Asymmetric reactions of 2-methoxy-1,4-benzoquinones with styrenyl systems: Enantioselective syntheses of 8-aryl-3-methoxybicyclo[4.2.0]oct-3- en-2,5-diones, 7-Aryl-3-hydroxybicyclo[3.2.1]oct-3-en-2,8-diones, 2-Aryl-6- methoxy-2,3-dihydrobenzofuran-5-ols, and pterocarpans

DOI: 10.1021/jo982164s

Source and publish data:

Journal of Organic Chemistry p. 2391 - 2405 (1999)

Update date:2022-07-31

Topics:

Authors:

Engler Engler

Letavic Letavic

Iyengar Iyengar

LaTessa LaTessa

Reddy Reddy

Read Full Text PDF DownLoad Join now for total 90,000,000 free articles

Article abstract of DOI:10.1021/jo982164s

Reactions of 2-methoxy-1,4-benzoquinones 2 and 3 with (E)- propenylbenzenes 1 promoted at -78 °C by Ti(IV)-TADDOLates prepared from diol-(+)-4 afford (1R,6R,7R,8R)-8-aryl-3-methoxy-7-methylbicyclo[4.2.0]oct- 3-en-2,5-diones 5/8 or (1R,5R,6R,7R)-7-aryl-3-hydroxy-6-methylbicyclo[3.2.1]- oct-3-en-2,8-diones 7/10 in good yield and high ee. (2S,3S)-2-Aryl-6-methoxy- 3-methyl-2,3-dihydrobenzofuran-5-ols 6/9 are also found, but in slightly lower ee. Cyclobutanes 5/8 cleanly and efficiently rearrange to the dihydrobenzofurans 6/9 without loss of enantiomeric purity upon treatment with the Ti-TADDOLates at higher temperatures. Reactions of (Z)- propenylbenzene 17 and of indene with 2 and 3 give products in moderate enantiomeric purity. Products obtained from reactions of 1-anisylcycloalkenes with 2 differ significantly in yield and enantiomeric purity. In the latter reactions, the ee's of the cyclobutane products are consistently much higher than those of the dihydrobenzofuran products. More significantly, products of different absolute configuration result from different cycloalkenes. With 1- anisylcyclopentene or 1-anisylcyclohexene, all of the products are of similar configuration and are obtained in comparable yields and ee's. However, 1- anisylcycloheptene affords products that are diastereomeric with those of the 1-anisylcyclopentene, and in lower ee's. A mechanistic model is proposed. Application of these reactions to the enantioselective synthesis of the pterocarpan class of isoflavonoid natural products is also reported.

Full text of DOI:10.1021/jo982164s

J . Org. Chem. 1999, 64, 2391-2405

2391

Asym m etr ic Rea ction s of 2-Meth oxy-1,4-ben zoqu in on es w ith

Styr en yl System s: En a n tioselective Syn th eses of

8-Ar yl-3-m eth oxybicyclo[4.2.0]oct-3-en -2,5-d ion es,

7-Ar yl-3-h yd r oxybicyclo[3.2.1]oct-3-en -2,8-d ion es,

2-Ar yl-6-m eth oxy-2,3-d ih yd r oben zofu r a n -5-ols, a n d P ter oca r p a n s

Thomas A. Engler,*^,†Michael A. Letavic, Rajesh Iyengar, Kenneth O. LaTessa,^‡and

J ayachandra P. Reddy

Department of Chemistry, University of Kansas, Lawrence, Kansas 66045

Received October 27, 1998

Reactions of 2-methoxy-1,4-benzoquinones 2 and 3 with (E)-propenylbenzenes 1 promoted at -78

°C by Ti(IV)-TADDOLates prepared from diol-(+)-4 afford (1R,6R,7R,8R)-8-aryl-3-methoxy-7-

methylbicyclo[4.2.0]oct-3-en-2,5-diones 5/8 or (1R,5R,6R,7R)-7-aryl-3-hydroxy-6-methylbicyclo[3.2.1]-

oct-3-en-2,8-diones 7/10 in good yield and high ee. (2S,3S)-2-Aryl-6-methoxy-3-methyl-2,3-

dihydrobenzofuran-5-ols 6/9 are also found, but in slightly lower ee. Cyclobutanes 5/8 cleanly and

efficiently rearrange to the dihydrobenzofurans 6/9 without loss of enantiomeric purity upon

treatment with the Ti-TADDOLates at higher temperatures. Reactions of (Z)-propenylbenzene

17 and of indene with 2 and 3 give products in moderate enantiomeric purity. Products obtained

from reactions of 1-anisylcycloalkenes with 2 differ significantly in yield and enantiomeric purity.

In the latter reactions, the ee’s of the cyclobutane products are consistently much higher than

those of the dihydrobenzofuran products. More significantly, products of different absolute configu-

ration result from different cycloalkenes. With 1-anisylcyclopentene or 1-anisylcyclohexene, all of

the products are of similar configuration and are obtained in comparable yields and ee’s. However,

1-anisylcycloheptene affords products that are diastereomeric with those of the 1-anisylcyclopentene,

and in lower ee’s. A mechanistic model is proposed. Application of these reactions to the

enantioselective synthesis of the pterocarpan class of isoflavonoid natural products is also reported.

Sch em e 1

In tr od u ction

The effect of molecular chirality on biological activity

has generated intense interest in asymmetric synthesis

over the last several decades.¹Much effort has been

expended on the development of asymmetric variants of

well-established reactions, and upon discovery of a new

reaction that creates new stereogenic centers focus often

quickly shifts to the development of enantioselective

versions using chiral auxiliaries, reagents, or catalysts.

We have found recently that reactions of various styrenyl

systems 1 with 1,4-benzoquinones 2/3 are extraordinarily

versatile processes, producing a variety of chiral cycload-

dition products.²In addition to products of Diels-Alder

reaction found under thermal conditions,³Lewis acid-

promoted reactions give up to three different types of

^†Current address: Lilly Corporate Center, Mail Drop 1523, Eli Lilly

and Company, Indianapolis, IN 46285. Phone: (317)433-2885. Fax:

(317)277-2035. E-mail: ENGLER_THOMAS@Lilly.com.

^‡Formerly Kenneth O. Lynch, J r.

(1) Gawley, R. E.; Aube´, J . Principles of Asymmetric Synthesis,

Tetrahedron Organic Chemistry Series Vol. 14; Pergamon: Tarrytown,

NY, 1996. Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of

Organic Compounds; Wiley-Interscience: New York, 1994.

products, 5-10 (Scheme 1).^2,4The latter three types of

products significantly extend the synthetic utility of

quinone-styrene reactions, which have since been used

to access racemic mixtures of a number of chiral biologi-

cally active natural products.²Products 5-10 are ob-

tained in Ti(IV)-promoted reactions of 1 with 2/3, with

mixtures of TiCl₄and Ti(Oi-Pr)₄generally the most

effective promoters. For this reason, our interest was

drawn to chiral Ti-TADDOLate complexes to potentially

generate 5-10 in nonracemic form.^5,6Herein, we describe

the full details of these studies.⁷

(2) (a) Engler, T. A.; Combrink, K. D.; Ray, J . E. J . Am. Chem. Soc.

1988 110, 7931-7933. (b) Engler, T. A.; Combrink, K. D.; Takusagawa,

F. J . Chem. Soc., Chem. Commun. 1989 1573-1576. (c) Engler, T. A.;

Reddy, J . P.; Combrink, K. D.; Vander Velde, D. J . Org. Chem. 1990,

55, 1248-1254. (d) Engler, T. A.; Letavic, M. A.; Combrink, K. D.;

Takusagawa, F. J . Org. Chem. 1990 56, 5810-5812. (e) Engler, T. A.;

Combrink, K. D.; Letavic, M. A.; Lynch, K. O., J r.; Ray, J . E. J . Org.

Chem. 1994, 59, 6567-6587. (f) Engler, T. A.; Wei, D.; Letavic, M. A.;

Combrink, K. D.; Reddy, J . P. J . Org. Chem. 1994 59, 6588-6599. (g)

Engler, T. A.; Gfesser, G. A.; Draney, B. W. J . Org. Chem. 1995, 60,

3700-06. (h) Engler, T. A.; LaTessa, K. O.; Iyengar, R.; Chai, W.;

Agrios, K. Bioorg. Med. Chem. 1996, 4, 1755-69. (i) Engler, T. A.;

Iyengar, R. J . Org. Chem. 1998, 63, 1929-1934.

Published on Web 03/18/1999

2392 J . Org. Chem., Vol. 64, No. 7, 1999

Engler et al.

Ta ble 1. Asym m etr ic Rea ction s of P r op en ylben zen e 1a

w ith Qu in on e 2

Resu lts

Initial studies employed propenylbenzene 1a and

quinone 2. Reactions promoted by mixtures of TiCl₄and

Ti(Oi-Pr)₄gave cyclobutane (()-5a and dihydrobenzofu-

ran (()-6a , with the latter usually as the major product.

Reactions of 1a with 2 employing Ti-TADDOLates

prepared from chiral diol-(+)-4^5,6by the protocols de-

scribed by Seebach,⁵Narasaka,⁶and Corey^8afor Diels-

Alder, 2+2 cycloaddition, and allylation reactions, among

others,^8b-ewere attempted with little success; either the

products were obtained with no enantiomeric enrichment,

or the Lewis acid was not an effective promoter.⁹Thus,

we surveyed empirically a number of other protocols⁵for

% yield (% ee)^a

promoter: ratio of

temp

(°C)

entry TiCl₄:Ti(OiPr)₄:4:2^b

(-)-5a

(-)-6a

0.5:0.5:0.5:1

0.5:0.5:1:1

1:1:1:1

2.5:2.5:2.5:1

-78

63 (60) 36 (6)^c

-78

52 (78) 42 (n.d.)^c,d

69 (77) 24 (n.d.)^c,d

-78

88 (92)

9 (41)^c

94 (71)^e

95 (82)^f

-78 to 0

-78 to rt

Determined by ¹H NMR with 11 as a chiral solvating agent

(see text). The number of equiv of 1a used was 1.1-2.6. ^cOnly

the trans isomer by ¹H NMR. n.d. ) not determined. ^etrans:cis

) 39:1 by NMR. ^ftrans:cis ) 15:1 by NMR.

preparing a Ti-TADDOLate that would afford the prod-

ucts in high ee. The results summarized in Table 1

revealed 2.5 equiv each with respect to the quinone 2 of

TiCl₄, Ti(Oi-Pr)₄, and the chiral diol-(+)-4, and low

reaction temperatures resulted in the formation of the

bicyclo[4.2.0]octenedione product (-)-5a in high enan-

tiomeric purity (Table 1, entry 4); smaller amounts of the

dihydrobenzofuran (-)-6a were also produced but in

lower ee. The Ti-(+)-4 complex was prepared by the

addition of TiCl₄to a dichloromethane solution of Ti(Oi-

Pr)₄at room temperature followed by the addition of a

dichloromethane solution of the diol. The solution of the

Ti-complex was then added to a dichloromethane solution

of 2 maintained at -78 °C followed by addition of the

propenylbenzene. Alternatively, a solution of the quinone

could be added to the Ti-complex. After the reactions were

judged to be complete by TLC, the mixtures were

subjected to a standard aqueous workup. Direct analysis

of the crude reaction mixtures by NMR or HPLC was

problematic. Thus, the diol and products were separated

by flash chromatography prior to determination of enan-

tiomeric purity. The chromatography introduces some

uncertainty into the degree of asymmetric induction,¹⁰

but the high yields and ee’s of the isolated products

impressively demonstrate the preparative value of the

reactions. Slowly warming the reaction mixtures pro-

duced dihydrobenzofuran (-)-6a in good yields, although

the enantiomeric excesses were not as high as those

found for (-)-5a (entries 5 and 6). Longer reaction times

at -78 °C had little effect on the ratio of products formed,

indicating that rearrangement²of the cyclobutane to the

dihydrobenzofuran was not occurring to a significant

extent at low temperatures.

(3) For leading references: (a) Lora-Tamayo, M. Tetrahedron 1958,

4, 17. (b) Inouye, Y.; Kakisawa, H. Bull. Chem. Soc. J pn. 1971, 44,

563-564. (c) Zhang, Z.-r.; Flachsmann, F.; Moghaddam, F. M.; Ru¨edi,

P. Tetrahedron Lett. 1994, 35, 2153-2156. (d) Willmore, N. K.; Hoic,

D. A.; Katz, T. J . J . Org. Chem. 1994, 59, 1889-1891. (e) Amatayakul,

T.; Cannon, J . R.; Dampawan, P.; Dechatiwongse, T.; Giles, R. G. F.;

Huntrakul, C.; Kusamrann, K.; Mokkhasamit, M.; Raston, C. L.;

Reutrakul, V.; White, A. H. Aust. J . Chem. 1979, 32, 71-88. (f) Tanga,

M. J .; Reist, E. J . J . Heterocyclic Chem. 1991, 28, 29-32. (g) Rosen, B.

I.; Weber, W. P. J . Org. Chem. 1977, 42, 3463-3468. (h) Manning, W.

B. Tetrahedron Lett. 1981, 22, 1571-1574. (i) Kelly, T. R.; Magee, J .

A.; Weibel, F. R. J . Am. Chem. Soc. 1980, 102, 798-799. (j) Kita, Y.;

Okunaka, R.; Honda, T.; Shindo, M.; Tamura, O. Tetrahedron Lett.

1989, 30, 3995-3998. (k) Schlu¨ter, A.-D.; Wegner, G.; Blatter, K.

Macromolecules 1989, 22, 3506.

(4) We, and others, have also extended Lewis acid-promoted reac-

tions of quinones to include other alkenyl systems. For leading ref-

erences, see: (a) Engler, T. A.; Agrios, K.; Reddy, J . P.; Iyengar, R.

Tetrahedron Lett. 1996, 37, 327-330. (b) Mukaiyama, T.; Sagawa, Y.;

Kobayashi, S. Chem. Lett. 1987, 2169-2172. (c) Murphy, W. S.; Neville,

D. Tetrahedron Lett. 1997, 38, 7933-7936. (d) Hosomi, A.; Sakurai,

H. Tetrahedron Lett. 1977, 4041-4044. (e) Hosomi, A.; Sakurai, H.

Tetrahedron Lett. 1978, 2589-2592. (f) Ipaktschi, J .; Heydari, A.

Angew. Chem., Int. Ed. Engl. 1992, 31, 313-314. (g) Ipaktschi, J .;

Heydari, A. Chem. Ber. 1992, 125, 1513-1515. (h) Naruta, Y. J . Am.

Chem. Soc. 1980, 102, 3774-3783. For reviews, see: (i) Perlmutter,

P. Conjugate Addition Reactions in Organic Synthesis, Tetrahedron

Organic Chemistry Series Vol. 9; Pergamon: Oxford, 1992; pp 322-

330. (j) Finley, K. T. In The Chemistry of the Quinonoid Compounds,

Vol. 2; Patai, S., Rappaport, Z., Eds.; Wiley: Chichester, 1988; p 537.

