁴These later

reactions are closely related to the cyclizations proposed.

Two examples are outlined in Scheme 1. Although

related, it is important to note that there is a significant

difference between these cyclizations and the ones pro-

posed in Figure 1. In the reactions outlined in Scheme

1, the initial oxidation involved the removal of an electron

from the electron-rich aromatic ring. The resulting

radical cation was then trapped by the olefin to form a

spirocyclic ring. The formation of the spirocyclic ring was

essential because it prevented reformation of the aro-

matic ring in the product and hence removed the elec-

troactive species from the reaction. There was no chance

for overoxidation of the product because the product did

not contain a functional group with a low enough oxida-

tion potential to compete with the initial oxidation of the

electron-rich phenyl ring. This would not be the case for

Surprisingly, cyclization reactions of this type had not

been studied previously, even though aromatic rings had

played critical roles in earlier anodic coupling reactions.

For example, intramolecular coupling reactions of aro-

matic rings have been used to synthesize a variety of

biphenyl ring skeletons³and intramolecular coupling

reactions of substituted phenols and olefins have been

^XAbstract published in Advance ACS Abstracts, February 15, 1996.

(1) Taken in part from the following: New, D. G. Ph.D. Thesis,

Washington University, St. Louis, MO, 1994, and Tesfai, Z. Ph.D.

Thesis, Washington University, St. Louis, MO, 1995. For preliminary

accounts of this work, see refs 2a,b.

(2) (a) Moeller, K. D.; New, D. G. Tetrahedron Lett. 1994, 35, 2857.

(b) Tesfai, Z.; Moeller, K. D. J . Electrochem. Soc. J pn. (Denki Kagaku)

1994, 62, 1115. (c) Hudson, C. M.; Moeller, K. D. J . Am. Chem. Soc.

1994, 116, 3347. (d) Tinao-Wooldridge, L. V.; Moeller, K. D.; Hudson,

C. M. J . Org. Chem. 1994, 59, 2381. (e) Moeller, K. D.; Tinao, L. V. J .

Am. Chem. Soc. 1992, 114, 1033. (f) Hudson, C. M.; Marzabadi, M. R.;

Moeller, K. D.; New, D. G. J . Am. Chem. Soc. 1991, 113, 7372.

(3) For a review, see: Yoshida, K. Electrooxidation in Organic

Chemistry: The Role of Cation Radicals as Synthetic Intermediates;

J ohn Wiley and Sons: New York, 1984; pp 136-151.

(4) (a) Swenton, J . S.; Carpenter, K.; Chen, Y.; Kerns, M. L.; Morrow,

G. W. J . Org. Chem. 1993, 58, 3308. (b) Maki, S.; Toyoda, K.;

Kosemura, S.; Yamamura, S. Chem. Lett. 1992, 1059. (c) Maki, S.;

Kosemura, S.; Yamamura, S.; Kawano, S.; Ohba, S. Chem Lett. 1992,

651. (d) Yamamura, S. In Electroorganic Synthesis; Little, R. D.,

Weinberg, N. L., Eds.; Marcel Dekker: New York, 1991; pp 309-315.

(e) Yamamura, S.; Shizuri, Y.; Shigemori, H.; Okuno, Y.; Ohkubo, M.

Tetrahedron 1991, 47, 635. (f) Morrow, G. W.; Chen, Y.; Swenton, J .

S. Tetrahedron 1991, 47, 655. (g) Callinan, A.; Chen, Y.; Morrow, G.

W.; Swenton, J . S. Tetrahedron Lett. 1990, 31, 4551. (h) Morrow, G.

W.; Swenton, J . S. Tetrahedron Lett. 1987, 28, 5445.

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

Sch em e 1

J . Org. Chem., Vol. 61, No. 5, 1996 1579

Sch em e 2

Sch em e 3

the cyclization reactions suggested in Figure 1. In these

cases, the aromatic ring would be preserved in the

product and still available for oxidation. In fact, the

cyclization reaction itself would add an alkyl substituent

to the aromatic ring and lower its oxidation potential. If

the initial oxidation step involved the aryl ring, then it

would be impossible to avoid overoxidation since the

product would oxidize more readily than the starting

material. Therefore, for the cyclizations proposed in

Figure 1 to be successful the nature of the initiating and

terminating groups in the cyclization reaction must be

opposite to that of the cyclization reactions outlined in

Scheme 1. The initial oxidation step must involve the

removal of an electron from the olefin coupling partner

(Scheme 2). Furthermore, if overoxidation is to be

avoided, then the oxidation potential of the olefin in the

substrate must be lower than the oxidation potential of

the aromatic ring in the cyclized product. This difference

in oxidation potential must be large enough to allow for

selectivity. We report herein our recent efforts to probe

the viability of these cyclization reactions.

the enol ether once generated. Substrate 4 was con-

structed in a straightforward fashion from 3-methoxy-

benzyl alcohol. A Peterson olefination reaction⁷was used

to put in the methoxy enol ether moiety instead of the

corresponding Wittig reaction because of the difficulty

encountered in trying to separate the triphenylphosphine

oxide byproduct from the desired enol ether.

The first attempt at cyclizing substrate 4 utilized

conditions that had been successful in a number of earlier

anodic olefin coupling reactions (Scheme 4). These

conditions utilized an undivided cell, a reticulated vitre-

ous carbon (RVC) anode, a carbon rod cathode, a 0.4 M

LiClO₄in 20% MeOH/CH₂Cl₂electrolyte solution, 2,6-

lutidine as a proton scavanger, and a constant current

of 11.1 mA. The reaction was continued until 2.0

faradays/mol of charge had been passed. This reaction

led to the formation of four cyclized products in a

combined yield of 57% and a ratio of 7.0:3.4:2.8:1. The

(5) All of the substrates reported here gave rise to irreversible waves

when examined by cyclic voltammetry. For this reason, all of the E_p/2

values reported were obtained using identical conditions. The values

were obtained using a BAS 100B electrochemical analyzer, a platinum

anode, a Ag/AgCl reference electrode (purchased from BAS), a sweep

rate of 25 mV/s, and a 0.1 N LiClO₄electrolyte solution. The quality

of the reference electrode was checked, and the potentials were

calibrated with the use of a simple methoxy enol ether.^2e

Initially, substrate 4 was selected for study (Scheme

3). This substrate was chosen because enol ethers have

a substantially lower oxidation potential (E_p/2) +1.40 V

vs Ag/AgCl)^2e,5than 3-methoxy-substituted phenyl rings

(E_p/2measured for methyl 3-methoxyphenylacetate )

+1.63 V vs Ag/AgCl)⁶and because the phenyl ring was

ideally substituted for “attacking” the radical cation of

(6) Moeller, K. D.; Wang, P. W.; Tarazi, S.; Marzabadi, M. R.; Wong,

P. L. J . Org. Chem. 1991, 56, 1058.

(7) For a review, see: Magnus, P. Aldrichimica Acta 1980, 13, 43.

1580 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

Sch em e 4

Sch em e 5

Sch em e 6

1

ratio of products was obtained by integration of the H

NMR obtained for the crude reaction product. This

analysis utilized the acetal protons of 5 and 6 and the

benzylic methine protons of 7 and 8. All four of these

signals were in the δ 4-5 range, clearly visible, and

cleanly resolved from each other. In addition to the

cyclized products, a 22% yield of recovered starting

material was obtained. The structures of products 7 and

8 were tentatively assigned on the basis of methine peaks

at δ 4.26 for product 7 and δ 4.56 for product 8 and

the product and that a chance for selectivity existed.

Along these lines, three potential solutions were appar-

ent. First, the inherent selectivity of the reaction could

be enhanced with the use of controlled potential elec-

trolysis conditions.⁹Second, the gap in oxidation poten-

tial between the aromatic ring and the electron-rich olefin

could be widened by finding an electron-rich olefin that

had a much lower oxidation potential than an enol ether.

Third, the gap in oxidation potential between the aro-

matic ring and the enol ether initiator could be widened

by raising the oxidation potential of the aromatic ring.

This alteration could be accomplished either by removing

the 3-methoxy substituent on the phenyl ring or by

replacing it with a less electron donating substituent.

1

methoxy peaks at δ 3.39 for 7 and δ 3.44 for 8 in the H

NMR spectrum for a mixture of 6, 7, and 8. Compounds

6, 7, and 8 were inseparable by chromatography. In

order to gain further evidence for the structure of

compounds 7 and 8, the crude product was treated with

pyridinium p-toluenesulfonate in methanol in order to

eliminate the benzylic methoxy substituents in 7 and 8

and form the corresponding cyclic styrene derivatives 9

and 10 (Scheme 5). At this point, the products could be

readily separated and characterized.

Minor products 7 and 8 were derived from overoxida-

tion of the initially formed products 5 and 6. When the

electrolysis was allowed to run until 4 faradays/mol had

been passed, compounds 7 and 8 were formed as the

major products. Apparently, a one-electron oxidation

of the cyclized product led to the radical cation of the

aromatic ring (Scheme 6). Elimination of the dimethoxy-

methyl substituent followed by a second one-electron

oxidation led to a methylated quinone methide⁸that in

turn trapped methanol to form the final product. The

fact that the overoxidized product was formed in a minor

amount in the 2 faradays/mol experimtent suggested that

the starting material did oxidize at a lower potential than

Experimentally, the simplest solution was to utilize a

controlled potential electrolysis. In order to maximize

selectivity, the potential was set at the lowest value that

would allow for a stable current flow. This value for the

potential was obtained by starting the electrolysis with

a potential of 0.0 V vs a Ag/AgCl reference electrode and

then slowly raising the potential until the flow of current

in the electrolysis was about the same as that used in

the constant current electrolysis experiment described

earlier. The potential derived in this fashion (+1.10 V

vs Ag/AgCl) was maintained throughout the course of the

(8) For leading references, please see: Angle, S. R.; Arnaiz, D. O.;

Boyce, J . P.; Frutos, R. P.; Louie, M. S.; Mattson-Arnaiz, H. L.; Rainier,

J . D.; Turnbull, K. D.; Yang, W. J . Org. Chem. 1994, 59, 6322 and

references therein.

(9) For a general description of electrochemical techniques and an

excellent review of background material, please see: Organic

Electrochemistry: An Introduction and A Guide, 2nd ed.; Baizer, M.

M., Lund. H., M. Dekker, Eds.; New York, 1983.

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

J . Org. Chem., Vol. 61, No. 5, 1996 1581

Sch em e 7

cyclization reaction. Except for the use of the controlled

potential, all of the reaction parameters were kept the

same as in the constant current electrolysis (Scheme 3).

In this experiment, a 57% isolated yield of cyclized

products was again obtained along with a 21% yield of

recovered starting material. However, the ratio of the

products formed was dramatically different than in the

previous constant current electrolysis. This time the

ratio of desired cyclized product to overoxidized product

was 16:1. Clearly, the controlled potential electrolysis

was useful for dramatically increasing the selectivity of

the reaction.

Although the use of a controlled potential electrolysis

helped with selectivity in the case of substrate 4, it was

difficult to forward controlled potential electrolyses as a

general solution for eliminating overoxidation in cycliza-

tion reactions involving electron-rich phenyl rings. In

fact, even the oxidation of 4 had pushed the limits of a

preparative scale controlled potential electrolysis. The

reaction took approximately 24 h to proceed to 80%

completion. Since the current drops off exponentially

with reaction time in a controlled potential electrolysis,

completion of the reaction would have required at least

an additional 12 h. What would happen if a more

electron-rich phenyl ring were to be used? To maintain

the same level of selectivity the potential for the reaction

would have to be lowered still further and the reaction

rate slowed even more. Of course, once the potential of

the aromatic ring became lower than that of the enol

ether, any selectivity would be impossible and overoxi-

dation would dominate. Clearly, an alternative way of

inducing selectivity needed to be found.

Sch em e 8

Sch em e 9

Because we hoped to maintain the versatility of the

reactions with respect to the aromatic ring participant,

initial efforts were aimed at identifying a suitable al-

ternative to the enol ether as the initiator for the re-

actions. To this end, substrates 11a and 11b were syn-

thesized and electrolyzed. The substrates were made in

a fashion identical to that outlined for substrate 4 above,

and their syntheses are detailed in the Experimental

Section. The vinyl sulfide substrate was selected as an

alternative to the enol ether because of its increased

hydrolytic stability and because of its lower oxidation

potential (E_p/2ca. +0.9 to +1.0 V vs Ag/AgCl reference

electrode/Pt working and auxiliary electrodes/0.1 N

LiClO₄in CH₃CN electrolyte solution/25 mV/s sweep

rate).

The anodic oxidation of substrate 11a at a retriculated

vitreous carbon anode in an undivided cell using a carbon

auxiliary electrode, 2,6-lutidine as a proton scavanger, a

0.4 N LiClO₄in 20% MeOH/CH₂Cl₂electrolyte solution,

and a constant current of 11.2 mA (2.0 faradays/mol) led

to the formation of a 64% yield of cyclized product along

with a 28% yield of recovered starting material (Scheme

7). In this example, the ratio of desired product to

overoxidized material was 1:1. Clearly, this would be a

more difficult oxidation to fix using controlled potential

conditions than was the oxidation of substrate 4.

However, the anodic oxidation of substrate 11b using

a vinyl sulfide initiator provided a nice solution to this

problem. In this example, no overoxidized material was

obtained. This reaction worked so well that it was

tempting to go back and try to fix the oxidation of

substrate 4 with the use of a vinyl sulfide. For this

reason, substrate 4b was synthesized and oxidized. The

cyclization of 4b was not as successful as the earlier

cyclization of 4a . Although no overoxidized material was

obtained, the mass balance of the reaction was very low.

Surprisingly, none of the cyclized product arising from

ortho substitution of the phenyl ring was obtained. With

such a poor mass balance for the reaction, it was

impossible to reach a definitive conclusion about this

experiment, other than that the earlier cyclization reac-

tion originating from oxidation of the vinyl sulfide group

had benefited strongly from the use of the more nucleo-

phillic dialkoxy phenyl ring (Scheme 8).

Having established that the cyclization reactions were

possible and that the reactions could be initiated suc-

cessfully in the presence of very electron-rich phenyl

rings, attention was turned toward understanding how

the nature of substituents on the phenyl ring would

influence the coupling reaction. For starters, substrate

14 was studied in order to determine if activation of the

phenyl ring was required for cyclization (Scheme 9). The

synthesis of the substrate 14 is detailed in the Experi-

mental Section. The extra methyl group was added to

the enol ether in order to lower the oxidation potential

of the substrate and stabilize the radical cation in an

effort to aid the cyclization reaction. However, the anodic

oxidation of 14 did not lead to the formation of any

cyclized product. Instead, the oxidation gave rise to an

unsaturated acetal. This acetal was most likely formed

from the elimination of a proton from the carbon R to

the radical cation of the enol ether before the cyclization

could occur.¹⁰The formation of this product was con-

firmed by hydrolysis of the acetal to form the R,â-

unsaturated ketone 15. Clearly, the 3-methoxy substit-

1582 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

Sch em e 10

Sch em e 11

Sch em e 12

uents utilized in the previous cyclizations were important

for cyclization.

Having established that the phenyl ring needed to be

activated for cyclization, we turned our attention toward

exploring if the reactions would be compatible with even

more electron-rich phenyl rings. For this reason, sub-

strate 16 was synthesized and oxidized. Once again, the

synthesis directly paralleled the synthesis of 4 and is

detailed in the Experimental Section. The anodic oxida-

tion of 16 (Scheme 10) was accomplished in an undivided

cell using the conditions outlined previously and led to

the formation of a 25% yield of the desired cyclized

product 17, a 27% yield of spirocyclic product 18, and a

20% yield of a product that has tentatively been assigned

as the uncyclized product 19 on the basis of its 300 MHz

¹H NMR spectrum. The presence of such a significant

amount of uncyclized material indicated that the tri-

methoxy-substituted phenyl ring was not as effective of

a coupling partner as either the 3-methoxy-substituted

phenyl ring or the 3,5-dimethoxy-substituted phenyl ring

studied earlier.¹¹

The formation of the spirocyclic product indicated a

“crossover” into the chemistry of Swenton and Yamamura

(vide infra) and appeared to be a direct result of the

electron-donating group on the 4-position of the phenyl

ring. In order to confirm this conclusion, the coupling

reaction utilizing a simple 4-methoxy-substituted phenyl

ring was examined (Scheme 11). As expected, this

cyclization led to the formation of a spirocyclic product.

No evidence for a fused bicyclic product was obtained.

From these results, it became clear that if 4-alkoxy-

substituted phenyl rings were to be used in the synthesis

of fused ring skeletons, then the alkoxy group would have

to be protected in a fashion that removed its electron-

donating ability. In the past, it has been shown that a

pivaloyloxy group can serve as an electronically neutral

substituent in anodic oxidation reactions.⁶In order to

test the effect of a 4-pivaloyloxy group on the current

cyclizations, substrates 22a and 22b were synthesized

in a fashion directly analogous to the earlier syntheses

of 4a and 4b. The oxidation of 22a led to a 42% yield of

cyclized products along with an 11% yield of recovered

starting material (Scheme 12). The cyclized material was

isolated as an inseparable 1.3 to 1 mixture of compounds

23a and 24. No spirocyclic product was observed.

Although the mass balance for the reaction was far lower

than desired, it was clear that the use of the 4-pivaloyloxy

substituent had a dramatic impact on the course of the

reaction. No longer was a spirocyclic product competitive

with the formation of the fused products.

Because the oxidation of 22a led to a significant

amount of overoxidized product, the vinyl sulfide sub-

strate 22b was studied. It was hoped that the use of the

vinyl sulfide would both eliminate the formation of the

overoxidation product 24 and increase the overall mass

balance of the reaction. Unfortunately attempts to

oxidize 22b have met with failure. While it is clear that

the vinyl sulfide group is cleanly oxidized (high yields of

the methanol trapping of the resulting radical cation can

be obtained), in all but one case the substrate did not

cyclize. In one instance a small amount (ca. 20%) of the

desired product was formed. In this example, none of

the overoxidized product was apparent by analysis of the

1

crude mixture by H NMR. However, in spite of exten-

sive efforts we have been unable to repeat this cyclization.

It is very clear that the 4-pivaloyloxy substituent badly

damages the substrates ability to afford cyclized products,

especially when compared to the 3,5-dimethoxy-substi-

tuted phenyl ring of substrate 11a and 11b. This result

when combined with the earlier described studies using

4-methoxy-substituted phenyl rings has led us to the

general conclusion that for the synthesis of fused bicyclic

ring systems 4-alkoxy-substituted phenyl rings need to

be avoided.

Having completed our study concerning the utility of

electron-rich phenyl rings for the coupling reactions, we

turned our attention to exploring the compatibility of

heteroatomic aromatic rings with the anodic oxidation

reactions. As a starting point for this work, furan rings

were selected for study. It was hoped that the use of

furan rings in the coupling reactions would lead to

bicyclic ring skeletons that could be either directly

(10) For an analogous chemical oxidation of enol ethers to unsatur-

ated carbonyls, see: Evans, P. A.; Longmire, J . M.; Modi, D. P.

Tetrahedron Lett. 1995, 36, 3985.

(11) For examples of the effect of ortho sustituents on the ability of

methoxy groups to donate electron density to the π-system of aromatic

rings, see: Rathore, R.; Kochi, J . K. J . Org. Chem. 1995, 60, 4399.

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

Ta ble 1

J . Org. Chem., Vol. 61, No. 5, 1996 1583

1

2

27a : R₁) SMe, R₂) H, n ) 1

28a : 54% (R₁) SMe, X ) OMe)

28b: 17% (R₁) OMe, X ) OMe)

28a : 67% (R₁) SMe, X ) OMe)

28b: 10% (R₁) OMe, X ) OMe)

28b: 75% (X ) OMe)

MeOH solvent

3

4

5

6

7

27b: R₁) OMe, R₂) H, n ) 1

27c: R₁) Ph, R₂) H, n ) 1

27d : R₁) CH₂TMS, R₂) H, n ) 1

27e: R₁) CH₂TMS, R₂) CH₃, n ) 1

27f: R₁) CH₃, R₂) CH₃, n ) 1

28c: 71% (X ) OMe)

28d : 49% (no X, R₁) CH₂)^a

28e: 51% (no X, R₁) CH₂)

28f: 24% (X ) OMe)

28e: 51% (no X, R₁) CH₂)

28g: 62% (X ) OMe)

8

9

10

11

12

27g: R₁) OMe, R₂) H, n ) 2

27h : R₁) Ph, R₂) H, n ) 2

27i: R₁) CH₃, R₂) CH₃, n ) 2

MeOH solvent

28g: 58% (X ) OMe)

28h : 41% (X ) OMe)^b

28h : 48% (X ) OMe)^b

28i:^c26% (no X, R₁) CH₂)^b

28ii: 6% (X ) OMe)^b

13

27j: R₁) OMe, R₂) H, n ) 3

28j:^d32% (X ) OMe)^b,e

The product formed cleanly but was volatile. Unoptimized yield. ^cThis reaction also led to an uncyclized product resulting from

a

b

d

oxidation of the furan ring (see text). This reaction also led to an uncyclized product resulting from oxidation of the enol ether (see text).

^eThis reaction used n-Bu₄NBF₄as the electrolyte. The yield using LiClO₄was 24%.

Sch em e 13

transformed into a variety of bicyclic furan-containing

natural products¹²or used in the synthesis of polycyclic

natural products by taking advantage of the utility of

furans as synthetic intermediates.¹³

Like the earlier electron-rich phenyl rings, furan rings

had previously served as substrates for anodic oxidation

reactions.¹⁴However, with one notable exception (the

trapping of the resulting radical cation with cyanide)

these reactions had not been used to synthesize either

carbon-carbon bonds or bicyclic ring systems. In order

to study the use of furans in anodic cyclization reactions,

a series of substituted furan substrates were synthesized.

A general route for the synthesis of the substrates is

outlined in Scheme 13 for the construction of 27a (R₁)

SMe) and 27b (R₁) OMe). In this synthesis, olefin 25

proved difficult to isolate. The overall yields of the

process benefited strongly from carrying the material all

the way through to alcohol 26 without purification. The

syntheses of the remaining furan-based substrates (Table

1, 27c-j) were essentially identical except for a change

in the anion used to generate 25 (vary the number of

methylene carbons in order to vary n), a change in the

final Wittig step (vary R₁and R₂), or both. The details

of these syntheses are described in the Experimental

Section.

The first furan substrate studied was 27a (Table 1,

entry 1). The vinyl sulfide was chosen as the initiator

to ensure that the furan ring in the desired product would

not compete with oxidation of the substrate. In this

manner, it was hoped that overoxidation of the product

would be avoided. The conditions used for the electrolysis

were identical to the constant current experiments

described above. In this experiment, the initial product

formed was not the desired bicyclic furan, but rather

appeared to be a mixture of products from initial cycliza-

tion of the radical cation to form a bicyclic radical cation

followed by trapping of this resulting radical cation with

methanol, a second oxidation, and then trapping of the

incipient cation with a second equivalent of methanol.

This initial product mixture could be readily converted

to the desired product 28a by treatment with p-toluene-

sulfonic acid in methanol. More conveniently, it was

found that the desired product 28a could be generated

directly by adding 5 equiv of p-toluenesulfonic acid to the

electrolysis cell following the oxidation and then allowing

the resulting mixture to stir at room temperature over-

night. After workup and purification, these conditions

(12) (a) Danheiser, R. L.; Stoner, E. J .; Koyama, H.; Yamashita, D.

S.; Klade, C. J . Am. Chem. Soc. 1989, 111, 4407. (b) Carte´, B.; Kernan,

M. R.; Barrabee, E. B.; Faulkner, D. J . J . Org. Chem. 1986, 51, 3528.

(c) Gopalan, A.; Magnus, P. J . Org. Chem. 1984, 49, 2317. (d) Tanis,

S. P.; Dixon, L. A. Tetrahedron Lett. 1987, 28, 2495. (e) Hiroi, K.; Sato,

H. Synthesis 1987, 811. (f) Padwa, A.; Ishida, M. Tetrahedron Lett.

1991, 32, 5673.

(13) For reviews, see: (a) Dean, F. M. Adv. Heterocycl. Chem. 1982,

30, 161. (b) Lipshutz, B. H. Chem. Rev. 1986, 86, 795.

(14) (a) Clauson-Kaas, N.; Limborg, F.; Dietrich, P. Acta Chem.

