Reaction of Resorcinol with α,β-Unsaturated Ketones

Article abstract of DOI:10.1021/jo9715478

Products of the acid-catalyzed reaction of resorcinol with α,β-unsaturated ketones have been found to fall into two classes. The first type of product is illustrated by 4-(3,4-dihydro-7-hydroxy-2,4,4-trimethyl-2H-1-benzopyran-2-yl)-1,3-benzenediol, 1, formed in 91% yield by the reaction of resorcinol with 4-methyl-3-penten-2-one (mesityl oxide). The second type of product is illustrated by (C₂-symmetric) 2,2′-spirobi(7-hydroxy-4,4-dimethylchroman), 4, formed in 85% yield by the reaction of resorcinol with 2,6-dimethyl-2,5-heptadien-4-one (phorone). A number of examples of reactions leading, in quite good yields, to products analogous to 1 are presented, as well as some reactions that fail. Flavan 1 is identical to the compound formed by the acid-catalyzed reaction of acetone with excess resorcinol. Compound 1 and a steroidal analogue of 1 have been found to be fluorescent.

Full text of DOI:10.1021/jo9715478

636
J . Org. Chem. 1998, 63, 636-641
Rea ction of Resor cin ol w ith r,â-Un sa tu r a ted Keton es
P. Livant* and Weizheng Xu1
Department of Chemistry, Auburn University, Auburn, Alabama 36849-5312
Received August 19, 1997
Products of the acid-catalyzed reaction of resorcinol with R,â-unsaturated ketones have been found
to fall into two classes. The first type of product is illustrated by 4-(3,4-dihydro-7-hydroxy-2,4,4-
trimethyl-2H-1-benzopyran-2-yl)-1,3-benzenediol, 1, formed in 91% yield by the reaction of resorcinol
with 4-methyl-3-penten-2-one (mesityl oxide). The second type of product is illustrated by (C2-
symmetric) 2,2′-spirobi(7-hydroxy-4,4-dimethylchroman), 4, formed in 85% yield by the reaction of
resorcinol with 2,6-dimethyl-2,5-heptadien-4-one (phorone). A number of examples of reactions
leading, in quite good yields, to products analogous to 1 are presented, as well as some reactions
that fail. Flavan 1 is identical to the compound formed by the acid-catalyzed reaction of acetone
with excess resorcinol. Compound 1 and a steroidal analogue of 1 have been found to be fluorescent.
Recently, we reported2 that the acid-catalyzed reaction
of acetone with excess resorcinol produced 1, rather than
the chemically reasonable isomeric alternative 2 (eq 1).
Because it is impossible to confidently distinguish 1 from
2 by spectroscopic means, an X-ray crystal structure
determination was required.
It seemed reasonable to suppose that the dehydrated
aldol product from acetone, mesityl oxide (4-methyl-3-
penten-2-one), might be an intermediate in the acetone
plus resorcinol reaction. As a step toward testing this
hypothesis, the reaction of mesityl oxide with resorcinol
was examined, and the results are reported herein.
Reactions closely analogous to this are known. To wit,
equations 2-9 give examples of reactions of resorcinol
with R,â-unsaturated acids, R,â-unsaturated esters, and
â-keto esters, the last being the classic von Pechmann
reaction. The particular heterocycle obtained is curiously
(1) Present address: Department of Chemistry and Biochemistry,
University of Delaware, Newark, DE 19716.
(2) Livant, P.; Webb, T. R.; Xu, W. J . Org. Chem. 1997, 62, 737-
742.
(3) (a) Offe, H. A.; Barkow, W. Chem. Ber. 1947, 80, 458-463. (b)
Mercier, C. C. R. Acad. Sci. Ser. C 1973, 276, 1053-1056.
(4) Talapatra, B.; Deb, T.; Talapatra, S. Indian J . Chem. 1986, 25B,
1122-1125.
(5) McGarry, L. W.; Detty, M. R. J . Org. Chem. 1990, 55, 4349-
4356.
(6) Colonge, J .; LeSech, E.; Marey, R. Bull. Soc. Chim. Fr. Ser. 5
1957, 776-779.
(7) Russell, A.; Frye, J . R. Organic Syntheses; Wiley: New York,
1955; Collect. Vol. III, pp 281-285.
(8) Sethna, S.; Phadke, R. Org. React. 1953, 7, 1-58.
(9) (a) Canter, F. W.; Curd, F. H.; Robertson, A. J . Chem. Soc. 1931,
1255-1265. (b) Chakravarti, D. J . Indian Chem. Soc. 1931, 8, 129-
136.