(k) Nishigaichi, Y.; Takuwa, A.; Naruta, Y.; Maruyama, K. Tetrahedron

1993, 49, 7395-7426. (l) Cintas, P. SYNLETT 1995, 1087-1109. (m)

Fleming, I.; Dunogue´s, J .; Smithers, R. Org. React. 1989 37, 57-575.

(n) Naruta, Y.; Nishigaichi, Y.; Maruyama, K. Org. Synth. 1992 71,

125-131. (o) Naruta, Y.; Maruyama, K. In The Chemistry of Quinonoid

Compounds, Vol. 2; Patai, S., Rappoport, Z., Eds.; Wiley-Interscience:

Chichester, 1988; Chapter 8. (p) Mukaiyama, T.; Iwasawa, N.; Yura,

T.; Clark, R. S. J . Tetrahedron 1987, 43, 5003-5017. (q) Nucleophilic

additions to quinones bearing electron-withdrawing groups are com-

mon; see references cited in the reviews.

(5) (a) Seebach, D.; Beck, A. K.; Imwinkelried, R.; Roggo, S.;

Wonnacott, A. Helv. Chim. Acta 1987, 70, 954-974. (b) Seebach, D.;

Dahinden, R.; Marti, R. E.; Beck, A. K.; Plattner, D. A.; Ku¨hnle, F. N.

M. J . Org. Chem. 1995, 60, 1788-1799. (c) Seebach, D.; Marti, R. E.;

Hintermann, T. Helv. Chim. Acta 1996, 79, 1710-1740.

Other attempts to prepare effective chiral Ti-complexes

for reactions of 1a with 2 were also examined.^5b,9The

cyclobutane product (-)-5a was generally always formed

in higher ee than the dihydrobenzofuran (-)-6a , and

efforts were thus focused on optimizing conditions for

generating the former. Deprotonation of diol-(+)-4 with

2 equiv of n-BuLi followed by the addition of TiCl₄

afforded a complex that gave (-)-5a in good ee (88%),

but low yield (29%); 6a was the major product (54%), but

found in poor ee (<5%). A complex prepared from TiCl₄/

Ti(Oi-Pr)₄/diethyl L-tartrate yielded racemic 6a as the

major product (74%). Systems involving TiCl₄and diol-

(+)-4, or TiCl₄/Ti(Oi-Pr)₄and racemic binaphthol, were

also studied; however, the major products were always

6a (52-72%), and these reactions were not pursued

further.

(6) (a) Narasaka, K.; Inoue, M.; Okada, N. Chem. Lett. 1986, 1109-

1112. (b) Narasaka, K.; Iwasawa, N.; Inoue, M.; Yamada, T.; Naka-

shima, M.; Sugimori, J . J . Am. Chem. Soc. 1989, 111, 5340-5345. (c)

Narasaka, K.; Hayashi, Y.; Shimadzu, H.; Niihata, S. J . Am. Chem.

Soc. 1992, 114, 8869-8885. (d) Narasaka, K.; Iwasawa, N. In Organic

Synthesis: Theory and Applications; Hudlicky, T., Ed.; J ai Press:

London, 1993; Vol. 2, pp 93-112. (e) Narasaka, K. Synthesis 1991,

1-11.

(7) For a preliminary account: Engler, T. A.; Letavic, M. A.; Reddy,

J . P. J . Am. Chem. Soc. 1991, 113, 5068-5070.

(8) (a) Corey, E. J .; Matsumura, Y. Tetrahedron Lett. 1991, 32,

6289-6292. For recent applications: (b) Yang, H. W.; Romo, D.

Tetrahedron Lett. 1998, 39, 2877-2880. (c) Gothelf, K. V.; Thomsen,

I.; J ørgensen, K. A. J . Am. Chem. Soc. 1996, 118, 59-64. (d) Gothelf,

K. V.; Hazell, R. G.; J ørgensen, K. A. J . Org. Chem. 1996, 61, 346-

355. (e) Gothelf, K. V.; J ørgensen, K. A. J . Org. Chem. 1994, 59, 5687-

5691. See also ref 21.

(9) To alleviate concerns about the stoichiometry, we reinvestigated

the asymmetric N-2-butenoyloxazolidinone/cyclopentadiene Diels-

Alder reaction (the “benchmark reaction” for evaluating new chiral

Diels-Alder catalysts) and closely reproduced the findings as described

in refs 5, 6, and 8. Utilization of these protocols for preparation of Ti-

TADDOLates failed for the quinone-propenylbenzene reactions, how-

ever, as did our recipe in the Diels-Alder reaction.

(10) Due to clathrate formation or enrichment, see discussion in ref

5b.

Asymmetric Reactions of 2-Methoxy-1,4-benzoquinones

J . Org. Chem., Vol. 64, No. 7, 1999 2393

Ta ble 2. Asym m etr ic Rea ction s of P r op en ylben zen es 1 w ith Qu in on es 2/3 P r om oted by th e Ti(IV)-(+)-4-TADDOLa te^a

products^b

(-)-5

% yield

(-)-6^c

% yield

(+)-7

% yield

entry

styrene

quinone

time (h)

temp (°C)

% ee

0.5

-78

-94

-78

-94

-78

products^b

(+)-9^c

% yield

(-)-8

(+)-10

entry

styrene

quinone

time (h)

temp (°C)

% yield

% ee

% yield

% ee

1.5

-94 f -78

-78

-78 f -30

-94

<10

-78

The Ti-TADDOLate promoter was composed of 2.5 equiv each of TiCl₄, Ti(OiPr)₄, and 4, with respect to quinones 2 and 3. Yields

and % ee’s of isolated products. ^ct/c g 45:1 by ¹H NMR.

Ta ble 3. Rea r r a n gem en ts of Cyclobu ta n es 5/8 to

Reactions of other propenylbenzenes 1 with quinones

Dih yd r oben zofu r a n s 6/9^a

2 and 3 were then examined employing the “optimized”

conditions (Table 1, entry 4). In reactions of 2, cyclobu-

cyclobutane,

entry

% ee

temp (°C)

product, t/c % yield^b% ee^b

tanes (-)-5a -c were formed as the major products with

propenylbenzenes 1a -c bearing electron-donating groups

on their aromatic rings, whereas bicyclo[3.2.1]-adducts

(+)-7d ,e were obtained with propenylbenzenes 1d ,e

(Table 2). Both types of products were isolated in high

ee. Reactions of quinone 3 at low reaction temperatures

also gave cyclobutanes (-)-8, but at higher reaction

temperatures, greater amounts of dihydrobenzofurans

(+)-9 were found. The latter reactions suggested an in

situ rearrangement of 8 to 9 under the reaction conditions

(vide infra and ref 2). Similar to reactions of 2, reactions

of 3 with the more neutral propenylbenzenes 1c,d gave

bicyclo[3.2.1]-adducts (+)-10c,d . All reactions were rou-

tinely subjected to flash chromatography to separate the

excess diol^5bprior to measurement of ee. Thus, the degree

of asymmetric induction implied by the enantiomeric pu-

rity of the isolated products should be viewed cautiously.^5b

Nevertheless, the reactions are effective for accessing

highly enantiomerically enriched products in reasonably

good yields.

(-)-5a , 100^c-78 to -10 (-)-6a , 93:1

100

100^c

(-)-5b, 90

(-)-5c, 86

-78 to 0

(-)-6b, 15:1

(-)-6c, 17:1

(-)-8a , 100^c-78 to -30 (+)-9a , 99:1^d

(-)-8b, 100^c-78 to -30 (+)-9b, 17:1

100^c

The Ti-TADDOLate promoter was composed of 0.6-0.8 equiv

each of TiCl₄, Ti(OiPr)₄, and 4, with respect to the cyclobutanes.

^bYield or ee of isolated product. ^cOnly one enantiomer observed in

the NMR/HPLC assays described in the text. Only trans by NMR.

be found in the Supporting Information accompanying

our preliminary communication⁷). For 10, the ee’s were

determined using the HPLC method.

Enantiomerically pure (within the limits of the NMR/

HPLC assays) cyclobutanes 5a /8a ,b and bicyclo[3.2.1]-

adducts 7d ,e were obtained by simple recrystallization

of enantiomerically enriched material. Resubjecting the

enantiomerically enriched, or pure, cyclobutanes at -78

°C to the chiral Ti-diol complex prepared from 0.63-0.78

equiv each of TiCl₄/Ti(Oi-Pr)₄/diol-(+)-4, followed by

warming the mixtures to -30 or 0 °C effected rearrange-

ment to the dihydrobenzofurans 6/9 with little, if any,

loss in enantiomeric purity (Table 3). Similar rearrange-

ments of 5/8 to 6/9 can be carried out with protic acids;²

however, the trans:cis ratio of the dihydrobenzofuran

products are lower (∼6-10:1) than those found with the

Ti(IV)-Lewis acid promoter.¹²In reactions of enantio-

merically enriched 5b,c, recrystallization of the products

6b/c afforded enantiomerically pure material. The step-

wise formation of the dihydrobenzofurans by first isolat-

ing the cyclobutanes and then effecting their rearrange-

ment offers a significant advantage over the direct for-

mation via allowing the cycloaddition reactions to warm.

The cyclobutanes readily recrystallize to enantiomerically

pure material, and their rearrangement to the dihydro-

benzofurans occurs without loss of enantiomeric purity.

The enantiomeric purities of the cycloadducts were

determined by H NMR utilizing (R)-(-)-2,2,2-trifluoro-

1-(9-anthryl)ethanol (11) as a chiral solvating agent¹¹

and/or by HPLC (Chiralcel OJ ). Racemic samples were

used as standards in both assays. In general, the signals

for the C-4 vinyl proton of the racemic cyclobutanes 5/8,

the C-7 aromatic proton of the racemic dihydrobenzo-

furans 6/9, and the C-4 hydrogen of the bicyclo[3.2.1]-

octenediones 7 were resolved in the presence of the

solvating agent (representative examples of spectra can

(12) The difference in ratios of cis:trans 6 is significant insofar as

their specific rotations are vastly different and of opposite sign (see

Experimental Section). Thus, attempts to monitor enantiomeric purity

of these mixtures by optical rotation can be very misleading.

(11) Pirkle, W. H.; Hoover, D. J . Top. Stereochem. 1982, 13, 263.

2394 J . Org. Chem., Vol. 64, No. 7, 1999

Engler et al.

It should be noted that the Ti-TADDOLate promoter

gave ratios of cyclobutane to dihydrobenzofuran products

considerably different from those found with simple

mixtures of TiCl₄/Ti(Oi-Pr)₄. With the former, the cy-

clobutane products were generally found as the major

products with electron-rich styrenes at low reaction

temperatures, whereas with TiCl₄/Ti(Oi-Pr)₄, the dihy-

drobenzofuran products were usually obtained as the

major products.²With the more neutral styrenes 1c-e,

the bicyclo[3.2.1]-products 7/10 were found as major

products with the Ti-TADDOLate promoter, whereas

with TiCl₄/Ti(Oi-Pr)₄, these products were found only in

minor amounts. These results suggest that the TAD-

DOLate Lewis acid varies considerably in Lewis acid

strength from simple mixtures of TiCl₄/Ti(Oi-Pr)₄.¹³

The structure and absolute configuration of the bicyclo-

[3.2.1]-adduct 7d was determined by conversion to heavy

atom derivative (+)-12 (eq 1) followed by single-crystal

X-ray analysis using the anomalous scattering of the

bromine atom (Bijvoet method).¹⁴The absolute configura-

tions of the other bicyclo[3.2.1]-adducts were then as-

signed by analogy.

methods described for similar degradations of 13 and 14

(eq 2),^15,16confirmed the (2S,3S) configuration. Further-

more, to verify that introduction of a 4-alkyl group on

the dihydrobenzofuran core of (-)-6a changed its sign of

optical rotation, it was converted to an allyl ether

followed by Claisen rearrangement to give (+)-16 (eq 3)

with dextrorotatory rotation. The relative configurations

of 5/8 were established previously, and their absolute

configuration is assigned on the basis of their conversions

to 6 and 9, respectively.

Reactions of (Z)-propenylbenzene 17 and of styrenyl

systems imbedded in ring systems were also studied. In

our original work on Lewis acid-promoted quinone-

styrene reactions, we observed that reactions of (Z)-

propenylbenzenes were in general slower and not as clean

as those of the corresponding (E)-isomers.^2eSimilarly,

reaction of 17 with the Ti-(+)-4-TADDOLate prepared

as described above afforded the cyclobutane product (-)-

18 in modest yield, but reasonably good ee (eq 4). Indene

reacted with quinones 2 and 3 to give products (-)-19

and (+)-20/21, respectively, in good yields, but modest

ee’s (eq 5). The enantiomeric purities of 18-21 were

again established via ¹H NMR using 11 and/or HPLC

(Chiralcel OJ ) with racemic mixtures as standards.

Recrystallization of 18 gave enantiomerically homoge-

neous material, but no attempts were made to obtain 19-

21 in higher enantiomeric enrichment. The absolute

configurations of the products are assigned tentatively

from mechanistic considerations (vide infra).

The absolute configurations of the dihydrobenzofuran

products (-)-6 were assigned by comparison of [R] and/

or ORD data with those of the natural products (2R,3R)-

(+)-obtusafuran,¹⁵13, and (2S,3S)-(-)-melanoxin, 14,¹⁶

of known configuration. The fact that the sign of the opti-

cal rotations of dihydrobenzofurans (-)-6 and (+)-9 were

opposite was of initial concern. However, degradation of

(+)-9a to dimethyl (R)-(+)-2-methyl succinate (15),¹⁷by

(13) The strength of the Lewis acid promoter in reactions of styrenyl

systems with 2-alkoxy- and 2-alkoxy-6-methyl-1,4-benzoquinones has

a significant influence on the ratio of the dihydrobenzofuran versus

the cyclobutane products. In general, with the strong Lewis acid TiCl₄,

mainly dihydrobenzofurans are found, whereas with mixtures of TiCl₄/

Ti(Oi-Pr)₄more of the cyclobutanes are produced, in many cases as

the major products. A further consideration in the Ti-TADDOLate-

promoted reactions is the large amount of Ti(IV) employed. Some of

the excess Ti(IV) may coordinate to the carbonyl group in 36, slowing

C-O bond formation to 6/9, but having a relatively much lesser effect

(if any) on C-C bond formation leading to 5/8. This stoichiometry may

also play a role in the difference in ee found between the cyclobutane

and dihydrobenzofuran products from reactions of 1 with 2/3. Coor-

dination of

a second Ti(IV) to 36 may allow a deprotonation-

reprotonation sequence to compete (see eq 11), resulting in lower ee

for 6/9. The rearrangements of 5/8 to 6/9, however, employ only 1 equiv

or less of Ti(IV), where 36-(Ti)₂complexes are not likely, and proceed

without loss of enantiomeric purity.