Scand. 1952, 6, 545. (b) Baggaley, A. J .; Brettle, R. J . Chem. Soc. C

1968, 969. (c) Yoshida, K.; Fueno, T. J . Org. Chem. 1971, 36, 1523. (d)

Panomarev, A. A.; Markushina, I. A. J . Gen. Chem. USSR 1963, 33,

3892. (e) Shono, T.; Matsumura, Y. Tetrahedron Lett. 1976, 17, 1363.

(f) Shono, T.; Matsumura, Y.; Hamaguchi, H.; Nakamura, K. Chem.

Lett. 1976, 1249. (g) Shono, T.; Hamaguchi, H.; Aoki, K. Chem. Lett.

1977, 1053. (h) Shono, T.; Matsumura, Y.; Tsubata, K.; Takata, J .

Chem. Lett. 1981, 1121.

1584 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

Sch em e 14

led to a 71% isolated yield of the bicyclic furan as a

mixture of acetals.

It is important to note that this cyclization also

proceeded nicely in pure methanol solvent (Table 1, entry

2). This observation has been characteristic of only the

best anodic olefin coupling reactions to date. For ex-

ample, intramolecular anodic olefin coupling reactions

between enol ethers were shown to lead to both five- and

six-membered ring formation in pure methanol,^2ewhile

analogous coupling reactions between either enol ethers

and allylsilanes or enol ethers and styrenes required the

use of a cosolvent.^2fIn pure methanol, these later cases

led to methanol trapping of the radical cation before

cyclization. In six-membered ring cases, no cyclized

material was obtained. However, in the current oxida-

tion no uncyclized material was observed.

Since the initial product from the oxidation and cy-

clization was not a furan, no possibility for overoxidation

of the product existed. This observation led us to

examine the anodic coupling reactions of furans with a

variety of electron-rich olefins. With respect to the

synthesis of six-membered rings, the furan ring could be

coupled with an enol ether, a styrene, a pair of allylsi-

lanes, and a simple trialkyl-substituted olefin (entries

3-7). All of the reactions proceeded smoothly and with

the exception of the allylsilane afforded good yields of

bicyclic furan products. In the case of the allylsilane

substrate (entry 5), the low yield appeared to be due to

the volatility of the product. Both TLC and NMR data

on the crude reaction mixture indicated that the product

had been formed cleanly.

trisubstituted olefin was untouched. No product having

the initial monosubstituted furan intact was observed in

the 300 MHz ¹H NMR of the crude reaction mixture. The

yield of 29 could be raised to 54% with the use of a carbon

rod anode in place of the retriculated vitreous carbon

anode used above. The formation of 29, the presence of

the trisubstituted olefin, and the complete absence of

products with the initial furan ring intact suggested that

in this example the furan ring was being oxidized and

serving as the initiator for the subsequent chemistry.

In the coupling reactions between furans and enol

ethers, the furan ring appeared to serve as the terminat-

ing group for the cyclization. For example, the anodic

oxidation of substrate 27j (entry 13) led to the formation

of both the eight-membered ring product 28j in a 32%

yield and the uncyclized product 30 in a 16% yield. This

The reactions also proved to be compatible with the

synthesis of seven-membered rings (entries 8-12). For

example, the cyclization of substrate 27g involving the

coupling of a furan ring and an enol ether led to a 62%

yield of the seven-membered ring product. No evidence

for an uncyclized product was obtained. This observation

also held for the cyclization of 27g in pure methanol

solvent (entry 9). In this case, a 58% yield of the cyclized

product was obtained. The formation of a seven-

membered ring in pure methanol solvent was surprising,

especially since the corresponding intramolecular cou-

pling of bis enol ether substrates (our most efficient

cyclizations to date) required the use of a cosolvent for

the formation of a seven-membered ring.^2eEven more

surprising was the seven-membered ring cyclization

resulting from the anodic coupling of a furan ring and a

styrene (Table 1, entries 10 and 11). Once again, this

cyclization did not require a cosolvent for formation of a

seven-membered ring. For comparison, the correspond-

ing anodic coupling reaction between an enol ether and

a styrene required the use of a cosolvent even when

forming a six-membered ring. Clearly, the furan ring

represents one of the most reactive coupling partners that

we have studied to date.

reaction was the first intramolecular anodic olefin cou-

pling reaction to give rise to an eight-membered ring.

Once again, the furan ring proved to be the most reactive

coupling partner that we have studied. The uncyclized

material in this example apparently arose from methanol

trapping of the radical cation derived from the enol ether.

The ¹H NMR of the crude reaction mixture gave no

indication of a product having the initial enol ether intact.

Apparently, the enol ether was oxidized in preference to

the furan ring. The formation of uncyclized products 29

and 30 suggested that the furan ring can serve as either

the terminator or the initiator for the reaction depending

on the nature of the coupling partner.

The coupling reaction of furans and enol ethers also

proved to be compatible with the construction of a

quaternary center (Scheme 14). The synthesis of sub-

strate 31 is detailed in the Experimental Section. In this

experiment, the trisubstituted enol ether 31 was oxidized

and the crude product treated with p-toluenesulfonic acid

in order to form product 32 in a 54% isolated yield. Once

again, this result was consistent with the furan ring

being an excellent coupling partner. For comparison, the

coupling of bis enol ether substrates was shown to be

compatible with the simultaneous formation of both a six-

membered ring and a quaternary center. However, the

coupling reaction of an enol ether and an allylsilane could

not overcome the difficulties associated with synthesizing

a six-membered ring and a quaternary center at the same

time.^2dThe allylsilane moiety was simply not reactive

enough as a trapping group for the enol ether radical

cation.

The anodic oxidation of substrate 27i (entry 12) also

proved to be interesting. In this case, the cyclization

afforded only a 32% yield of cyclized product. In addition,

a 25% yield of the uncyclized product 29 was obtained.

Finally, we were curious as to whether the chemistry

developed above for cyclization reactions involving furans

could be extended to additional heteroatomic aryl rings.

Apparently, this product was formed from methanol

trapping of a radical cation derived from the furan. The

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

J . Org. Chem., Vol. 61, No. 5, 1996 1585

Sch em e 15

phenyl ring. Furan rings were found to be excellent

coupling partners for the reactions. Cyclizations involv-

ing furans were shown to be compatible with the forma-

tion of both six- and seven-membered rings, the genera-

tion of a quaternary carbon, and the use of a variety of

electron-rich olefins ranging from enol ethers and vinyl

sulfides to simple alkyl-substituted olefins. It appears

that the furans will serve as either the initiating or the

terminating group for the oxidative coupling reaction

depending on the nature of the electron-rich olefin.

Finally, the intramolecular anodic olefin coupling reac-

tion was shown to be compatible with the use of a pyrrole

ring as one of the participants. Work aimed at elucidat-

ing the overall utility of these reactions for organic

synthesis is currently underway.

Exp er im en ta l Section ¹⁶

4-(3-Meth oxyp h en yl)-1-bu ten e (2). To a round-bottom

flask cooled in an ice bath were added 11.12 g (80.5 mmol) of

3-methoxybenzyl alcohol, 40 mL of THF, and 7.20 g (26.6

mmol) of phosphorus tribromide. After 30 min the reaction

mixture was checked by TLC. The TLC still showed some of

the starting alcohol so 0.54 g (2.0 mmol) more of the phospho-

rus tribromide was added. The reaction mixture was checked

after another 30 min. This TLC showed that the reaction

mixture still contained starting material, so an additional 2.31

g (8.5 mmol) of phosphorus tribromide was added. The

reaction mixture was checked again after 30 min showing that

all of the starting material had been consumed. To the

reaction mixture was added 30.0 mL of THF followed by 161.0

mL (161.0 mmol) of a 1.0 M solution of allylmagnesium

bromide in ether. After 1 h, the reaction mixture still showed

the presence of some of the intermediate by TLC so an

additional 20.0 mL (20.0 mmol) of the allylmagnesium bromide

solution was added. The reaction mixture was stirred for an

additional 2 h. The reaction was quenched with 10 N H₂SO₄,

and the mixture was diluted with H₂O and then extracted

three times with ether. The ether layers were washed once

with brine, dried over MgSO₄, filtered, and concentrated in

vacuo. The crude product was chromatographed through 800

mL of silica gel using a gradient elution from straight hexane

to 10% ether/hexane (1% steps every 750 mL of eluent) to

afford 10.66 g (82%) of the desired product: ¹H NMR (300

MHz, CDCl₃) δ 7.17-7.23 (m, 1 H), 6.71-6.80 (m, 3 H), 5.86

(ddt, J ) 17.0 Hz, J ) 10.3 Hz, J ) 6.6 Hz, 1 H), 5.05 (dq, J

) 15.0 Hz, J ) 1.7 Hz, 1 H), 4.98 (d of multiplets, J ) 9.0 Hz,

1 H), 3.79 (s, 3 H), 2.69 (t, J ) 7.9 Hz, 2 H), 2.37 (dt, J ) 8.3

Hz, J ) 7.0 Hz, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 143.5,-

138.0, 129.2, 120.8, 114.9, 114.2, 111.0, 55.1, 35.4; IR (neat,

NaCl) 3076, 3001, 2977, 2938, 2856, 2835, 1611, 1601, 1594,

1585, 1492, 1465, 1455, 1436, 1416, 1335, 1312, 1290, 1259,

1190, 1165, 1152,1052, 996, 913, 862, 778, 741 cm^-1; GCMS

(PCI) m/ e (rel intensity) 121 (100), 163 (62), 149 (21), 122 (19),

161 (14), 164 (13), 162 (13), 191 (4), 203 (2); HRMS (EI) m/ e

calcd for C₁₁H₁₄O 162.1041, found 162.1045.

Sch em e 16

For this reason, the pyrrole-based substrate 36 was

synthesized (Scheme 15). The starting material for the

synthesis was made using the published procedure.¹⁵

A

vinyl sulfide was used as the initiator and the pyrrole

protected as the pivaloyl amide in order to ensure initial

oxidation of the electron-rich olefin and to avoid potential

overoxidation problems.

The oxidation of 36 was again conducted in a fashion

that was directly analogous to the earlier constant

current electrolyses (Scheme 16). A 66% isolated yield

of cyclized product was obtained as a mixture of two

diastereomers (37). Unlike the furan cases, the majority

of the bicyclic product obtained directly from the elec-

trolysis without acid treatment was the bicyclic aromatic

compound. Only a tiny amount of methanol-trapped

product could be observed. This product could be elimi-

nated to the pyrrole product 37 in a fashion directly

analogous to the furan cases.

In conclusion, we have found that intramolecular

anodic olefin coupling reactions utilizing electron-rich

olefins and aryl rings can provide an effective means for

synthesizing fused, bicyclic ring skeletons. Reactions

involving phenyl rings benefited strongly from a 3-meth-

oxy substituent on the phenyl ring. Although overoxi-

dation of the cyclized products in these cases was

observed, this problem could be addressed with the use

of controlled potential electrolysis conditions when a

monomethoxy phenyl ring was used and with the use of

a vinyl sulfide moiety as the initiator when a more

electron-rich phenyl ring was used. When the phenyl

rings contained a 4-methoxy substituent, spirocyclic

product formation began to compete with the desired

fused bicyclic products. Although it appeared that this

side reaction could be controlled with the use of a less

electron donating 4-pivaloyoxy substituent, yields for this

process were not synthetically viable and it is currently

best to avoid the use of a 4-alkoxy substituent on the

4-(3-Meth oxyp h en yl)-1-bu ta n ol (3). To a round-bottom

flask cooled in an ice bath were added 50.0 mL of THF and

6.25 mL (62.5 mmol) of borane-dimethyl sulfide complex. To

this solution was added 8.529 g (121.7 mmol) of 2-methyl-2-

butene. The reaction mixture was stirred for 1 h, and then

5.064 g (31.2 mmol) of 6 was added. The reaction mixture was

stirred for 1 h, and then the reaction was quenched SLOWLY

with a solution of 62.4 mL of 3 N NaOH and 28.8 mL of 30%

H₂O₂that had been cooled in an ice bath. The resulting

mixture was stirred for 1 h, diluted with H₂O, and then

extracted with ether until TLC analysis showed that no more

product was being extracted. The organic layer was dried over

MgSO₄, filtered, and concentrated in vacuo. The resulting oil

was chromatographed through 400 mL of silica gel with a

gradient elution of hexane to 50% ether in hexane (5% steps

(15) Bray B. L.; Mathies, P. H.; Naef, R.; Solas, D. R.; Tidwell, T.

T.; Artis, D. R.; Muchowski, J . M. J . Org. Chem. 1990, 55, 6317.

(16) For a description of general experimental details, please see:

Wong, P. L.; Moeller, K. D. J . Am. Chem. Soc. 1993, 115, 11434.

1586 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

every 500 mL of eluent) to afford 5.307 g (94%) of the desired

product: ¹H NMR (300 MHz, CDCl₃) δ 7.14-7.19 (m, 1 H),

6.69-6.77 (m, 3 H), 3.75 (s, 3 H), 3.58 (t, J ) 6.4 Hz, 2 H),

2.59 (t, J ) 7.5 Hz, 2 H), 2.46 (br s, 1 H), 1.51-1.71 (m, 4 H);

¹³C NMR (75 MHz, CDCl₃) δ 159.6, 144.1, 129.3, 120.9, 114.2,

111.0, 62.5, 55.1, 35.7, 32.3, 27.5; IR (neat, NaCl) 3300 br,

3000, 2936, 2885, 2861, 2837, 1610, 1601, 1594, 1585, 1488,

1465, 1454, 1436, 1380, 1369, 1333, 1314, 1261, 1190, 1164,

1151, 1043, 996, 984, 874, 778, 738, 696 cm^-1; GCMS (PCI)

m/ e (rel intensity) 163 (100), 121 (37), 164 (21), 181 (8), 122

(8); HRMS (EI) m/ e calcd for C₁₁H₁₆O₂180.1150, found

180.1157.

(E,Z)-1-Meth oxy-5-(3-m eth oxyph en yl)-1-pen ten e (4). To

a round-bottom flask were added 3.501 g (19.4 mmol) of 3,

1.820 g (23.3 mmol) of dimethyl sulfoxide, and 20 mL of

dichloromethane. The solution was cooled to -78 °C. After

15 min, 2.716 g (21.4 mmol) of oxalyl chloride was added and

the reaction mixture was allowed to warm to between -50 and

-60 °C. After 15 min, 9.837 (97.2 mmol) of triethylamine was

added and the reaction mixture allowed to warm to room

temperature. Water was added and most of the dichlo-

romethane evaporated in vacuo. The remaining solution was

extracted three times with ether. The ether layers were then

washed once with brine and dried over MgSO₄. The resulting

aldehyde was taken up in 40 mL of THF and then used in the

Peterson olefination below.

A round-bottom flask was charged with 6.885 g (58.2 mmol)

of (methoxymethyl)trimethylsilane and 40 mL of THF. The

mixture was cooled to -78 °C and then 44.8 mL (58.2 mmol)

of sec-butyllithium was added. The reaction mixture was

allowed to warm to between -25 and -35 °C for 30 min. The

reaction mixture was then recooled to -78 °C and the aldehyde

generated above added to the flask. This mixture was allowed

to warm slowly to room temperature and stirred for 16 h. At

that time, the reaction was quenched with H₂O, and the

mixture was extracted with ether. The ether layers were

washed with brine, dried over MgSO₄, and filtered, and the

solvent was removed in vacuo. To the Peterson adduct was

then added 20 mL of THF. In a separate flask 8 g of potassium

hydride (35 wt % in mineral oil) was washed 2× with hexane.

THF (5 mL) was added to the potassium hydride, and then

the Peterson adduct was added to the flask. The reaction was

stirred for 1 h and then checked by TLC to see if all of the

Peterson adduct was gone. Upon completion, the reaction

mixture was cooled to 0 °C and slowly quenched with metha-

nol. After quenching, the reaction mixture was diluted with

H₂O and extracted three times with ether. The ether layer

was dried over MgSO₄and filtered and then the solvent

removed in vacuo. The crude product was then chromato-

graphed through 300 mL of silica gel using a gradient elution

of hexane to 15% ether in hexane (1.5% steps every 500 mL of

eluent) to afford a 2.133 g (53%) yield of the desired product:

¹H NMR (300 MHz, CDCl₃) δ 7.14-7.20 (m, 1 H), 6.69-6.77

(m, 3 H), 6.28 (d, J ) 12.6 Hz, 0.47 H), 5.88 (dt, J ) 6.2 Hz, J

) 1.5 Hz, 0.53 H), 4.72 (dt, J ) 12.6 Hz, J ) 7.3 Hz, 0.47 H),

4.34 (dt, J ) 6.2 Hz, J ) 7.3 Hz, 0.53 H), 3.76 (s, 3 H), 3.55 (s,

1.59 H), 3.48 (s, 1.41 H), 2.55-2.61 (m, 2 H), 2.11 (dt, J ) 7.1

Hz, J ) 7.7 Hz, 1.14 H), 1.95 (dt, J ) 7.14 Hz, J ) 7.4 Hz,

0.86 H), 1.60-1.70 (m, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 159.6,

147.4, 146.4, 144.4, 144.2, 129.2, 129.1, 120.9, 114.2, 110.9,

106.4, 102.5, 59.4, 55.8, 55.0, 35.6, 35.3, 32.4, 31.6, 27.3, 23.6;

IR (neat, NaCl) 3032, 3000, 2936, 2932, 2856, 2835, 1601, 1594,

1584, 1487, 1465, 1456, 1437, 1390, 1314, 1261, 1209, 1190,

1180, 1164, 1152, 1132, 1109, 1044, 935, 873, 778, 738, 696

cm^-1; GCMS (PCI) m/ e (rel intensity) 175 (100), 121 (35), 147

(23), 71 (19), 122 (13), 203 (10), 174 (8), 149 (7), 171 6); HRMS

(EI) m/ e calcd for C₁₃H₁₈O₂206.1307, found 206.1313. Anal.

Calcd for C₁₃H₁₈O₂: C, 75.69; H, 8.80. Found: C, 75.70; H,

8.93.

until 449.0 C had been passed. The current was then reduced

to 10.3 mA current until a total of 467.1 C had passed (2.0

faradays/mol). Upon completion, the reaction mixture was

diluted with H₂O and extracted with ether. The ether layer

was washed once with brine, dried over MgSO₄, and filtered,

and the solvent was removed in vacuo. The crude product was

then chromatographed using 100 mL of silica gel that had been

slurry packed with a solution of 1% triethylamine in hexane.

A gradient elution from hexane to 20% ether in hexane (2%

steps every 100 mL of eluent) afforded 0.111 g (22%) of the

recovered starting material and 0.328 g (57%) of a mixture of

the desired products. From the 300 MHz ¹H NMR a 2.8:1 ratio

of product (5 and 6) to overoxidized material (7 and 8) and a

2.2:1 ratio of major products (where cyclization occurred at

the position para to the methoxy group) 5 and 7 to minor

products (where cyclization occurred at the position ortho to

the methoxy group) 6 and 8 was obtained.

Con tr olled P oten tia l Electr olysis. To a 100 mL round-

bottom flask were added 0.321 g (1.6 mmol) of 4, 0.836 g (7.8

mmol) of 2,6-lutidine, 2.656 g (25.0 mmol) of lithium perchlo-

rate, 12.4 mL of methanol, and 49.8 mL of dichloromethane.

This mixture was degassed by sonication for 45 min and then

oxidized at a voltage of +1.10 V (vs Ag/AgCl) until a total of

250.5 C had passed (1.66 faradays/mol). Upon completion, the

reaction mixture was diluted with H₂O and extracted with

ether. The ether layer was washed once with brine, dried over

MgSO₄, and filtered, and the solvent was removed in vacuo.

The crude product was then chromatographed on a column of

45 mL of silica gel that had been slurry packed with a solution

of 1% triethylamine in hexane. A gradient elution from hexane

to 20% ether in hexane (2% steps every 75 mL of eluent) led

to the isolation of 0.0685 g (21%) of the recovered starting

material and 0.213 g (57%) of a mixture of the desired

1

products. From the 300 MHz H NMR a 16:1 ratio of product

to overoxidized product and 1.6:1 ratio of major products

(where cyclization occurred at the position para to the methoxy

group) 5 and 7 to minor products (where cyclization occurred

at the position ortho to the methoxy group) 6 and 8 was

obtained.

6-Met h oxy-1,2,3,4-t et r a h yd r on a p h t h a len eca r b oxa l-

1

d eh yd e d im eth yl a ceta l (5): H NMR (300 MHz, CDCl₃) δ

7.26 (d, J ) 8.6 Hz, 1 H), 6.69 (dd, J ) 8.6 Hz, J ) 2.8 Hz, 1

H), 6.61 (d, J ) 2.8 Hz, 1 H), 4.36 (d, J ) 6.5 Hz, 1 H), 3.76 (s,

3 H), 3.34 (s, 6 H), 2.98 (dt, J ) 5.5 Hz, J ) 5.7 Hz, 1 H),

2.71-2.79 (m, 2 H), 1.65-2.03 (m, 4 H); ¹³C NMR (75 MHz,

CDCl₃) δ 157.7, 139.2, 131.4, 128.2, 113.4, 111.6, 108.2, 55.8,

55.7, 55.1, 55.0, 53.7, 53.6, 39.9, 30.0, 24.1, 20.1; IR (neat,

NaCl) 2933, 2914, 2833, 1608, 1584, 1576, 1501, 1465, 1254,

1233, 1190, 1154, 1115, 1075, 971, 938, 913, 871, 835, 813,

732 cm^-1; GCMS (PCI) m/ e (rel intensity) 205 (100), 173 (68),

237 (54), 204 (45), 206 (42), 233 (19), 203 (19) 171 (18), 201

(15); HRMS (EI) m/ e calcd for C₁₄H₂₀O₃236.1412, found

236.1418. Anal. Calcd for C₁₄H₂₀O₃: C, 71.16; H, 8.53.

Found: C, 71.31; H, 8.52.

8-Met h oxy-1,2,3,4-t et r a h yd r on a p h t h a len eca r b oxa l-

1

d eh yd e d im eth yl a ceta l (6): H NMR (300 MHz, CDCl₃) δ

7.09 (t, J ) 7.9 Hz, 1 H), 6.71 (d, J ) 7.8 Hz, 1 H), 6.67 (d, J

) 7.8 Hz, 1 H), 4.64 (d, J ) 3.1 Hz, 1 H), 3.82 (s, 3 H), 3.38-

3.43 (m, 4 H includes s 3.43), 3.14 (s, 3 H), 2.60-2.81 (m, 2

H), 2.03-2.26 (m, 2 H), 1.53-1.68 (m, 2 H);

¹³C NMR (75 MHz,

CDCl₃) δ 157.5, 140.8, 126.5, 124.9, 121.6, 108.4, 107.4, 57.0,

55.6, 55.2, 35.4, 29.7, 21.8, 20.5; IR (neat, NaCl) 2934, 2833,

1599, 1584, 1467, 1439, 1373, 1337, 1322, 1293, 1265, 1251,

1212, 1189, 1148, 1123, 1096, 1068, 1002, 962, 893, 833, 804,

784, 774, 751, 734, 701 cm^-1; GCMS (PCI) m/ e (rel intensity)

173 (100), 205 (95), 204 (47), 75 (29), 174 (24), 203 (24), 206

(22), 175 (12), 201 (11), 172 (10), 237 (1); HRMS (EI) m/ e calcd

for C₁₄

H₂₀O₃, 236.1412, found 236.1414. Anal. Calcd for

C₁₄H₂₀O₃: C, 71.16; H, 8.53. Found: C, 71.48; H, 8.40.

2-Meth oxy-5,6-d ih yd r on a p h th a len e (9) a n d 1-Meth -

oxy-7,8-d ih yd r on a p h th a len e (10). In order to help identify

the overoxidized material, a mixture of the four compounds

was treated with PPTS in methanol to eliminate the overoxi-

dized material. To a round-bottom flask were added 2.403 g

of a mixture of 5, 6, 7, and 8 (the ratio of the mixture was

7.3:3.5:3:1, respectively), 5.125 g (20.4 mmol) of PPTS, and 50

An odic Oxidation of (E,Z)-1-Meth oxy-5-(3-m eth oxyph e-

n yl)-1-p en ten e (4). Con sta n t Cu r r en t Electr olysis. To

a 100 mL round-bottom flask were added 0.500 g (2.4 mmol)

of 4, 1.299 g (12.1 mmol) of 2,6-lutidine, 4.120 g (38.7 mmol)

of lithium perchlorate, 19.4 mL of methanol, and 77.4 mL of

dichloromethane. This mixture was degassed by sonication

for 45 min and then oxidized at a constant current of 11.1 mA

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

J . Org. Chem., Vol. 61, No. 5, 1996 1587

mL of methanol. The reaction mixture was refluxed for 4 h

and then gently heated for 16 h. Upon completion, the reaction

mixture was diluted with H₂O and extracted with ether. The

ether layer was washed once with brine, dried over MgSO₄,

and filtered, and the solvent was removed in vacuo. The crude

product was then chromatographed on a column of 500 mL of

silica gel using a gradient elution from hexane to 20% ether

in hexane (3% steps every 300 mL of eluent) in order to afford

42.8 mg (0.26 mmol) of 10, 271.6 mg (1.70 mmol) of 9, 232.9

mg (1.23 mmol) of the aldehyde derived from the hydrolysis

of 6 (which could be returned to acetal 6 with 0.5 g Montmo-

rillonite Clay K-10 and 0.8 mL of trimethyl orthoformate

followed by filtration through Celite), and 1.185 g (5.02 mmol)

of a mixture of 5 and 6. Compound 9 was contaminated with

a small amount of an unidentified product.