(10) (a) Pillon, D. Bull. Soc. Chim. Fr. 1952, 19, 324-330. (b)
Mentzer, C.; Molho, D.; Vercier, P. Compt. Rendu 1951, 232, 1488-
1490. (c) Mentzer, C.; Molho, D.; Vercier, P. Bull. Soc. Chim. Fr. 1952,
19, 91-93.
variable: compare eq 3, which gives a chromanone, and
eq 4, which gives a dihyrocoumarin. Also, eq 5 gives a
coumarin (note 5-hydroxy rather than 7-hydroxy) while
eq 6 gives a chromone. Even the von Pechmann reaction
(eq 8), which under conditions of catalysis by H2SO4
always gives a coumarin, was once erroneously thought
to give a chromone under catalysis with P2O5 (the so-
S0022-3263(97)01547-8 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/22/1998

Reaction of Resorcinol with R,â-Unsaturated Ketones
J . Org. Chem., Vol. 63, No. 3, 1998 637
called Simonis reaction).9 In fact, it does give the Simonis
chromone at high temperature in the absence of catalyst
(the Mentzer reaction, eq 9).10
In contrast to these well-studied reactions, our search
of the literature revealed very few reports of the acid-
catalyzed reaction of resorcinol with R,â-unsaturated
ketones, and of those, most dealt with mesityl oxide. A
variety of products has been postulated to arise from the
reaction of resorcinol with mesityl oxide. One report
claimed that 2 was the product of the reaction of
resorcinol with mesityl oxide.11 Another alleged 3 was
the product of reaction of resorcinol with mesityl oxide.12
A patent claimed that 4 was obtained from the acid-
catalyzed reaction of resorcinol, mesityl oxide, and ac-
etone.13
A patent gave 5 as the product of base-catalyzed
reaction of resorcinol and mesityl oxide.14 Yet another
patent claimed 1 resulted from acid-catalyzed reaction
of resorcinol and mesityl oxide.15 Proof of structure was
unavailable in this last case, and it is unclear how 2 was
ruled out. As for a reaction of resorcinol with an R,â-
unsaturated ketone other than mesityl oxide, to the best
of our knowledge, there exist only two examples. Namely,
benzylidenacetophenone in acid medium and in the
presence of an oxidizing agent was reported to give 6,16
and benzylideneacetone and resorcinol gave either 7 or
8, depending on solvent and temperature.17 In view of
this somewhat checkered record, it was decided to expand
our investigation to include the acid-catalyzed reactions
of resorcinol with R,â-unsaturated ketones in a more
general sense.
with excess resorcinol gives 1 in 91% isolated yield (Table
1, entry 1). That the product was indeed 1 rather than
the alternatives 2,11 3,12 or 4,13 suggested previously, was
confirmed by the exact match of all spectroscopic data of
this product with those of the product of the acetone plus
resorcinol reaction, shown previously2 to be 1 by means
of an X-ray structure determination.
Certain R,â-unsaturated ketones other than mesityl
oxide react with resorcinol in the same way that mesityl
oxide does. Before discussing these examples, it is
necessary to address briefly the problem of structure
proof, since uncertainties akin to the problem of distin-
Resu lts a n d Discu ssion
R ea ct ion s of R esor cin ol w it h r,â-Un sa t u r a t ed
Keton es. The acid-catalyzed reaction of mesityl oxide
Ta ble 1. Acid -Ca ta lyzed Rea ction s of r,â-Un sa tu r a ted Keton es w ith Excess Resor cin ol
a
b
Ar ) 2,4-dihydroxyphenyl. Isolated.

638 J . Org. Chem., Vol. 63, No. 3, 1998
Livant and Xu
guishing 1 and 2 could arise in other cases as well.
Fortunately, there is a telltale signal in the 13C NMR of
1 that appears at 79.4 ppm and is due to the aliphatic
quaternary carbon bound to oxygen, i.e., C2 according to
flavan numbering. Compounds analogous to 2 would not
be expected to have a quaternary carbon signal near this
chemical shift position, and therefore, we consider the
observation of such a signal to be diagnostic of a structure
like 1. In fact, most of the compounds reported below
have a signal in the range 79-83 ppm.
appeared at 101.0 ppm, rather than the expected 79-83
ppm region. Comparing 10 and 12, C2 in 10 is bonded
to 2,4-dihydroxyphenyl, oxygen, and two secondary car-
bons, while C2 in 12 is bonded to 2,4-dihydroxyphenyl,
oxygen, and two tertiary carbons. The difference is the
addition in 12 of two substituents â to C2. Using the
venerable Grant and Paul scheme19 for estimating carbon
chemical shifts, the two extra â substituents should cause
a downfield shift of 18.8 ppm. The chemical shift of C2
in 10 is 83.2 ppm. Therefore, the predicted value for C2
of 12 is 102.0 ppm, which is tolerably close to the
observed value of 101.0 ppm. (The prediction is even
better for the corresponding acetate derivatives: 99.7
ppm predicted versus 99.6 ppm actual.) We cannot state
with certainty whether 12-exo or 12-endo is formed. (Exo
and endo refer to the pendent 2,4-dihydroxyphenyl sub-
stituent.) Molecular mechanics calculations shed no light
here since the calculated difference in energy between
12-exo and 12-endo is too small to be relied upon.