(14) See ref 1b, pp 25 and 113, and Hamilton, W. C. Acta Crystallogr.

1965, 18, 502-510.

(15) Gregson, M.; Ollis, W. D.; Redman, B. T.; Sutherland, I. O.;

Dietrichs, H. H.; Gottlieb, O. R. Phytochemistry 1978, 17, 1395-1400.

(16) Donnelly, B. J .; Donnelly, D. M. X.; O’Sullivan, A. M.; Pren-

dergast, J . P. Tetrahedron 1969, 25, 4409-4414.

(17) Hayashi, T.; Yamamoto, A.: Hagihara, T. J . Org. Chem. 1986,

51, 723-727.

Reactions of 1-anisylcycloalkenes with quinone 2 were

far more complex. As with styrenes 1, reactions of 22/23

Asymmetric Reactions of 2-Methoxy-1,4-benzoquinones

J . Org. Chem., Vol. 64, No. 7, 1999 2395

Ta ble 4. Asym m etr ic Rea ction s of 1-An isylcycloa lk en es w ith Qu in on e 2 P r om oted by Ti(IV)-TADDOLa tes

The number of equiv of alkene used was 1.0-1.1. ^bIsolated yields. ^cDetermined by NMR and/or HPLC on isolated samples. Estimated

from optical rotation. ^eNot determined.

afforded cyclobutanes (-)-24/26 in high ee accompanied

by dihydrobenzofurans (+)-25/27 in considerably lower

ee (Table 4). Contrary to reactions of the styrenes,

however, was that the dihydrobenzofuran products 25/

27 were often found in quantities greater than the

cyclobutanes, even at low reaction temperatures. Reac-

tions of 22 with 2 were conducted under a variety of

conditions in efforts to optimize the formation of (-)-24,

with some improvement. In addition, reactions of the

cycloheptene derivative 28 gave a cyclobutane product

(-)-29 in moderate yield but with ee significantly lower

than those found for analogous products 24/26; dihy-

drobenzofuran (+)-30 was also found in moderate yield

and low ee (eq 7). More importantly, initial structural

analysis (¹H NMR, ¹H-¹H NOE) suggested that cyclobu-

tane (-)-29 possessed a different relative stereochemistry

than products 24/26. As a result, the absolute stere-

ochemistries of all of the products were in question;

rearrangements of cyclobutanes with the relative stere-

ochemistries of 24/26 and 29 would give dihydrobenzo-

furans with opposite absolute configuration.

Detailed structural analyses of the products and

further examination of reaction conditions for their

formation were undertaken. The relative cis-anti-cis

stereochemistry for cyclobutanes 24/26 and the cis-syn-

cis stereochemistry around the cyclobutane ring in (-)-

29 were assigned from ¹H-¹H NOE data (summarized

in the Experimental Section). The enantiomeric excesses

of (-)-24/26 were determined by ¹H NMR analysis in the

presence of 11, in which signals at 5.94 and 5.88 ppm

(H-3) for (-)-24 and (-)-26, respectively, split into pairs

of well-resolved singlets. To verify the accuracy of these

NMR experiments, considerable effort was expended in

obtaining racemic samples of 24, a previously unknown

compound.^2eCareful analysis of reactions of 22 with 2

promoted by various mixtures of TiCl₄/Ti(OiPr)₄at low

temperature revealed that cyclobutane rac-24 was pro-

duced in low, but serviceable yield; as in our previous

studies, dihydrobenzofuran rac-25 was usually found as

the major product (Table 5).^2eAttempts to isolate cy-

clobutane rac-26 from similar reactions of 23 with 2 were

not successful. However, promotion of reactions of both

22 and 23 with 2 using a Ti-TADDOLate formed from

diol ent-(-)-4 gave cyclobutane products ent-(+)-24/26

(Table 4, entry 11), whose NMR spectra in the presence

2396 J . Org. Chem., Vol. 64, No. 7, 1999

Engler et al.

Ta ble 5. Ti(IV)-P r om oted Rea ction s of

1-An isylcyclop en ten e 22 w ith Qu in on e 2

% yield^a

TiCl₄:Ti(OiPr)₄

(equiv)^b

temp

(°C)

time

(min)

entry

rac-24

rac-25

1:1 (0.5)

2:1 (0.5)

1:1 (1.0)

2:1 (1.0)

1:1 (2.5)

-78

-85

-90

-78

120

62^c

33^c

F igu r e 1. Alignment for four-membered ring closure.

Isolated yields. Equiv of Ti/quinone. ^cBased on recovered 22

(+)-27, and (+)-30 in good yields (>94%) and with no ap-

parent loss of enantiomeric purity. As before, the enanti-

omeric purities were determined by the NMR and HPLC

analyses described above utilizing racemic samples^2eas

(54%).

of 11 and [R] identified them as enantiomers of (-)-24/

26. rac-24 and ent-24/26 were then used as standards to

validate the NMR assay developed to measure ee, as well

as to develop an HPLC method in which a Chiralcel OD

column worked well.

standards. In the H NMR experiments, singlets at 6.44,

6.49, and 6.48 ppm for (+)-25, (+)-27, and (+)-30,

respectively, all split into well-resolved signals upon

addition of (-)-11. H-¹H NOE experiments¹⁸ruled out

Cyclobutane (-)-29 was also analyzed by both the H

the possibility of trans-fused diastereomers. Further-

more, to double-check the assignment of absolute stere-

ochemistry, dihydrobenzofuranol (+)-25 was converted

to a p-bromobenzoate (eq 10), the structure of which was

NMR method with 11 and by chiral HPLC (Chiralcel OD

column). The NMR experiment showed that a signal at

6.12 ppm split into a pair of well-resolved singlets upon

addition of the chiral solvating agent, and the ratio of

these signals was consistent with the ratio of two signals

observed in the chiral HPLC experiment.

Again, simple recrystallization of enantiomerically

enriched samples of both (-)-24/26 gave enantiomerically

pure material, within the limits of the NMR/HPLC

assays described above. Similarly, recrystallization of (-)-

29 gave a sample that was homogeneous in these assays

and was thus taken to be enantiomerically pure. In all

cases, the only peaks observed in the analyses of the

recrystallized samples were those corresponding to the

major isomers from the cycloaddition reactions.

To confirm their structures, enantiomerically homo-

geneous cyclobutanes (-)-24 and (-)-29 were selectively

reduced with Red-Al and the alcohol products converted

to carbamates (-)-31 and (-)-32, respectively (eqs 8 and

9), which were subjected to single-crystal X-ray analyses.

confirmed by single-crystal X-ray analysis.¹⁴The absolute

stereochemistries of (+)-27 and (+)-30 were then as-

signed from the mechanism likely operative in their

formation from (-)-26/29. That the diastereomeric cy-

clobutanes 24/26 and 29 from the cycloaddition reactions

with the Ti-(+)-4 TADDOLate all have the same sign

of optical rotation as do the diastereomeric dihydroben-

zofurans 25/27 and 30 is apparently fortuitous.

Discu ssion

We have suggested that the mechanism for the qui-

none-propenylbenzene reactions involves bidentate co-

ordination of Ti(IV) to the quinone followed by a 5+2

(4π+2π) cycloaddition of the pentadienyl carbocation

moiety of complex 34¹⁹with the propenylbenzene to

(18) ¹H-¹H NOE data for (()-25 and (()-27 can be found in ref 2e.

Similar ¹H-¹H NOEs are also found in (()-30.

(19) Similar 5+2 cycloadditions are observed in carbocations formed

in solvolysis of quinone monoketals or oxidation of phenols, see: (a)

Bu¨chi, G.; Mak, C.-P. J . Am. Chem. Soc. 1977, 99, 8073-8075. (b)

Bu¨chi, G.; Chu, P.-S. J . Org. Chem. 1978, 43, 3717-3719. (c) Mak,

C.-P.; Bu¨chi, G. J . Org. Chem. 1981, 46, 1-3. (d) Mortlock, S. V.;

Seckington, J . K.; Thomas, E. J . J . Chem. Soc., Perkin Trans. 1 1988,

2305-2307. (e) Angle, S. R.; Turnbull, K. D. J . Org. Chem. 1993, 58,

5360-5369. (f) Gates, B. D.; Dalidowicz, P.; Tebben. A.; Wang, S.;

Swenton, J . S. J . Org. Chem. 1992, 57, 2135-2143. (g) Yamamura, S.;

Shizuri, Y.; Shigemori, H.; Okuno, Y.; Ohkubo, M. Tetrahedron 1991,

47, 635-644. (h) Takakura, H.; Yamamura, S. Tetrahedron Lett. 1998,

39, 3717-3720. (i) Shizuri, Y.; Shigemori, H.; Suyama, K.; Nakamura,

K.; Okuno, Y.; Ohkubo, M.; Yamamura, S. In Studies in Natrural

Product Chemistry, Vol. 8; Atta-ur-Rahman, Ed.; Elsevier: Amsterdam,

1991; pp 159-173. (j) Grieco, P. A.; Walker, J . K. Tetrahedron 1997,

53, 8975-8996. (k) Collins, J . L.; Grieco, P. A., Walker, J . K.

Tetrahedron Lett. 1997, 38, 1321-1324. An analogous process is the

perezone to pipitzol rearrangement, see: (l) J oseph-Nathan, P.; San-

tillan, R. L. In Studies in Natrural Product Chemistry, Vol. 5; Atta-

ur-Rahman, Ed.; Elsevier: Amsterdam; 1989; pp 763-813. Intramo-

lecular carbocation activation of quinones to intermolecular 5+2

cycloaddition with styrene has been reported; see: (m) Mamont, P.

Bull. Soc. Chim. Fr. 1970, 1557-1564.

The structures revealed and the absolute configurations

indicated by the anomalous scattering¹⁴were consistent

with the original assignments of (-)-24 and (-)-29. The

absolute configuration of (-)-26 was then assigned due

to the similarity of H-¹H NOE data collected from it

and from (-)-24 (Figure 1).

Enantiomerically homogeneous cyclobutanes (-)-24/26/

29 underwent protic acid-catalyzed rearrangement to en-

antiomerically homogeneous dihydrobenzofurans (+)-25,

Asymmetric Reactions of 2-Methoxy-1,4-benzoquinones

J . Org. Chem., Vol. 64, No. 7, 1999 2397

Sch em e 2

afford bicyclo[3.2.1]octenyl carbocation 35 (Scheme 2; for

clarity, the other ligands on Ti are not shown). In this

cycloaddition, the stereoselectivity is a result of a prefer-

ence for the aryl group of the propenylbenzene to occupy

an endo orientation with respect to the carbocation

moiety. The regioselectivity is directed by bond formation

between the more nucleophilic C-â of the propenylben-

zene and the electrophilic C-5 atom of the Ti-quinone

complex. From 35, dealkylation (a) affords the bicyclo-

[3.2.1]-adducts 7/10. Alternatively, C-1/C-7 bond cleavage

(b) gives stabilized benzylic carbocation 36, and C-C

bond formation between the carbocation center and the

Ti-enolate produces 5/8, whereas C-O bond formation

with loss of H⁺yields 6/9. In the latter processes, collapse

to 5/8 is best described as the product of kinetic control;

that is, yields of 5/8 relative to 6/9 are highest at low

reaction temperature. Formation of 6/9 is thermodynami-

cally preferred, as evidenced by its exclusive formation

at higher cycloaddition temperatures and by rearrange-

ment of the cyclobutanes 5/8 to 6/9 under either Ti(IV)-

or H⁺-promoted processes. The kinetic preference for 5/8

is likely stereoelectronic in origin. The carbocation center

can approach the enolate from a direction close to the

preferred perpendicular alignment with little distortion

(Figure 1); however, considerable twisting is required for

the carbocation to be placed in a position to bond to an

electron lone pair on the carbonyl oxygen.¹³

The effects of substituents, on either the quinone or

the propenylbenzene, on the relative yields of the various

products are readily explained by this mechanism. Reac-

tions of propenylbenzenes 1 bearing electron-donating

OCH₃substitutents produce no 5+2 products, suggesting

that the rate of dealkylation is slow relative to collapse

to stabilized benzylic carbocation 36. That the neutral

propenylbenzene 1e gives only the 5+2 products indicates

that the dealkylation competes effectively in this case

with collapse to 36. Reaction of 1d is particularly instruc-

tive; only the 5+2 product is found, which indicates again

that collapse of 35 to 36 is slow compared to the dealkyl-

ation, probably due to steric inhibition of resonance

stabilization of the developing benzylic carbocation by the

o-methyl group. The higher yields of dihydrobenzofurans

relative to the cyclobutanes in reactions of 3, compared

to reactions of 2, are likely due either to slower formation

of the latter from intermediate 36 or to faster rearrange-

ment of the cyclobutane to the dihydrobenzofuran. Either

can be attributed to steric interaction between the methyl

group-R and the aryl ring.

reactions. Applying this model to the quinone-propenyl-

benzene reactions, the observed attack on the C-3 si/C-5

re face²²is represented as A in Figure 2. In this model,

one of the quasiaxial phenyl rings of the TADDOL ligand

blocks the C-3 re/C-5 si face of the quinone. The alterna-

tive binding mode B places the carbonyl group to be

activated in an unfavorable position cis rather than trans

to a Cl ligand.²⁰In addition, upon approach of a prope-

nylbenzene with the aryl group in an endo position, a

steric interaction develops between the aryl group and

Cl* in B (see arrow) that is not found in A. The more

sluggish reactions and lower ee’s found with (Z)-prope-

nylbenzene 17 and with indene reflect the higher steric

expense in placing both the aryl and methyl groups in

an endo orientation.

Caution is recommended, however, in relying strongly

on this preliminary model. The experimental “recipes”

used for preparation of Ti-TADDOLate promoter(s) vary

The cycloaddition model proposed in Scheme 2 is useful

for rationalizing the enantioselectivities found in the

products from reactions of the quinones with propenyl-

benzenes 1. Although it may appear that the chiral

ligands of the Ti-TADDOLate-quinone complex would

be far from the center of action, the reactions do provide

products in high ee; thus the ligands obviously have

considerable influence. Seebach⁵and DiMare²⁰have

suggested a model²¹for the Ti-TADDOLate-promoted

(20) (a) Haase, C.; Sarko, C. R.; DiMare, M. J . Org. Chem. 1995,

60, 1777-1787. (b) This model is further supported by quantum

chemical calculations, see: Garc´ıa, J . I.; Mart´ınez-Merino, V.; Mayoral,

J . A. J . Org. Chem. 1998, 63, 2321-2324.

(21) For other models, see ref 8 and Gothelf, K. V.; J ørgensen, K.