Sp ectr a l d a ta for 2-m eth oxy-5,6-d ih yd r on a p h th a len e

(9): ¹H NMR (300 MHz, CDCl₃) δ 6.92-6.94 (m, 1 H), 6.65-

6.67 (m, 2 H), 6.40 (d, J ) 9.6 Hz, 1H), 5.88 (dt, J ) 9.6 Hz, J

) 4.4 Hz, 1 H), 3.77 (s, 3 H), 2.76 (t, J ) 8.2 Hz, 2 H), 2.23-

2.31 (m, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 158.5, 137.2, 127.3,

127.1, 126.8, 125.9, 113.8, 111.0, 55.2, 28.0, 23.0; GCMS (PCI)

m/ e (rel intensity) 161 (100), 160 (53), 57 (46), 103 (30), 189

(29), 88 (25), 83 (20), 115 (20), 66 (19), 85 (18), 94 (13), 55 (13),

146 (8), 201 (5); HRMS (EI) m/ e calcd for C₁₁H₁₂O 160.0888,

found 160.0880.

) 2.2 Hz, 1 H), 3.75 (s, 6 H), 3.60 (t, J ) 6.3 Hz, 2 H), 2.56 (t,

J ) 7.3 Hz, 2 H), 2.53 (br s, 1 H), 1.52-1.71 (m, 4 H); ¹³C

NMR (75 MHz, CDCl₃) δ 160.7, 144.8, 106.5, 97.6, 62.5, 55.2,

36.0, 32.3, 27.4; IR (neat, NaCl) 3300 br, 2998, 2938, 2863,

2839, 1604, 1596, 1593, 1462, 1428, 1348, 1324, 1314, 1293,

1204, 1157, 1152, 1147, 1058, 939, 921, 833, 696 cm^-1; GCMS

(PCI) m/ e (rel intensity) 211 (100), 193 (67), 212 (22), 151 (17),

194 (15), 210 (11) 152 (8), 239 (7), 209 (7), 179 (3), 191 (3),

251, (3); HRMS (EI) m/ e calcd for C₁₂H₁₈O₃210.1255, found

210.1253.

(E,Z)-5-(3,5-Dim et h oxyp h en yl)-1-m et h oxy-1-p en t en e

(11a ). The reaction sequence was carried out in a similar

fashion to that described above for the synthesis of 4. In this

example, 1.421 g (6.8 mmol) of the alcohol, 1.029 g (8.1 mmol)

of oxalyl chloride, 0.680 g (8.7 mmol) of dimethyl sulfoxide,

3.390 g (33.5 mmol) of triethylamine, and 13 mL of dichlo-

romethane were used. After workup the aldehyde was diluted

with 10 mL of THF and added to the ylide synthesized below.

The ylide was formed from 6.952 g (20.3 mmol) of (methoxym-

ethyl)triphenylphosphonium chloride, 11.9 mL (20.3 mmol) of

tert-butyllithium, and 50 mL of THF. Workup and chroma-

tography afforded 0.815 g (51%) of the desired product:

¹H

NMR (300 MHz, CDCl₃) δ 6.27-6.36 (m, 3.67 H), 5.89 (d, J )

6.2 Hz, 0.33 H), 4.73 (dt, J ) 12.6 Hz, J ) 7.3 Hz, 0.67 H),

4.35 (dt, J ) 6.3 Hz, J ) 7.3 Hz, 0.33 H), 3.76 (s, 6 H), 3.57 (s,

1 H), 3.50 (s, 2 H), 2.55 (t, J ) 7.7 Hz, 2 H), 2.11 (dt, J ) 7.5

Hz, J ) 7.3 Hz, 0.66 H), 1.97 (dt, J ) 7.6 Hz, J ) 7.2 Hz, 1.34

H), 1.60-1.70 (m, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 160.7,

160.6, 147.3, 146.4, 145.2, 144.9, 106.4, 106.3, 102.5, 97.6, 59.4,

55.8, 55.2, 35.8, 35.5, 32.2, 31.4, 27.2, 23.6; IR (neat, NaCl)

3055, 3032, 2999, 2940, 2929, 2855, 2838, 1656, 1605, 1596,

1461, 1454, 1429, 1346, 1324, 1310, 1292, 1265, 1205, 1149,

1134, 1109, 1067, 1061, 933, 836, 747, 689 cm^-1; GCMS (PCI)

m/ e (rel intensity) 205 (100), 237 (44), 152 (33), 71 (28), 57

(25), 223 (22), 177 (21), 206 (16), 204 (14), 61 (13), 55 (13), 251

(13), 265 (5); HRMS (EI) m/ e calcd for C₁₄H₂₀O₃236.1412,

found 236.1433. Anal. Calcd for C₁₄H₂₀O₃: C, 71.16; H, 8.53.

Found: C, 71.18; H, 8.55.

Sp ectr a l d a ta for 1-m eth oxy-7,8-d ih yd r on a p h th a len e

(10): ¹H NMR (300 MHz, CDCl₃) δ 7.01 (t, J ) 7.8 Hz, 1 H),

6.84 (d of multiplets, J ) 9.9 Hz, 1 H), 6.72 (d, J ) 7.6 Hz, 1

H), 6.71 (d, J ) 8.4 Hz, 1 H), 6.03 (dt, J ) 9.8 Hz, J ) 4.4 Hz,

1 H), 3.81 (s, 3H), 2.75 (t, J ) 8.3 Hz, 2 H), 2.23-2.31 (m, 2

H); ¹³C NMR (75 MHz, CDCl₃) δ 154.6, 137.0, 127.6, 127.3,

122.8, 121.5, 120.7, 108.7, 55.5, 27.7, 22.8.

4-(3,5-Dim eth oxyp h en yl)-1-bu ten e. To a round-bottom

flask was added 4.201 g (24.9 mmol) of 3,5-dimethoxybenzyl

alcohol in 15 mL of THF. The mixture was warmed in order

to melt the alcohol, and then 2.281 g (12.5 mmol) of phosphorus

tribromide was added. After 30 min, the reaction mixture was

diluted with H₂O and then extracted with ether. The ether

layers were dried over MgSO₄and filtered and the solvent

removed in vacuo. The crude mixture was then diluted with

24 mL of THF and 35.4 mL (35.4 mmol) of allylmagnesium

bromide (1.0 M solution in ether) added. The reaction mixture

was stirred for 12 h, then the reaction was quenched with 25

mL of 10 N H₂SO₄, and the mixture was diluted with H₂O and

extracted with ether. The resulting ether layer was washed

once with brine, dried over MgSO₄, filtered, and concentrated

in vacuo. The crude reaction mixture was chromatographed

through 250 mL of silica gel using a gradient elution of hexane

to 21% ether in hexane (3% steps every 500 mL of eluent) to

afford 3.551 g (74%) of the desired product: ¹H NMR (300 MHz,

CDCl₃) δ 6.36 (d, J ) 2.2 Hz, 2 H), 6.31 (t, J ) 2.2 Hz, 1 H),

5.86 (ddt, J ) 17.0 Hz, J ) 10.3 Hz, J ) 6.5 Hz, 1 H), 5.05

(dq, J ) 17.1 Hz, J ) 1.7 Hz, 1 H), 4.98 (d of multiplets, J )

10.2 Hz, 1 H), 3.78 (s, 6 H), 2.65 (t, J ) 7.8 Hz, 2 H), 2.36 (dt,

J ) 8.3 Hz, J ) 6.9 Hz, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ

160.7, 144.3 138.0, 114.9 106.5 97.8, 55.2, 35.7, 35.3; IR (neat,

NaCl) 3079, 2999, 2937, 2855, 2838, 1641, 1607, 1596, 1462,

1429, 1350, 1319, 1295, 1206, 1194, 1158, 1150, 1066, 1061,

993, 920, 830, 695 cm^-1; GCMS (PCI) m/ e (rel intensity) 193

(100), 194 (22), 151 (18), 221 (14), 192 (14), 191 (7), 179 (5),

233 (5); HRMS (EI) m/ e calcd for C₁₂H₁₆O₂192.1150, found

192.1153.

4-(3,5-Dim eth oxyp h en yl)-1-bu ta n ol. The hydroboration

was carried out in a similar fashion to that described above

for the synthesis of 3. In this example, 2.502 g (13.0 mmol) of

4-(3,5-dimethoxyphenyl)-1-butene, 3.558 g (50.8 mmol) of

2-methyl-2-butene, 2.6 mL (26.0 mmol) of the borane-dimethyl

sulfide complex, and 100 mL of THF were used. The reaction

mixture was worked up with 15 mL of 3 N NaOH and 7.5 mL

of H₂O₂. The crude reaction mixture was washed once with a

saturated solution of sodium sulfite, dried over MgSO₄, filtered,

and concentrated in vacuo. The product was then chromato-

graphed through 400 mL of silica gel using a gradient elution

from hexane to 30% ether in hexane (5% steps every 300 mL

of eluent) to afford 2.640 g (97%) of the desired product: ¹H

NMR (300 MHz, CDCl₃) δ 6.34 (d, J ) 2.2 Hz, 2 H), 6.29 (t, J

An odic Oxidation of 11a: For m ation of 6,8-Dim eth oxy-

1,2,3,4-tetr a h yd r on a p h th a len eca r boxa ld eh yd e Dim eth yl

Acetal (12a) an d 1,6,8-Tr im eth oxy-1,2,3,4-tetr ah ydr on aph -

t h a len e (13). The oxidation of 11a was done in a fashion

similar to the constant current electrolysis of substrate 4. In

this case 0.243 g (1.0 mmol) of 11a , 0.548 g (5.2 mmol) of 2,6-

lutidine, 1.767 g (16.5 mmol) of lithium perchlorate, 8.2 mL

of methanol, and 33.0 mL of dichloromethane were oxidized

at a constant current of 11.4 mA until 198.8 C (2.00 faradays/

mol) of charge had been passed. Workup and chromatography

afforded 0.0608 g (25%) of the recovered starting material,

0.0856 g (31%) of the desired product (12a ), and 0.0752 g (33%)

of the overoxidized material (13). The spectral data for 12a

are as follows:

¹H NMR (300 MHz, CDCl₃) δ 6.30 (d, J ) 2.2

Hz, 1 H), 6.26 (s, 1 H), 4.60 (d, J ) 3.1 Hz, 1 H), 3.80 (s, 3 H),

3.78 (s, 3 H), 3.43 (s, 3 H), 3.16 (s, 3 H), 3.28-3.31 (m, 1 H),

2.57-2.78 (m, 2 H), 2.00-2.24 (m, 2 H), 1.53-1.66 (m, 2 H);

13C NMR (75 MHz, CDCl₃) δ 158.5, 141.3, 117.4, 108.6, 104.7,

96.2, 96.1, 57.1, 57.0, 55.7, 55.6, 55.4, 55.3, 55.2, 55.1, 35.1,

30.2, 29.8, 21.8, 20.6, 20.5; IR (neat, NaCl) 2937, 2935, 2857,

1652, 1635, 1606, 1593, 1559, 1540, 1506, 1488, 1464, 1457,

1436, 1424, 1355, 1339, 1297, 1276, 1214, 1201, 1144, 1123,

1103, 1084, 1053, 964, 826 cm^-1

; GCMS (PCI) m/ e (rel

intensity) 203 (100), 235 (92), 234 (86), 233 (34), 220 (33), 75

(18), 236 (15), 221 (9), 202 (9), 217 (8), 189 (8), 263 (6); HRMS

(EI) m/ e calcd for C₁₅H₂₂O₄266.15180, found 266.15141. The

spectral data for 13 (mixed with ca. 25% of 12a ) are as

follows: ¹H NMR (300 MHz, CDCl₃) δ 6.30 (s, 1 H), 6.22 (s, 1

H), 4.47 (bd s, 1 H), 3.82 (s, 3 H), 3.77 (s, 3 H), 3.42 (s, 3 H),

2.68 (m, 2 H), 2.21 (bd d, J ) ca. 14 Hz, 1 H), 1.94 (m, 1 H),

1.69 (m, 1 H), 1.47 (tt, J _t) 14 Hz, J _t) 3 Hz, 1 H); ¹³C NMR

(75 MHz, CDCl₃) δ 159.9, 159.5, 139.9, 118.2, 104.3, 96.4, 70.0,

56.7, 55.6, 55.3, 29.7, 26.7, 17.3; IR (neat, NaCl) 2935, 2836,

1600, 1463, 1347, 1304, 1280, 1217, 1206, 1192, 1156, 1142,

1125, 1103, 1083, 1048 cm^-1; HRMS (EI) m/ e calcd for

C₁₃H₁₈O₃222.1256, found 222.1245.

(E,Z)-5-(3,5-Dim et h oxyp h en yl)-1-(m et h ylt h io)-1-p en -

ten e (11b). A solution of 5.273 g (25.1 mmol) of the alcohol

1588 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

and 2.548 g (32.6 mmol) of dimethyl sulfoxide in 50 mL of

dichloromethane was cooled to -78 °C. After 15 min, 3.822 g

(30.1 mmol) of oxalyl chloride was added and the reaction

mixture allowed to warm to between -50 and -60 °C. After

another 15 min, 12.701 g (125.5 mmol) of triethylamine was

added and the reaction allowed to warm to room temperature.

Water was added to the mixture and then the dichloromethane

evaporated in vacuo. The remaining solution was extracted

with ether, and then the combined ether layers were washed

with brine, dried over MgSO₄, and concentrated in vacuo. As

this was happening, a solution of 17.4 g (48.5 mmol) of

[(methylthio)methyl]triphenylphosphonium chloride in 150 mL

of THF was cooled to 0 °C and treated with 26.9 mL (48.5

mmol) of a phenyllithium in a 70:30 cyclohexane:ether solu-

tion. The reaction mixture was allowed to warm to room

temperature and stirred for 1 h. The aldehyde from above was

then diluted with THF and transferred into the flask contain-

ing the ylide. After 16 h, the reaction was quenched with

water and ether. The layers were separated, and then the

aqueous layer was extracted with ether. The combined organic

layers were dried over MgSO₄and concentrated in vacuo. The

crude product was chromatographed through silica gel using

a gradient elution from hexane to 18% ether in hexane by

adding an additional 3% of ether for every 250 mL of eluent

used in order to afford 3.798 g (60%) of the desired vinyl

sulfide: ¹H NMR (300 MHz, CDCl₃) δ 6.29-6.35 (m, 3 H), 5.98

(dt, J ) 15.1 Hz, J ) 1.5 Hz, 0.4 H), 5.90 (d, J ) 9.4 Hz, 0.6

H), 5.54 (dt, J ) 9.4 Hz, J ) 7.1 Hz, 0.4 H), 5.44 (dt, J ) 14.9

Hz, J ) 7.0 Hz, 0.6 H), 3.77 (s, 6 H), 2.53-2.60 (m 2 H), 2.26

(s, 1.2 H), 2.21 (s, 1.8 H), 2.09-2.20 (m, 2 H), 1.65-1.75 (m 2

H); ¹³C NMR (75 MHz, CDCl₃) δ 160.7, 144.8, 144.7, 128.3,

127.2, 126.9, 124.2, 106.50, 106.45, 97.75, 97.70, 55.2, 35.8,

35.6, 32.6, 30.9, 30.6, 28.8, 17.1, 15.1; IR (neat, NaCl) 2999,

2936, 2923, 2920, 2857, 2837, 1608, 1604, 1594, 1461, 1428,

1351, 1323, 1310, 1292, 1205, 1193, 1154, 1068, 1059, 940, 831,

696 cm^-1; GCMS (PCI) m/ e (rel intensity) 253 (100), 205 (90),

206 (58), 281 (57), 152 (49), 267 (41), 191 (37), 177 (34), 219

(34), 204 (33), 254 (31), 87 (23), 233 (21), 63 (16), 151 (14), 255

(14), 282 (10), 295 (10), 293 (8); HRMS (EI) m/ e calcd for

C₁₄H₂₀O₂S 252.1184, found 252.1186. Anal. Calcd for

C₁₄H₂₀O₂S: C, 66.63; H, 7.99. Found: C, 66.54; H, 8.05.

An odic Oxidation of 11b: For m ation of 6,8-Dim eth oxy-

1,2,3,4-tetr a h yd r on a p h th a len eca r boxa ld eh yd e Dim eth yl

Th ioa ceta l (12b). Again the anodic oxidation was done in a

fashion similar to the constant current electrolysis of substrate

4. In this example, 0.418 g (1.7 mmol) of 11b, 0.890 g (8.3

mmol) of 2,6-lutidine, 2.826 g (26.6 mmol) of lithium perchlo-

rate, 13.3 mL of methanol, and 53.1 mL of dichloromethane

were oxidized at a constant current of 11.3 mA until 320.9 C

(2.0 faradays/mol) of charge had been passed. Workup and

chromatography afforded 0.338 g (72%) of the desired product

(12b) as a mixture of two diastereomers: ¹H NMR (300 MHz,

CDCl₃) δ 6.24-6.28 (m, 2 H), 4.89 (d, J ) 4.8 Hz, 0.5 H), 4.79

(d, J ) 4.3 Hz, 0.5 H), 3.785 (s, 1.5 H), 3.780 (s, 1.5 H), 3.76 (s,

3 H), 3.61-3.69 (m, 0.5H), 3.40-3.47 (m, 2 H includes 1.5H

s), 3.22 (s, 1.5 H), 2.56-2.82 (m, 2 H), 1.47-2.26 (m, includes

2.17 (s), 1.89 (s), 7 H); ¹³C NMR (75 MHz, CDCl₃) δ 158.7,

158.6, 158.5, 158.4, 141.4, 140.7, 118.5, 117.5, 104.7, 104.5,

96.2, 96.0, 94.6, 93.2, 57.3, 57.0, 55.25, 55.20, 55.1, 37.0, 30.5,

29.8, 24.1, 23.8, 21.1, 20.1, 15.1, 14.1; IR (neat, NaCl) 2933,

2836, 1605, 1595, 1488, 1463, 1424, 1353, 1302, 1277, 1216,

workup and chromatography, the reaction mixture afforded

0.885 g (59%) of the desired product: ¹H NMR (300 MHz,

CDCl₃) δ 7.14-7.19 (m, 1 H), 6.69-6.77 (m, 3 H), 5.96 (dt, J

) 14.9 Hz, J ) 1.6 Hz, 0.57 H), 5.88 (dt, J ) 9.3 Hz, J ) 1.3

Hz 0.43 H), 5.52 (dt, J ) 9.4 Hz, J ) 7.1 Hz, 0.43 H), 5.42 (dt,

J ) 14.9 Hz, J ) 7.0 Hz, 0.57 H), 3.75 (s, 3 H), 2.55-2.62 (m,

2 H), 2.22 (s, 1.29 H), 2.19 (s, 1.71 H), 2.07-2.16 (m, 2 H),

1.64-1.75 (m, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 159.6, 144.0,

143.9, 129.2, 128.3, 127.3, 126.8, 124.2, 120.9, 114.2, 114.1,

111.05, 111.00, 55.1, 54.9, 35.7, 35.3,32.7, 31.1, 30.7, 28.8, 17.0,

15.0; IR (neat, NaCl) 3023, 3000, 2935, 2921, 2857, 2834, 1610,

1594, 1584, 1487, 1465, 1456, 1436, 1313, 1291, 1260, 1165,

1152, 1043, 938, 873, 778, 735, 696 cm^-1; GCMS (PCI) m/ e

(rel intensity) 223 (100), 175 (98), 251 (83), 224 (55), 121 (46),

147 (44), 203 (40), 176 (34), 174 (26), 225 (24), 222 (23), 161

(21); HRMS (EI) m/ e calcd for C₁₃H₁₈OS 222.1078, found

222.1091. Anal. Calcd for C₁₃H₁₈OS: C, 70.23; H, 8.16.

Found: C, 70.35; H, 8.23.

6-Met h oxy-1,2,3,4-t et r a h yd r on a p h t h a len eca r b oxa l-

d eh yd e Dim eth yl Th ioa ceta l (5b). Compound 4b was

oxidized in a manner similar to the constant current oxidation

of compound 4a . In this experiment, 0.364 g (1.64 mmol) of

4b, 0.879 g (8.2 mmol) of 2,6-lutidine, 2.792 g (26.2 mmol) of

lithium perchlorate, 13.1 mL of methanol, and 52.4 mL of

dichloromethane were electrolyzed at a constant current of

12.0 mA until 319.2 C (2.02 faradays/mol) of charge had been

passed. Workup and chromatography afforded 0.159 g (38%)

of the desired product as a mixture of two diastereomers: ¹H

NMR (300 MHz, CDCl₃) δ 7.31 (d, J ) 8.4 Hz, 0.67 H), 7.19 (J

) 8.6 Hz, 0.33 H), 6.59-6.70 (m, 2 H), 4.52 (d, J ) 4.9 Hz,

0.67 H), 4.39 (d, J ) 7.4 Hz, 0.33 H), 3.755, 3.750 (2 s, 3 H),

3.37 (s, 2 H), 3.34 (s, 1 H), 3.08-3.22 (m, 1 H), 2.64-2.84 (m,

2 H), 1.62-2.14 (m, including 2.09 (s), 2.02 (s), 7 H); ¹³C NMR

(75 MHz, CDCl₃) δ 157.8, 157.7, 139.3, 138.8, 131.3, 130.4,

129.1, 128.9, 113.5, 113.4, 111.4, 94.5, 93.4, 56.8, 56.5, 55.1,

55.0, 47.6, 41.1, 29.9, 25.8, 25.3, 20.5, 20.1, 14.1, 11.7; IR (neat,

NaCl) 2989, 2983, 2977, 2930, 2883, 2866, 2833, 1464, 1457,

1447, 1436, 1428, 1419, 1326, 1318, 1298, 1273, 1255, 1238,

1187, 1154, 1123, 1099, 1080, 1041, 946, 871, 835, 816, 787,

774 cm^-1; GCMS (PCI) m/ e (rel intensity) 147 (100), 173 (78),

221 (74), 113 (69), 63 (68), 148 (59), 61 (27), 175 (23), 121 (21),

249 (20), 174 (19), 95 (19); HRMS (EI) m/ e calcd for C₁₆H₂₀O₂S

252.1184, found 252.1175.

(E,Z)-2-Meth oxy-6-p h en yl-2-h exen e (14). The oxidation

of 4-phenyl-1-butanol (1.931 g, 12.9 mmol) was done under the

same Swern oxidation conditions described above for the

syntheses of 4 and 4b. In this case, 1.794 g (14.1 mmol) of

oxalyl chloride, 1.209 g (15.6 mmol) of dimethyl sulfoxide, 6.527

g (64.5 mmol) of triethylamine, and 20 mL of dichloromethane

were used. After the usual workup, the reaction mixture was

added to the reaction mixture described below.

Lithium diisopropylamide (LDA) was synthesized by adding

15.5 mL (38.7 mmol) of n-butyllithium to 3.916 g (38.7 mmol)

of diisopropylamine in 10.0 mL of THF at -78 °C. After 1 h

at -78 °C, the LDA was added to a -78 °C solution of 10.066

g (38.7 mmol) of (methoxyethyl)diphenylphosphine oxide in 20

mL of THF. To the resulting mixture was added the aldehyde

made above. The reaction was held at -78 °C for 1 h and

then allowed to warm slowly to room temperature. After

stirring for 12 h, the reaction was quenched with H₂O.