Methyl vinyl ketone and ethyl vinyl ketone (Table 1,
entries 2 and 3) both work well in this reaction, giving a
single product (9 and 10, respectively) cleanly and in good
yield. (Purification of these, and, to a certain extent, all
the products reported in Table 1 was rendered more
challenging than usual by their tendency to retain solvent
tenaciously. Drying under vacuum at an elevated tem-
perature was usually required to obtain sharp-melting
samples that gave correct elemental analysis. To cir-
cumvent this problem, acetate derivatives were prepared
for almost all the products listed.) A sample of 5-methyl-
3-hexen-2-one (sold as a mixture with ca. 20 mol %
5-methyl-4-hexen-2-one) was tried, in the hope that the
nonconjugated enone contaminant would not interfere.
However, a product that was probably polymeric, judging
from the NMR spectra, was obtained. Benzylideneac-
etone (entry 4) gave 11 in 78% isolated yield as a mixture
of two diastereomers in a ratio of about 5:1. Molecular
mechanics calculations18 indicate 11-cis is lower in energy
than 11-trans by 1.1 kcal/mol. (Taking, for convenience,
this ∆E to be ∆G, and assuming the product distribution
results from thermodynamic control, the calculated en-
ergy difference of 1.1 kcal/mol at 25 °C corresponds to a
predicted ratio of 6.4:1). These calculations suggest, but
certainly do not prove, that 11-cis is the major product
in this case.
Cases in which the enone π-system is endocyclic gave
more variable results. 2-Cyclohexenone and 4,4-di-
methyl-2-cyclohexenone gave high yields of 13a and 13b,
respectively (entry 6). Also, a steroidal A-ring R,â-
unsaturated ketone, cholest-1-en-3-one, produced the
analogous 14 in quite good yield (entry 7). We did not
determine whether C2 (steroid numbering) in 14 is syn
or anti to C19; however, molecular mechanics calcula-
tions predict the syn to be favored by more than 5 kcal/
mol. By contrast, the two steroidal R,â-unsaturated
ketones 15 and 16 failed to react with resorcinol at all
under our standard conditions.
The cases presented so far involve R,â-unsaturated
ketones in which the enone π-system is not part of a ring.
The reaction of resorcinol with 3-methylene-2-norbor-
nanone (entry 5) provides a case in which the π-system
is exocyclic to a ring. The reaction provides a single
product, 12, cleanly. Curiously, the C2 signal in 12
2-Cyclopentenone and 2-cycloheptenone reacted with
resorcinol but did not give products similar to any
discussed thus far. The products were water-soluble,
which made separation from excess resorcinol, also
water-soluble, a difficult task. In the 2-cycloheptenone
case, basifying the aqueous product mixture caused a
solid to precipitate. This solid gave a parent peak at m/z
202 in its mass spectrum, suggesting it is the condensa-
tion product of one resorcinol (MW 110) and one 2-cyclo-
heptenone (MW 110) with loss of one water (MW 18). Its
13C NMR spectrum showed no carbonyl carbon. In view
of the difficulty in obtaining a pure sample, definitive
evidence that would lead to a proof of structure is lacking.
(11) Kondrat’eva, G. G.; Volkotrub, M. N. Sin. Issled. Eff. Khim-
Dobavok., Polim. Mater. 1969, No. 2, 373-376; Chem. Abstr. 1972, 76,
59369z.
(12) Liska, K. J . J . Med. Chem. 1972, 15, 1177-1179.
(13) Harada, H.; Usui, M. J pn Patent J P 62,111,988, 1985; Chem.
Abstr. 1988, 109, 149374b.
(14) Kumiai Chemical Industry Co., Ltd., Mitsubishi Petrochemical
Co., Ltd. J pn. Patent 82,109,779, 1982; Chem. Abstr. 1982, 97, 198111f.
(15) (a) Mitsui Petrochemical Industries, Ltd. J pn Patent J P 82
16,877, 1982; Chem. Abstr. 1982, 97, 6153b. (b) Mitsui Petrochemical
Industries, Ltd. J pn Patent J P 82,114,585, 1982; Chem. Abstr. 1982,
97, 216003d.