A. J . Org. Chem. 1995, 60, 6847-6851.

(22) The same facial selectivity is also found in Diels-Alder reac-

tions of quinones promoted by Ti-TADDOLates: Engler, T. A.; Letavic,

M. A.; Lynch, K. O., J r.; Takusagawa, F. J . Org. Chem. 1994 59, 1179-

1183.

F igu r e 2. Possible transition states for Ti-TADDOLate-

promoted reactions of 1 with 2 based on models^5,20,21proposed

by Seebach and DiMare.

2398 J . Org. Chem., Vol. 64, No. 7, 1999

Engler et al.

Sch em e 3

significantly for different applications,^5band structural

data on the complexes are lacking for most.^20,21The

reactions described herein are a case in point. The

stoichiometry required to generate an effective promoter,

devised empirically after substantial effort, differs con-

siderably from those used by Seebach,⁵Narasaka,⁶and

Corey⁸for N-2-butenoyloxazolidinone/cyclopentadiene Di-

els-Alder reactions and other applications; that is, best

results were found with a complex prepared from a 2.5:

2.5:2.5:1 combination of TiCl₄/Ti(Oi-Pr)₄/diol-4/quinone

(i.e., 2 equiv of Ti/diol and 5 equiv of Ti/quinone).

Recently, Seebach^5csuggested that Ti-TADDOLate-

catalyzed reactions may proceed via cationic trigonal-

bipyramidal complexes such as C (Figure 2), formed by

abstraction of Cl^-from complex A by a second Ti(IV)

entity. This suggestion may account for the stoichiometry

required in the quinone-propenylbenzene reactions.

Of obvious concern are the significant differences in

ee’s found for the cyclobutane products in comparison to

the dihydrobenzofurans from the same reactions. This

concern is heightened in reactions of the arylcycloalkenes

in which the major products are usually the dihydroben-

zofurans, formed in low ee, whereas the minor cyclobu-

tane products are found in high ee. In the latter reactions,

bicyclo[3.2.1]-adducts are not found;²³thus the involve-

ment of bicyclic cationic intermediates is questionable.

However, it is not necessary to distinguish between the

direct formation of benzylic carbocations 41-44 (Scheme

3) from their formation via the cycloaddition mechanism

wherein bicyclo[3.2.1]carbocations 37-40 are formed

initially followed by fragmentation as shown. Both mech-

anisms can lead to cyclobutane products. In principle, the

arylcycloalkene can approach the C-5 carbon of the

quinone-Ti complex from the si or re face and in orien-

tations with the aryl group endo or exo with respect to

the pentadienyl moiety (i.e., face selectivity with respect

to the alkene; the exo or endo terms are used for

convenience and need not imply a cycloaddition mecha-

nism). After complexation and either cycloaddition or

alkylation, four bicyclic carbocations 37-40 or benzylic

cations 41-44 are possible, which then lead to products.

Since the ring fusions in the 2+2 adducts are cis in all

cases (trans fused systems are apparently prohibitively

strained), each of the carbocations 37-40, or 41-44, gives

rise to a cyclobutane of unique diastereochemistry and

chirality. This is not the case with the dihydrobenzofuran

products; either of two of the carbocationic intermediates,

or cyclobutanes, can produce the same enantiomeric

dihydrobenzofuran. Thus, the stereochemistry of the

cyclobutane products provides the most information

regarding the quinone facial selectivity and orientation

of the approaching arylcycloalkene.

The structures of (-)-24/26 from reactions promoted

by the Ti-(+)-4 complex suggest a pathway leading via

intermediates 38 or 42, i.e., C-5 re attack, aryl group exo

(Scheme 3). On the other hand, the structure of the

arylcycloheptene 2+2 adduct, (-)-29, indicates a route

via 37 or 41, i.e., C-5 re attack, aryl group endo. These

results again establish re face selectivity with respect to

the C-5 carbon of the quinone.²²Reasons for the apparent

switch from an endo-aryl orientation in reactions of the

propenylbenzenes 1 and arylcycloheptene 28 to an exo

orientation in reactions with arylcyclopentene 22 and

arylcyclohexene 23 are not apparent. In the absence of

structural information for the chiral Ti-promoter or its

complex with the quinone, explanations for this change

in orientation would be highly speculative. The routes

to the dihydrobenzofuran enantiomers (+)-25/27 and (+)-

30 each could involve two different cations in the series

37-40 or 41-44. They could also result from rearrange-

ment of two different cyclobutane products.

(23) Bicyclo[3.2.1]-adducts have however been found in reactions of

arylcycloalkenes with quinones promoted by simple mixtures of TiCl₄/

Ti(Oi-Pr)₄; see ref 2e.

As previously mentioned, the difference in enantio-

meric purities of the cyclobutanes versus the dihydroben-

Asymmetric Reactions of 2-Methoxy-1,4-benzoquinones

J . Org. Chem., Vol. 64, No. 7, 1999 2399

Ta ble 6. Rea r r a n gem en t of (-)-24^ato (+)-25 P r om oted by Ti-TADDOLa tes

entry

reactants: ratio of TiCl₄:Ti(Oi-Pr)₄:(+)-4:24

quench temp (°C)^b

time (min)

(-)-24 % consumption^c

(+)-25 % ee^d

2.5:2.5:2.5:1

1.6:1.6:1.6:1

-78

-75

-60

-40

-20

-5

>99

100

100^e

entry

reactants: ratio of TiCl₄:Ti(OiPr)₄:(-)-4:24

quench temp (°C)^b

time (min)

(-)-24 % consumption^c

(+)-25 % ee^d

2.5:2.5:2.5:1

1.6:1.6:1.6:1

-78

-75

-60

>99

100

100^e

Enantiomerically pure. Reactants were combined at -78 °C and allowed to warm to this temperature over the time indicated. ^cBased

on disappearance of (-)-24 by HPLC. By HPLC; 100 indicates that only one enantiomer was observed. ^eBy HPLC and NMR (see text)

after workup and flash chromatography.

zofurans from the same reaction mixtures raises ques-

tions about a common mechanism for their formation.

Possible explanations include (1) entirely different mech-

anisms for formation of the two types of products, (2)

formation of cationic intermediates 37-40 or 41-44, in

some unknown proportion, followed by a kinetic parti-

tioning in their collapse to the products, or (3) racem-

ization during rearrangement of one or more of the

cyclobutanes to the dihydrobenzofurans. With respect to

the possibility of entirely different mechanisms, there is

no evidence for other intermediates. Thus, this possibility

remains open, but will not be discussed further. An

explanation based on kinetic partitioning involves a

complex set of kinetic expressions, and information

required for a complete analysis is lacking. To explain

the amount of any one product requires knowing some-

thing about the relative rates of formation of each

intermediate 37-40 and/or 41-44 and the relative rates

of their respective collapse to two different products. The

possibility that some of the individual steps in this

scheme may be reversible (see below) further complicates

matters.

Racemization during conversion of the cyclobutanes to

dihydrobenzofurans seems unlikely on initial consider-

ation, since enantiomerically pure cyclobutanes (-)-24/

26/29 undergo protic acid-catalyzed rearrangement to

dihydrobenzofurans (+)-25/27/30 without loss of enan-

tiomeric purity. However, these rearrangements are done

at room temperature using protic acid (p-TsOH) while

the cycloaddition experiments involve a Ti(IV)-Lewis

acid at -78 °C. Thus, a rearrangement under the

cycloaddition reaction conditions must occur at low

temperatures and likely involve the chiral Lewis acid.

In the latter event, it is possible that the rearrangement

process is slowed enough for a racemization via a depro-

tonation-protonation sequence on the carbocationic in-

termediates (eq 11) to compete with dihydrobenzofuran

formation.

tane (-)-24 was chosen because (a) it could be readily

prepared in sufficient quantities and in high enantio-

meric excess from the cycloaddition reaction, (b) a single

recrystallization then afforded enantiomerically pure

material, and (c) its ee and that of (+)-25 from the same

cycloaddition reaction were dramatically different. These

experiments were carried out with Ti-TADDOLates

prepared from both (+)-4 and ent-(-)-4. Thus, (-)-24 was

treated with 1:1:1 mixtures of TiCl₄/Ti(Oi-Pr)₄/(+)- or (-)-

diols 4, 2.5 equiv of each with respect to the cyclobutane,

and the reactions were monitored by chiral HPLC. After

complete disappearance of the cyclobutane, the reactions

were worked up and chromatographed to isolate (+)-25.

The relative rates of disappearance of (-)-24 with both

enantiomeric Ti(IV)-TADDOLates were very high even

at low temperature (Table 6, entries 1 and 8). To explain

the sizable amounts of the cyclobutanes isolated from the

original cycloaddition reactions, it was postulated that

the number of equivalents of the free chiral Lewis acid

with respect to the cyclobutane in the original reaction

mixture was lower than that in the rearrangement

experiments now described. Some of the Lewis acid was

likely bound to other species present in the reaction

mixtures (the dihydrobenzofuran, the quinone, and/or the

alkene).

Efforts were then made to modify the conditions to slow

the rearrangement and perhaps to more accurately mimic

the cycloaddition conditions. Thus, rearrangements of

cyclobutane (-)-24 to (+)-25 promoted by complexes

formed from 1.6 equiv of each TiCl₄/Ti(Oi-Pr)₄/TAD-

DOLs-4 were studied (Table 6). The results clearly

demonstrated that the rearrangement of (-)-24 promoted

by Ti-(+)-4 proceeded more slowly than with Ti-ent-

(-)-4. For example, at -60 °C the rearrangement of (-)-

24 with Ti-ent-(-)-4 was almost complete in 20 min

(>99% consumption of the cyclobutane), while the rear-

rangement with Ti-(+)-4 stood at 32% conversion (Table

6, entries 10 and 3) and was at 87% conversion only after

40 min at -20 °C (entry 5). The faster reaction with Ti-

ent-(-)-4 was clean and free of ent-(-)-25 by HPLC

analysis. The slower reaction promoted by Ti-(+)-4 was

not as clean, and although at longer reaction times the

major product was clearly (+)-25, sizable peaks with

retention times similar to ent-(-)-25 were apparent in

the crude reaction mixture. A co-inject with authentic

rac-25 did not discount the presence of small amounts

of ent-(-)-25. However, in both reactions the dihydroben-

zofuran product (+)-25 was isolated chromatographically

To study the possibility of an in situ cyclobutane to

dihydrobenzofuran rearrangement, attempts were made

to resubject enantiomerically pure cyclobutane (-)-24 to

the reaction conditions, with interesting results. Cyclobu-

2400 J . Org. Chem., Vol. 64, No. 7, 1999

Engler et al.

and found to be a single enantiomer (∼80% yields) by

the chiral HPLC and NMR assays.

rearrangements are not catalytic in Ti(IV) since a Ti(IV)-

phenoxide results with loss of H⁺. The protic acid

generated could alter the nature of the Ti-TADDOLate

or effect the rearrangement of each enantiomer of 24 to

the respective enantiomers of 25 in an enantiospecific

manner, thus leading to racemic product. These potential

kinetic resolutions were not explored further due to the

limited availability of rac-24.

This study established that (1) Ti-TADDOLate mix-

tures do promote the rearrangement of the cyclobutanes

to the dihydrobenzofurans apparently with little or no

loss of enantiomeric purity, and (2) the rearrangement

proceeds at significantly different rates with the enan-

tiomeric Ti-TADDOLate’s, i.e., Ti-(+)-4 versus Ti-ent-

(-)-4.

Taken as a whole, the data indicate that the facial or

exo/endo selectivities in the formation of 37-40 or 41-

44 (Scheme 3) are probably not in fact as high as implied

by the ee’s of the isolated cyclobutane products. For

example, in reactions involving the Ti-(+)-4 TADDOL-

ate, if one assumes that (1) both enantiomers of cyclobu-

tane 24 are formed, with (-)-24 major as a result of some

facial selectivity and complete exo or endo selectivity, (2)

under the reaction conditions the minor isomer ent-(+)-

24 rearranges faster than (-)-24, and (3) these processes

are occurring exclusively, then the dihydrobenzofuran

product isolated should be enriched in ent-(-)-25. But

this is not the case. The yield of the isolated product (+)-

25 coupled with what is known about the rearrangement

of (-)-24 to (+)-25 with Ti-(+)-4 (slow) as compared to

that of ent-(+)-24 to ent-(-)-25 with Ti-(+)-4 [fast, i.e.,

the same as (-)-24 with Ti-ent-(-)-4] indicates that

dihydrobenzofuran (+)-25 must be formed by some mech-

anism other than rearrangement of (-)-24. If this alter-

nate route gives mainly (+)-25, such a process occurring

simultaneously with rearrangement of the minor cyclo-

butane isomer ent-(+)-24 results in enrichment of iso-

lated (-)-24 and low ee of (+)-25. A similar argument

can be made assuming complete facial selectivity and

some exo/endo selectivity.

Ap p lica tion s

The 2-aryl-2,3-dihydrobenzofuran framework is imbed-

ded in the structure of the pterocarpan family of natural

products. Pterocarpans and related isoflavonoids exhibit

a variety of biological activities, particularly as antifungal

and antibacterial agents.²⁴Some pterocarpans also show

significant anti-HIV activity, as HIV-1 reverse tran-

scriptase inhibitors,^2hand represent a unique and un-

derexplored class of such agents. Because of these

biological activities, there has been considerable interest

recently in asymmetric syntheses of pterocarpans.²⁵

In the present study, reactions of 2H-chromenes 45a /b

with quinones 2 and 3 were developed as efficient and

expedient asymmetric syntheses of pterocarpans (Scheme

4 and Table 7), particularly those with oxygen substit-

Sch em e 4

Thus, some type of kinetic partitioning from intermedi-

ates 37-40 or 41-44 (Scheme 3) likely explains the

formation of cyclobutanes 24/26 and dihydrobenzofurans

25/27 in different ee’s. It is not clear whether a similar

argument can be made to explain the difference in the

ee’s found in products (-)-29 and (+)-30 from reactions

of arylcylcoheptene 28 with quinone 2. The low yields

and modest ee’s found for (-)-29 (diastereomeric with 24/

26) coupled with a lack of a supply of ent-29 discouraged

the exploration of relative rates of rearrangement of these

cyclobutanes with the enantiomeric Ti-TADDOLates.

The cyclobutane to dihydrobenzofuran rearrangements

mediated by the chiral Ti-TADDOLates described above

establish that the rates of these processes are consider-

ably different for isomeric cyclobutanes. Different rear-

rangement rates are not unexpected for diastereomeric

cyclobutanes, and the case of 24 versus ent-24 represents

a kinetic resolution.¹To explore the latter, rac-24 was

treated with the chiral Ti-TADDOLate mixtures under

various conditions (0.25-2.5 equiv of the Ti-TAD-

DOLate, -85 to -10 °C, 15-60% conversion), after which

the reactions were worked up and recovered starting

material and products isolated and analyzed for enan-

tiomeric enrichment. There was little evidence of kinetic

resolution (% ee of product 25 < 6%; % ee of recovered

24 < 5%). There are, however, significant differences in

the conditions of these experiments and those of the

original cycloaddition of 22 and 2. The latter reactions

likely contain, in addition to the Ti-TADDOLate and

chiral products 24/25, unreacted quinone and arylcy-

cloalkene, which could alter the nature of the Lewis acid.