Triethylamine (10 mL) was added to the mixture before

completing the workup in order to minimize hydrolysis of the

enol ether. The resulting solution was extracted three times

with ether, and then the organic layers were dried over MgSO₄

and concentrated in vacuo. It is important to note that the

crude reaction mixture was concentrated without warming the

bath since the product proved to be sensitive to heat. The

crude product was chromatographed through 300 mL of

activity III basic alumina with a gradient elution of hexane to

15% ether in hexane (3% steps every 200 mL of eluent) to

afford 1.383 g (61%) of the slightly impure (<5% impurity by

NMR) desired product: ¹H NMR (300 MHz, CDCl₃) δ 7.13-

7.31 (m, 5 H), 4.44 (t, J ) 7.3 Hz, 0.35 H), 4.38 (t, J ) 7.3 Hz,

0.65 H), 3.49 (s, 1.05 H), 3.53 (s, 1.95 H), 2.58-2.68 (m, 2 H),

1.98-2.13 (m, 2 H), 1.83 (s, 1.05 H), 1.76 (s, 1.95 H), 1.58-

1.73 (m, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 154.1, 151.7, 143.5,

1200, 1148, 1100, 1079, 1052, 947, 910, 848, 827, 732 cm^-1

;

GCMS (PCI) m/ e (rel intensity) 251 (100), 250 (66), 203 (59),

252 (34), 204 (25), 236 (17), 249 (14), 253 (8), 279 (8), 205 (7),

237 (6), 231 (5); HRMS (EI) m/ e calcd for C₁₅H₂₂O₃S 282.1290,

found 282.1277.

(E,Z)-5-(3-Met h oxyp h en yl)-1-(m et h ylt h io)-1-p en t en e

(4b). Compound 4b was made in a fashion directly analogous

to 11b above. In this case, 1.217 g (6.8 mmol) of 3, 0.686 g

(8.8 mmol) of dimethyl sulfoxide, 13.0 mL of dichloromethane,

1.028 g (8.1 mmol) of oxalyl chloride, and 3.416 g (33.8 mmol)

of triethylamine were used in the Swern oxidation. The

resulting aldehyde was added to the ylide derived from 4.845

g (13.5 mmol) of [(methylthio)methyl]triphenylphosphonium

chloride in 10 mL of THF and 7.5 mL (13.5 mmol) of a 1.8 M

solution of phenyllithium in 70:30 cyclohexane:ether. After

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

J . Org. Chem., Vol. 61, No. 5, 1996 1589

143.3, 129.0, 128.85, 128.80, 126.2, 126.1, 111.8, 108.6, 96.6,

55.9, 54.3, 46.5, 35.9, 35.6, 32.8, 32.0, 26.7, 24.6, 17.6, 16.3;

IR (neat, NaCl) 3062, 3027, 2998, 2931, 2856, 1669, 1603, 1496,

1463, 1453, 1439, 1391, 1274, 1261, 1224, 1181, 1140, 1083,

1030, 806, 745, 699 cm^-1; GCMS (PCI) m/ e (rel intensity) 117

(100), 159 (56), 191 (53), 85 (22), 91 (21), 145 (20), 192 (12),

118 (12), 189 (10), 81 (10), 160 (9); HRMS (EI) m/ e calcd for

(E,Z)-1-(Meth ylth io)-5-(3,4,5-tr im eth oxyp h en yl)-1-p en -

ten e (16). The reaction sequence was carried out in a similar

fashion to that described above for the synthesis of 4b. In this

example, 2.603 g (10.8 mmol) of the alcohol, 1.645 g (13.0

mmol) of oxalyl chloride, 1.097 g (14.0 mmol) of dimethyl

sulfoxide, 5.465 g (54.0 mmol) of triethylamine, and 22 mL of

dichloromethane were used. After workup the aldehyde was

diluted with 75 mL of THF and added to a solution of the ylide

formed from 7.752 g (21.6 mmol) of [(methylthio)methyl]-

triphenylphosphine chloride, 12.0 mL (21.6 mmol) of phenyl-

lithium, and 12 mL of THF. Workup and chromatography

afforded 2.131 g (70%) of the desired product: ¹H NMR (300

MHz, CDCl₃) δ 6.41 (s, 0.9 H), 6.39 (s, 1.1 H), 6.01 (d, J )

15.1 Hz, 0.55 H), 5.92 (d, J ) 9.3 Hz, 0.45 H), 5.55 (dt, J ) 9.4

Hz, J ) 7.2 Hz, 0.45 H), 5.46 (dt, J ) 14.9 Hz, J ) 7.0 Hz,

0.55 H), 3.85 (s, 6 H), 3.82 (s, 3 H), 2.54-2.61 (m, 2 H), 2.27

(s, 1.35 H), 2.23 (s, 1.65 H), 2.12-2.21 (m, 2 H), 1.66-1.77 (m,

2 H); ¹³C NMR (75 MHz, CDCl₃) δ 153.0, 138.2, 138.1 128.2,

127.2, 126.8, 124.2, 105.2, 60.8, 56.0, 35.9, 35.7, 32.7, 31.2, 30.8,

28.8, 17.0, 15.1; IR (neat, NaCl) 2994, 2935, 2925, 2856, 2837,

1589, 1559, 1508, 1457, 1420, 1349, 1332, 1319, 1238, 1183,

1131, 1126, 1039, 1011, 958, 939, 824, 780, 668 cm^-1; GCMS

(PCI) m/ e (rel intensity) 282 (100), 235 (83), 284 (26), 61 (26),

181 (25), 79 (24), 63 (22), 87 (22), 282 (20), 207 (15), 236 (14),

115 (12); HRMS (EI) m/ e calcd for C₁₅H₂₂O₃S 282.1290, found

282.1294. Anal. Calcd for C₁₅H₂₂O₃S: C, 63.80; H, 7.85.

Found: C, 63.84; H, 7.63.

C

₁₃H₁₈O 190.1358, found 190.1345.

(E)-6-P h en yl-3-h exen -2-on e (15). In a fashion similar to

the oxidation of compound 4, 0.256 g (1.3 mmol) of 14, 0.780

g (7.3 mmol) of 2,6-lutidine, 2.468 g (23.2 mmol) of lithium

perchlorate, 11.6 mL of methanol, and 46.4 mL of dichlo-

romethane were oxidized at a constant current of 11.3 mA until

285.0 C (2.03 faradays/mol) of charge had been passed. After

workup, the products were taken up in 15 mL of acetone and

1.011 g (4.0 mmol) of PPTS was added. The reaction mixture

was allowed to stir for 72 h and then worked up by being

diluted with H₂O and then extracting the resulting mixture

three times with ether. The ether layers were dried over

MgSO₄, filtered, and concentrated in vacuo. The crude oil was

then chromatographed through 45 mL of silica gel using a

gradient elution from hexane to 30% ether/hexane (3% steps

every 75 mL of eluent) to afford 0.120 g (51%) of the desired

product. NMR analysis showed that the product contained a

few percent of an unidentified impurity. 15: ¹H NMR (300

MHz, CDCl₃) δ 7.12-7.28 (m, 5 H), 6.78 (dt, J ) 16.0 Hz, J )

6.8 Hz, 1 H), 6.05 (d, J ) 16.0 Hz, 1 H), 2.75 (t, J ) 7.6 Hz, 2

H), 2.50 (dt, J ) 7.1 Hz, J ) 7.4 Hz, 2 H), 2.18 (s, 3 H); ¹³C

NMR (75 MHz, CDCl₃) δ 198.6, 147.1, 140.6, 131.7, 128.5,

128.3, 126.2, 34.4, 34.1, 26.9; IR (neat, NaCl) 3085, 3062, 3028,

3003, 2927, 2857, 1697, 1675, 1626, 1602, 1496, 1454, 1429,

1360, 1317, 1292, 1254, 1187, 1161, 1088, 1078, 1029, 1021,

976, 748, 700 cm^-1; GCMS (PCI) m/ e (rel intensity) 175 (100),

71 (94), 91 (35), 117 (28), 119 (18), 176 (17), 203 (17), 105 (16),

157 (12), 116 (11), 131 (10), 174 (6); HRMS (EI) m/ e calcd for

C₁₂H₁₄O 174.1045, found 174.1051.

An od ic Oxid a tion of Su bstr a te 16: 6,7,8-Tr im eth oxy-

1,2,3,4-tetr a h yd r on a p h th a len eca r boxa ld eh yd e Dim eth yl

Th ioa ceta l (17) a n d 7,9-Dim eth oxy-8-oxosp ir o[4.5]d eca -

6,9-dien ecar boxaldeh yde Dim eth yl Th ioacetal (18). Com-

pound 16 was oxidized in a manner similar to the constant

current oxidation of compound 4. In this experiment, 0.210 g

(0.74 mmol) of 16, 0.399 g (3.72 mmol) of 2,6-lutidine, 1.260 g

(11.8 mmol) of lithium perchlorate, 5.9 mL of methanol, and

23.7 mL of dichloromethane were used. The oxidation was

conducted with a constant current of 11.0 mA until 130.4 C

(1.8 faradays/mol) of charge had been passed. Workup and

chromatography afforded 58.7 mg (25%) of the desired product

(17) as a mixture of diastereomers and 59.7 mg (27%) of the

spirocyclic compound (18) as a mixture of diastereomers. In

addition, a 21% yield of material that was tentatively assigned

4-(3,4,5-Tr im eth oxyp h en yl)-1-bu ten e. The desired olefin

product was synthesized in a fashion directly analogous to the

previously described synthesis of olefin 2. In this example,

5.006 g (25.3 mmol) of 3,4,5-trimethoxybenzyl alcohol, 3.420

g (12.6 mmol) of phosphorus tribromide, and 100 mL of THF

were used. Upon completion, 50.5 mL (50.5 mmol) of a 1.0 M

solution of allylmagnesium bromide in ether was added.

Workup and chromatography afforded 2.777 g (49%) of the

desired product: ¹H NMR (300 MHz, CDCl₃) δ 6.41 (s, 2 H),

5.86 (ddt, J ) 17.0 Hz, J ) 10.3 Hz, J ) 6.6 Hz, 1 H), 5.06

(dq, J ) 17.2 Hz, J ) 1.7 Hz, 1 H), 4.99 (d of multiplets, J )

11.9 Hz, 1 H), 3.84 (s, 6 H), 3.82 (s, 3 H), 2.65 (t, J ) 7.9 Hz,

2 H), 2.37 (dt, J ) 7.0 Hz, J ) 7.7 Hz, 2 H); ¹³C NMR (75 MHz,

CDCl₃) δ 153.1, 138.1, 137.8, 136.1, 115.1, 105.3, 60.9, 56.1,

56.0, 35.9, 35.7; IR (neat, NaCl) 3074, 2997, 2939, 2838, 1640,

1589, 1509, 1457, 1420, 1345, 1325, 1237, 1183, 1151, 1133,

1128, 1042, 1011, 966, 913, 842, 822, 781 cm^-1; GCMS (PCI)

m/ e (rel intensity) 223 (100), 181 (91), 222 (24), 208 (18), 209

(17), 221 (15), 224 (15), 182 (10); HRMS (EI) m/ e calcd for

C₁₃H₁₈O₃222.1256, found 222.1262.

1

by H NMR as the uncyclized product 19 was obtained. The

1

spectral data for compound 17 are as follows: H NMR (300

MHz, CDCl₃) δ 6.42 (s, 1 H), 4.87 (d, J ) 3.8 Hz, 0.33 H), 4.79

(d, J ) 5.7 Hz, 0.67 H), 3.90 (s, 2H), 3.85 (s, 1 H), 3.83 (s, 6 H)

3.55-3.61 (m, 0.67 H), 3.33-3.38 (m, 0.33 H), 3.43 (s, 2 H),

3.23 (s, 1 H), 2.57-2.78 (m, 2 H), 2.21 (s, 1 H), 1.52-2.08 (m,

including 1.97 s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 151.6,

139.8, 134.6 123.1, 122.4, 107.4, 94.3, 94.2, 60.8, 60.7, 57.8,

57.1, 55.7, 38.2, 37.2, 29.7, 29.6, 24.1, 23.7, 20.5, 20.4, 14.5,

14.2; IR (neat, NaCl) 2977, 2934, 2879, 2834, 1599, 1581, 1493,

1463, 1457, 1432, 1407, 1349, 1271, 1193, 1141, 1122, 1094,

1082, 1034, 1022, 1009, 946, 910, 823, 796 cm^-1; GCMS (PCI)

m/ e (rel intensity) 265 (100), 233 (84), 264 (80), 263 (29), 266

(24), 234 (21), 218 (16), 249 (13), 237 (9), 219 (9), 191 (8), 293

(7); HRMS (EI) m/ e calcd for C₁₆H₂₄O₄S 312.1395, found

312.1365. The spectral data for compound 18 are as follows:

¹H NMR (300 MHz, CDCl₃) δ 5.97 (d, J ) 1.8 Hz, 0.25 H),

5.91 (d, J ) 0.9 Hz, 0.75 H), 5.81 (d, J ) 1.8 Hz, 0.75 H), 5.69

(d, J ) 0.9 Hz, 0.25 H), 3.89 (d, J ) 5.7 Hz, 0.5 H), 3.80-3.85

(m, 0.5 H), 3.70 (s, 3 H), 3.68 (s, 3 H), 3.46 (s, 0.25 H), 3.22 (s,

0.75 H), 3.06 (s, 2 H), 2.54-2.61 (m, 1 H), 1.67-2.32 (m,

includes 1.97(s), 1.96 (s), 1.94 (s), 9H); ¹³C NMR (75 MHz,

CDCl₃) δ 178.7, 176.9, 176.8, 151.0, 150.8, 149.6, 149.6, 125.0,

122.6, 122.5, 1220.5, 117.4, 115.3, 89.2, 86.4, 84.7, 58.1, 57.4,

55.3, 55.3, 55.1, 54.9, 54.7, 49.5, 48.9, 40.5, 39.9, 29.5, 28.4,

5.6, 25.5, 22.3, 21.7, 18.2, 14.7, 11.6, 9.2; IR (neat, NaCl) 2953,

2936, 2926, 2883, 2878, 2854, 2831, 1663, 1659, 1616, 1591,

1463, 1457, 1265, 1229, 1192, 1114, 1080, 1008, 938, 913, 900,

876, 866, 730 cm^-1; GCMS (PCI) m/ e (rel intensity) 219 (100),

251 (67), 63 (58), 95 (55), 221 (49), 97 (46), 267 (41), 250 (25),

253 (22), 167 (22), 91 (21), 57 (19), 299 (2); HRMS (EI) m/ e

calcd for C₁₅H₂₂O₄S 298.12387, found 298.12251. The proton

NMR data for tentatively assigned 19 (contaminated with the

corresponding aldehyde product and another small impurity)

4-(3,4,5-Tr im eth oxyp h en yl)-1-bu ta n ol. The hydrobora-

tion was carried out in a similar fashion to that described

above for the synthesis of 3. In this example, 20.960 g (94.3

mmol) of the olefin, 25.789 g (367.9 mmol) of 2-methyl-2-

butene, 18.9 mL (188.7 mmol) of the borane-dimethylsulfide

complex, and 300 mL of THF were used. The reaction was

worked up with 120 mL of 3 N NaOH and 60 mL of H₂O₂.

Workup and chromatography afforded 9.659 g (43%) of the

desired product: ¹H NMR (300 MHz, CDCl₃) δ 6.40 (s, 2 H),

3.85 (s, 6 H), 3.82 (s, 3 H), 3.66 (t, J ) 6.2 Hz, 2 H), 2.59 (t, J

) 7.4 Hz, 2 H), 1.87 (br peak, 1 H), 1.59-1.74 (m, 4 H); ¹³C

NMR (75 MHz, CDCl₃) δ 153.0, 138.2, 135.9, 105.2, 62.7, 60.8,

56.0, 36.1, 32.3, 27.7; IR (neat, NaCl) 3354 br, 2993, 2935,

2861, 2840, 1589, 1509, 1460, 1420, 1382, 1368, 1343, 1333,

1323, 1237, 1183, 1150, 1129, 1124, 1065, 1031, 1009, 973, 832,

828, 780, 735 cm^-1; GCMS (PCI) m/ e (rel intensity) 241 (100),

223 (75), 181 (31), 79 (25), 239 (24), 240 (24), 221 (16), 242

(14), 193 (13), 224 (12), 151 (11), 61 (11); HRMS (EI) m/ e calcd

for C₁₃H₂₀O₄240.1361, found 240.1360.

1590 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

are as follows: ¹H NMR (300 MHz, CDCl₃) δ 6.42 (s, 2 H),

4.30 (d, J ) 6 Hz, 1 H), 3.86 (s, 6 H), 3.81 (s, 3 H), 3.45 (s with

shoulder, 6 H for 19 (the signal integrates low because of the

presence of the aldehyde)), 3.39 (m, 1 H) 2.59 (m, 2 H), 2.13

and 2.12 (two singlets, 3 H (the signal integrates low because

of the presence of the aldehyde)), 1.80 (m, 2 H), 1.67 (m, 2 H).

(E,Z)-1-Meth oxy-5-(4-m eth oxyp h en yl)-1-p en ten e (20).

Substrate 20 was synthesized in a fashion identical to that

described above for the synthesis of substrate 4. In this

example, 1.524 g (8.5 mmol) of 4-(4′-methoxyphenyl)-1-butanol,

1.287 g (10.1 mmol) of oxalyl chloride, 0.859 g (11.0 mmol) of

dimethyl sulfoxide, 4.276 g (42.3 mmol) of triethylamine, and

16 mL of dichloromethane were used. After workup the

aldehyde was diluted with 75 mL of THF and added to the

ylide. The ylide was formed from 8.690 g (25.4 mmol) of

(methoxymethyl)triphenylphosphonium chloride, 14.9 mL (25.4

mmol) of a 1.7 M solution of tert-butyllithium in pentane, and

75 mL of THF. Workup and chromatography afforded 1.302

g (75%) of the desired product: ¹H NMR (300 MHz, CDCl₃) δ

7.07 (dd, J ) 8.6 Hz, J ) 2.2 Hz, 2 H), 6.79 (dd, J ) 8.6 Hz, J

) 2.2 Hz, 2 H), 6.27 (d, J ) 12.7 Hz, 0.55 H), 5.86 (d with fine,

J ) 6.3 Hz, J ) 1.5, 0.45 H), 4.71 (dt, J ) 12.6 Hz, J ) 7.4 Hz,

0.55 H), 4.34 (dt, J ) 6.5 Hz, J ) 7.2 Hz, 0.45 H), 3.73 (s, 3

H), 3.53 (s, 0.45 H), 3.48 (s, 0.55 H), 2.50-2.57 (m, 2 H), 2.09

(dtd, J ) 7.3 Hz, J ) 7.5 Hz, J ) 1.4 Hz, 0.9 H), 1.93 (dt, J )

7.1 Hz, J ) 7.4 Hz, 1.1 H), 1.57-1.67 (m, 2 H); ¹³C NMR (75

MHz, CDCl₃) δ 157.65, 157.60, 147.3, 146.3, 134.7, 134.5, 129.3,

113.65, 113.60, 106.4, 102.5, 59.4, 55.7, 55.1, 34.6, 34.3, 32.7,

31.9, 27.2, 23.6; IR (neat, NaCl) 3029, 2999, 2932, 2854, 2834,

1663, 1653, 1635, 1612, 1584, 1513, 1464, 1457, 1441, 1390,

1300, 1245, 1209, 1177, 1151, 1132, 1108, 1037, 934, 831, 810,

749, 739, 700 cm^-1; GCMS (PCI) m/ e (rel intensity) 175 (100)

193 (77), 176 (56), 121 (56), 203 (38), 177 (34), 142 (24), 174

(23), 122 (19), 207 (16), 192 (16), 206 (15), 120 (11); HRMS

(EI) m/ e calcd for C₁₃H₁₈O₂206.1307, found 206.1316. Anal.

Calcd for C₁₃H₁₈O₃: C, 70.24; H, 8.16. Found: C, 70.14; H,

7.98.

35.9, 35.6; IR (neat, NaCl) 3463 br, 3074, 2999, 2975, 2937,

2840, 1459, 1455, 1428, 1364, 1342, 1320, 1289, 1269, 1239,

1214, 1207, 1151, 1123, 1118, 1110, 1106, 1042, 1017, 998, 911,

824, 803, 793 cm^-1; GCMS (PCI) m/ e (rel intensity) 209 (100),

210 (45), 237 (32), 207 (26), 205 (18), 195 (12), 249 (11), 208

(6); HRMS (EI) m/ e calcd for C₁₂H₁₆O₃208.1099, found

208.1099.

4-(3,5-Dim eth oxy-4-(pivaloyloxy)ph en yl)-1-bu tan ol. Pro-

tection of the phenol derivative synthesized above was ac-

complished at 0 °C using 3.705 g (17.8 mmol) of the phenol,

9.660 g (80.1 mmol) of trimethylacetyl chloride, 8.106 g (80.1

mmol) of triethylamine, and 75 mL of dichloromethane. The

reaction mixture was stirred for 1 h, then the reaction was

quenched with water, and the mixture was diluted with ether,

extracted once with 3 N NaOH, and then washed three times

with H₂O. The ether layers were washed once with brine,

dried over MgSO₄, filtered, and concentrated in vacuo, and

then the resulting oil was filtered through 150 mL of silica

gel. This material was then used directly in a hydroboration

reaction that was done in a fashion directly analogous to the

procedure described for the synthesis of 3. In this example,

the 4-(3,5-dimethoxy-4-(pivaloyloxy)phenyl)-1-butene synthe-

sized in the first step, 7.362 g (105.2 mmol) of 2-methyl-2-

butene, 5.34 mL (53.4 mmol) of the borane-dimethyl sulfide

complex, and 150 mL of THF were used. The reaction was

quenched with 25 mL of 3 N NaOH and 12.5 mL of H₂O₂.

Workup and chromatography afforded 4.212 g (76%) of the

desired product: ¹H NMR (300 MHz, CDCl₃) δ 6.41 (s, 2 H),

3.77 (s, 6 H), 3.62 (t, J ) 6.3 Hz, 2 H), 2.59 (t, J ) 7.4 Hz, 2

H), 1.86 (s, 1 H), 1.53-1.73 (m, 4 H), 1.37 (s, 9 H); ¹³C NMR

(75 MHz, CDCl₃) δ 176.7, 151.9, 140.5, 127.0, 105.0, 62.6, 56.1,

39.1, 32.3, 27.6, 27.2; IR (neat, NaCl) 3316 br, 2971, 2939,

2870, 2842, 1753, 1603, 1510, 1479, 1463, 1423, 1397, 1368,

1345, 1325, 1281, 1242, 1204, 1190, 1130, 1050, 1029 cm^-1

;

GCMS (PCI) m/ e (rel intensity) 227 (100), 209 (98), 85 (94),

207 (65), 210 (54), 103 (50), 177 (38), 226 (32), 167 (31), 208

(28), 293 (24), 87 (23), 311 (22), 79 (21), 310 (10); HRMS (EI)

m/ e calcd for C₁₇H₂₆O₅310.1780, found 310.1792.

An od ic Oxid a tion of 20: 8-Oxosp ir o[4.5]d eca -6,9-d i-

en eca r boxa ld eh yd e Dim eth yl Aceta l (21). Compound 20

was oxidized in a manner similar to the constant current

oxidation of compound 4. In this experiment, 0.124 g (0.6

mmol) of 20, 0.323 g (3.0 mmol) of 2,6-lutidine, 1.027 g (9.7

mmol) of lithium perchlorate, 4.8 mL of methanol, and 19.3

mL of dichloromethane were used. The oxidation was con-

ducted with a constant current of 11.2 mA until 116.6 C (2.0

faradays/mol) of charge had been passed. Workup and chro-

matography afforded 68.2 mg (51%) of the desired product: ¹H

NMR (300 MHz, CDCl₃) δ 7.04 (dd, J ) 10.0 Hz, J ) 2.9 Hz,

1 H), 6.83 (dd, J ) 9.8 Hz, J ) 2.9 Hz, 1 H), 6.30 (dd, J ) ca.

10.2 Hz, J ) 1.9 Hz, 1 H), 6.26 (dd, J ) 9.8 Hz, J ) 1.9 Hz, 1

H), 4.05 (d, J ) 7.4 Hz, 1 H), 3.28 (s, 3 H), 3.18 (s, 3 H), 2.43-

2.51 (m, 1 H), 1.69-2.09 (m, 6 H); ¹³C NMR (75 MHz, CDCl₃)

δ 186.8, 156.5, 150.1, 128.3, 127.8, 105.5, 54.9, 53.6, 52.8, 50.5,

38.8, 26.1, 22.6; IR (neat, NaCl) 2952, 2881, 2832, 1623, 1598,

1456, 1448, 1408, 1394, 1387, 1368, 1361, 1263, 1246, 1189,

1145, 1132, 1094, 1062, 1035, 1023, 985, 953, 932, 908, 861,

772 cm^-1; GCMS (PCI) m/ e (rel intensity) 191 (100), 159 (90),

187 (20), 190 (18), 160 (14), 192 (13), 161 (13), 219 (11), 189

(8); HRMS (EI) m/ e calcd for C₁₃H₁₈O₃222.1256, found

222.1241. Anal. Calcd for C₁₃H₁₈O₃: C, 70.24; H, 8.16.

Found: C, 70.14; H, 7.98.