(16) Robinson, R.; Walker, J . J . Chem. Soc. 1934, 1435-40.
(17) Clayton, D. W.; Elstow, W. E.; Ghosh, R.; Platt, B. C. J . Chem.
Soc. 1953, 581-586.
(18) The software used was PCModel (Serena Software), with the
“MMX” force field.
(19) Quoted in: Stothers, J . B. Carbon-13 NMR Spectroscopy;
Academic: New York, 1972; p 58.

Reaction of Resorcinol with R,â-Unsaturated Ketones
J . Org. Chem., Vol. 63, No. 3, 1998 639
We also tried more extended π-systems. â-Ionone, an
R,â,γ,δ-unsaturated ketone, gave an apparent polymer
(eq 10). However, phorone (2,6-dimethyl-2,5-heptadien-
12)21 to give 3,4-dimethylphenol in 92% yield. The
reaction of 4H-pyran-4-one with resorcinol gave an
intractable product mixture.
4-one), the “double” aldol condensation product from
acetone, did not give 1, but rather gave an 85% yield of
spirochroman 4 (Table 1, entry 8). (This material was
previously prepared12,20 in 26% yield by the reaction of
resorcinol with mesityl oxide catalyzed by FeCl3). The
fact that, under our conditions, mesityl oxide gave 1 and
phorone gave 4 leads one to speculate that in the
reactions of acetone with resorcinol in which both 1 and
4 are formed,2 the amounts of 1 and 4 reflect the amounts
of mesityl oxide and phorone, respectively, formed from
aldol self-condensations of acetone (eq 11). In support
of this suggestion is the observation2 that acetone and
resorcinol tend to give more 4 as the ratio of acetone to
resorcinol increases.
F lu or escen ce Stu d ies. Steroid 14 is fluorescent. At
concentrations in ethyl acetate of 0.0061, 0.061, 0.24, and
0.61 mM, 14 exhibited an emission peak at 308 nm. This
should be compared to the emission found for resorcinol
itself (0.7 mM in EtOAc) at 305 nm. The excitation
spectra of 14 (308 nm emission) over this series of
concentrations showed a single feature at 280 nm at the
two lower concentrations, a broad multipeak band be-
tween 274 and 286 nm at 0.24 mM, and two resolved
features at 267 and 292 nm at 0.61 mM. At 0.61 mM,
the absorption spectrum of 14 shows one band at 280 nm.
Emission spectra and excitation spectra were recorded
for 1 at concentrations of 0.06 and 0.6 mM in EtOAc, and
they were consistent with what was found for 14:
emission at 308 nm, a single excitation band at 283 nm
at the lower concentration and two excitations, at 267
and 292 nm, at the higher concentration. In 14 or 1, the
splitting of the excitation band into two bands at higher
concentrations is the result of intermolecular association
between the resorcinol moieties of 14 or 1. These
spectroscopic results suggest that the reaction of resor-
cinol with steroidal R,â-unsaturated ketones could be
used to append a fluorescent label to a steroid.
Exp er im en ta l Section
Gen er a l P r oced u r e for Rea ction of Resor cin ol w ith
r,â-Un sa tu r a ted Keton es. A mixture of 1 equiv of ketone,
6 equiv of resorcinol, and 1 equiv of HCl (as 10% HCl) dissolved
in 1:1 (v/v) Et2O:CH2Cl2 was stirred and refluxed for 24 h.
Solvents were removed on the rotary evaporator, water was
added, and the mixture was extracted repeatedly with CH2-
Cl2. The extracts were washed with warm water and dried
over anhydrous Na2SO4. After filtration and removal of
solvent, the crude product was purified by column chroma-
tography on silica gel with 30:70 (v/v) ethyl acetate:hexanes.
4-(3,4-Dih ydr o-7-h ydr oxy-2,4,4-tr im eth yl-2H-1-ben zopy-
r a n -2-yl)-1,3-ben zen ed iol, 1. Following the general proce-
dure, the reaction of 0.436 g (4.45 mmol) mesityl oxide, 2.940
g (26.70 mmol) of resorcinol, and 1.5 mL of 10% aqueous HCl
(4.1 mmol) in 70 mL of 1:1 (v/v) Et2O:CH2Cl2 gave 1.215 g
(91%) of a colorless solid, mp 231-232 °C. 1H NMR, 13C, NMR
and IR spectra were identical with those obtained for 1
Analogously, the reaction of trans,trans-dibenzylidene-
acetone with resorcinol gave spiro compound 17 in 76%
yield. 1H NMR of the crude product showed a small
amount of what we presume to be another diastereomer
of 17. However, after purification only one diastereomer
was obtained, almost certainly the one in which both
pendent phenyls are in pseudoequatorial positions. MM2
calculations predict this to be the lowest-energy diaste-
reomer (1.5 kcal/mol lower than di-axial; 2.1 kcal/mol
1
lower than axial-equatorial), and the H NMR coupling
constants of H4 (chroman numbering), 5.7 and 12.9 Hz,
are consistent with the diequatorial diastereomer. It is
interesting to note that just as mesityl oxide gave 1 while
phorone gave 4, benzylideneacetone gave a product
analogous to 1 (viz. 11) while dibenzylideneactone gave
a product analogous to 4 (viz. 17).