A second, perhaps more significant issue is that the

uents at C-3 and C-9.²⁶As observed in the reactions

promoted by simple mixtures of TiCl₄/Ti(Oi-Pr)₄,^2creac-

tions of 45b, bearing a C-7 methoxy substituent, provided

the pterocarpan products (+)-50/51, whereas reactions

of 45a at low temperatures yielded the cyclobutane

(24) See references cited in ref 2c,h and (a) van Aardt, T. G.; van

Heerden, P. S.; Ferreira, D. Tetrahedron Lett. 1998, 39, 3881-3884.

For reviews, see: (b) Dean, F. M. In Total Synthesis of Natural

Products, Vol 1; ApSimon, J ., Ed.; Wiley-Interscience: New York, 1973;

pp 467-562. (c) J ain, A.C; Tuli, D. K. J . Sci. Ind. Res. 1978, 37, 287-

304. (d) Ingham, J . L. In Progress in the Chemistry of Organic Natural

Products, Vol. 43; Herz, W., Grisebach, H., Kirby, G. W., Eds.; Springer-

Verlag: Wien, 1983; pp 1-266. (e) Dewick, P. M. In The Flavonoids,

Advances in Research Since 1986; Harborne, J . B., Ed.; Chapman &

Hall: London, 1994; Chapter 5. (f) Boland, G. M.; Donnelly, D. M. X.

Nat. Prod. Rep. 1998, 15, 241-260, and previous reviews in these

series.

(25) For recent examples, see: (a) Versteeg, M.; Bezuidenhoudt, B.

C. B.; Ferreira, D.; Swart, K. J . J . Chem. Soc., Chem. Commun. 1995,

1317-1318. (b) Pinard, E.; Gaudry, M.; He´not, F.; Thellend, A.

Tetrahedron Lett. 1998, 39, 2739-2742. (c) Mori, K.; Kisida, H. Liebigs

Ann. Chem. 1988, 721-723. For enantiomeric separation of pterocar-

pans by HPLC, see: (d) Antus, S.; Bauer, R.; Gottsegen, A.; Wagner,

H. J . Chromatogr. 1990, 508, 212-216. For a recent racemic synthe-

sis: (e) Miki, Y.; Fujita, R.; Matsushita, K.-i. J . Chem. Soc., Perkin

Trans. 1 1998, 2533-2536.

(26) Limited SAR studies indicate that C-3/-9 oxygenation may be

important for biological activity; see refs 2c,h.

Asymmetric Reactions of 2-Methoxy-1,4-benzoquinones

J . Org. Chem., Vol. 64, No. 7, 1999 2401

Ta ble 7. Ti-TADDOLa te-P r om oted Rea ction s of

Qu in on es 2/3 w ith 2H-Ch r om en es 45a /b^a

3: Reaction of 1a with 2 Is Repr esen tative. (1R,6R,7R,8R)-

3-Meth oxy-8-(3,4-dim eth oxyph en yl)-7-m eth ylbicyclo[4.2.0]-

oct-3-en -2,5-d ion e (5a ) a n d (2S,3S)-6-Meth oxy-2-(3,4-d i-

m et h oxyp h en yl)-3-m et h yl-2,3-d ih yd r o-5-b en zofu r a n ol

(6a ). To a solution of Ti(Oi-Pr)₄(2.63 mL, 8.84 mmol) in CH₂-

Cl₂(5 mL) at room temperature was added TiCl₄(966 µL, 8.81

mmol) followed rapidly by the addition of a solution of diol

(+)-4 (4.71 g, 8.91 mmol) in CH₂Cl₂(10 mL). The mixture was

then added to a solution of 2 (0.495 g, 3.58 mmol) in CH₂Cl₂

(30 mL) at -78 °C (or -94 °C, see Table 2: the order of

addition was not important; in some reactions, a solution of

the quinone in CH₂Cl₂was added to the Ti(IV)-diol-4 solu-

tion). The reaction mixture was stirred at -78 °C (or -94 °C)

for 15 min, and then 1a (644 µL, 3.81 mmol) was added

dropwise. The reaction mixture was stirred for 2 h at -78 °C

(or the time and temperature indicated in Tables 2 and 7),

and then NaHCO₃(8.06 g) and i-PrOH (20 mL) were added at

-78 °C. The mixture was diluted with water, filtered through

Celite, and extracted with CH₂Cl₂(3 × 250 mL). The organic

extracts were combined, dried (Na₂SO₄), and concentrated.

Chromatography of the resulting oil on flash silica with 20%,

60%, and then 70% EtOAc/hexanes as eluents gave recovered

diol (+)-4 (routinely >90%, which was pooled for recycling),

6a (0.105 g, 9%, 41% ee), and 5a (0.977 g, 86%, 92% ee).

Recrystallization of 5a from EtOAc/hexanes provided enan-

chromene^b

quinone

time (h) product

% yield^c% ee^d

45a

45b

1.5

0.5

All reactions were conducted in CH₂Cl₂at -78 °C. The Ti-

TADDOLate promoter was prepared in the same manner as

described for reactions of 1. 1.1-1.5 equiv with respect to the

quinone. ^cIsolated yields. By ¹H NMR in the presence of 11.

products (+)-46/47. All products were found in 60-84%

ee. The enantiomeric purities were determined by ¹H

NMR analysis in the presence of 11.^7,11

Recrystallization of enantiomerically enriched cyclobu-

tane (+)-46 gave enantiomerically homogeneous material.

Treatment of the latter with H₂SO₄effected its rear-

rangement to pterocarpan (+)-48 without loss of enan-

tiomeric purity. Similarly, H⁺-promoted rearrangement

of a sample of (+)-47 which was of 61% ee gave ptero-

carpan (+)-49 with the same enantiomeric purity. The

absolute configuration of synthetic (+)-50 was assigned

on the basis of comparison with its naturally occurring

enantiomer.^27aThe stereochemistry of the other ptero-

carpans prepared herein was then assigned by analogy

and by comparison of their optical rotations^27b(at several

different wavelengths) and NMR spectra in the presence

of (-)-11 to those of (+)-50. The relative stereochemistry

in racemic cyclobutanes (+)-46/47 has been determined

previously by ¹H-¹H NOE experiments,^2cand their

rearrangement to pterocarpans (+)-48/49 establishes the

absolute stereochemistry of 46/47.

tiomerically homogeneous product; mp 137-138 °C, [R]₅₈₉

)

-295 (c 12.8 mg/mL, CHCl₃). Repeated recrystallizations (up

to five) did not increase the specific rotation.

The following reactions were conducted in a similar manner

with the same ratios but variable quantities of reactants (for

reaction temperatures and times, see Tables 2 and 7).

1b with 2 gave a 20.3:1 mixture of 5b (0.255 g, 86%, 90%

ee) and an unidentified isomer and 6b (0.031 g, 11%, 53% ee);

the optical rotation of this sample of 5b was [R]₅₉₈) -273 (c

10 mg/mL, CHCl₃).

1c with 2 gave a 19.9:1 mixture of 5c (0.101 g, 71%) and an

unidentified isomer; the major isomer was 86% ee by ¹H NMR

and the optical rotation of this sample was [R]₅₈₉) -225 (c 10

mg/mL, CHCl₃).

1d with 2 gave 7d (0.089 g, 64%, 84% ee).

1e with 2 gave 7e (0.150 g, 61%, 90% ee).

1a with 3 gave 8a (0.407 g, 35%, 88% ee) and 9a (0.564 g,

48%, 83% ee).

1b with 3 gave 8b (0.754 g, 72%, 87% ee) and 9b (0.162 g,

15%, 78% ee).

2-Aryl-3-methyl-2,3-dihydrobenzofuran and bicyclo-

[3.2.1]octendione structures are also found in numerous

other biologically active natural products.²⁸The enanti-

oselective reactions described herein offer a convenient

asymmetric route to many of these systems and hold

considerable potential for additional applications as well.

1c with 3 gave 8c (0.014 g, 10%, 70% ee) and 10c (0.084 g,

61%, 92% ee); the optical rotation of this sample of 10c was

[R]₅₈₉) +52, [R]₅₇₈) +55, [R]₅₄₆) +64, [R]₄₃₆) +65 (c 1.8

mg/mL CHCl₃).

1d with 3 gave 10d (0.107 g, 79%, 96% ee); the optical

rotation of this sample of 10d was [R]₅₈₉) +381, [R]₅₇₈) +401,

[R]₅₄₆) +478, [R]₄₃₆) +1080, [R]₃₆₅) +2914 (c 1.5 mg/mL

CHCl₃).

17 with 2 gave a 10:1 ratio of 18 and 5e (0.057 g, 42%, 78%

ee for the major isomer).

Indene with 2 gave 19 (0.115 g, 85%, 63% ee); the optical

rotation of this sample was [R]₅₈₉) -6.8 (c 11 mg/mL, CHCl₃).

Indene with 3 gave gave 20 (0.033 g, 24%, 34% ee) and 21

Exp er im en ta l Section

Gen er a l P r oced u r es. Diols-4,^5,6propenylbenzenes 1/17,^2e

1-anisylcycloalkenes,^2eand 2H-chromenes 45a /b^2cwere pre-

pared as previously described. All reactions were conducted

under an atmosphere of N₂and were monitored by TLC.

Melting points are uncorrected. NMR spectra were recorded

on samples dissolved in CDCl₃, unless indicated otherwise, and

data are reported in δ (ppm) relative to TMS or residual CHCl₃

as internal standards. Coupling constants are reported in hertz

(Hz). IR data are reported in cm^-1. Spectral data for (-)-5a -

c, (-)-6a -c, (+)-7d ,e, (-)-8a -c, (+)-9a ,b, (+)-10c,d , (-)-18,19,

(+)-20, -21, -25, -27, -30, -46-51, all white solids unless stated

otherwise, were identical to those of their racemates previously

reported.^2c,eOptical rotations were taken at ambient temper-

ature.

(0.090 g, 68%, 51% ee) as oils. For this sample of 20, [R]₅₈₉

)

+4, [R]₅₇₈) +2, [R]₅₄₆) +4, [R]₄₃₆) +1 (c 2.1 mg/mL CHCl₃);

for 21, [R]₅₈₉) +177, [R]₅₇₈) +186, [R]₅₄₆) +218, [R]₄₃₆

+241 (c 9.2 mg/mL CHCl₃).

45a with 2 gave 46 (0.12 g, 86%, 84% ee).

45a with 3 gave 47 (0.093 g, 64%, 61% ee).

)

Gen er al P r ocedu r e for Cycloaddition s of P r open ylben -

zen es 1 a n d 2H-Ch r om en es 45a /b w ith Qu in on es 2 a n d

45b with 2 gave 50 (0.116 g, 77%, 75% ee); the optical

rotation for this sample was [R]₅₈₉) +124 (c 23.4 mg/mL,

CHCl₃).

45b with 3 gave 51 (0.135 g, 83%, 80% ee); the optical

rotation for this sample was [R]₅₈₉) +113 (c 14 mg/mL,

CHCl₃).

Recrystallization of the following products from EtOAc/

hexanes gave enatiomerically homogeneous samples: 7d , mp

212-214 °C, [R]₅₈₉) +368, [R]₅₄₆) +463, [R]₄₃₆) +1046, [R]₃₆₅

) +2940 (c 2.9 mg/mL, CHCl₃). Anal. Calcd for C₁₆H₁₆O₃: C,

74.98; H, 6.29. Found: C, 74.70; H, 6.25. 7e, mp 200-202 °C,

(27) (a) Bezuidenhoudt, B. C. B.; Brandt, E. V.; Ferreira, D.

Phytochemistry 1987, 26, 531-535. A sample of (-)-50 was kindly

provided by Professor Ferreira: natural (-)-50, [R]₅₈₉) -155 (c 5.3

mg/mL, CHCl₃); synthetic (+)-50 (75% ee by ¹H NMR analysis with

11), [R]₅₈₉) +124 (c 23 mg/mL). (b) In the pterocarpan series, a

dextrorotatory [R] is indicative of (6aS,11S) configuration, see refs 27

and 24e.

(28) For reviews: (a) Ward, R. S. Nat. Prod. Rep. 1997, 14, 43-74,

and previous reviews in this series. (b) Gottlieb, O.; Yoshida, M. In

Natural Products of Woody Plants I; Rowe, J . W., Ed.; Springer-

Verlag: Berlin, 1989; Chapter 7.3.

2402 J . Org. Chem., Vol. 64, No. 7, 1999

Engler et al.

[R]₅₈₉) +353, [R]₅₇₈) +374, [R]₅₄₆) +446, [R]₄₃₆) +1019,

[R]₃₆₅) +2953 (c 4.1 mg/mL, CHCl₃). 8a , mp 133.5-134.0 °C,

[R]₅₈₉) -166, [R]₅₇₈) -179, [R]₅₄₆) -225, [R]₄₃₆) -918 (c

5.8 mg/mL, CHCl₃). Anal. Calcd for C₁₉H₂₂O₅: C, 71.98; H,

6.71. Found: C, 71.78; H, 6.87. 8b, mp 134.0-134.2 °C, [R]₅₈₉

) -172, [R]₅₇₈) -184, [R]₅₄₆) -229, [R]₄₃₆) -835 (c 12.5

mg/mL CHCl₃). Anal. Calcd for C₁₈H₂₀O₄: C, 69.08; H, 6.71.

Found: C, 68.89; H, 6.88. 18, mp 127.5-129 °C, [R]₅₈₉) -229

(c 5.4 mg/mL, CHCl₃). 46, mp 166-167 °C, [R]₅₈₉) +44.4 (c

8.6 mg/mL, CHCl₃).

extracts were washed with brine (50 mL), dried (Na₂SO₄), and

concentrated to a yellow oil. Chromatography using 10%

EtOAc/hexanes as eluent afforded (+)-12 (0.044 g, 63%) as a

clear, colorless oil. Crystallization from i-PrOH/hexanes gave

clear, colorless needles; mp 126-127 °C; TLC R_f) 0.54 (30%

EtOAc/hexanes); ¹H NMR (500 MHz) 7.17-7.12 (m, 3H), 6.78-

6.76 (m, 1H), 6.14 (s, 1H), 3.86 (dd, J ) 1.7, 6.9 Hz, 1H), 3.48

(dd, J ) 5.3, 6.9 Hz, 1H), 3.38 (d, J ) 1.7 Hz, 1H), 2.84 (dq, J

) 6.9, 7.0, 1H), 2.37 (s, 3H), 1.28 (d, J ) 7.0 Hz, 1H); ¹³C NMR

(125 MHz) 196.9, 187.3, 148.3, 137.0, 135.5, 130.8, 127.7, 126.7,

126.4, 118.3, 66.2, 64.2, 46.3, 40.7, 21.7, 20.0; HRMS m/z

334.0209 (calcd for C₁₆H₁₅BrO₃, 334.0205); [R]₅₈₉) +300° (c

1.9 mg/mL, CHCl₃).