4-(3,5-Dim eth oxy-4-h ydr oxyph en yl)-1-bu ten e. The alky-

lation was accomplished in the same way as the synthesis of

2. In this example, 7.2 g (39.1 mmol) of the 3,5-dimethoxy-

4-hydroxybenzyl alcohol, 5.3 g (19.6 mmol) of phosphorus

tribromide, and 100 mL of THF were used. Upon completion

of the bromination, 156 mL (156 mmol) of a 1.0 M solution of

allylmagnesium bromide in ether was added. Workup and

chromatography afforded 4.68 g (58%) of a slightly impure (5%

impurity by NMR) product that was carried on without further

purification: ¹H NMR (300 MHz, CDCl₃) δ 6.41 (s, 2 H), 5.85

(ddt, J ) 17.0 Hz, J ) 10.3 Hz, J ) 6.6 Hz, 1 H), 5.05 (d of d,

J ) 17.9 Hz, J ) 1.8 Hz, 1 H), 4.98 (d of multiplets, J ) 10.3

Hz, 1 H), 4.55 (br peak, 1 H), 3.85 (s, 6 H), 2.63 (t, J ) 7.8 Hz,

2 H), 2.34 (dt, J ) 8.2 Hz, J ) 6.9 Hz, 2 H); ¹³C NMR (75

MHz, CDCl₃) δ 146.9, 138.1, 133.0, 132.8, 115.0, 105.0, 56.2,

(E,Z)-5-(3,5-Dim eth oxy-4-(p iva loyloxy)p h en yl)-1-m eth -

oxy-1-p en ten e (22a ). The reaction sequence was carried out

in a similar fashion to the procedure described above for the

synthesis of 4. In this example, 1.640 g (5.3 mmol) of the

alcohol synthesized in the previous experiment, 0.804 g (6.3

mmol) of oxalyl chloride, 0.536 g (6.9 mmol) of dimethyl

sulfoxide, 2.672 g (26.4 mmol) of triethylamine, and 11 mL of

dichloromethane were used. After workup, the aldehyde was

diluted with 75 mL of THF and added to the ylide. The ylide

was formed from 5.430 g (15.8 mmol) of (methoxymethyl)-

triphenylphosphonium chloride, 9.3 mL (15.8 mmol) of tert-

butyllithium, and 75 mL of THF. Workup and chromatogra-

phy afforded 1.360 g (76%) of the desired product: ¹H NMR

(300 MHz, CDCl₃) δ 6.42 (s, 0.42 H), 6.40 (s, 0.58 H), 6.29 (d,

J ) 12.6 Hz, 0.58 H), 5.90 (d, J ) 6.3 Hz, 0.42 H), 4.74 (dt J

) 12.6 Hz, J ) 7.3 Hz, 0.58 H), 4.36 (dt, J ) 6.3 Hz, J ) 7.5

Hz, 0.42 H), 3.77 (s, 6 H), 3.58 (s, 1.26 H), 3.51 (s, 1.76 H),

2.57 (t, J ) 7.7 Hz, 2H), 2.12 (dt, J ) 6.7 Hz, J ) 7.8 Hz, 0.84

H), 1.97 (dt, J ) 7.2 Hz, J ) 7.2 Hz, 1.16 H), 1.58-1.71 (m, 2

H), 1.37 (s, 9 H); ¹³C NMR (75 MHz, CDCl₃) δ 176.6, 152.05,

152.00, 147.4, 146.5, 141.0, 140.7, 106.4, 105.2, 102.6, 59.6,

56.2, 56.1, 39.1, 36.1, 35.9, 32.5, 31.7, 27.3, 23.6; IR (neat,

NaCl) 2962, 2936, 2870, 2857, 2841, 1754, 1654, 1602, 1510,

1479, 1462, 1422, 1338, 1279, 1242, 1207, 1125, 1029, 936, 822,

757, 737 cm^-1; GCMS (PCI) m/ e (rel intensity) 221 (100), 253

(80), 71 (70), 85 (62), 337 (62), 239 (49), 103 (40), 222 (24), 87

(23), 167 (22), 252 (21), 305 (19), 365 (16), 86 (16); HRMS (EI)

m/ e calcd for C₁₉H₂₈O₅336.1937, found 336.1917.

An od ic Oxid a tion of 22a : 6,8-Dim eth oxy-7-(p iva loyl-

oxy)-1,2,3,4-tetr a h yd r on a p h th a len eca r boxa ld eh yd e Di-

m et h yl Acet a l (23a ) a n d 1,6,8-Tr im et h oxy-7-(p iva loyl-

oxy)-1,2,3,4-tetr a h yd r on a p h th a len e (24). Compound 22a

was oxidized in a manner similar to the constant current

oxidation of compound 4. In this experiment, 0.155 g (0.5

mmol) of 22a , 0.246 g (2.3 mmol) of 2,6-lutidine, 0.781 g (7.3

mmol) of lithium perchlorate, 3.6 mL of methanol, and 14.7

mL of dichloromethane were used. The reaction was oxidized

with a constant current of 12.1 mA until 90.3 C (2.04 faradays/

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

J . Org. Chem., Vol. 61, No. 5, 1996 1591

mol) of charge had been passed. Workup and chromatography

afforded an inseparable mixture of 71.0 mg (42%) of product

23a and the overoxidized material 24. The ratio of product to

overoxidized material was determined to be approximately

1.3:1 by integration of the acetal proton in 23a and the benzylic

1415, 1395, 1352, 1304, 1280, 1243, 1231, 1215, 1191, 1116,

1083, 1029, 952, 736 cm^-1; GCMS (PCI) m/ e (rel intensity) 57

(100), 63 (35), 95 (22), 335 (21), 97 (20), 303 (19), 334 (17), 251

(12), 85 (11), 219 (10), 91 (9), 51 (7); HRMS (EI) m/ e calcd for

C₂₀H₃₀O₅S 382.18138, found 382.18037.

1

methine proton of 24 in the H NMR of the crude product. The

4-(3′-F u r yl)-1-bu ta n ol (26). To a solution of 5.001 g (51.0

mmol) of 3-furanmethanol in 49.0 mL of THF at 0 °C was

added 4.832 g (17.8 mmol) of phosphorus tribromide. The

reaction was stirred until the starting alcohol disappeared as

monitored by thin layer chromatography. The reaction was

quenched with water, and the mixture was extracted with

ether. The ether layers were washed with saturated sodium

bicarbonate and brine, dried over MgSO₄, filtered, and con-

centrated in vacuo. The crude product was then taken up in

150 mL of THF and cooled to 0 °C. To this solution was added

102 mL (102 mmol) of a 1.0 M allylmagnesium bromide in

ether solution. The reaction mixture was allowed to warm to

room temperature and was stirred for 18 h. The reaction was

then quenched with 50 mL of 10.0 M sulfuric acid at 0 °C.

The crude reaction mixture was diluted with water and

extracted with ether. The combined organic extracts were

washed with saturated sodium bicarbonate and brine, dried

over MgSO₄, and filtered, and the solvent was removed at

atmospheric pressure. The crude olefin obtained in this

fashion was added to a solution of disiamylborane prepared

as follows: To a stirred solution of 10.2 mL (102 mmol) of

borane-dimethyl sulfide complex in 60 mL of THF at 0 °C

was added 14.300 g (204 mmol) of 2-methyl-2-butene. The

reaction mixture was stirred for 1 h and then the crude product

from above added. The resulting mixture was allowed to stir

for an additional 1 h, and then the reaction was quenched

carefully and slowly with 73 mL of 3 N sodium hydroxide

followed by 27 mL of 30% hydrogen peroxide that had been

cooled in an ice bath. The reaction was stirred for an

additional 1 h, diluted with water, and extracted with ether.

The combined organic layers were dried over MgSO₄, filtered,

and concentrated in vacuo. The crude product was loaded on

a silica gel column that was slurry-packed using a 20% ethyl

acetate in hexane solution that contained 1% triethylamine.

Elution with 35% ethyl acetate in hexane afforded 3.499 g

(49%) of the desired alcohol 26 over three steps starting from

3-furanmethanol: ¹H NMR (300 MHz, CDCl₃) δ 7.34 (d, 1H),

7.21 (s, 1H), 6.26 (s, 1H), 3.63 (t, J ) 6.2 Hz, 2H), 2.44 (t, J )

6.8 Hz, 2H), 2.38 (s, 1H), 1.57-1.65 (m, 4H); ¹³C NMR (300

MHz, CDCl₃) δ 142.7, 138.8, 125.0, 110.0, 62.5, 32.3, 26.2, 24.6;

IR (neat, NaCl) 3346 br, 3084, 2936, 2863, 1501, 1472, 1458,

1446, 1419, 1382, 1344, 1163, 1105, 1063, 992, 873, 779, 730

cm^-1; GCMS (PCI) m/ e (rel intensity) 13 (100), 81 (49), 124

(27), 141 (19), 95 (14), 109 (13), 151 (10), 82 (9), 122 (9), 121

(8), 121 (6), 125 (6), 163 (3); HRMS (EI) m/ e calcd for C₈H₁₂O₂

140.0837, found 140.0838.

products were identified by ¹H NMR in direct analogy to

products 5 and 7. In addition to the oxidation products, 17.5

mg (11%) of the starting material was recovered. The follow-

ing spectral data are for a 60:40 mixture of the two products.

Integrals are reported as 0.6 H for one proton in the major

product and 0.4 H for one proton in the minor product: ¹H

NMR (300 MHz, CDCl₃) δ 6.45 (s, 0.6 H), 6.44 (s, 0.4 H), 4.58

(d, J ) 3.8 Hz, 0.6 H), 4.48 (t, J ) 2.7 Hz, 0.4 H), 3.85 (s, 1.2

H), 3.78 (s, 1.8 H), 3.75 (s, 3 H), 3.43 (s, 1 H) 3.40 (s, 1.5 H),

3.23-3.27 (m, 0.6 H), 3.17 (s, 1.5 H), 2.58-2.82 (m, 2 H), 1.56-

2.27 (m, 4 H), 1.40 (s, 9H); ¹³C NMR (75 MHz, CDCl₃) δ 176.1,

151.9, 151.7, 151.1, 150.2, 137.3, 136.0, 131.1, 123.2, 121.9,

108.6, 107.7, 107.6, 107.5, 70.6, 69.0, 63.8, 61.8, 61.0, 56.75,

56.70, 56.65, 56.60, 56.1, 56.0, 55.95, 55.90, 53.4, 39.1, 39.0,

35.7, 29.7, 29.4, 27.3, 27.2, 26.2, 25.7, 21.7, 20.3, 17.2, 15.8;

IR (neat, NaCl) 2973, 2966, 2957, 2938, 2875, 2838, 1747, 1609,

1396, 1362, 1349, 1329, 1303, 1282, 1222, 1210, 1190, 1152,

1115, 1081, 973, 736 cm^-1; GCMS for the desired prodcut (PCI)

m/ e (rel intensity) 75 (100), 57(82), 335 (36), 251 (14), 303 (13),

334 (10), 85 (9), 336 (7), 219 (7), 333 (5); HRMS for the desired

product (EI) m/ e calcd for C₂₀H₃₀O₆366.2043, found 366.2025.

(E,Z)-5-(3,5-Dim eth oxy-4-(p iva loyloxy)p h en yl)-1-m eth -

oxy-1-p en ten e (22b). The reaction sequence was carried out

in a similar fashion to the procedure described above for the

synthesis of 4b. In this example, 1.640 g (5.3 mmol) of 4-(3,5-

dimethoxy-4-(pivaloyloxy)phenyl)-1-butanol, 0.804 g (6.3 mmol)

of oxalyl chloride, 0.536 g (6.9 mmol) of dimethyl sulfoxide,

2.672 g (26.4 mmol) of triethylamine, and 11 mL of dichlo-

romethane were used. After workup, the aldehyde was diluted

with 75 mL of THF and added to the ylide. The ylide was

formed from 5.430 g (15.8 mmol) of (methoxymethyl)triph-

enylphosphonium chloride, 9.3 mL (15.8 mmol) of tert-butyl-

lithium, and 75 mL of THF. Workup and chromatography

1

afforded 1.360 g (76%) of the desired product: H NMR (300

MHz, CDCl₃) δ 6.42 (s, 0.42 H), 6.40 (s, 0.58 H), 6.29 (d, J )

12.6 Hz, 0.58 H), 5.90 (d, J ) 6.3 Hz, 0.42 H), 4.74 (dt J )

12.6 Hz, J ) 7.3 Hz, 0.58 H), 4.36 (dt, J ) 6.3 Hz, J ) 7.5 Hz,

0.42 H), 3.77 (s, 6 H), 3.58 (s, 1.26 H), 3.51 (s, 1.76 H), 2.57 (t,

J ) 7.7 Hz, 2H), 2.12 (dt, J ) 6.7 Hz, J ) 7.8 Hz, 0.84 H), 1.97

(dt, J ) 7.2 Hz, J ) 7.2 Hz, 1.16 H), 1.58-1.71 (m, 2 H), 1.37

(s, 9 H); ¹³C NMR (75 MHz, CDCl₃) δ 176.6, 152.05, 152.00,

147.4, 146.5, 141.0, 140.7, 106.4, 105.2, 102.6, 59.6, 56.2, 56.1,

39.1, 36.1, 35.9, 32.5, 31.7, 27.3, 23.6; IR (neat, NaCl) 2962,

2936, 2870, 2857, 2841, 1754, 1654, 1602, 1510, 1479, 1462,

1422, 1338, 1279, 1242, 1207, 1125, 1029, 936, 822, 757, 737

cm^-1; GCMS (PCI) m/ e (rel intensity) 221 (100), 253 (80), 71

(70), 85 (62), 337 (62), 239 (49), 103 (40), 222 (24), 87 (23), 167

(22), 252 (21), 305 (19), 365 (16), 86 (16); HRMS (EI) m/ e calcd

for C₁₉H₂₈O₅336.19366, found 336.19171.

An od ic Oxid a tion of 22b: 6,8-Dim eth oxy-7-(p iva loyl-

oxy)-1,2,3,4-tetr a h yd r on a p h th a len eca r boxa ld eh yd e Di-

m eth yl Th ioa ceta l (23b). Compound 22b was oxidized in a

manner similar to the constant current oxidation of compound

4b and under a variety of conditions. In the experiment that

led to cyclization, 0.351 g (1.0 mmol) of 22b, 0.533 g (5.0 mmol)

of 2,6-lutidine, 1.685 g (15.8 mmol) of lithium perchlorate, 7.9

mL of methanol, and 31.7 mL of dichloromethane were used.

The oxidation was conducted with a constant current of 10.9

mA until 191.5 C (2.0 faradays/mol) of charge had been passed.

Workup and chromatography afforded 76.4 mg (20%) of the

desired product as a mixture of diastereomers. As mentioned

in the text, efforts to repeat this experiment have not been

successful and for the most part the cyclizations led to

methoxylation of the vinyl sulfide without cyclization in

analogy to the formation of product 19. 23b: ¹H NMR (300

MHz, CDCl₃) δ 6.45, (s, 1 H), 4.88 (d, J ) 3.0 Hz, 0.5 H), 4.72

(d, J ) 5.6 Hz, 0.5 H), 3.78 (s, 3 H), 3.76 (s, 3 H), 3.55-3.61

(m, 0.5 H), 3.41 (s, 1.5 H), 3.30-3.35 (m, 0.5 H), 3.19 (s, 1.5

H), 2.59-2.79 (m, 3 H), 1.37-2.09 (m, 15 H); IR (neat, NaCl)

2957, 2933, 2871, 1756, 1734, 1604, 1506, 1491, 1480, 1458,

(E,Z)-5-(3-F u r yl)-1-(m eth ylth io)-1-p en ten e (27a ). The

reaction sequence was carried out in a similar fashion to that

described above for the synthesis of 11b. In this example,

1.106 g (7.3 mmol) of 26, 1.104 (8.2 mmol) of oxalyl chloride,

0.736 g (9.4 mmol) of dimethyl sulfoxide, 3.669 g (36.3 mmol)

of triethylamine, and 21 mL of dichloromethane was used.

After workup the aldehyde was diluted with 100 mL of THF

and added to the ylide. The ylide was formed from 7.806 g

(21.8 mmol) of [(methylthio)methyl]triphenylphosphine chlo-

ride, 12.1 mL (21.8 mmol) of phenyllithium, and 20 mL of THF.

Workup and chromatography afforded 0.931 g (70%) of the

desired product: ¹H NMR (300 MHz, CDCl₃) δ 7.34, 7.33 (2 s,

1 H), 7.22 (s, 0.45 H), 7.20 (s, 0.55 H), 6.27 (d, J ) 1.1 Hz,

0.45 H), 6.25 (d, J ) 1.1 Hz, 0.55 H), 5.98 (d of multiplets, J

) 15.0, Hz, 0.55 H), 5.90 (dt, J ) 9.4 Hz, J ) 1.5 Hz, 0.45 H),

5.53 (dt, J ) 9.4 Hz, J ) 7.2 Hz, 0.45 H), 5.43 (dt, J ) 14.9

Hz, J ) 7.0 Hz, 0.55 H), 2.39-2.46 (m, 2 H), 2.26 (s, 1.35 H),

2.22 (s, 1.65 H), 2.09-2.19 (m, 2 H), 1.59-1.70 (m, 2 H); ¹³C

NMR (75 MHz, CDCl₃) δ 142.7, 142.6, 138.8, 128.2, 127.2,

126.7, 124.8, 124.7, 124.1, 111.2, 110.9, 32.6, 29.7, 29.3, 28.7,

24.3, 24.0, 17.0, 15.0; IR (neat, NaCl) 3132, 3005, 2986, 2980,

2921, 2857, 1609, 1560, 1501, 1455, 1437, 1382, 1351, 1310,

1244, 1163, 1065, 1027, 960, 938, 873, 779, 725, 703, 690 cm^-1

;

GCMS (PCI) m/ e (rel intensity) 135 (100), 61 (58), 183 (54),

136 (51), 81 (48), 107 (39), 211 (20), 163 (18), 134 (18), 100

1592 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

(16), 87 (15), 82 (12), 184 (11); HRMS (EI) m/ e calcd for C₁₀H₁₄

OS 182.07653, found 182.07434. Anal. Calcd for C₁₀H₁₄OS:

C, 65.89; H, 7.74. Found: C, 65.89; H, 7.72.

-

166.09949. Anal. Calcd for C₁₀H₁₄O₂: C, 72.26; H, 8.49.

Found: C, 71.95; H, 8.37.

An od ic Oxid a tion of Su bstr a te 27b: 4,5,6,7-Tetr a h y-

d r ob en zo[b]fu r a n -7-ca r b oxa ld eh yd e Dim et h yl Acet a l

(28b). Compound 27b was oxidized in a manner similar to

the constant current electrolysis of compound 4 described

above. In this experiment, 0.203 g (1.2 mmol) of 27b, 0.655 g

(6.1 mmol) of 2,6-lutidine, 2.077 g (19.5 mmol) of lithium

perchlorate, 9.8 mL of methanol, and 39.0 mL of dichlo-

romethane were used. The oxidation was conducted with a

constant current of 11.3 mA until 235.4 C (2.0 faradays/mol)

of charge had been passed. Upon completion of the oxidation,

1.160 g (6.1 mmol) of TsOH was added and the resulting

mixture stirred overnight. Workup and chromatography af-

forded 0.179 g (75%) of the desired product: ¹H NMR (300

MHz, CDCl₃) δ 7.29 (d, J ) 1.9 Hz, 1 H), 6.19 (d, J ) 1.8 Hz,

1 H), 4.46 (d, J ) 6.2 Hz, 1 H), 3.43 (s, 3 H), 3.36 (s, 3 H),

3.01-3.07 (m, 1 H), 2.35-2.48 (m, 2 H), 1.77-1.92 (m, 3 H),

1.60-1.72 (m, 1 H); ¹³C NMR (75 MHz, CDCl₃) δ 149.4, 141.1,

118.9, 110.3, 106.0, 54.6, 54.3, 37.3, 24.2, 22.2, 21.2; IR (neat,

NaCl) 2933, 2854, 2832, 1505, 1457, 1447, 1374, 1303, 1236,

1217, 1189, 1159, 1122, 1106, 1077, 1059, 991, 976, 966, 954,

892, 730, 691 cm^-1; GCMS (PCI) m/ e (rel intensity) 75 (100),

165 (95), 166 (12), 164 (11), 74 (9), 133 (7), 135 (6), 76 (5), 61

(5) 163 (5), 195 (4); HRMS (EI) m/ e calcd for C₁₁H₁₆O₃

196.10994, found 196.10805. Anal. Calcd for C₁₁H₁₆O₃: C,

67.31; H, 8.22. Found: C, 67.14; H, 8.20.

(E,Z)-5-(3′-Fu r yl)-1-ph en yl-1-pen ten e (27c). This styrene-

based substrate was prepared in a fashion directly analogous

to the synthesis of substrate 4a . In this experiment, a solution

of 1.982 g (14.2 mmol) of the alcohol (26), 1.220 g (15.6 mmol)

of dimethyl sulfoxide, 35.0 mL of dichloromethane, 2.163 g

(17.0 mmol) of oxalyl chloride, and 4.311 g (42.6 mmol) of

triethylamine were used in the Swern oxidation. The resulting

aldehyde was added to an ylide derived from 18.445 g (42.6

mmol) of benzyltriphenylphosphonium bromide in 84.0 mL of

THF and 32.8 mL (42.6 mmol) of a 1.3 M solution of sec-

butyllithium in cyclohexane. After workup and chromatog-

raphy, the reaction mixture afforded 1.600 g (53%) of the

desired product (27c) as a mixture of cis and trans olefins.

The spectral data for this mixture were as follows: ¹H NMR

(300 MHz, CDCl₃) δ 7.16-7.35 (m, 7H), 6.17-6.46 (m, 2.6 H),

5.64-5.69 (dt, J ) 7.3 Hz, J ) 11.7 Hz, 0.4 H), 2.33-2.49 (m,

3 H), 2.21-2.28 (m, 1 H), 1.6-1.78 (m, 2 H); ¹³C NMR (75 MHz,

CDCl₃) δ 142.7, 142.6, 138.8, 132.4, 130.4, 130.2, 129.2, 128.7,

128.5, 128.1, 126.8, 126.5, 125.9, 110.9, 32.5, 30.2, 29.5, 28.0,

24.3, 24.2; IR (neat, NaCl) 3059, 3027, 2925, 2859, 1621, 1500,

1446, 1168, 1070, 1030, 974, 890, 780, 760, cm^-1; GCMS (PCI)

m/ e (rel intensity) 131 (100), 213 (50), 132 (21), 121 (17), 82

(16), 214 (14), 159 (14), 169 (14), 107 (12), 83 (7); HRMS (EI)

m/ e calcd for C₁₅H₁₆O 212.1201, found 212.1206.