1
prepared from resorcinol and acetone.2 In addition, H NMR
(acetone-d6, 250 MHz) δ: 0.78 (s, 3H), 1.20 (s, 3H), 1.65 (s,
3H), 1.84 (d, J ) 13.9 Hz, 1H), 3.01 (d, J ) 13.9 Hz, 1H), 6.18
(dd, J ) 8.5, 2.4 Hz, 1H), 6.35 (d, J ) 2.4 Hz, 1H), 6.37 (dd, J
) 8.4, 2.6 Hz, 1H), 6.43 (d, J ) 2.5 Hz, 1H), 7.01 (d, J ) 8.4
Hz, 1H), 7.02 (d, J ) 8.5 Hz, 1H).
2,2′-Sp ir ob i(7-h yd r oxy-4,4-d im et h ylch r om a n ), 4, was
prepared in 85% yield as described previously2 from 2,6-
dimethyl-2,5-heptadien-4-one and resorcinol, mp 199-200 °C
(lit.12,20 mp 199-200 °C).
4-(3,4-Dih yd r o-7-h yd r oxy-2-m eth yl-2H-1-ben zop yr a n -
2-yl)-1,3-ben zen ed iol, 9. The reaction of 1.03 g (14.7 mmol)
of methyl vinyl ketone, 9.71 g (88.2 mmol) of resorcinol, and
5.4 mL of 10% aqueous HCl (15 mmol) produced 3.12 g (78%)
of a colorless solid. A 521 mg portion was subjected to silica
A cyclic analogue of phorone is 4,4-dimethyl-2,5-
cyclohexadienone. Under our standard reaction condi-
tions, this compound did not react with resorcinol, but
rather underwent a dienone-phenol rearrangement (eq
(20) Liska assigned structure 3 to his product. However, after we
repeated his procedure and obtained a material with identical mp, 13
C
NMR made it clear that it was indeed 4 that was produced. In
particular, the quaternary carbon signal at 99.95 ppm was consistent
with the spiro carbon of 4 but inconsistent with any carbon in Liska’s
proposal.
(21) (a) Waring, A. J . Adv. Alicycl. Chem. 1966, 1, 129-256. (b)
Miller, B. Mechs. Mol. Migr. 1968, 1, 247-313. (c) Perkins, M. J .; Ward,
P. Mechs. Mol. Migr. 1971, 4, 55-112.

640 J . Org. Chem., Vol. 63, No. 3, 1998
Livant and Xu
gel chromatography followed by drying at 110 °C under
vacuum to give 318 mg of pure 9, mp 174-176 °C. Anal.
Calcd for C16H16O4: C, 70.57; H, 5.92. Found: C, 69.98; H,
6.00. HRMS: m/z calcd for C16H16O4 272.1048, found 272.1054.
MS (EI): m/z (rel intens) 272 (M+, 17), 257 (11), 150 (45), 137
(28), 123 (100). 1H NMR (methanol-d4, 250 MHz) δ: 1.63 (s,
3H), 1.81 (m, 1H), 2.28 (m, 1H), 2.43 (m, 1H), 2.87 (m, 1H),
6.12 (dd, J ) 8.5, 2.4 Hz, 1H), 6.25 (m, 2H), 6.33 (d, J ) 2.3
C, 73.60; H, 6.79. Found: C, 72.84; H, 6.74. HRMS: m/z calcd
for C20H22O4 326.1518, found 326.1515. MS (EI): m/z (rel
intens) 326 (M, 27), 270 (52), 269 (21), 255 (100), 161 (29). H
1
NMR (CDCl3, 400 MHz; assignments from COSY and C-H
correlation spectra) δ: 0.870 (s, 3H), 1.185 (s, 3H), 1.21 (m,
1H; H10a or H10e), 1.367 (dt, J ) 13.6, 4.6 Hz, 1H, H10e or
H10a), 1.916 (dt, J ) 14.0, 5.0 Hz, 1H; H9a), 2.009 (dt, J )
13.3, 3.0 Hz, 1H; H3e), 2.280 (m, 1H; H9e), 2.505 (ap-
proximately quintet, J ) 3.0 Hz, 1H; H4), 2.664 (dd, J ) 13.5,
3.1 Hz, 1H; H3a), 4.493 (s, 1H; OH), 4.817 (s, 1H; OH), 6.35-
6.38 (m, 3H), 6.424 (d, J ) 2.5 Hz, 1H), 6.864 (d, J ) 8.7 Hz,
1H), 7.005 (d, J ) 8.4 Hz, 1H), 8.207 (s, 1H; OH). A long-
range coupling, H3e/H9e, was evident in the COSY spectrum.