Gen er a l P r oced u r e for Rea r r a n gem en ts of 5/8 to 6/9

w ith Ti(IV)-d iol 4: Rea r r a n gem en t of 5a to 6a Is

Rep r esen ta tive. To a solution of Ti(Oi-Pr)₄(450 µL, 1.51

mmol) in CH₂Cl₂(1 mL) at room temperature was added TiCl₄

(165 µL, 1.50 mmol) followed rapidly by the addition of a

solution of diol (+)-4 (0.793 g, 1.50 mmol) in CH₂Cl₂(1 mL).

The mixture was added to a solution of enantiomerically

homogeneous cyclobutane (-)-5a (0.720 g, 2.28 mmol) in CH₂-

Cl₂(10 mL) at -78 °C. The reaction mixture was stirred at

-78 °C for 2 h and then allowed to slowly warm to -10 °C (or

the temperature indicated in Table 3) over 4 h. NaHCO₃(2.51

g) and i-PrOH (5 mL) were added, and the mixture was diluted

with water, filtered through Celite, and extracted with CH₂-

Cl₂(3 × 100 mL). The organic extracts were combined, dried

(Na₂SO₄), and concentrated. Chromatography of the resulting

oil on flash silica with 20%, 35%, and then 50% EtOAc/hexanes

as eluents gave a >50:1 mixture of 6a (0.676 g, 94%, 100% ee)

and its cis isomer.

Dim eth yl (R)-2-Meth ylsu ccin a te (15).^15-17A solution of

dihydrobenzofuran (+)-9a (1.01 g, 3.057 mmol) in AcOH (50

mL) was added to 10% Pd/C (0.522 g). The mixture was stirred

under an atmosphere of H₂for 48 h, then poured into saturated

aqueous NaHCO₃, and filtered. The filtrate was extracted with

CH₂Cl₂(3 × 200 mL). The organic extracts were combined,

dried (MgSO₄), and concentrated. Chromatography of the

resulting oil on flash silica with 20% and then 40% EtOAc/

hexanes as eluents gave a mixture of the quinone and the

hydroquinone (0.769 g), which was dissolved in AcOH (30 mL).

Ozone was bubbled through the solution for 4 h. The resulting

mixture was concentrated, 30% aqueous H₂O₂was added, and

the mixture was heated to reflux for 1 h. The solvent was

removed under reduced pressure, THF was added to dissolve

the oil, and the mixture was added to a solution of CH₂N₂in

diethyl ether at 0 °C. The mixture was stirred for 1 h and was

poured into saturated aqueous NH₄Cl and extracted with CH₂-

Cl₂(3 × 150 mL). The organic extracts were combined, dried

(Na₂SO₄), and concentrated. Chromatography of the resulting

oil on flash silica with 15% EtOAc/hexanes as eluent gave 15

(0.054 g, 11%), TLC R_f(35% EtOAc/hexanes, visualized with

Protic acid-catalyzed rearrangements²afforded 6a in 8-12:1

trans:cis ratios, and from the latter experiments, the trans and

cis isomers were isolated by HPLC (µporasil, 0.5% 2-propanol/

hexanes). Physical and spectral data for trans-6a : mp 131-

132 °C (EtOAc/hexanes); H NMR (500 MHz) 1.35 (d, J ) 7,

3H), 3.39 (dq, J ) 7, 9, 1H), 3.87 (s, 3H), 3.88 (s, 3H), 5.04 (d,

J ) 9, 1H), 5.25 (s, 1H), 6.49 (s, 1H), 6.73 (d, J ) 1, 1H), 6.86

(d, J ) 8, 1H), 6.9-7.0 (m, 2H); ¹³C NMR (125 MHz) 17.9, 45.3,

55.89, 55.93, 56.2, 93.1, 94.2, 109.1, 109.4, 110.9, 118.9, 123.0,

iodine) 0.49; H NMR (500 MHz) 1.20 (d, J ) 7, 3H), 2.39 (d,

J ) 17, 6, 1H), 2.73 (dd, J ) 17, 8, 1H), 2.90 (sextet, J ) 7,

1H), 3.66 (s, 3H), 3.68 (s, 3H); ORD (c 10.7 mg/mL, CHCl₃);

[R]₅₈₉) +4.4; lit.²⁹[R]₅₈₉) +4.1 (c 4.1 mg/mL, CHCl₃).

133.1, 139.9, 146.2, 149.0, 149.2, 152.3; [R]₅₈₉) -32, [R]₅₇₈

)

-33, [R]₅₄₆) -40, [R]₄₃₆) -97, [R]₃₆₅) -282 (c 2.6 mg/mL,

CHCl₃). Spectral data for cis-6a (92% ee): ¹H NMR (500 MHz)

0.78 (d, J ) 7, 3H), 3.55 (dq, J ) 7, 9, 1H), 3.87 (s, 3H), 3.88

(s, 3H), 3.89 (s, 3H), 5.24 (s, 1H), 5.72 (d, J ) 9, 1H), 6.54 (s,

1H), 6.75 (s, 1H), 6.85 (s, 3H); ¹³C NMR (125 MHz) 17.1, 41.2,

55.87, 55.88, 56.2, 88.0, 94.1, 109.5, 110.1, 110.8, 118.7, 123.9,

(2S,3S)-4-Allyl-2-(3,4-d im eth oxyp h en yl)-6-m eth oxy-3-

m eth yl-2,3-d ih yd r o-5-ben zofu r a n ol (16). To a solution of

dihydrobenzofuran (-)-6a (0.668 g, 2.11 mmol) in acetone (30

mL) was added K₂CO₃(1.57 g) and allyl bromide (0.75 mL,

8.7 mmol). The mixture was heated to reflux for 48 h and then

cooled and poured into saturated aqueous NaHCO₃. The

mixture was extracted with CH₂Cl₂(3 × 100 mL), and the

organic extracts were combined, dried (Na₂SO₄), and concen-

trated. Chromatography of the resulting oil on flash silica with

15% and then 20% EtOAc/hexanes as eluents gave the allyl

ether of 6a (0.601 g, 80%); mp 68.0-68.5 °C (EtOAc/hexanes);

130.7, 139.7, 146.1, 148.4, 148.7, 152.2; [R]₅₈₉) +140, [R]₅₇₈

)

+148, [R]₅₄₆) +172, [R]₄₃₆) +336, [R]₃₆₅) +548 (c 3.1 mg/

mL, CHCl₃).

The following reactions were conducted in a similar manner

(for reaction temperatures, see Table 3).

A sample of enantiomerically enriched 5b (90% ee) afforded

a 14.5:1 ratio of 6b and its cis isomer (0.197 g, 79%, 93% ee

for the major isomer).

A sample of enantiomerically enriched 5c (86% ee) afforded

a 17:1 ratio of 6c and its cis isomer (0.098 g, 93%, 82% ee for

the major isomer).

Enantiomerically homogeneous 8a afforded pure trans-9a

(0.111 g, 100%, 100% ee) as an oil; [R]₅₈₉) +10.3 (c 11.1 mg/

mL, CHCl₃).

TLC R_f(35% EtOAc/hexanes) 0.34; H NMR (500 MHz) 1.36

(d, J ) 7, 3H), 3.39 (m, 1H), 3.84 (s, 3H), 3.88 (s, 3H), 3.89 (s,

3H), 4.55 (d, J ) 6, 2H), 5.06 (d, J ) 9, 1H), 5.26 (dd, J ) 1,

11, 1H), 5.39 (dd, J ) 1, 17, 1H), 6.09 (ddd, J ) 6, 11, 17, 1H),

6.52 (s, 1H), 6.74 (s, 1H), 6.86 (d, J ) 8, 1H), 6.95 (dd, J ) 2,

8, 1H), 6.97 (d, J ) 2, 1H); ¹³C NMR (125 MHz) 17.9, 45.3,

55.8, 55.9, 56.1, 71.6, 93.2, 95.2, 109.1, 110.9, 111.2, 117.5,

118.8, 122.0, 133.0, 133.9, 142.4, 149.0, 149.1, 150.3, 153.7;

HRMS m/z 356.1628 (calcd for C₂₁H₂₄O₅356.1624); ORD (c 22.0

mg/mL, CHCl₃) [R]₅₈₉) -4.2, [R]₅₇₈) -4.9, [R]₅₄₆) -6.3, [R]₄₃₆

) -22.1, [R]₃₆₅) -100. Anal. Calcd for C₂₁H₂₄O₅: C, 70.77;

H, 6.79. Found: C, 70.88; H, 6.80.

Enantiomerically homogeneous 8b afforded a 17.0:1 ratio

of 9b and its cis isomer (0.094 g, 83%, 100% ee) as a colorless

oil; the optical rotation of this sample was [R]₅₈₉) +31.7 (c

9.4 mg/mL, CHCl₃).

A solution of the allyl ether prepared as described above

(0.270 g, 0.758 mmol) in N,N-dimethylaniline (5 mL) was

heated to reflux for 4 h. The mixture was cooled to room tem-

perature and poured into aqueous HCl and ice. The mixture

was shaken and then extracted with CHCl₃(3 × 100 mL). The

organic extracts were combined, dried (Na₂SO₄), and concen-

trated. Chromatography of the resulting oil on flash silica with

10% and then 15% EtOAc/hexanes as eluents gave 16 (0.220

g, 81%) as a colorless oil; TLC R_f(35% EtOAc/hexanes) 0.39;

¹H NMR (500 MHz) 1.40 (d, J ) 7, 3H), 3.3-3.5 (m, 3H), 3.84

Recrystallization of 6b and 6c from EtOAc/hexanes provided

stereo- and enantiomerically homogeneous material: trans-

6b, mp 112.5-114.0 °C; [R]₅₈₉) -31.4 (c 6.6 mg/mL, CHCl₃);

trans-6c, mp 141-141.5 °C; [R]₅₈₉) -47.1 (c 36.2 mg/5 mL,

CHCl₃).

(+)-(1R,5R,6R,7R)-4-Br om o-3-h yd r oxy-6-m et h yl-7-(2-

m eth ylp h en yl)bicyclo[3.2.1]oct-3-en -2,8-d ion e (12). Bicy-

clo[3.2.1]-adduct (+)-7d (0.053 g, 0.21 mmol) was dissolved in

THF (2 mL) at 0 °C, and N-bromosuccinimide (0.040 g, 0.22

mmol) was added in small portions. The yellow reaction

mixture was stirred for 5 min and then treated with saturated

aqueous NaHCO₃(20 mL). The aqueous layer was separated

and extracted with CH₂Cl₂(3 × 20 mL). The combined organic

(29) Cohen, S. G.; Milovanovic, A. J . Am. Chem. Soc. 1968, 90,

3495-3502.

Asymmetric Reactions of 2-Methoxy-1,4-benzoquinones

J . Org. Chem., Vol. 64, No. 7, 1999 2403

(s, 3H), 3.85 (s, 3H), 3.86 (s, 3H), 4.99 (m, 1H), 5.02 (dd, J )

2, 2, 1H), 5.08 (d, J ) 6, 1H), 5.29 (s, 1H), 5.99 (m, 1H), 6.82

3.47 (dd, J ) 7.2, 7.6 Hz, 1H), 3.34 (d, J ) 7.2 Hz, 1H), 3.25

(bt, J ) 7.6 Hz, 1H), 2.29 (bd, J ) 13 Hz, 1H), 1.96 (dt, J )

3.0, 13 Hz, 1H), 1.88-1.85 (m, 1H), 1.80-1.67 (m, 2H), 1.65-

1.52 (m, 2H), 1.50-1.46 (m, 1H); ¹³C NMR (125 MHz) 198.6,

191.0, 162.7, 158.0, 134.8, 128.2, 113.2, 111.6, 56.1, 55.1, 53.0,

48.3, 44.3, 42.9, 40.0, 24.7, 21.4, 20.3; [R]₅₈₉) -227 (c 1.1 mg/

mL, CHCl₃). Anal. Calcd for C₂₀H₂₂O₄: C, 73.59; H, 6.81.

(d, J ) 8, 1H), 6.85 (d, J ) 2, 1H), 6.88 (dd, J ) 2, 8, 1H); ¹³

NMR (125 MHz) 20.2, 30.7, 44.7, 55.68, 55.71, 55.9, 91.7, 92.1,

108.8, 110.9, 114.8, 118.1, 121.3, 121.8, 134.2, 135.7, 137.6,

146.1, 148.8, 149.0, 151.7; HRMS m/z - 356.1627 (calcd for

₂₁H₂₄O₅356.1624); ORD (c 22.0 mg/mL, CHCl₃) [R]₅₈₉) +66,

[R]₅₇₈) +69, [R]₅₄₆) +80, [R]₄₃₆) +147, [R]₃₆₅) +230.

Gen er a l P r oced u r e for Cycloa d d ition s of Ar ylcyclo-

a lk en es: Rea ction of 22 w ith 2 Is Rep r esen ta tive. (-)-

(4a R,4bR,7a S,7bR)-4,4a ,4b,5,6,7,7a ,7b-Octa h yd r o-2-m eth -

oxy-7a -(4-m eth oxyp h en yl)-1H-ben zo[3,4]cyclobu ta [1,2]-

cyclop en ta n -1,4-d ion e (24) a n d (+)-(3a R,8bR)-2,3,3a ,8b-

t et r a h yd r o-6-m et h oxy-3a -(4-m et h oxyp h en yl)-1H -cyclo-

p en ta [b]ben zofu r a n -7-ol (25). TiCl₄(270 µL, 2.46 mmol) and

a solution of (+)-4 (1.320 g, 2.50 mmol) in CH₂Cl₂(2 mL) were

added rapidly, in sequence, to a solution of Ti(Oi-Pr)₄(740 µL,

2.49 mmol) in CH₂Cl₂(2 mL) at room temperature. The dark

mixture was stirred for 15 min and cooled to -78 °C, and a

solution of 2 (0.139 g, 1.01 mmol) in CH₂Cl₂(2 mL) was added.