Electr olysis of Com p ou n d 27c: Syn th esis of 7-(1′-

Meth oxy-1′-p h en ylm eth yl)-4,5,6,7-tetr a h yd r oben zo[b]fu -

r a n (28c). A 100 mL round bottom flask equipped with a

reticulated vitreous carbon anode, a carbon rod cathode, and

a nitrogen inlet was charged with 0.181 g (0.85 mmol) of 27c,

0.455 g (4.2 mmol) of 2,6-lutidine, 1.064 g (10.0 mmol) lithium

perchlorate, 5.0 mL of methanol, and 20.0 mL of dichlo-

romethane. The mixture was degassed by sonication for 30

min and then oxidized at a constant current of 11.6 mA until

164.4 C (2.0 faradays/mol) of charge had been passed. Upon

completion of the oxidation, 1.522 g (8.0 mmol) of p-toluene-

sulfonic acid was added, and the solution was stirred for 30

min. The reaction was diluted with ether and washed with

water, saturated sodium bicarbonate, and brine. The organic

layer was dried over MgSO₄, filtered, and concentrated in

vacuo. The crude product was chromatographed through a

silica gel column that was slurry-packed using a 1% triethy-

lamine/hexane solution. The column was eluted with hexane

to afford 0.146 g (71%) of the desired product (28c) as a 1:1

mixture of diastereomers. The spectral data for the mixture

of diastereomers are as follows: ¹H NMR (300 MHz, CDCl₃) δ

7.18-7.34 (m, 6 H), 6.17 (d, J ) 1.8 Hz, 1 H), 4.64 (d, J ) 4.5

Hz, 0.5 H), 4.43 (d, J ) 6.8 Hz, 0.5 H), 3.27 (s, 1.5 H), 3.18-

3.22 (m, 2 H), 3.05 (m, 0.5 H), 2.36-2.46 (m, 1 H), 2.24-2.34

(m, 1 H), 1.67-1.87 (m, 1 H), 1.38-1.64 (m, 3 H); ¹³C NMR

An od ic Oxid a tion of (E,Z)-5-(3-F u r yl)-1-(m eth ylth io)-

1-p en ten e (27a ): Syn th esis of 4,5,6,7-Tetr a h yd r oben zo-

[b]fu r a n -7-ca r boxa ld eh yd e Dim eth yl Th ioa ceta l (28a )

a n d 4,5,6,7-Tetr a h yd r oben zo[b]fu r a n -7-ca r boxa ld eh yd e

Dim eth yl Acetal (28b). Solven t System A: 20% Meth an ol:

80% Dich lor om eth a n e. Compound 27a was oxidized in a

manner similar to the constant current electrolysis of com-

pound 4. In this experiment, 0.212 g (1.2 mmol) of 27a , 0.623

g (5.82 mmol) of 2,6-lutidine, 1.975 g (18.6 mmol) of lithium

perchlorate, 9.3 mL of methanol, and 37.1 mL of dichlo-

romethane were used. The oxidation was conducted with a

constant current of 11.3 mA until 224.0 C (2.0 faradays/mol)

of charge had been passed. For the optimal set of conditions,

1.107 g (5.8 mmol) of TsOH was added to the solution after

completion of the electrolysis. The mixture was stirred

overnight. Workup and chromatography as described above

afforded 0.132 g (54%) of the mixed acetal product 28a as a

1.2:1 mixture of diastereomers and 0.0391 g (17%) of the

dimethoxy acetal product 28b.

Solven t System B: 100% Meth a n ol. In a similar fashion,

compound 27a was oxidized in methanol solvent. In this

experiment, 0.203 g (1.1 mmol) of 27a , 0.597 g (5.57 mmol) of

2,6-lutidine, 1.890 g (17.8 mmol) of lithium perchlorate, and

44.4 mL of methanol were used. The oxidation was conducted

with a constant current of 11.3 mA until 220.7 C (2.06

faradays/mol) of charge had been passed. Upon completion

of the oxidation, 1.059 g (5.6 mmol) of TsOH was added and

the mixture stirred overnight. Workup and chromatography

afforded 0.159 g (67%) of the mixed acetal product 28a as an

approximately 1:1 mixture of diastereomers and 0.022 g (10%)

of the dimethoxy acetal product 28b: ¹H NMR (300 MHz,

CDCl₃) δ 7.29 (d, J ) 2.1 Hz, 1 H), 6.19-6.21 (m, 1 H), 4.61

(d, J ) 4.7 Hz, 0.37 H), 4.55 (d, J ) 6.7 Hz, 0.63 H), 3.46 (s,

1.89 H), 3.40 (s, 1.11 H), 3.15-3.25 (m, 1 H), 2.36-2.48 (m, 2

H), 1.60-2.20 (m, 7 H, includes 2.13 (s) and 2.07 (s)); ¹³C NMR

(75 MHz, CDCl₃) δ 150.2, 149.7, 141.0, 140.8, 119.0, 118.7,

110.4, 110.3, 91.5, 90.4, 56.8, 56.3, 40.3, 39.1, 25.9, 25.0, 22.25,

22.20, 21.5, 13.0, 12.2; IR (neat, NaCl) 2984, 2925, 2855, 2822,

1684, 1575, 1569, 1559, 1503, 1444, 1304, 1288, 1237, 1217,

1191, 1182, 1152, 1136, 1099, 1074, 1037, 1028, 961, 938, 730

cm^-1; GCMS (PCI) m/ e (rel intensity) 165 (100), 164 (94), 91

(62), 133 (45), 163 (31), 150 (28), 166 (27), 181 (24), 105 (12),

193 (11), 161 (8), 149 (8); HRMS (EI) m/ e calcd for C₁₁H₁₆O₂S

212.08709, found 212.08757. Anal. Calcd for C₁₁H₁₆O₂S: C,

62.24; H, 7.60. Found: C, 62.13; H, 7.59.

(E,Z)-5-(3-F u r yl)-1-m eth oxy-1-p en ten e (27b). The reac-

tion sequence was carried out in a similar fashion to that

described above for the synthesis of 4. In this example, 1.010

g (7.2 mmol) of 26, 1.098 g (8.7 mmol) of oxalyl chloride, 0.732

g (9.4 mmol) of dimethyl sulfoxide, 3.198 g (31.6 mmol) of

triethylamine, and 21 mL of dichloromethane were used. After

workup the aldehyde was diluted with 75 mL of THF and

added to the ylide. The ylide used was generated from 7.415

g (21.6 mmol) of (methoxymethyl)triphenylphosphonium chlo-

ride, 12.7 mL (21.6 mmol) of a 1.7 M solution of tert-

butyllithium in pentane, and 20 mL of THF. Workup and

chromatography afforded 0.618 g (52%) of the desired prod-

uct: ¹H NMR (300 MHz, CDCl₃) δ 7.34 (s, 1 H), 7.20 (s, 1 H),

6.26-6.31 (m, 1.78 H), 5.89 (d, J ) 6.2 Hz, 0.22 H), 4.72 (dt,

J ) 12.6 Hz, J ) 7.4 Hz, 0.78 H), 4.34 (dt, J ) 6.3 Hz, J ) 7.4

Hz, 0.22 H), 3.57 (s, 0.66 H), 3.50 (s, 2.34 H), 2.41 (t, J ) 7.6

Hz, 2 H), 2.11 (dt, J ) 7.3 H, J ) 7.5 Hz, 0.44 H), 1.96 (dt, J

) 7.6 Hz, 7.2 Hz, 1.56 H), 1.54-1.64 (m, 2 H); ¹³C NMR (75

MHz, CDCl₃) δ 147.4, 146.4, 142.7, 142.6, 138.8, 125.2, 125.0,

111.4, 111.15, 111.10, 106.3, 102.55, 102.50, 59.5, 55.9, 31.0,

30.2, 27.3, 24.3, 24.0, 23.6; IR (neat, NaCl) 3056, 3033, 3001,

2931, 2926, 2857, 2832, 1733, 1725, 1717, 1700, 1696, 1684,

1675, 1671, 1652, 1636, 1628, 1575, 1569, 1559, 1540, 1521,

1516, 1501, 1457, 1443, 1390, 1382, 1265, 1239, 1209, 1182,

1161, 1131, 1109, 1065, 1024, 934, 873, 782, 725 cm^-1; GCMS

(PCI) m/ e (rel intensity) 135 (100), 107 (76), 81 (45), 167 (26),

163 (24), 136 (20), 71 (19), 79 (13), 84 (13), 117 (12), 82 12), 83

(12); HRMS (EI) m/ e calcd for C₁₀H₁₄O₂166.09937, found

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

J . Org. Chem., Vol. 61, No. 5, 1996 1593

(75 MHz, CDCl₃) δ 140.8, 140.4, 128.0, 127.4, 127.2, 127.0,

110.3, 110.2, 85.5, 83.6, 57.3, 57.0, 41.6, 40.4, 24.9, 23.5, 22.2,

22.1, 21.8, 21.1; IR (neat, NaCl) 2920, 3051, 3020, 2925, 2857,

triethylamine in hexane solution. Elution of the column with

hexane afforded 0.079 g (49%) of the desired product 28d : ¹H

NMR (300 MHz, CDCl₃) δ 7.25 (bs, 1 H), 6.19 (bs, 1 H), 5.79-

5.91 (m, 1 H), 5.06-5.11 (m, 2 H), 3.34-3.43 (m, 1 H), 2.38-

2.43 (m, 2 H), 1.64-1.98 (m, 4 H); ¹³C NMR (75 MHz, CDCl₃)

δ 151.1, 140.8, 139.5, 117.4, 115.4, 110.3, 38.9, 30.1, 22.2, 21.1;

IR (neat, NaCl) 2927, 2922, 2852, 1646, 1503, 1465, 1446, 1437,

1150, 1105, 890, 726 cm^-1; GCMS (PCI) m/ e (rel intensity)

149 (100), 121 (19), 147 (15), 148 (14), 81 (4), 119 (4), 131 (3),

107 (3); HRMS (EI) m/ e calcd for C₁₀H₁₂O 148.0888, found

148.0876.

1496, 1443, 1165, 1153, 1104, 1064, 1024, 874, 780, 750 cm^-1

;

GCMS (PCI) m/ e (rel intensity) 121 (100), 122 (32), 211 (26),

165 (10), 212 (7), 107 (5), 91 (5), 123 (4), 149 (2), 75 (2); HRMS

(EI) m/ e calcd for C₁₆H₁₈O₂242.1307, found 242.1355.

(E,Z)-6-(3′-F u r yl)-1-(tr im eth ylsilyl)-2-h exen e (27d ). A

stirred solution of 1.254 g (8.9 mmol) of the alcohol (26), 0.765

g (9.8 mmol) of dimethyl sulfoxide, and 23.0 mL of dichlo-

romethane was treated with 1.356 g (10.7 mmol) of oxalyl

chloride at -78 °C. The reaction mixture was stirred for 15

min and the temperature then raised to between -50 and -60

°C. After an additional 15 min, 2.702 g (26.7 mmol) of

triethylamine was added and the reaction mixture was allowed

to warm to room temperature. The reaction was diluted with

ether and washed with water, saturated sodium bicarbonate,

and brine. The organic layer was then dried over MgSO₄,

filtered, and concentrated in vacuo. The resulting aldehyde

was added to the following silyl-containing ylide without

further purification. To a suspension of 9.538 g (26.7 mmol)

of methyltriphenylphosphonium bromide in 54 mL of terahy-

drofuran at 0 °C was added 10.7 mL (26.7 mmol) of a 2.5 M

n-butyllithium in hexane solution. The mixture was allowed

to warm to room temperature, allowed to stir for 1 h, and then

cooled to 0 °C. To this mixture was added dropwise 5.717 g

(26.7 mmol) of (iodomethyl)trimethylsilane. The mixture was

again allowed to warm to room temperature. After 1 h, the

reaction mixture was cooled to -78 °C and treated with an

additional 10.7 mL (26.7 mmol) of the 2.5 M solution of

n-butyllithium in hexane. The dark red solution was allowed

to warm to room temperature and stirred for 1.5 h. The

reaction was recooled to -78 °C, and then a solution of the

crude aldehyde prepared above in 25 mL of THF added with

the use of a cannula. The reaction mixture was allowed to

slowly warm to room temperature and then stirred for 17 h.

The reaction was quenched with saturated aqueous am-

monium chloride and the aqueous layer extracted with ether.

The combined organic layers were washed with saturated

sodium bicarbonate and brine, dried over MgSO₄, filtered, and

concentrated in vacuo. The crude reaction product was taken

up in 50 mL of hexane to precipitate most of the triph-

enylphosphine oxide. The hexane was removed in vacuo. The

crude product was eluted with hexane through a silica gel

column that was slurry-packed using a 1% triethylamine in

hexane solution to afford 0.790 g (40%) of the desired product

27d : ¹H NMR (300 MHz, CDCl₃) δ 7.33 (s, 1 H), 7.20 (s, 1 H),

6.26 (s, 1 H), 5.38-5.44 (m, 1 H), 5.27-5.32 (m, 1 H), 2.43 (t,

J ) 7.6 Hz, 2 H), 2.02-2.06 (m, 2 H), 1.58-1.66 (m, 2 H), 1.44-

1.57 (m, 2 H), 0.00-0.04 (s, 9 H); ¹³C NMR (75 MHz, CDCl₃)

δ 142.6, 138.8, 128.3, 126.9, 126.6, 125.9, 125.1, 110.9, 32.3,

30.3, 30.1, 26.6, 24.4, 24.1, 22.6, 18.5, -1.8, -1.9; IR (neat,

NaCl) 3008, 2954, 2873, 2858, 1642, 1503, 1457, 1247, 1152,

1027, 852, 841, 773 cm^-1; GCMS (PCI) m/ e (rel intensity) 73

(100), 207 (28), 133 (20), 74 (9), 208 (5), 223 (4), 131 (4), 134

(2), 91 (2); HRMS (EI) m/ e calcd for C₁₃H₂₂OSi 222.1440, found

222.1441.

(E ,Z)-6-(3′-F u r yl)-2-m e t h yl-1-(t r im e t h ylsilyl)-2-h e x-

en e (27e). This allylsilane was prepared in a fashion directly

analogous to the synthesis described above for 27d . In this

experiment, the desired aldehyde was made using 2.623 g (18.7

mmol) of the alcohol 26, 1.611 g (20.6 mmol) of dimethyl

sulfoxide, 47.0 mL of dichloromethane, 2.855 g (22.5 mmol) of

oxalyl chloride, and 5.689 g (56.2 mmol) of triethylamine. The

resulting aldehyde was added to the following silyl-containing

ylide without further purification. To a 0 °C suspension of

20.872 g (56.2 mmol) of ethyltriphenylphosphonium bromide

in 112 mL of terahydrofuran was added 22.5 mL (56.2 mmol)

of a 2.5 M n-butyllithium in hexane solution. The mixture

was raised to room temperature, allowed to stir for 1 h, and

then cooled to 0 °C. To this mixture was added dropwise

12.033 g (56.2 mmol) of (iodomethyl)trimethylsilane. The

mixture was again allowed to warm to room temperature.

After 1 h, the reaction mixture was cooled to -78 °C and

treated with additional 10.7 mL (26.7 mmol) of a 2.5 M

n-butyllithium in hexane solution. The dark red solution was

allowed to warm to room temperature and stirred for 1.5 h.

The solution was then recooled to -78 °C and a solution of

the crude aldehyde prepared above in 50 mL of THF added

by cannulation. The resulting mixture was allowed to slowly

warm to room temperature and stirred for 17 h. Workup and

chromatography as described above for previous Wittig reac-

tions afforded 2.255 g (51%) of the desired product 27e which

showed a small amount (5% by NMR) of an unidentified

impurity: ¹H NMR (300 MHz, CDCl₃) δ 7.3 (s, 1 H), 7.18 (s, 1

H), 6.25 (s, 1 H), 4.93-5.01 (m, 1 H), 2.40 (t, J ) 7.6 Hz, 2 H),

1.98-2.05 (m, 2.5 H), 1.48-1.67 (m, 6.5 H), 0.02-0.09 (s, 9

H); ¹³C NMR (75 MHz, CDCl₃) δ 142.5, 138.8, 138.7 133.3,

133.0, 125.2, 122.1, 121.7, 110.9, 30.5, 30.3, 29.8, 29.7, 28.1,

27.7, 26.2, 24.5, 24.3, 23.2, 18.6, -0.73, -1.25; IR (neat, NaCl)

3008, 2954, 2873, 2858, 1642, 1503, 1457, 1247, 1152, 1027,

852, 841, 773 cm^-1; GCMS (PCI) m/ e (rel intensity) 73 (100)

74 (18), 161 (9), 75 (8), 235 (7), 236 (2), 251 (2); HRMS (EI)

m/ e calcd for C₁₄H₂₄OSi 236.1596, found 236.1596.

Electr olysis of Com p ou n d 27e. Syn th esis of 7-(1′-

Meth yleth yliden e)-4,5,6,7-tetr ah ydr oben zo[b]fu r an (28e).

The anodic oxidation of 27e was performed in a similar fashion

to that described for the oxidation of 27c. In this experiment,

a 100 mL round-bottom flask equipped with a reticulated

vitreous carbon anode, a carbon rod cathode, and a nitrogen

inlet was charged with 0.234 g (1.0 mmol) of 27e, 0.536 g (5.0

mmol) of 2,6-lutidine, 1.064 g (10.0 mmol) of lithium perchlo-

rate, 5.0 mL of methanol, and 20.0 mL of dichloromethane.

This mixture was degassed by sonication for 30 min and then

oxidized at a constant current of 11.6 mA until 429.9 C (4.5

faradays/mol) of charge had been passed. Upon completion

of the oxidation, 1.536 g (8.0 mmol) of p-toluenesulfonic acid

was added to the mixture and the reaction mixture stirred for

30 min. Workup and chromatography afforded 0.082 g (51%)

of slightly impure (<5% impurity by NMR) 28e: ¹H NMR (300

MHz, CDCl₃) δ 7.25 (s, 1 H), 6.18 (s, 1 H), 4.88 (s, 1 H), 4.65

(s, 1 H), 3.40-3.45 (m, 1 H), 2.43-2.46 (m, 2 H), 1.61-1.92

(m, 7 H); ¹³C NMR (75 MHz, CDCl₃) δ 151.1, 146.3, 140.7,

118.2, 112.4, 110.1, 42.3, 28.9, 22.2, 21.3, 20.0; IR (neat, NaCl)

2923, 2921, 2853, 1647, 1504, 1464, 1457, 1447, 1436, 1374,

1149, 1105, 892, 725 cm^-1; GCMS (PCI) m/ e (rel intensity)

163 (100), 164 (26), 121 (25), 162 (21), 161 (20), 135 (9), 191

(7), 133 (7), 107 (6), 147 (5), 122 (5); HRMS (EI) m/ e calcd for

C₁₁H₁₄O, 162.1044, found 162.1048.

Electr olysis of Com p ou n d 27d . Syn th esis of 7-Eth -

ylid en e-4,5,6,7-tetr a h yd r oben zo[b]fu r a n (28d ). The con-

stant current electrolysis of 27d was performed in a similar

manner to that described above for the oxidation of 27c. In

this experiment, a 100 mL round-bottom flask equipped with

a reticulated vitreous carbon anode, a carbon rod cathode, and

a nitrogen inlet was charged with 0.242 g (1.1 mmol) of 27d ,

0.589 g (5.5 mmol) of 2,6-lutidine, 1.491 g (14.0 mmol) of

lithium perchlorate, 7.0 mL of methanol, and 28.0 mL of

dichloromethane. This mixture was degassed by sonication

for 30 min and then oxidized at a constant current of 11.6 mA

until 477.7 C (4.5 faradays/mol) of charge had been passed.

Upon completion of the oxidation, 1.674 g (8.8 mmol) of

p-toluenesulfonic acid was added and the mixture stirred for

30 min. The reaction was diluted with ether and washed with

water, saturated sodium bicarbonate, and brine. The organic

layer was then dried over MgSO₄, filtered, and concentrated

in vacuo. The crude product was chromatographed through

(E,Z)-6-(3′-F u r yl)-2-m eth yl-2-h exen e (27f). The trisub-

stituted olefin substrate was prepared in an analogous fashion

to the synthesis of substate 4 described above. In this

a

silica gel column that was slurry-packed using a 1%

1594 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

experiment, 1.989 g (14.2 mmol) of the alcohol (26), 1.220 g

(15.6 mmol) of dimethyl sulfoxide, 36 mL of dichloromethane,

2.163 g (20.0 mmol) of oxalyl chloride, and 4.311 g (42.6 mmol)

of triethylamine were used in the Swern oxidation. The

resulting aldehyde was added without further purification to

the ylide generated from 18.416 g (42.6 mmol) of isopropylt-

riphenylphosphonium iodide in 84 mL of THF and 32.8 mL

(42.6 mmol) of a 1.3 M solution of sec-butyllithium in cyclo-

hexane. Workup and purification (chromatography followed

by Kugelrohr distillation) afforded 1.118 g (48%) of the desired

product 27f without any triphenylphosphine oxide: ¹H NMR

(300 MHz, CDCl₃) δ 7.33 (s, 1 H), 7.19 (s, 1 H), 6.26 (s, 1 H),

5.14 (m, 1 H), 2.38-2.43 (t, J ) 7.7 Hz, 2 H), 2.00-2.05 (m, 2

H), 1.98 (s, 3 H), 1.56-1.70 (m, 5 H); ¹³C NMR (75 MHz, CDCl₃)

δ 142.6, 138.8, 131.6, 125.1, 124.2, 110.9, 30.1, 27.6, 25.7, 24.3,

17.7; IR (neat, NaCl) 2965, 2929, 2858, 1605, 1502, 1447, 1377,

1163, 1065, 1025, 873, 776 cm^-1; GCMS (PCI) m/ e (rel

intensity) 199 (100), 81 (100), 165 (55), 82 (45), 147 (35), 109

(33), 83 (27), 107 (23), 95 (23); HRMS (EI) m/ e calcd for

C₁₁H₁₆O 164.11201, found 164.1210.

Electr olysis of Com p ou n d 27f. Syn th esis of 7-(1′-

Meth yleth ylid en e)-4,5,6,7-tetr a h yd r oben zo[b]fu r a n (28e)

a n d 7-(1′-Meth oxy-1′-m eth yleth yl)-4,5,6,7-tetr a h yd r oben -

zo[b]fu r a n (28f). The anodic oxidation of 27f was performed

in a similar manner to that of described for the electrolysis of

substrate 27c. In this experiment, a 100 mL round-bottom

flask equipped with a reticulated vitreous carbon anode, a

carbon rod cathode, and a nitrogen inlet was charged with

0.191 g (1.2 mmol) of 22, 0.624 g (5.8 mmol) of 2,6-lutidine,

1.064 g (10.0 mmol) of lithium perchlorate, 5.0 mL of methanol,

and 20.0 mL of dichloromethane. This mixture was degassed

by sonication for 30 min and then oxidized at a constant

current of 11.6 mA until 223.9 C (2 faradays/mol) of charge

had been passed. Upon completion of the oxidation, 1.765 g

(9.3 mmol) of p-toluenesulfonic acid was added and the mixture

stirred for 30 min. Workup and chromatography then afforded

0.0961 g (51%) of 28e and 0.0541 g (24%) of 28f. The spectral

data for compound 28e were the same as those reported above.

The spectral data for compound 28f are as follows: ¹H NMR

(300 MHz, CDCl₃) δ 7.26 (d, J ) 1.6 Hz, 1 H), 6.18 (d, J ) 1.7

Hz, 1 H), 3.26 (s, 3 H), 2.98-3.00 (m, 1 H), 2.38-2.43 (m, 2

H), 1.91-1.94 (m, 2 H), 1.51-1.68 (m, 2 H), 1.34 (s, 3 H), 1.09

(s, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ 150.9, 140.4, 119.3, 110.2,

49.0, 42.3, 25.6, 24.3, 22.8, 22.6, 22.4; IR (neat, NaCl) 2973,

2942, 2937, 2933, 2928, 2853, 1506, 1457, 1380, 1366, 1216,

1140, 1131, 1104, 1080, 1068, 893, 720 cm^-1; GCMS (PCI) m/ e

(rel intensity) 73 (100), 163 (91), 164 (19), 74 (14), 161 (4), 179

(4), 121 (4), 193 (3), 122 (2), 165 (2), 123 (2); HRMS (EI) m/ e

calcd for C₁₂H₁₈O₂194.1307, found 194.1372.

bined organic extracts were washed with saturated sodium

bicarbonate and brine, dried over MgSO₄, and filtered. The

solvent was removed by distillation at atmospheric pressure.

The crude olefin obtained was added to disiamylborane

without further purification.

To a stirred solution of 10.0 mL (100 mmol) of borane-

dimethyl sulfide complex in 50.0 mL of THF at 0 °C was added

14.028 g (200 mmol) of 2-methyl-2-butene. The reaction

mixture was stirred for 1 h and then the olefin prepared above

added. The reaction mixture was stirred for an additional 1

h and then the reaction quenched carefully with 71.4 mL of 3

N NaOH followed by 26.0 mL of 30% hydrogen peroxide. The

mixture was stirred for an additional 1 h, diluted with

water, and extracted with ether. The combined organic layers

were dried over MgSO₄, filtered, and concentrated in vacuo.

Chromatography through a silica gel column that was slurry-

packed using a 20% ethyl acetate in hexane solution containing

1% triethylamine using 30% ethyl acetate in hexane as eluent

afforded 4.850 g (63%) of the desired alcohol over the three

steps from 3-furanmethanol: ¹H NMR (300 MHz, CDCl₃) δ

7.32 (s, 1 H), 7.19 (s, 1 H), 6.24 (s, 1 H), 3.57 (t, J ) 6.7 Hz, 2

H), 3.12 (s, 1 H), 2.40 (t, J ) 7.5 Hz, 2 H), 1.51-1.61 (m, 4 H),

1.34-1.42 (m, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 142.4, 138.5,

121.9, 110.8, 52.2, 32.2, 29.6, 25.2, 24.5; IR (neat, NaCl) 3352

br, 2932, 2858, 1501, 1464, 1350, 1323, 1160, 1060, 1027, 870,

777 cm^-1; GCMS (PCI) m/ e (rel intensity) 81 (100), 137 (86),

155 (29), 109 (21), 82 (15), 119 (14), 95 (9), 138 (8), 93 (6), 71

(5), 55 (4); HRMS (EI) m/ e calcd for C₉H₁₄O₂154.0994, found

154.0979.