1H NMR (CD3CN, 400 MHz; assignments from COSY and
C-H correlation spectra) δ: 0.845 (s, 3H), 1.119 (m, 1H; H10a
or H10e), 1.224 (s, 3H), 1.367 (dt, J ) 13.3, 4.8 Hz, 1H; H10e
or H10a), 1.584 (dt, J ) 13.5, 2.9, 1H; H3e), 1.733 (m, 1H;
H9e), 2.404 (approx quintet, J ) 2.7 Hz, 1H; H4), 2.630 (dt, J
) 13.9, 5.2 Hz, 1H; H9a), 2.129 (dd, J ) 13.6, 3.0 Hz, 1H; H3a),
6.24-6.30 (m, 3H), 6.343 (d, J ) 2.5 Hz, 1H), 6.781 (d, J ) 8.2
Hz, 1H), 7.326 (d, J ) 9.2 Hz, 1H). H3e/H9e coupling was
evident in the COSY spectrum. 13C NMR (CDCl3, 100.6 MHz)
δ: 156.61 (quat), 156.24 (quat), 155.39 (quat), 154.66 (quat),
131.46 (CH), 126.10 (CH), 122.39 (quat), 117.61 (quat), 107.36
(CH), 106.93 (CH), 104.94 (CH), 102.24 (CH), 82.02 (quat.; C2),
42.71 (CH; C4), 35.81 (CH2; C9), 33.86 (quat; C11), 31.49 (CH2;
C10), 30.77 (CH2; C3), 29.55 (CH3), 25.05 (CH3).
Hz, 1H), 6.71 (d, J ) 8.2 Hz, 1H), 6.96 (d, J ) 8.5 Hz, 1H). 13
C
NMR (methanol-d4, 62.9 MHz) δ: 158.40 (quat), 157.68 (quat),
156.11 (quat), 156.07 (quat), 130.88 (CH), 129.26 (CH), 122.80
(quat), 115.02 (quat), 108.90 (CH), 107.16 (CH), 104.53 (CH),
104.10 (CH), 80.58 (quat), 32.40 (CH2), 28.17 (CH3), 23.38
(CH2).
4-(3,4-Dih yd r o-7-h yd r oxy-2-eth yl-2H-1-ben zop yr a n -2-
yl)-1,3-ben zen ed iol, 10. The reaction of 0.52 g (6.2 mmol)
of ethyl vinyl ketone, 4.08 g (37.1 mmol) of resorcinol, and 2.3
mL of 10% aqueous HCl (6.3 mmol) generated 1.50 g (85%) of
a colorless solid. A 352 mg portion was subjected to silica gel
chromatography followed by drying at 110 °C under vacuum
to give 275 mg of pure 10, mp 162-168 °C. HRMS: m/z calcd
for C17H18O4 286.1205, found 286.1207. MS (EI): m/z (rel
1
intens) 286 (M+, 6), 257 (21), 149 (15), 147 (22), 123 (100). H
NMR (methanol-d4, 250 MHz) δ: 0.81 (t, 3H), 1.87 (m, 2H),
2.20 (m, 2H), 2.41 (m, 1H), 2.80 (m, 1H), 6.12 (dd, J ) 8.5, 2.5
Hz, 1H), 6.25 (m, 2H), 6.33 (d, J ) 2.5 Hz, 1H), 6.71 (d, J )
8.2 Hz, 1H), 6.92 (d, J ) 8.5 Hz, 1H). 13C NMR (methanol-d4,
62.9 MHz) δ: 158.40 (quat), 157.66 (quat), 156.23 (quat),
156.15 (quat), 130.83 (CH), 130.64 (CH), 120.90 (quat), 115.35
(quat), 108.79 (CH), 107.05 (CH), 104.42 (CH), 104.11 (CH),
83.23 (quat), 33.43 (CH2), 30.96 (CH2), 23.22 (CH2), 8.45 (CH3).