After 15 min, the reaction mixture was treated with a solution

of 22 (0.176 g, 1.01 mmol) in CH₂Cl₂(1 mL). The reaction mix-

ture was stirred for 20 min, and solid NaHCO₃(1 g) and

i-PrOH (3 mL) were then added. The mixture was poured

rapidly into H₂O (20 mL) and CH₂Cl₂(10 mL), swirled

vigorously, and filtered through Celite (CH₂Cl₂wash). The

aqueous layer was separated and extracted with CH₂Cl₂(20

mL). The combined organic extracts were washed with brine

(50 mL), dried (Na₂SO₄), decanted, and concentrated. Chro-

matography of the resultant brown oil with 10%, 15%, 30%,

and then 50% EtOAc/hexanes gave recovered diol-4 (typically

>95%, which was pooled for recycling), (+)-25 (0.134 g, 42%,

20% ee) as a colorless oil, and (-)-24 (0.151 g, 48%, 96% ee)

as a white solid.

Recrystallization of (-)-24 from EtOAc/hexanes afforded

enantiomerically homogeneous material as colorless needles:

mp 174-175 °C; TLC R_f) 0.25 (50% EtOAc/hexanes); ¹H NMR

(500 MHz) 7.03 (d, J ) 8.8 Hz, 2H), 6.75 (d, J ) 8.8 Hz, 2H),

5.94 (s, 1H), 3.73 (s, 3H), 3.54 (s, 3H), 3.33 (d, J ) 8.4 Hz,

1H), 3.29 (dd, J ) 3.7, 7.1 Hz, 1H), 2.85 (dd, J ) 3.7, 8.4 Hz,

1H), 2.41 (dd, J ) 6.1, 13 Hz, 1H), 2.08-1.95 (m, 3H), 1.85-

1.77 (m, 1H), 1.64-1.58 (m, 1H); ¹³C NMR (75 MHz) 199.1,

192.2, 163.8, 158.2, 133.8, 128.1, 113.6 (2C), 58.6, 56.0, 55.1,

51.1, 46.2, 44.4, 42.1, 33.0, 25.2; [R]₅₈₉) -149 (c 2.3 mg/mL,

CHCl₃). Anal. Calcd for C₁₉H₂₀O₄: C, 73.05; H, 6.47. Found:

C, 72.60; H, 6.30. NOE data for (-)-24: irradiation of H-4a

(2.85 ppm) resulted in 10% enhancement of the H-7b signal

(3.33 ppm); however, no enhancement of the H-4b signal (3.29

ppm) was observed. Likewise, irradiation of the ortho protons

on the aryl ring (7.03 ppm) caused a 6% enhancement of the

H-4b signal (3.29 ppm), but no significant enhancement of the

H-7b signal.

Found: C, 73.49; H, 6.60. NOE data for (-)-26: a 6%

enhancement of the signal for H-4b (3.25 ppm) was observed

upon irradiation of the aryl ortho protons (7.12 ppm), with no

significant enhancement of the H-8b signal (3.34 ppm), but a

7% enhancement of the H-8b signal was observed upon

irradiation of H-4a (3.47 ppm).

(-)-29: mp 249-250 °C; TLC R_f) 0.16 (30% EtOAc/

hexanes); HPLC (Chiralcel OD; 20% 2-propanol/hexanes; 1 mL/

min; 254 nm), (-)-29 rt 23-24 min, (+)-29 rt 19-20 min; H

NMR (500 MHz) 7.37 (d, J ) 8.8 Hz, 2H), 6.90 (d, J ) 8.8 Hz,

2H), 6.12 (s, 1H), 3.83 (s, 3H), 3.80 (s, 3H), 3.61 (d, J ) 9.5

Hz, 1H), 3.47-3.39 (m, 2H), 2.09-0.82 (m, 10H); ¹³C NMR (125

MHz) 198.2, 192.8, 162.9, 157.7, 140.8, 128.3, 115.0, 113.5,

56.3, 55.3, 53.1, 51.9, 44.0, 40.7, 37.0, 31.0, 30.4, 28.2, 24.6;

HRMS m/z 341.1751 (M⁺+ 1); calcd for C₂₁H₂₄O₄, 341.1753);

[R]₅₈₉) -296 (c 1.6 mg/mL, CHCl₃). NOE data for (-)-29:

irradiation of the C-9a aryl ring protons (7.37 ppm) ortho to

the attachment to the cyclobutane caused 7% enhancements

of both the H-9b doublet (3.62 ppm) and the H-4a/b multiplet

(3.4-3.5 ppm, these signals overlap); in addition, irradiation

of H-9b gave a 5% enhancement of the H-4a/4b multiplet.

r el-(4a R,4bR,7a S,7bR)-4,4a ,4b,5,6,7,7a ,7b-Octa h yd r o-2-

m eth oxy-7a -(4-m eth oxyp h en yl)-1H-ben zo[3,4]cyclobu ta -

[1,2]cyclop en ta n -1,4-d ion e [(+)-24] a n d r el-(3a R,8bR)-

2,3,3a ,8b-Tetr a h yd r o-6-m eth oxy-3a -(4-m eth oxyp h en yl)-

1H-cyclop en ta [b]ben zofu r a n -7-ol [(+)-25]. TiCl₄(0.3 mL,

2.7 mmol) was added to a solution of Ti(Oi-Pr)₄(0.8 mL, 2.7

mmol) in CH₂Cl₂(3 mL) at room temperature. The reaction

mixture was stirred for 15 min and cooled to -85 °C, and a

solution of 2 (0.750 g, 5.4 mmol) in CH₂Cl₂(5 mL) was added.

After 15 min, a solution of 22 (0.940 g, 5.4 mmol) in CH₂Cl₂(3

mL) was added. The reaction mixture was stirred for 1.5 h at

-85 °C and then quenched with solid NaHCO₃(1 g) and

i-PrOH (5 mL). The mixture was partitioned between H₂O/

CH₂Cl₂(10 mL each) and filtered through Celite. The aqueous

layer was separated and extracted with CH₂Cl₂(6 × 20 mL).

The combined organic extracts were washed with brine (10

mL), dried (Na₂SO₄), and concentrated. Chromatography of the

resultant yellow-brown oil with 10%, 15%, 30%, and 50%

EtOAc/hexanes gave starting 22 (0.510 g, 54% recovery), (()-

25 (0.470 g, 28%) as a colorless oil, and (()-24 (0.250 g, 15%)

as a white solid, mp 147-149 °C. ¹H NMR (500 MHz) analyses

in the presence 11 (5 equiv by weight with respect to substrate)

in CDCl₃showed that signals at 5.96 and 6.43 in rac-24 and

rac-25, respectively, split up into two distinct signals. Chiral

HPLC (Chiralcel OD, 20% 2-propanol/hexanes; 1 mL/min; 254

nm) separated the two enantiomers of each compound with

the following retention times: (-)-24, 27-28 min; (+)-24, 22-

23 min; (+)-25, 11-12 min; (-)-25, 9-10 min.

Enantiomeric purities were determined by (1) ¹H NMR (500

MHz) analyses in the presence of 11 (5 equiv by wt with

respect to the substrate) in CDCl₃and/or (2) HPLC (Chiralcel

OD, 20% 2-propanol/hexanes; 1 mL/min; 254 nm) in which the

two enantiomers of each compound appeared as well-separated

peaks. Racemic mixtures of 24 were used as standards and

were prepared by either (1) combination of (-)-24 and (+)-24,

which was made from an identical reaction to the one described

above, but with a complex of Ti(IV) with diol-(-)-4, or (2) from

reactions of the quinone with anisylcyclopentene catalyzed by

simple mixtures of TiCl₄/Ti(iOPr)₄; see the following experi-

mentals.

(+)-(4a S,4bS,7a R,7bS)-4,4a ,4b,5,6,7,7a ,7b-Octa h yd r o-2-

m eth oxy-7a -(4-m eth oxyp h en yl)-1H-ben zo[3,4]cyclobu ta -

[1,2]cyclop en ta n -1,4-d ion e [(+)-24] a n d (-)-(3a S,8bS)-

2,3,3a ,8b-Tetr a h yd r o-6-m eth oxy-3a -(4-m eth oxyp h en yl)-

1H-cyclop en ta [b]ben zofu r a n -7-ol [(-)-25]. This reaction

was carried out in a manner exactly analogous to reaction of

22 with 2 promoted by a complex of Ti(IV) and (+)-4. Thus,

the complex prepared from TiCl₄(0.093 mL, 0.85 mmol), (-)-4

(0.45 g, 0.85 mmol), and Ti(Oi-Pr)₄(0.253 mL, 0.85 mmol)

promoted a reaction of 2 (0.05 g, 0.34 mmol) in CH₂Cl₂(1 mL)

with 22 (0.06 g, 0.34 mmol) to afford after chromatography

diol (-)-4 (0.44 g, 98% recovery); (-)-25 (0.05 g, 48%, 15%ee)

as a colorless oil, [R]₅₈₉) -22 (c 5.0 mg/mL, CHCl₃); and (+)-

24 (0.03 g, 31%, 87%ee) as a white solid. Recrystallization of

enantiomerically enriched (+)-24 from EtOAc/hexanes gave

enantiomerically homogeneous material as colorless needles;

mp 170-171 °C; [R]₅₈₉) +150 (c 3.0 mg/mL, CHCl₃). The

products were identified by comparison of their ¹H and ¹³C

spectra to those of (-)-24 and (+)-25.

The following reactions were conducted in a similar manner

(for reaction temperatures and times, see Table 4).

23 with 2 gave gave (+)-27 (0.131 g, 40%, 20% ee) as a

colorless oil and (-)-26 (0.066 g, 20%, 92% ee).

28 with 2 gave starting quinone (0.021 g, 15%), (+)-30 (0.203

g, 59%, 20% ee), and (-)-29 (0.087 g, 25%, 57% ee).

Recrystallization of (-)-26 and (-)-29 from EtOAc/hexanes

afforded enantiomerically homogeneous material as colorless

needles: (-)-26, mp 217-219 °C; TLC: R_f) 0.11 (30% EtOAc/

hexanes); H NMR (500 MHz) 7.12 (d, J ) 8.8 Hz, 2H), 6.82

(d, J ) 8.8 Hz, 2H), 5.88 (s, 1H), 3.76 (s, 3H), 3.64 (s, 3H),

2404 J . Org. Chem., Vol. 64, No. 7, 1999

Engler et al.

P r otic Acid -Ca ta lyzed Rea r r a n gem en ts of Cyclobu -

ta n es (-)-24, (-)-26, a n d (-)-29: Rea ction of (-)-24 to (+)-

25 Is Rep r esen ta tive. p-Toluenesulfonic acid (p-TsOH, 0.009

g) was added to a solution of (-)-24 (0.068 g, 0.22 mmol) in

CH₂Cl₂(2 mL) at room temperature. The reaction mixture was

stirred for 5 min and then treated with saturated aqueous

NaHCO₃(20 mL). The aqueous layer was separated and

extracted with CH₂Cl₂(20 mL). The combined organic extracts

were washed with H₂O (40 mL) and brine (40 mL), dried (Na₂-

SO₄), filtered, and concentrated. The resultant colorless oil was

chromatographed (30% EtOAc/hexanes) to afford (+)-25 (0.064

[1,2]cyclohepten-1-one (0.032, 91%) as a colorless oil: TLC R_f

0.2 (50% EtOAc/hexanes); ¹H NMR (500 MHz) 7.45 (d, J )

7.6, 2H), 6.87 (d, J ) 8.8, 2H), 5.50 (d, J ) 1.2, 1H), 4.73 (d, J

) 8.0, 1H), 3.89 (s, 3H), 3.80 (s, 3H), 3.38-3.31 (m, 2H), 3.03

(t, J ) 9.9, 1H), 2.92 (d, J ) 1.7, 1H), 2.15-2.03 (m, 2H), 1.92-

1.80 (m, 2H), 1.69-1.63 (m, 2H), 1.37-1.23 (m, 3H), 1.08-

1.04 (m, 1H); ¹³C NMR (125 MHz) 198.8, 175.9, 157.0, 143.7,

128.6, 113.1, 102.2, 65.7, 56.6, 55.2, 53.2, 42.9, 42.8, 41.0, 33.9,

31.6, 29.9, 27.4, 25.2.

A solution of the alcohol prepared as described in the

preceding paragraph (0.035 g, 0.1 mmol) and 4-bromophenyl

isocyanate (0.030 g, 0.15 mmol) in dry toluene (10 mL) was

refluxed for 65 h, cooled, and concentrated to a white solid.

Chromatography (30% EtOAc/hexanes) afforded (-)-32 (0.040

g, 98%) as a white solid, mp 197-199 °C (toluene/hexanes):

TLC R_f0.4 (50% EtOAc/hexanes); ¹H NMR (400 MHz) 7.47

(d, J ) 8.6, 2H), 7.38 (d, J ) 8.2, 2H), 7.22 (d, J ) 6.8, 2H),

6.9 (br s, 1H), 6.84 (d, J ) 8.2, 2H), 5.99 (d, J ) 7.7, 1H), 5.55

(s, 1H), 3.82 (s, 3H), 3.78 (s, 3H), 3.42-3.31 (m, 2H), 3.01 (t, J

) 10.0, 1H), 2.17-2.11 (m, 1H), 1.96-1.84 (m, 2H), 1.69-1.65

(m, 1H), 1.41-1.24 (m, 5H), 1.11-1.07 (m, 1H); ¹³C NMR (100

MHz) 197.6, 173.3, 157.3, 152.5, 142.6, 136.6, 132.1, 128.0,

120.5, 116.7, 113.5, 103.5, 66.9, 56.8, 55.3, 52.7, 42.9, 42.0, 40.5,

34.2, 31.4, 29.7, 27.3, 25.1; HRMS m/z 540.1379 (M⁺, calcd for

g, 94%, >99% ee), mp 74-75 °C (EtOAc/hexanes); [R]₅₈₉

+110 (c 1.9 mg/mL, CHCl₃).

)

In a similar manner, reaction of enantiometically pure (-)-

26 (0.018 g, 0.055 mmol) initially at 0 °C and then room

temperature (1 h) gave (+)-27 (0.018 g, 100%, >99% ee) as a

colorless oil; [R]₅₈₉) +104 (c 2.3 mg/mL, CHCl₃).

Reaction of enantiomerically pure (-)-29 (0.017 g, 0.050

mmol) at room temperature (1 h) gave (+)-30 (0.017 g, 100%,

>99% ee), mp 115-116 °C (EtOAc/hexanes); [R]₅₈₉) +81 (c

1.5 mg/mL, CHCl₃).

(-)-(3a S,3b R,4S,7a R,7b R)-(4-Br om op h en yl)ca r b a m -

ic Acid , 5-Meth oxy-3a -(4-m eth oxyp h en yl)-7-oxo-2,3,3a ,-

3b,4,7,7a ,7b-octa h yd r o-1H-cyclop en ta [3,4]cyclobu ta [1,2]-

b en zen -4-yl E st er (31). A solution of enantiomerically

homogeneous (-)-24 (0.047, 0.15 mmol) in THF (6 mL) was

added to a solution of Red-Al (0.030 mL of a 65+ wt % solution

in toluene, 0.15 mmol) in Et₂O (7 mL) maintained at -5 °C.