(E,Z)-6-(3′-F u r yl)-1-m eth oxy-1-h exen e (27g). This enol

ether was prepared in an analogous fashion to substrate 4a .

In this experiment, 3.251 g (21.1 mmol) of 5-(3′-furyl)-1-

pentanol, 0.842 g (10.8 mmol) of dimethyl sulfoxide, 53.0 mL

of dichloromethane, 1.492 g (11.8 mmol) of oxalyl chloride, and

6.405 g (63.3 mmol) of triethylamine were used in the Swern

oxidation. The resulting aldehyde was added to the ylide

generated from 21.699 g (63.3 mmol) of (methoxymethyl)-

triphenylphosphonium chloride in 126 mL of THF and 48.7

mL (63.3 mmol) of a 1.3 M solution of sec-butyllithium in

cyclohexane. Workup and chromatography afforded 2.051 g

(54%) of the desired product 27g as a mixture of cis and trans

olefin isomers: ¹H NMR (300 MHz, CDCl₃) δ 7.31 (s, 1 H),

7.19 (s, 1 H), 6.24-6.29 (m, 1.7 H), 5.84 (d, J ) 4.4 Hz, 0.3 H),

4.69 (dt, J _d) 9.6 Hz, J _t) 12.3 Hz, 0.7 H), 4.31 (dt, J _t) 6.5

Hz, J _d) 7.3 Hz, 0.3 H), 3.53 (s, 0.8 H), 3.46 (s, 2.2 H), 2.39 (t,

J ) 7.5 Hz, 2 H), 2.07-2.09 (m, 0.5 H), 1.89-1.91 (m, 1.5 H),

1.53-1.58 (m, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 147.0, 146.1,

142.5, 138.6, 125.1, 125.0, 110.9, 106.5, 102.6, 59.2, 55.6, 30.3,

29.4, 29.3, 27.4, 24.5, 23.5; IR (neat, NaCl) 2929, 2856, 1655,

1501, 1462, 1209, 1162, 1130, 1109, 1025, 873, 778 cm^-1

;

5-(3′-F u r yl)-1-p en ta n ol. Into a flame-dried 500 mL three-

necked flask equipped with a condenser, an addition funnel,

and a nitrogen inlet was added 2.552 g (105.0 mmol) of

magnesium powder, 100 mL of THF, and half of the total

required amount of 13.501 g (100.0 mmol) of 4-bromo-1-butene.

The reaction was stirred vigorously until the Grignard forma-

tion was initiated. The other half of the bromo compound was

diluted with 150 mL of THF and added slowly to the reaction

mixture over a period of 4 h. In a separate flask, 3-furylmethyl

bromide was prepared by treating 4.905 g (50.0 mmol) of

3-furanmethanol with 4.737 g (17.5 mmol) of phosphorus

tribromide in 50.0 mL of THF at 0 °C. The reaction was

stopped when no more of the alcohol was present as evidenced

by thin layer chromatography. The reaction was quenched

with water and the mixture extracted with ether. The ether

layer was washed with saturated sodium bicarbonate and

brine, dried over MgSO₄, filtered, and concentrated in vacuo.

The crude product was diluted with 100 mL of THF and was

placed in the addition funnel used above during the generation

of the Grignard reagent. The Grignard reagent solution

prepared above was cooled to a range of -30 to -50 °C and

then treated with 1.904 g (10.0 mmol) of anhydrous copper

iodide. The cold temperature was maintained and the bromide

added from the addition funnel over a 2 h period. The reaction

was allowed to warm to room temperature and stirred for 18

h. The reaction was quenched with 10 M sulfuric acid at 0

°C, diluted with water, and extracted with ether. The com-

GCMS (PCI) m/ e (rel intensity) 149 (100), 81 (54), 121 (40),

150 (26), 181 (20), 109 (10), 82 (10), 107 (9), 148 (9), 147 (81),

122 (7); HRMS (EI) m/ e calcd for C₁₁H₁₆O₂180.1150, found

180.1163.

E lect r olysis of com p ou n d 27g. Syn t h esis of 8-(1′,1′-

Dim eth oxym eth yl)cycloh epta[b]fu r an (28g). Solven t Sys-

tem A: 20% Meth a n ol:80% Dich lor om eth a n e. The con-

stant current electrolysis of 27g was performed in a fashion

that was directly analogous to the electrolysis of 27c described

above. In this experiment, a 100 mL round-bottom flask

equipped with a reticulated vitreous carbon anode, a carbon

rod cathode, and a nitrogen inlet was charged with 0.110 g

(0.61 mmol) of 27g, 0.328 g (3.1 mmol) of 2,6-lutidine, 1.064 g

(10.0 mmol) of lithium perchlorate, 5.0 mL of methanol, and

20.0 mL of dichloromethane. This mixture was degassed by

sonication for 30 min and then oxidized at a constant current

of 11.6 mA until 177.7 C (2 faradays/mol) of charge had been

passed. Upon completion of the oxidation, 0.928 g (4.9 mmol)

of p-toluenesulfonic acid was added to the mixture and the

reaction mixture was stirred for 30 min. Following workup

and chromatography, 0.0794 g (62%) of the desired product

28g was obtained.

Solven t System B: 100% Meth a n ol. In this experiment,

a 100 mL round-bottom flask equipped with a reticulated

vitreous carbon anode, a carbon rod cathode, and a nitrogen

inlet was charged with 0.110 g (0.61 mmol) of 27g, 0.328 g (5

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

J . Org. Chem., Vol. 61, No. 5, 1996 1595

mmol) of 2,6-lutidine, 1.064 g (10.0 mmol) of lithium perchlo-

rate, and 25.0 mL of methanol. This mixture was degassed

by sonication for 30 min and then oxidized at a constant

current of 11.6 mA until 177.7 C (2 faradays/mol) of charge

had been passed. Upon completion of the oxidation, 0.928 g

(8 mmol) of p-toluenesulfonic acid was added to the mixture

and the reaction mixture stirred for 30 min. Following workup

and chromatography, the reaction mixture afforded 0.074 g

(58%) of the desired product 28g: ¹H NMR (300 MHz, CDCl₃)

δ 7.16 (d, J ) 1.8 Hz, 1 H), 6.11 (d, J ) 1.8 Hz, 1 H), 4.56 (d,

J ) 7.6 Hz, 1 H), 3.35 (s, 3 H), 3.28 (s, 3 H), 3.25 (m, 1 H),

2.46-2.49 (m, 2 H), 1.41-2.1.82 (m, 6 H); ¹³C NMR (75 MHz,

CDCl₃) δ 151.8, 139.2, 121.5, 113.0, 104.4, 54.3, 53.2, 41.3, 28.4,

26.8, 25.6; IR (neat, NaCl) 2933, 2854, 1505, 1457, 1447, 1374,

3.23 (1.5 H), 3.11 (br s, 2.5 H), 2.47-2.53 (m, 2.35 H), 2.15-

2.83 (m, 0.65 H), 1.73-1.84 (m, 2 H), 1.41-1.70 (m, 3 H); ¹³C

NMR (75 MHz, CDCl₃) δ 140.9, 140.2, 139.1, 138.7, 128.2,

127.9, 127.6, 127.5, 127.3, 127.1, 126.7, 121.5, 120.7, 113.0,

112.8, 84.9, 82.7, 56.9, 56.8, 46.4, 45.2, 28.3, 28.3, 27.2, 26.8,

26.1, 25.7, 25.5; IR (neat, NaCl) 3052, 2934, 2850, 1490, 1446,

1102, 1056, 1026, 873, 776, 693 cm^-1; GCMS (PCI) m/ e (rel

intensity) 121 (100), 135 (39), 122 (27), 225 (25), 179 (4), 136

(2); HRMS (EI) m/ e calcd for C₁₆H₁₈O (M⁺- CH₂O) 226.1357,

found 226.1358.

(E,Z)-7-(3′-F u r yl)-2-m eth yl-2-h ep ten e (27i). Substrate

27i was prepared using chemistry that was directly analogous

to the synthesis of substrate 4a . In this experiment, 1.989 g

(8.2 mmol) of 5-(3′-furyl)-1-pentanol, 0.705 g (9.0 mmol) of

dimethyl sulfoxide, 21.0 mL of dichloromethane, 1.249 g (9.8

mmol) of oxalyl chloride, and 2.489 g (24.6 mmol) of triethy-

lamine were used for the Swern oxidation. The resulting

aldehyde was added to an ylide generated from 10.634 g (24.6

mmol) of isopropyltriphenylphosphonium iodide in 49.0 mL of

THF and 18.9 mL (24.6 mmol) of a 1.3 M solution of sec-

butyllithium in cyclohexane. When the reaction was complete,

workup and purification (chromatography and Kulgerohr

distillation) afforded 0.844 g (58%) of the desired product 27i:

¹H NMR (300 MHz, CDCl₃) δ 7.32 (s, 1 H), 7.19 (s, 1 H), 6.25

(s, 1 H), 5.08-5.14 (m, 1 H), 2.4 (t, J ) 7.41 Hz, 2 H), 1.96-

2.03 (m, 2 H), 1.68 (s, 3 H), 1.50-1.59 (m, 5 H), 1.34-1.41 (m,

2 H); ¹³C NMR (75 MHz, CDCl₃) δ 142.6, 138.7, 131.4, 125.2,

124.6, 110.0, 29.7, 29.5, 25.7, 24.7, 17.6; IR (neat, NaCl) 2962,

2930, 2920, 2857, 1631, 1502, 1448, 1378, 1162, 1064, 1027,

873, 775 cm^-1; GCMS (PCI) m/ e (rel intensity) 179 (100), 81

(16), 123 (12), 161 (11), 82 (8), 109 (5), 178 (0.5); HRMS (EI)

m/ e calcd for C₁₂H₁₈O 178.1358, found 178.1360.

1303, 1233, 1218, 1156, 1108, 1068, 986, 953, 890, 730 cm^-1

;

GCMS (PCI) m/ e (rel intensity) 179 (100), 75 (87), 147 (45),

180 (40), 76 (38), 177 (19), 74 (14), 121 (13), 209 (8), 178 (7),

77 (5); HRMS (EI) m/ e calcd for C₁₁H₁₅O₂(M⁺- OCH₃)

179.1072, found 179.1064.

(E,Z)-6-(3′-Fu r yl)-1-ph en yl-1-h exen e (27h ). This styrene-

based substrate was prepared in a fashion analogous to the

synthesis of substrate 4a . In this experiment, the Swern

oxidation was performed with 1.234 g (8.0 mmol) of 5-(3′-furyl)-

1-pentanol, 0.688 g (8.8 mmol) of dimethyl sulfoxide, 20.0 mL

of dichloromethane, 1.218 g (9.6 mmol) of oxalyl chloride, and

2.429 g (24.0 mmol) of triethylamine. The resulting aldehyde

was added to a ylide generated from 10.392 g (24.0 mmol) of

benzyltriphenylphosphonium bromide in 48.0 mL of THF and

18.5 mL (24.0 mmol) of a 1.3 M solution of sec-butyllithium in

cyclohexane. After the reaction was complete, workup and

chromatography as before afforded 1.010 g (56%) of the desired

product 27h as a mixture of cis and trans olefin isomers. The

spectral data for the mixture of isomers are as follows: ¹H

NMR (300 MHz, CDCl₃) δ 7.10-7.49 (m, 7 H), 6.41 (d, J )

11.7 Hz, 1 H), 6.23 (s, 1 H), 5.64 (dt, J ) 7.3 Hz, J ) 11.7, 1

H), 2.30-2.41 (m, 4 H), 1.45-1.62 (m, 4 H); ¹³C NMR (75 MHz,

CDCl₃) δ 142.6, 138.7, 132.8, 128.9, 128.7, 128.6, 128.4, 128.1,

127.6, 126.4, 124.9, 110.9, 29.5, 29.4, 28.3, 24.5; IR (neat, NaCl)

3092, 3015, 2940, 2856, 1615, 1503, 1491, 1443, 1162, 1062,

1025, 962, 872, 778, 745 cm^-1; GCMS (PCI) m/ e (rel intensity)

91 (100), 227 (87); HRMS (EI) m/ e calcd for C₁₆H₁₈O 226.1358,

found 226.1352.

Electr olysis of Com p ou n d 27h . Syn th esis of 8-(1′-

Meth oxy-1′-ph en ylm eth yl)cycloh epta[b]fu r an (28h ). Sol-

ven t System A: 20% Meth a n ol:80% Dich lor om eth a n e.

The anodic oxidation of 27h was performed in a manner that

was directly analogous to the electrolysis of 27c described

above. In this experiment, a 100 mL round-bottom flask

equipped with a reticulated vitreous carbon anode, a carbon

rod cathode, and a nitrogen inlet was charged with 0.217 g

(0.96 mmol) of 27h , 0.514 g (4.8 mmol) of 2,6-lutidine, 1.064 g

(10.0 mmol) of lithium perchlorate, 5.0 mL of methanol, and

20.0 mL of dichloromethane. This mixture was degassed by

sonication for 30 min and then oxidized at a constant current

of 11.6 mA until 185.3 C (2 faradays/mol) of charge had been

passed. Upon completion of the oxidation, 1.461 g (7.7 mmol)

of p-toluenesulfonic acid was added to the reaction and the

mixture then stirred for 30 min. Workup and chromatography

as described earlier afforded 0.143 g (58%) of the desired

product (28h ) as a 1:1 mixture of diastereomers.

Electr olysis of Com p ou n d 27i. Syn th esis of 8-(1′-

Meth yleth ylid en e)cycloh ep ta [b]fu r a n (28i), 7-(2′-Meth -

oxy-2′-p r op yl)cycloh ep ta [b]fu r a n (28ii), a n d 7-[3′-(2′,5′-

Dim eth oxy-2′,5′-d ih ydr ofu r n yl)]-2-m eth yl-2-h epten e (29).

The anodic oxidation of 27i was performed in a fashion that

was directly analogous to the electrolysis of 27c described

above. In this experiment, a 100 mL round-bottom flask

equipped with a reticulated vitreous carbon anode, a carbon

rod cathode, and a nitrogen inlet was charged with 0.192 g

(1.1 mmol) of 27i, 0.589 g (5.5 mmol) of 2,6-lutidine, 1.064 g

(10.0 mmol) of lithium perchlorate, 5.0 mL of methanol, and

20.0 mL of dichloromethane. This mixture was degassed by

sonication for 30 min and then oxidized at a constant current

of 11.6 mA until 212.3 C (2 faradays/mol) of charge had been

passed. Upon completion of the oxidation, 1.674 g (98.8 mmol)

of p-toluenesulfonic acid was added to the mixture and the

reaction mixture stirred for 30 min. Workup and purification

then afforded 0.050 g (26%) of 28i, 0.0014 g (6%) of 28ii, and

0.062 g (25%) of 29. Another reaction was carried out under

the same set of electrolysis conditions except that a carbon

rod was employed as an anode in place of the RVC anode used

above. This experiment led to the formation of 0.142 g (54%)

of 29. The spectral data for compound 28i are as follows: ¹H

NMR (300 MHz, CDCl₃) δ 7.18 (d, J ) 1.7 Hz), 6.14 (d, J )

1.7 Hz, 1 H), 4.87 (s, 1 H), 4.42 (s, 1 H), 3.55-3.57 (m, 1 H),

2.44-2.49 (m, 2 H), 1.96-1.99 (m, 1 H), 1.55-1.86 (m, 8 H);

¹³C NMR (75 MHz, CDCl₃) δ 145.2, 139.2, 121.4, 112.4, 112.9,

112.4, 46.6, 30.2, 28.5, 25.6, 21.2; IR (neat, NaCl) 2927, 2920,

2852, 1647, 1503, 1465, 1447, 1435, 1374, 1150, 1106, 893, 724

cm^-1; GCMS (PCI) m/ e (rel intensity) 177 (100), 135 (30), 178

(23), 121 (22), 175 (21), 175 (20), 81 (14), 95 (11), 109 (10), 149

(9), 57 (8); HRMS (EI) m/ e calcd for C₁₂H₁₆O 176.1201, found

176.1206. The spectral data for compound 28ii are as fol-

lows: ¹H NMR (300 MHz, CDCl₃) δ 7.17 (d, J ) 1.7 Hz, 1 H),

6.13 (d, J ) 1.7 Hz), 3.21 (s, 3 H), 3.02-3.04 (m, 1 H), 2.53-

2.65 (m, 1.44 H), 2.38-2.45 (m, 1.36 H), 1.70-2.01 (m, 5.20

H), 1.27 (s, 3 H), 1.17 (s, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ

152.6, 138.7, 121.6, 112.9, 48.9, 47.0, 27.7, 26.5, 26.0, 24.6, 24.3,

23.6; IR (neat, NaCl) 2973, 2972, 2941, 2930, 2854, 1507, 1380,

Solven t System B: 100% Meth a n ol. In this experiment,

a 100 mL round-bottom flask equipped with a reticulated

vitreous carbon anode, a carbon rod cathode, and a nitrogen

inlet was charged with 0.238 g (0.098 mmol) of 27h , 0.523 g

(4.9 mmol) of 2,6-lutidine, 1.064 g (10.0 mmol) of lithium

perchlorate, and 25.0 mL of methanol. This mixture was

degassed by sonication for 30 min and then oxidized at a

constant current of 11.6 mA until 189.1 C (2 faradays/mol) of

charge had been passed. Upon completion of the oxidation,

1.491 g (7.8 mmol) of p-toluenesulfonic acid was added to the

reaction and the mixture then stirred for 30 min. Workup and

chromatography afforded 0.108 g (43%) of the desired product

28h as a 1:1 mixture of diastereomers. The spectral data for

compound 28h are as follows: ¹H NMR (300 MHz, CDCl₃) δ

7.22-7.33 (m, 5 H), 7.09 (m, 1 H), 6.84 (s, 0.5 H), 6.15 (s, 0.5

H), 4.45 (d, J ) 8.4 Hz, 0.5 H), 4.41 (d, J ) 8.8 Hz, 0.5 H),

1367, 1216, 1164, 1153, 1130, 1080, 1064, 1025, 893, 721 cm^-1

GCMS (PCI) m/ e (rel intensity) 73 (100), 177 (32), 74 (14),

178 (8), 135 (3), 121 (2), 193 (1); HRMS (EI) m/ e calcd C₁₂H₁₆

;

O

[M - CH₃OH], 176.1200, found 176.1205. The spectral data

for compound 29 are as follows: ¹H NMR (300 MHz, CDCl₃) δ

1596 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

5.82 (d, J ) 1.1 Hz, 0.5 H), 5.80 (d, J ) 2.6 Hz, 0.5 H), 5.69 (s,

1 H), 5.53 (s, 0.5 H), 5.42 (s, 0.5 H), 5.09-5.10 (m, 1 H), 3.36-

3.41 (m, 6 H), 2.09-2.14 (m, 2 H), 1.95-2.00 (m, 2 H), 1.68 (s,

3 H), 1.60 (s, 3 H), 1.49-1.57 (m, 2 H), 1.35-1.42 (m, 2 H);

¹³C NMR (75 MHz, CDCl₃) δ 146.1, 145.6, 131.5, 124.2, 123.8,

123.8, 123.5, 108.9, 107.6, 106.8, 54.1, 54.0, 53.5, 53.5, 29.5,

29.4, 27.6, 26.7, 26.4, 26.3, 25.6, 17.6; IR (neat, NaCl) 2956,

2880, 2830, 1615, 1480, 1386, 1180, 1230, 1125, 1110, 1010,

1003,980, 875 cm^-1; LRFAB 209.2 [M + H - CH₃OH]; HRFAB

Calcd for C₁₃H₂₁O₂[M + H - CH₃OH] 209.1546, found

209.1555.

flask equipped with a reticulated vitreous carbon anode, a

carbon rod cathode, and a nitrogen inlet was charged with

0.092 g (0.47 mmol) of 27j, 0.252 g (2.35 mmol) of 2,6-lutidine,

3.293 g (10.0 mmol) of tetrabutylammonium tetrafluoroborate,

5.0 mL of methanol, and 20.0 mL of dichloromethane. The

mixture was degassed by sonication for 30 min and then

oxidized at a constant current of 11.6 mA until 117.6 C (2.6

faradays/mol) of charge had been passed. Upon completion

of the oxidation, 0.723 g (3.8 mmol) of p-toluenesulfonic acid

was added to the mixture and the reaction stirred for 30 min.

Workup and chromatography as described above afforded

0.152 g (32%) of 28j. Compound 28j was contamintated with

an inseparable side product. In addition, 0.018 g (16%) of 30

was obtained. The spectral data for compound 28j are as

follows: ¹H NMR (300 MHz, CDCl₃) δ 7.26 (d, J ) 1.7 Hz, 1

H), 6.42 (d, J ) 1.7 Hz, 1 H), 4.80 (d, J ) 7.4 Hz, 1 H), 3.40 (s,

3 H), 3.37 (s, 3 H), 3.29 (m, 1 H), 2.43-2.48 (m, 2 H), 1.72-

1.82 (m, 4 H), 1.49-1.61 (m, 4 H); ¹³C NMR (75 MHz, CDCl₃)

δ 151.4, 140.0, 138.6, 113.2, 112.0, 105.5, 54.1, 53.9, 39.6, 30.5,

29.5, 26.3, 25.3, 24.7; IR (neat, NaCl) 2951, 2863, 2840, 1507,

6-(3′-F u r yl)-1-h exa n ol. The desired alcohol was synthe-

sized in a manner similar to the preparation of compound 26

described above. In this experiment, the Grignard reagent was

prepared from 2.551 g (105.0 mmol) of magnesium powder,

110 mL of THF, and 14.904 g (100 mmol) of 5-bromo-1-pentene.

As described above, 3-furylmethyl bromide was prepared using

4.905 g (50 mmol) of 3-furanmethanol, 4.738 g (17.5 mmol) of

phosphorus tribromide, and 53.0 mL of THF. The crude

3-furylmethyl bromide was diluted with 150 mL of THF and

then added to a -30 to -50 °C solution of the Grignard reagent

and 1.904 g (10.0 mmol) of anhydrous copper iodide over a

period of 3 h. The reaction mixture was then allowed to warm

to room temperature and stirred for 18 h. Workup as

described earlier afforded the desired olefin which was then

used crude in the hydroboration reaction described below.

To a stirred solution of 10.0 mL (100 mmol) of borane-

dimethyl sulfide complex in 50 mL of THF at 0 °C was added

14.027 g (200 mmol) of 2-methyl-2-butene. The reaction

mixture was stirred for 1 h and the olefin prepared above

added. The reaction mixture was stirred for an additional 1

h. At that time, the reaction was quenched carefully with 71.4

mL of a 3 N solution of sodium hydroxide followed by 26.5 mL

of a 30% aqueous hydrogen peroxide solution. Column chro-

matography through a silica gel column that was slurry-packed

using a 25% ethyl acetate in hexane solution containing 1%

triethylamine using 25% ethyl acetate in hexane as the eluant

afforded 4.032 g (48%) of the desired alcohol over three steps

from 3-furanmethanol: ¹H NMR (300 MHz, CDCl₃) δ 7.33 (d,

J ) 1.8 Hz, 1 H), 7.19 (s, 1 H), 3.59 (t, J ) 6.6 Hz, 2 H), 2.51

(s, 1 H), 2.40 (t, J ) 7.6 Hz, 2 H), 1.50-1.58 (m, 4 H), 1.34-

1.37 (m, 4 H); ¹³C NMR (75 MHz, CDCl₃) δ 142.4, 138.5, 125.0,

110.9, 62.5, 32.5, 29.8, 28.9, 25.4, 24.5; IR (neat, NaCl) 3300

br, 2940, 2860, 1510, 1492, 1370, 1280, 1200, 1080, 1050, 890,

810 cm^-1; GCMS (PCI) m/ e (rel intensity) 82 (100), 95 (24),

168 (20), 67 (8); HRMS (EI) m/ e calcd for C₁₀H₁₆O₂168.1150,

found 168.1153.