4-(3,4-Dih yd r o-7-h yd r oxy-2-m eth yl-4-p h en yl-2H-1-ben -
zop yr a n -2-yl)-1,3-ben zen ed iol, 11. The reaction of 623 mg
(4.23 mmol) of trans-4-phenyl-3-buten-2-one, 2.81 g (25.5
mmol) of resorcinol, and 2.3 mL of 10% aqueous HCl (6.3
mmol) afforded 1.34 g (90%) of product. Silica gel chroma-
tography (ethyl acetate/hexane) generated 705 mg (64%) of
major isomer and 175 mg (12%) of minor isomer. The following
data refer to the major isomer. HRMS: m/z calcd for C22H20O4
348.1361, found 348.1350. MS (EI): m/z (rel intens) 348 (M+,
1), 199 (34), 175 (25), 151 (100), 150 (43), 137 (26), 135 (44),
123 (31), 107 (24), 77 (28).
2′,4′,7-Tr ih yd r oxy-2,3-n or bor n a n ofla va n , 12. The reac-
tion of 938 mg (7.45 mmol) of 3-methylene-2-norbornanone
(97% purity), 5.08 g (46.2 mmol) of resorcinol, and 2.8 mL of
10% aqueous HCl (7.6 mmol) gave 1.96 g (81%) of product. A
511 mg portion was recrystallized from CHCl3 to afford 488
mg of pure 12, mp 190-192 °C. HRMS: m/z calcd for C20H20O4
324.1362, found 324.1362. MS (EI): m/z (rel intens) 324 (M+,
20), 201 (100), 147 (21), 123 (69), 67 (24). 1H NMR (CD3CN,
400 MHz, assignments from COSY and C-H correlation
spectra) δ: 0.78-0.81 (m, 1H), 1.21-1.32 (m, 2H), 1.43-1.49
(m, 2H), 1.75-1.79 (m, 1H), 2.393 (m, 1H; H9), 2.451 (m, 1H;
H12), 2.847 and 2.937 (AB quartet, J ) 15.0 Hz, 2H; H4a and
H4e), 3.267 (d, J ) 5.0 Hz, 1H; H3), 6.16-6.24 (m, 4H), 6.742
(dd, J ) 8.0, 0.9 Hz, 1H), 6.946 (d, J ) 8.2 Hz, 1H). 13C NMR
(CD3CN, 100.6 MHz) δ: 162.52 (quat), 158.44 (quat), 157.73
(quat), 157.18 (quat), 134.32 (CH), 126.47 (CH), 120.92 (quat),
115.88 (quat), 108.21 (CH), 107.91 (CH), 103.78 (CH), 100.99
(quat; C2), 99.58 (CH), 52.65 (CH; C3), 47.22 (CH; C9), 42.32
(CH; C12), 39.41 (CH2), 39.22 (CH2), 24.29 (CH2), 22.72 (CH2).
2′,4′,7-tr ih yd r oxy-2,4-p r op a n ofla va n , 13a . The reaction
of 0.272 g (2.83 mmol) of 2-cyclohexenone, 1.825 g (16.57 mmol)
of resorcinol, and 1.0 mL of 10% aqueous HCl (2.7 mmol)
afforded 0.711 g (87%) of a colorless solid, mp 113-130 °C. A
portion was subjected to silica gel chromatography (30:70
EtOAc/hexanes) followed by drying at 110 °C under vacuum
to give analytically pure 13a , mp 189-191 °C. Analytical and
spectroscopic data were reported previously.2
2′,4′,7-Tr ih yd r oxy-2,4-(1,3-ch olesta n o)fla va n , 14. The
reaction of 0.173 g (0.451 mmol) of cholest-1-en-3-one, 0.496 g
(4.50 mmol) of resorcinol, and 0.2 mL of 10% aqueous HCl
according to the general procedure gave 0.239 g (91%) of crude
product. Recrystallization from EtOAc/hexanes and drying
under vacuum at 110 °C afforded 0.218 g (83%) of analytically
pure 14, mp 265-266 °C. Anal. Calcd for C39H54O4: C, 79.82;
H, 9.28. Found: C, 79.72; H, 9.24. HRMS: m/z calcd for
C
39H54O4 586.4022, found 586.4020. MS (EI): m/z (rel intens)
586 (M, 7), 271 (21), 270 (100), 255 (97), 161 (36), 123 (79). 1H
NMR (CDCl3, 400 MHz) δ: 8.300 (s, 1H, OH), 7.002 (d, J )
8.6 Hz, 1H), 6.914 (d, J ) 8.2 Hz, 1H), 6.55-6.35 (m, 4H), 4.963
(s, 1H, OH), 4.919 (s, 1H, OH), 2.956 (approx quintet, J ) 2.5
Hz, 1H), 2.616 (dd, J ) 13.6, 2.9 Hz, 1H), 2.03-0.70 (m), 1.634