Over 1 h the temperature increased to 5 °C, and the reaction

was then quenched by addition of saturated aqueous NH₄Cl

(5 mL). The mixture was extracted with CH₂Cl₂(3 × 10 mL),

dried (MgSO₄), and concentrated. Chromatography of the

resultant colorless oil (50% EtOAc/hexanes) gave (3aR,3bR,7S,-

7aR,7bS)-7-hydroxy-6-methoxy-7b-(4-methoxyphenyl)-1,2,3,-

3a,3b,7,7a,7b-octahydrocyclopenta[3,4]cyclobuta[1,2]benzen-4-

one (0.040, 85%) as a white solid, mp 134-137 °C (EtOAc/

hexanes): TLC R_f0.1 (50% EtOAc/hexanes); ¹H NMR (500

MHz) 7.25 (d, J ) 8.8, 2H), 6.81 (d, J ) 8.8, 2H), 5.31 (d, J )

1.5, 1H), 4.50 (dt, J ) 8.0, 7.7, 1.5, 1H), 3.78 (s, 3H), 3.53 (s,

3H), 3.07 (dd, J ) 8.0, 3.5, 1H), 3.01 (t, J ) 8.0, 1H), 2.60 (dd,

J ) 8.8, 3.5, 1H), 2.37 (dd, J ) 12.5, 5.0, 1H), 2.03-1.92 (m,

3H), 1.73-1.61 (m, 3H); ¹³C NMR (75 MHz) 199.1, 175.5, 158.0,

135.4, 130.1, 113.3, 100.7, 66.8, 56.2, 55.2, 53.5, 47.2, 44.4, 42.6,

42.3, 33.2, 25.2; HRMS m/z 315.1591 (M⁺+ 1, calcd for

₂₈H₃₀O₅BrN: 540.1385); [R]₅₈₉) -107° (c 3.5 mg/mL, CHCl₃).

(+)-[(3a R,8bR)-2,3,3a ,8b-Tetr a h yd r o-6-m eth oxy-3a -(4-

m eth oxyph en yl)-1H-cyclopen ta[b]ben zofu r an -7-yl]-4-br o-

m oben zoa te (33). NaH (60% dispersion in mineral oil; 0.027

g, 0.68 mmol) was washed with hexanes and THF (10 mL) was

added. The slurry was treated with a solution of enantiomeri-

cally pure (+)-25 (0.046 g, 0.15 mmol) in THF (10 mL), and

the reaction mixture stirred at room temperature for 20 min.

A solution of p-bromobenzoyl chloride (0.033 g, 0.15 mmol) in

THF (5 mL) was added dropwise; the reaction mixture was

stirred for 15 min and then diluted with H₂O (20 mL) and

poured into CH₂Cl₂(20 mL). The aqueous layer was separated

and extracted with CH₂Cl₂(2 × 20 mL). The combined organic

extracts were washed with H₂O (50 mL) and brine (50 mL),

dried (Na₂SO₄), filtered, and concentrated. Chromatography

of the resultant colorless oil using 20% EtOAc/hexanes fur-

nished (+)-33 (0.069 g, 94%) as a white solid. Recrystallization

from EtOAc/hexanes gave colorless prisms, mp 120-121 °C:

TLC R_f) 0.42 (30% EtOAc/hexanes); ¹H NMR (500 MHz) 8.05

(d, J ) 8.5 Hz, 2H), 7.63 (d, J ) 8.5 Hz, 2H), 7.38 (d, J ) 8.8

Hz, 2H), 6.88 (d, J ) 8.8 Hz, 2H), 6.87 (s, 1H), 6.53 (s, 1H),

3.80 (s, 3H), 3.77 (s, 3H), 3.76 (d, J ) 8.6 Hz, 1H), 2.41 (dd, J

) 5.7, 14 Hz, 1H), 2.18-2.05 (m, 2H), 1.94-1.84 (m, 2H), 1.74

(apparent nonet, J ) 6.2 Hz, 1H); ¹³C NMR (125 MHz) 164.6,

158.8, 158.3, 151.2, 136.9, 133.3, 131.8, 131.7, 128.6, 128.5,

125.9, 122.1, 118.3, 101.3, 94.2, 56.1, 55.3, 54.7, 42.6, 36.2, 25.1;

HRMS: m/z 494.0717 (calcd for C₂₆H₂₃O₅Br, 494.0729); [R]₅₈₉

) +44 (c 7.62 mg/mL, CHCl₃).

₁₉H₂₃O₄: 315.1596). Anal. Calcd for C₁₉H₂₂O₄: C, 72.59; H,

7.05. Found: C, 72.20; H, 7.08.

A solution of the alcohol prepared as described in the pre-

ceding paragraph (0.025 g, 0.08 mmol) and 4-bromophenyl iso-

cyanate (0.032 g, 0.16 mmol) in dry benzene (10 mL) was

refluxed for 24 h, cooled, and concentrated to a white solid.

Chromatography (30% EtOAc/hexanes) afforded (-)-31 (0.040

g, 98%) as a white solid, mp 219-222 °C (EtOAc/hexanes):

TLC R_f0.36 (50% EtOAc/hexanes); H NMR (500 MHz) 7.41

Gen er a l P r oced u r e for th e Rea r r a n gem en t of (-)-24

w ith Ti(IV)-TADDOLa tes. TiCl₄(0.005 mL, 0.05 mmol) and

a solution of 4 (0.025 g, 0.05 mmol) in CH₂Cl₂(1 mL) were

added rapidly, in sequence, to a solution of Ti(Oi-Pr)₄(0.014

mL, 0.05 mmol) in CH₂Cl₂(1 mL) at room temperature. The

dark-red reaction mixture was stirred for 15 min and cooled

to -78 °C, and a solution of (-)-24 (0.010 g, 0.032 mmol) in

CH₂Cl₂(1 mL) was added. Aliquots (0.1-0.2 mL) were drawn

at the times and temperatures indicated in Table 6 and

quenched with a slurry of i-PrOH (2 mL)/solid NaHCO₃(1 g)

maintained at the quench temperature. The resultant mix-

tures were partitioned between H₂O/CH₂Cl₂(5 mL each) and

filtered through Celite, and the aqueous layer was separated

and extracted with CH₂Cl₂(3 × 5 mL). The combined organic

extracts were washed with brine (5 mL), dried (Na₂SO₄), and

concentrated. The residue was dissolved in i-PrOH, quinone

2 was added, and the mixture was analyzed by HPLC (Chi-

ralcel OD; 20% 2-propanol/hexanes; 1 mL/min; 254 nm). The

relative response factors for each compound were determined

by analysis of a mixture of 2, rac-24, and rac-25 by both ¹H

NMR and HPLC. After complete conversion of (-)-24 (as indi-

cated by HPLC) the reaction was worked up as described above

and chromatographed to give (+)-25 (0.008 g, 82% from reac-

(d, J ) 8.8, 2H), 7.20 (d, J ) 7.7, 4H), 6.79 (d, J ) 7.6, 2H),

6.12 (br s, 1H), 5.63 (dd, J ) 7.7, 1.0, 1H), 5.41 (s, 1H), 3.77 (s,

3H), 3.54 (s, 3H), 3.15-3.10 (m, 2H), 2.66 (dd, J ) 9.0, 5.0,

1H), 2.26 (dd, J ) 10.0, 5.0, 1H), 2.03-1.94 (m, 3H), 1.77-

1.64 (m, 2H); ¹³C NMR (125 MHz) 198.8, 172.9, 157.8, 151.9,

136.5, 132.0, 129.9 (2C), 120.0, 116.1, 112.4, 102.3, 67.1, 56.2,

55.2, 53.5, 47.9, 43.5, 42.5, 40.8, 33.1, 25.3; [R]₅₈₉) -96 (c 2.0

mg/mL, CHCl₃).

(-)-(1S,4a S,4b R,9a R,9b R)-(4-Br om op h en yl)ca r b a m ic

Acid , 2-Meth oxy-9a -[4-m eth oxyp h en yl]-4-oxo-4,4a ,4b,5,-

6,7,8,9,9a ,9b -d eca h yd r o-1H -b en zo[3,4]cyclob u t a [1,2]cy-

cloh ep ten -1-yl Ester (32). A solution of enantiomerically

homogeneous (-)-29 (0.035, 0.1 mmol) in THF (5 mL) was

added to a solution of Red-Al (0.020 mL of a 65+ wt % solution

in toluene, 0.1 mmol) in THF (5 mL) maintained at -5 °C.

The reaction temperature increased to 0 °C over 20 min, and

the reaction was then quenched by addition of saturated

aqueous NH₄Cl (5 mL). The mixture was extracted with CH₂-

Cl₂(3 × 10 mL), dried (MgSO₄), and concentrated. Chroma-

tography of the resultant colorless oil (50% EtOAc/hexanes)

gave (4S,4aR,4bR,9aS,9bR)-4-hydroxy-3-methoxy-4b-[4-meth-

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

Asymmetric Reactions of 2-Methoxy-1,4-benzoquinones

J . Org. Chem., Vol. 64, No. 7, 1999 2405

tion with (-)-4, and 0.006 g, 79% from (+)-4, yields adjusted

to take into account the aliquots removed). The enantiomeric

purity was determined by HPLC and by ¹H NMR as described

above.

In the reaction involving (+)-4, complete conversion of (-)-

24 was observed after 60 min, when the reaction temperature

had reached 10 °C. In reactions involving (-)-4, complete

conversion of (-)-24 was observed after 20 min, when the

reaction temperature had reached -60 °C.

P r otic Acid -Ca ta lyzed Rea r r a n gem en ts of (+)-46/47 to

(+)-48/49. Both reactions were conducted in a similar manner.

Thus, concentrated H₂SO₄(2 drops) was added to a solution

of enantiomerically pure (+)-46 (0.029 g, 0.11 mmol) in CH₂-

Cl₂(3.6 mL). The reaction mixture was stirred for 5 min and

then poured into saturated aqueous NaHCO₃(30 mL). The

aqueous layer was separated and extracted with CH₂Cl₂(3 ×

20 mL), and the combined organic extracts were washed with

brine (50 mL) and concentrated to give (+)-48 (0.026 g, 90%,

>99% ee). The % ee was determined by ¹H NMR (500 MHz)

analysis in the presence of 11 in CDCl₃(the singlet at δ 6.49

in the presence of 11; the specific rotation of this sample was

[R]₅₈₉) +136 (c 7.2 mg/mL, CHCl₃).

Ack n ow led gm en t. We thank Professor Daneel Fer-

reira for a sample of (-)-50, and Professor Fusao

Takusagawa and Mr. Larry Seib for X-ray structure

determinations. Financial support was provided by NIH

(GM 39820) and NSF (CHE-9116576, OSR-9255223,

OSR-9550487) and the University of Kansas General

Research and J .R. and Inez J ay Funds. Fellow- and

granteeships from the Alfred P. Sloan Foundation

(T.A.E.), NIH (M.A.L., GM 07775), and Eli Lilly and

Company (T.A.E.) are acknowledged with gratitude.

Su p p or tin g In for m a tion Ava ila ble: ORTEP drawings,

crystallographic data, tables of atomic coordinates, bond

lengths and angles, and anisotropic thermal parameters for

(+)-12, (-)-31, (-)-32, and (+)-33; IR and mass spectral data

for new compounds; copies of NMR spectra of all new com-

pounds characterized by HRMS. This material is available free

of charge via the Internet at http://pubs.acs.org.

ppm in the racemate split into well-resolved signals); [R]₅₈₉

+163 (c 16 mg/mL, CHCl₃).

)

In a similar manner, enantiomerically enriched (+)-47 (61%

ee, 0.083 g) gave, after chromatography (30% EtOAc/hexanes),

(+)-49 (0.072 g, 86%), which was 61% ee by H NMR analysis

J O982164S

Products guided by the article

Product name:(6aS,11aS)-3,9-Dimethoxy-6a,11a-dihydro-6H-benzo[4,5]furo[3,2-c]chromen-8-ol

Cas No:224446-69-9

R&D Labs maybe for 224446-69-9

JIANGSU GRAND XINYI PHARMACEUTICAL CO.,LTD

website:http://www.xypharm.com/

Contact:+86-515-84383317

Address:Chenli Road,Costal Chemical Area Binhai,Yancheng, Jiangsu,China
Shanghai Sunwise Chemical Co., Ltd

website:http://www.sunwisechem.com

Contact:86-021-33883180

Address:Room 10E, Building G, Westlink Center, No. 2337 Gudai Road, Minhang District, Shanghai, China PC: 201100
Brichem ( Haikou ) Co. Ltd.

Contact:86-898-65311214

Address:Room 102, BLDG. 68 Jiangnan City, No. 66 Heping Road,Haikou, Hainan, China
Jiangsu Wanlong Chemical Co., Ltd.

website:http://www.wanlongchem.com

Contact:+86-511-86810993 0086-511-8681 0888;

Address:Quanzhou Town, Danyang City, Jiangsu
Guangzhou PI & PI Biotech Inc. Ltd.

Contact:+86-20-81716320

Address:Suite 501,198 Kaiyuan Ave.,Science City, Guangzhou,China

Relevant to this article

Doi:10.1016/S0277-5387(00)86543-2
(1987)
Liquid Crystalline Compounds Bearing Pyridine Ring: Azoxy Compounds with Alkyl Substituents

Doi:10.1246/bcsj.60.2905
(1987)
Asymmetric total synthesis of the immunosuppressant (-)-pironetin

Doi:10.1002/anie.200802423
(2008)
Spiro Derivatives of Tetrahydrothiophene. Synthesis of the Quinolizidine<3-spiro-2'>tetrahydrothiophene System Using Solid/Liquid or Liquid/Liquid Phase-Transfer Catalysis

Doi:10.1055/s-1987-33430
(1987)
Catalytic Enantioselective Allylations of Acetylenic Aldehydes via 2-Propanol-Mediated Reductive Coupling

Doi:10.1021/acs.orglett.8b01776
(2018)
Ultrasound-assisted synthesis, anticonvulsant activity, and docking study of indole-appended thiazolidin-4-ones

Doi:10.1002/ardp.201400148
(2014)

Article Doi

Asymmetric reactions of 2-methoxy-1,4-benzoquinones with styrenyl systems: Enantioselective syntheses of 8-aryl-3-methoxybicyclo[4.2.0]oct-3- en-2,5-diones, 7-Aryl-3-hydroxybicyclo[3.2.1]oct-3-en-2,8-diones, 2-Aryl-6- methoxy-2,3-dihydrobenzofuran-5-ols, and pterocarpans

DOI: 10.1021/jo982164s

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jo982164s

Full text of DOI:10.1021/jo982164s

Products guided by the article

R&D Labs maybe for 224446-69-9

Relevant to this article

Hot Product