1462, 1443, 1320, 1234, 1160, 1102, 1060, 970, 940, 880 cm^-1

;

GCMS (PCI) m/ e (rel intensity) 75 (100), 193 (40), 76 (13),

194 (9), 74 (3), 223 (2), 195 (1); HRMS (EI) m/ e calcd for

C₁₃H₂₀O₃[M⁺- CH₃OH], 192.1150, found 192.1148. The

spectral data for compound 30 are as follows: ¹H NMR (300

MHz, CDCl₃) δ 7.34 (s, 1 H), 7.20 (s, 1 H), 6.26 (s, 1 H), 4.20

(d, J ) 5.5 Hz, 1 H), 3.46 (s, 6 H), 3.42 (s, 3 H), 3.12-3.18 (m,

1 H), 2.41 (t, J ) 7.4 Hz, 2 H), 1.34-1.59 (m, 8 H); ¹³C NMR

(75 MHz, CDCl₃) δ 142.4, 138.4, 124.5, 105.4, 103.2, 81.2, 58.6,

55.2, 54.6, 29.5, 29.2, 28.6, 25.4, 24.8; IR (neat, NaCl) 2940,

2852, 1508, 1425, 1275, 1233, 1187, 1131, 1080, 1050, 1025,

1020, 1015, 1008, 97, 843 cm^-1; GCMS (PCI) m/ e (rel intensity)

193 (100), 75 (23), 194 (22), 161 (15), 121 (6), 191 (6), 133 (6),

221 (4), 162 (4), 195 (3); HRMS (EI) m/ e calcd for C₁₄H₂₄O₄

[M⁺- CH₃OH] 224.1412, found 224.1409.

5-(3-F u r yl)-2-p en ta n ol. A stirred solution of 0.906 g (11.6

mmol) of 26, 0.997 g (12.8 mmol) of dimethyl sulfoxide, and

30.0 mL of dichloromethane was cooled to -78 °C. After 15

min, 1.767 g (13.9 mmol) of oxalyl chloride was added and the

reaction mixture was allowed to warm to a temperature range

of -50 to -60 °C. After 20 min, the reaction was quenched

with 3.521 g (34.8 mmol) of triethylamine and mixture brought

to room temperature. After being diluted with ether, the crude

aldehyde was washed with water, saturated sodium bicarbon-

ate, and brine. The organic layer was dried over MgSO₄,

filtered, and concentrated in vacuo. The concentrated crude

aldehyde was placed in a round-bottom flask with 50 mL of

THF and cooled to -78 °C. To this mixture was slowly added

11.6 mL (34.8 mmol) of a 3.0 M solution of methylmagnesium

bromide in ether. The reaction mixture was stirred for 15 h

and then the reaction quenched with 26.0 mL of 10 M sulfuric

acid at 0 °C. This mixture was diluted with water and

extracted with ether. The combined ether layers were washed

with saturated sodium bicarbonate and brine, dried over

MgSO₄, filtered, and concentrated at reduced pressure. The

crude material was chromatographed through a silica gel

column that was slurry-packed using a 1% triethylamine in

hexane solution. The column was eluted with a 15% ethyl

acetate in hexane solution to afford 0.625 g (35%) of the desired

product and 0.372 g (23%) of the recovered starting alcohol.

The spectral data for the desired alcohol are as follows: ¹H

NMR (300 MHz, CDCl₃) δ 7.34 (s, 1 H), 7.21 (s, 1 H), 6.27 (s,

1 H), 3.78-3.84 (m, 1 H), 2.41-2.46 (t, J ) 7.6 Hz, 2 H), 1.41-

1.72 (m, 5 H), 1.18-1.2 (d, J ) 9.2 Hz, 3 H); ¹³C NMR (75

MHz, CDCl₃) δ 142.7, 138.8, 124.9, 110.9, 67.9, 38.8, 26.1, 24.7,

23.5; IR (neat, NaCl) 3430 br, 2984, 2956, 2881, 1508, 1450,

1390, 1180, 1150, 1108, 1105, 960, 800 cm^-1; LRMS (EI) m/ e

(rel intensity) 94 (100), 95 (27), 154 (23), 107 (9), 121 (8), 136

(6); HRMS (EI) m/ e calcd for C₉H₁₄O₂154.0994, found

154.0996.

(E,Z)-7-(3′-F u r yl)-1-m eth oxy-1-h ep ten e (27j). This enol

ether substrate was prepared in a fashion directly analogous

to the preparation of substrate 4a . In this experiment, 2.245

g (13.4 mmol) of 6-(3′-furyl)-1-hexanol, 1.152 g (14.7 mmol) of

dimethyl sulfoxide, 34.0 mL of dichloromethane, 2.041 g (16.1

mmol) of oxalyl chloride, and 4.068 g (63.3 mmol) of triethy-

lamine were used in the Swern oxidation. The resulting

aldehyde was added to a ylide generated from 13.781 g (40.2

mmol) of (methoxymethyl)triphenylphosphonium chloride in

80.4 mL of THF and 31.0 mL (40.2 mmol) of a 1.3 M solution

of sec-butyllithium in cyclohexane. When the reaction was

complete, workup and chromatography afforded 1.326 g (51%)

of the desired product 27j as a mixture of cis and trans

isomers: ¹H NMR (300 MHz, CDCl₃) δ 7.32 (d, J ) 1.6 Hz, 1

H), 7.19 (s, 1 H), 6.25-6.29 (m, 1.6 H), 5.85 (d, J ) 6.3 Hz, 0.4

H), 4.71 (dt, J _t) 7.3 Hz, J _d) 12.6 Hz, 0.6 H), 4.32 (dt, J _d

)

6.4 Hz, J _t) 7.3 Hz, 0.4 H), 3.55 (s, 1.2 H), 3.48 (s, 1.8 H), 2.39

(t, J ) 7.6 Hz, 2 H), 2.04-2.07 (m, 0.6 H), 1.90-1.92 (m, 1.4

H), 1.52-1.56 (m, 2 H), 1.33-1.36 (m, 4 H); ¹³C NMR (75 MHz,

CDCl₃) δ 146.9, 145.9, 142.5, 142.4, 138.6, 125.2, 125.1, 110.9,

106.8, 102.8, 59.3, 55.7, 30.5, 29.7, 29.5, 28.8, 28.5, 27.5, 24.6,

23.7; IR (neat, NaCl) 2940, 2865, 1655, 1506, 1480, 1350, 1210,

1158, 1120, 1030, 880, 770 cm^-1; GCMS (PCI) m/ e (rel

intensity) 163 (100), 81 (91), 71 (50), 121 (29), 195 (38), 135

(28), 95 (26), 109 (23), 164 (21), 82 (18), 145 (13); HRMS (EI)

m/ e calcd for C₁₂H₁₈O₂194.1306, found 194.1323.

(E,Z)-5-(3′-F u r yl)-1-m eth oxy-2-m eth yl-1-p en ten e (31).

This enol ether substrate was prepared in a fashion that was

directly analogous to the synthesis used for preparing sub-

strate 4. In this experiment, the Swern oxidation utilized

1.272 g (7.1 mmol) of 5-(3-furyl)-2-pentanol, 0.609 g (7.8 mmol)

of dimethyl sulfoxide, 18.0 mL of dichloromethane, 1.081 g (8.5

Electr olysis of Com p ou n d 27j. Syn th esis of 9-(1′,1′-

Dim eth oxym eth yl)cycloocta [b]fu r a n (28j) a n d 7-(3′-F u -

r yl)-1,1,2-tr im eth oxyh ep ta n e (30). A 100 mL round-bottom

Anodic Olefin Coupling and Use of Electron-Rich Aryl Rings

J . Org. Chem., Vol. 61, No. 5, 1996 1597

mmol) of oxalyl chloride, and 2.155 g (21.3 mmol) of triethy-

lamine. The resulting aldehyde was added to the ylide

prepared from 7.302 g (21.3 mmol) of (methoxymethyl)-

triphenylphosphonium chloride in 43.0 mL of THF and 16.4

mL (21.3 mmol) of a 1.3 M solution of sec-butyllithium in

cyclohexane. After the reaction was complete, workup and

chromatography afforded 0.741 g (58%) of 31 as a mixture of

cis and trans olefin isomers: ¹H NMR (300 MHz, CDCl₃) δ

7.34 (s, 1 H), 7.21 (s, 1 H), 6.27 (s, 1 H), 5.77 (s, 1 H), 3.54 (s,

1.3 H), 3.52 (s, 1.7 H), 2.35-2.42 (m, 2 H), 2.11 (t, 1.3 H), 1.89

(t, 0.7 H), 1.89-1.58 (m, 5 H); ¹³C NMR (75 MHz, CDCl₃) δ

142.6, 142.5, 142.5, 142.0, 141.9, 138.7, 125.2, 113.8, 111.0,

110.9, 59.1, 33.4, 28.5, 28.1, 27.8, 245.4, 17.1, 12.6; IR (neat,

NaCl) 2957, 2874, 1648, 1505, 1483, 1348, 1215, 1161, 1130,

1035, 1020, 984, 886, 770 cm^-1; HRMS (EI) m/ e calcd for

C₁₁H₁₆O₂180.1150, found, 180.1150.

hexane to 25% ether in hexane by adding an additional 2% of

ether for every 100 mL of eluent to afford 0.380 g (71%) of the

desired product 34: ¹H NMR (300 MHz, CDCl₃) δ 6.68 (t, J )

2.4 Hz, 1 H), 6.52 (s, 1 H), 6.15 (s, 1 H), 4.57-4.59 (m, 1 H),

3.72-3.91 (m, 2 H), 3.37-3.53 (m, 2 H), 1.21-1.88 (m, 13 H),

1.10 (s, 9 H), 1.07 (s, 9 H); ¹³C NMR (75 MHz, CDCl₃) δ 126.1,

124.0, 121.1, 110.7, 98.8, 67.6, 62.3, 30.9, 29.6, 27.8, 26.9, 25.6,

19.7, 18.0, 17.9, 17.8, 11.8; IR (neat, NaCl) 2943, 2928, 2894,

2867, 1479, 1464, 1384, 1367, 1352, 1270, 1200, 1185, 1157,

1136, 1119, 1101, 1076, 1033, 1021, 994, 972, 920, 906, 691,

658 cm^-1; GCMS (PCI) m/ e (rel intensity) 296 (100), 122 (51),

85 (40), 252 (37), 297 (33), 295 (29), 278 (27), 294 (23), 236

(15), 324 (13), 253 (12), 279 (10); HRMS (EI) m/ e calcd for

C₂₂H₄₁NO₂Si 379.2906, found 379.2907.

4-(N-P iva loyl-3-p yr r olyl)-1-bu ta n ol (35). A round-bot-

tom flask was charged with 1.838 g (4.8 mmol) of 34, 40 mL

of THF, and 9.7 mL (9.7 mmol) of tetrabutylammonium

fluoride (1.0 M in THF). After 30 min, the reaction was

checked by TLC at which time the reaction was complete. The

reaction was quenched with H₂O, the mixture was diluted with

ether, and the layers were separated. The combined organic

layers were then dried over MgSO₄and concentrated in vacuo.

The resulting oil was chromatographed through 75 mL of silica

gel with 10 g of activity 3 neutral alumina on top. A gradient

elution from hexane to 30% ether in hexane was performed

by adding an additional 3% of ether for every 100 mL of eluent

in order to afford 1.127 g (>100%) of the desired product. This

product was utilized directly for the next step without further

purification. In a sample procedure, 1.080 g (4.8 mmol) of the

product was diluted with 20 mL of THF, the mixture cooled

to 0 °C, and 8.1 mL (14.5 mmol) of phenyllithium added. The

reaction mixture was stirred for 15 min, and then 1.605 g (13.3

mmol) of trimethylacetyl chloride was added. The reaction

mixture was allowed to warm to room temperature and stirred

for 12 h, and then the reaction was quenched with H₂O. The

crude mixture was stirred for an additional 10 min and then

extracted with ether. The combined ether layers were washed

with brine, dried over MgSO₄, and then concentrated in vacuo.

The resulting oil was chromatographed through 75 mL of silica

gel using a gradient elution from hexane to 30% ether in

hexane by adding an additional 3% of ether for every 100 mL

of eluent used in order to afford 1.281 g (86% over two steps)

of the desired product. This product was then carried directly

on to the next step, where in a 25 mL 14/20 round-bottom flask

the 1.281 g (4.2 mmol) of diprotected pyrrole was diluted with

50 mL of ethanol, cooled to 0 °C, and treated with 0.198 g (1.0

mmol) of p-toluenesulfonic acid. The reaction mixture was

stirred overnight and then diluted with ether and water. The

layers were separated, and the aqueous layer was extracted

with ether. The combined ether layers were then dried over

MgSO₄and concentrated in vacuo. The resulting oil was

chromatographed through 75 mL of silica gel with a gradient

elution from hexane to 30% ether/hexane by adding an

additional 3% of ether for every 100 mL of eluent used to afford

E lect r olysis of Com p ou n d 31. Syn t h esis of 7-(1′,1′-

Dim eth oxym eth yl)-7-m eth yl-4,5,6,7-tetr a h yd r oben zo[b]-

fu r a n (32). To a 100 mL round-bottom flask equipped with

a reticulated vitreous carbon anode, a carbon rod cathode, and

a nitrogen inlet were added 0.193 g (0.92 mmol) of 31, 0.494

g (2.4 mmol) of 2,6-lutidine, 1.064 g (10.0 mmol) of lithium

perchlorate, 5.0 mL of methanol, and 20.0 mL of dichlo-

romethane. This mixture was degassed by sonication for 30

min and then oxidized at a constant curent of 11.6 mA until

177.5 C (2 faradays/mol) of charge had been passed. Upon

completion of the oxidation, 1.401 g (7.4 mmol) of p-toluene-

sulfonic acid was added. The mixture was stirred for 30 min.

Workup and chromatography as described earlier afforded

0.100 g (52%) of the desired product 32: ¹H NMR (300 MHz,

CDCl₃) δ 7.26 (s, 1 H), 6.17 (s, 1 H), 4.32 (s, 1 H), 3.45 (s, 3 H),

3.34 (s, 3 H), 2.38-2.43 (m, 2 H), 1.97-2.07 (m, 1 H), 1.78-

1.94 (m, 1 H), 1.63-1.74 (m, 1 H), 1.45-1.54 (m, 1 H), 1.24 (s,

1 H); ¹³C NMR (75 MHz, CDCl₃) δ 153.6, 140.5, 117.7, 111.2,

110.2, 57.9, 57.7, 40.9, 31.2, 22.4, 21.5, 20.1; IR (neat, NaCl)

2943, 2852, 2835, 1506, 1454, 1446, 1374, 1301, 1240, 1231,

1185, 1121, 1105, 1074, 1054, 992, 954, 889, 730, 689 cm^-1

;

GCMS (PCI) m/ e (rel intensity) 179 (100), 75 (55), 147 (8), 61

(7), 74 (6), 145 (4), 203(3); HRMS (EI) m/ e calcd for C₁₂H₁₈O₃

210.1256, found 210.1255.

4-(N-(Tr iisop r op ylsilyl)-3-p yr r olyl)-1-(2′-tetr a h yd r op y-

r a n oxy)bu ta n ol (34). To a round-bottom flask was added

0.771 g (2.6 mmol) of 3-bromo-1-triisopropylpyrrole¹⁴in 2 mL

of THF. The solution was cooled to -78 °C and after 15 min

1.6 mL (2.6 mmol) of a 1.6 M n-butyllithium in hexane solution

added. The mixture was then stirred for 30 min and then the

THP ether of 4-hydroxybutanal added. The aldehyde was

synthesized from the monoprotected 1,4-butanediol by a Swern

oxidation in a fashion directly analogous to the synthesis of

compound 4 described above. This experiment was done using

0.871 g (5.1 mmol) of the alcohol, 0.777 g (6.1 mmol) of oxalyl

chloride, 0.518 g (6.1 mmol) of dimethyl sulfoxide, 2.581 g (25.5

mmol) of triethylamine, and 10 mL of dichloromethane. After

addition of the aldehyde to the anion generated from the

bromide, the reaction mixture was slowly warmed to room

temperature and stirred for 16 h. The reaction was then

quenched with H₂O and the mixture extracted with ether. The

combined ether layers were dried over MgSO₄and concen-

trated in vacuo. The crude product was chromatographed

through 100 mL of silica gel using a gradient elution from

hexane to 25% ether in hexane by adding an additional 2% of

ether for every 100 mL of eluent to afford 0.510 g (51%) of the

pure, alkylated product. For convenience, the alcohol on the

carbon next to the pyrrole ring was immediately removed by

diluting the purified product with 10 mL of THF, cooling the

mixture to 0 °C, and then adding 1 mL (10 mmol) of borane-

dimethyl sulfide. The reaction mixture was allowed to warm

to room temperature and monitored by TLC. After 3 h, the

reaction mixture still showed starting material so the reaction

mixture was placed in warm tap water. After an additional

30 min, the reaction was complete. The reaction was then

quenched with H₂O, ether was added, and the layers were

seperated. The aqueous layer was extracted with ether, and

the combined ether layers were dried over MgSO₄and con-

centrated in vacuo. The resulting oil was chromatographed

through 50 mL of silica gel using a gradient elution from

1

0.791 g (85%) of the desired product 35: H NMR (300 MHz,

CDCl₃) δ 7.35 (t, J ) ca. 3 Hz, 1 H), 7.19 (s, 1 H), 6.12 (dd, J

) ca. 3 Hz, J ) 1.6 Hz, 1 H), 3.66 (t, J ) 6.2 Hz, 2 H), 2.44-

2.48 (m, 2 H), 1.92 (s, 1 H), 1.57-1.68 (m, 4 H), 1.44 (s, 9 H);

¹³C NMR (75 MHz, CDCl₃) δ 175.6, 127.6, 120.7, 117.7, 113.0,

62.7, 40.4, 32.4, 28.6, 26.6, 26.3; IR (neat, NaCl) 3310 br, 2975,

2936, 2875, 2863, 1701, 1478, 1461, 1406, 1371, 1341, 1310,

1165, 1070, 1038, 908, 811, 784, 756 cm^-1; GCMS (PCI) m/ e

(rel intensity) 224 (100), 57 (89), 122 (48), 206 (24), 140 (23),

85 (19), 225 (17), 58 (15), 223 (11), 120 (9), 252 (8); HRMS

(EI) m/ e calcd for C₁₃H₂₁NO₂223.1572, found 223.1574.

(E,Z)-5-(N-P iva loyl-3-p yr r olyl)-1-(m et h ylt h io)-1-p en -

ten e (36). The reaction sequence was carried out in a similar

fashion to that described above for the synthesis of 11b. In

this example, 0.505 g (2.3 mmol) of 35, 0.345 g (2.7 mmol) of

oxalyl chloride, 0.230 g (2.9 mmol) of dimethyl sulfoxide, 1.144

g (11.3 mmol) of triethylamine, and 5 mL of dichloromethane

were used. After workup the aldehyde was diluted with 10

mL of THF and added to the ylide. The ylide was formed from

2.0278 g (5.7 mmol) of [(methylthio)methyl]triphenylphos-

phonium chloride, 3.14 mL (5.65 mmol) of a 1.8 M phenyl-

lithium in 70/30 cyclohexane/ether solution, and 20 mL of THF.

1598 J . Org. Chem., Vol. 61, No. 5, 1996

New et al.

Workup and chromatography afforded 0.382 g (64%) of the

desired product: ¹H NMR (300 MHz, CD₃COCD₃) δ 7.41 (br

s, 1 H), 7.28 (br s, 1 H), 6.15 (br s, 1 H), 6.03 (d, J ) 14.9 Hz,

0.6 H), 5.96 (dt, J ) 9.4 Hz, J ) 1.3 Hz, 0.4 H), 5.54 (dt, J )

9.3 Hz, J ) 7.2 Hz, 0.4 H), 5.44 (dt, J ) 14.9 Hz, J ) 7.0 Hz,

0.6 H), 2.41-2.47 (m, 2 H), 2.23 (s, 1.2 H), 2.19 (s, 1.8 H), 2.14-

2.19 (m, 2 H), 1.60-1.71 (m, 2 H), 1.41 (s, 9 H); ¹³C NMR (75

MHz, CD₃COCD₃) δ 175.7, 128.4, 128.2, 128.0, 127.0, 125.1,

121.4, 121.3, 118.05, 118.00, 113.4, 40.9, 33.3, 30.7, 29.3, 28.6,

26.9, 26.7, 16.7, 14.7; IR (neat, NaCl) 2977, 2931, 2922, 2877,

2856, 1700, 1661, 1477, 1466, 1438, 1405, 1370, 1354, 1309,

1165, 1073, 1038, 938, 811, 793, 755, 703 cm^-1; GCMS (PCI)

m/ e (rel intensity) 57 (100), 218 (86), 134 (69), 266 (39), 165

(19), 294 (18), 219 (14), 182 (12), 58 (10), 87 (10), 135 (8); HRMS

(EI) m/ e calcd for C₁₅H₂₃NOS 265.1500, found 265.1442. Anal.

Calcd for C₁₅H₂₃NOS: C, 67.89; H, 8.74; N, 5.28. Found: C,

67.65; H, 8.74; N, 5.09.

1-N-P iva loyl-4,5,6,7-t et r a h yd r oin d ole-7-ca r b oxa ld e-

h yd e Dim eth yl Th ioa ceta l (37). Compound 36 was oxidized

in a manner similar to the constant current oxidation of

compound 4. In this experiment, 0.196 g (0.7 mmol) of 36,

0.375 g (3.5 mmol) of 2,6-lutidine, 1.277 g (12.0 mmol) of

lithium perchlorate, 5.6 mL of methanol, and 22.4 mL of

dichloromethane were used. A constant current of 11.7 mA

until 143.0 C (2.0 faradays/mol) of charge had been passed.

Workup and chromatography afforded 0.145 g (66%) of the

desired product as a mixture of diastereomers: ¹H NMR (300

MHz, CD₃COCD₃) δ 7.35 (d, J ) 3.4 Hz, 0.31 H), 7.24 (d, J )

3.4 Hz, 0.69 H), 5.98 (d, J ) 3.5 Hz, 0.31 H), 5.93 (d, J ) 3.4

H, 0.69 H), 4.44 (d, J ) 5.5 Hz, 0.31 H), 4.20 (d, J ) 8.9 Hz,

0.69 H), 3.95-3.97 (m, 0.69 H), 3.81-3.85 (m, 0.31 H), 3.26

(s, 0.93 H), 3.25 (s, 2.07 H), 2.42-2.46 (m, 2 H), 2.14-2.18 (m,

1 H), 2.03 (s, 0.93 H), 1.96 (s, 2.07 H), 1.65-1.76 (m, 3 H),

1.43 (s, 2.79 H), 1.40 (s, 6.21 H); ¹³C NMR (75 MHz, CD₃-

COCD₃) δ 180, 130.8, 122.7, 121.5, 121.1, 111.4, 110.2, 93.3,

91.8, 56.9, 56.6, 41.9, 38.8, 36.5, 29.0, 28.9, 28.6, 26.1, 25.4,

23.6, 20.1, 19.7, 13.3, 9.8; IR (neat, NaCl) 3100, 2975, 2930,

2869, 2843, 2822, 1708, 1487, 1479, 1461, 1444, 1428, 1398,

1368, 1356, 1294, 1275, 1244, 1217, 1191, 1175, 1141, 1095,

1074, 940, 915, 865, 789, 755, 716, 696 cm^-1; GCMS (PCI) m/ e

(rel intensity) 248 (100), 247 (71), 164 (56), 216 (38), 264 (26),

246 (24), 249 (21), 85 (20), 95 (18), 276 (17), 56 (17), 97 (16),

103 (14); HRMS (EI) m/ e calcd for C₁₆H₂₅NO₂S 295.1606,

found 295.1620.

Ack n ow led gm en t. We thank the National Science

Foundation (CHE-9023698) and Washington University

for their generous support of this work. DGN thanks

the Department of Education for a GAANN fellowship,

and Z.T. thanks Washington University for a Wheeler

Fellowship. We also gratefully acknowledge the Wash-

ington University High Resolution NMR Facility, par-

tially supported by NIH grants RR02004, RR05018, and

RR07155, and the Washington University Mass Spec-

trometry Resource Center, partially supported by NI-

HRR00954, for their assistance.

Su p p or tin g In for m a tion Ava ila ble: Proton and carbon

NMR data for all new compounds (127 pages). This material

is contained in libraries on microfiche, immediately follows this

article in the microfilm version of the journal, and can be

ordered from the ACS; see any current masthead page for

ordering information.

J O9518359

Article Doi

Intramolecular anodic olefin coupling reactions and the use of electron-rich aryl rings

DOI: 10.1021/jo9518359

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jo9518359

Full text of DOI:10.1021/jo9518359

Products guided by the article

R&D Labs maybe for 112468-44-7

Relevant to this article

Hot Product