(s, 3H), 1.033 (s, 3H) 0.941 (d, J ) 6.5 Hz, 3H), 0.874 (d, J )
6.6 Hz, 3H), 0.871 (d, J ) 6.6 Hz, 3H), 0.704 (s, 3H). 13C NMR
(CDCl3, 100.6 MHz) δ: 156.49 (quat), 156.21 (quat), 155.34
(quat), 155.22 (quat), 130.29 (CH), 126.15 (CH), 122.31 (quat),
116.82 (quat), 107.65 (CH), 106.96 (CH), 104.92 (CH), 102.32
(CH), 82.77 (quat, C3), 56.41 (CH, C14 or C17), 56.24 (CH,
C17 or C14), 48.17 (CH, C9), 42.86 (quat, C13), 42.50 (CH,
C1?), 39.74 (CH2, C4 or C12), 39.69 (CH2, C12 or C4), 39.55
(CH2, C24), 39.42 (CH, C5?), 37.01 (quat, C10), 36.19 (CH,
C20), 35.93 (CH2, C22), 35.80 (CH, C8), 31.13 (CH2, C7 or C2),
30.64 (CH2, C2 or C7), 28.19 (CH2, C16), 28.02 (CH, C25), 27.64
(CH2, C6?), 24.02 (CH2, C15), 23.82 (CH2, C23), 22.78 (CH3,
C26 or C27), 22.55 (CH3, C27 or C26), 21.92 (CH2, C11), 18.71
(CH3, C21), 14.15 (CH3, C19), 12.42 (CH3, C18).
2,2′-Sp ir obi(7-h yd r oxy-4-p h en ylch r om a n ), 17. A mix-
ture of 0.474 g (2.02 mmol) of trans,trans-dibenzylideneac-
etone, 1.339 g (12.2 mmol) of resorcinol, 0.2 mL of 36% HCl,
25 mL of ether, and 25 mL of CH2Cl2 was stirred and refluxed
24 h. After removal of solvents, addition of water afforded a
precipitate, which was collected and washed repeatedly with
water. After vacuum-drying, the product was subjected to
silica gel column chromatography (30:70 (v/v) EtOAc:hexanes)
to yield 0.671 g (76%) of 17 as a 3:1 mixture of diastereomers,
mp 270-283 °C. After recrystallization from CH2Cl2/CH3CN,
the major diastereomer was obtained, mp 299-302 °C.
HRMS: m/z calcd for C29H24O4 436.1675, found 436.1676. MS
(EI): m/z (rel intens) 436 (M, 4), 237 (10), 270 (100), 199 (100),
197 (18). 1H NMR (CD3COCD3, 250 MHz) δ: 7.38-7.28 (m,
5H), 6.511 (d, J ) 8.5 Hz, 1H), 6.36-6.32 (m, 2H), 4.499 (dd,
J ) 5.7, 12.9 Hz, 1H), 2.406 (dd, J ) 5.7, 13.4 Hz, 1H), 2.196
(app t, J ) 13.2 Hz, 1H). 13C NMR (CD3,COCD3, 62.9 MHz)
δ: 157.97 (quat), 153.86 (quat), 145.56 (quat), 130.64 (CH),
129.77 (CH), 129.59 (CH), 127.65 (CH), 118.26 (quat), 109.80
(CH), 104.29 (CH), 97.75 (quat), 41.72 (CH2), 38.17 (CH).
11,11-Dim et h yl-2′,4′,7-t r ih yd r oxy-2,4-p r op a n ofla va n ,
13b. Following the general procedure, 1.72 g (13.8 mmol) of
4,4-dimethyl-2-cyclohexenone, 8.86 g (80.5 mmol) of resorcinol,
and 4.8 mL of 10% aqueous HCl gave 4.01 g (92%) of product.
A 501 mg portion was recrystallized from CHCl3 to afford 408
mg of pure 13b, mp 187-188 °C. Anal. Calcd for C20H22O4:

Reaction of Resorcinol with R,â-Unsaturated Ketones
J . Org. Chem., Vol. 63, No. 3, 1998 641
P r ep a r a tion of Aceta te Der iva tives. The following
general procedure was used: To a solution of 2 mmol of
hydroxy compound in 10 mL of dry pyridine was added 32
mmol of acetic anhydride dropwise with stirring. The reaction
mixture was stirred for 1.5 h at 40-50 °C, allowed to cool to
room temperature, and poured onto 60 mL of ice-water. The
precipitate that formed was collected by filtration, recrystal-
lized from EtOAc/hexanes, and dried at 110 °C in vacuo.
Usually, one further recrystallization provided an analytically
pure acetate. Complete spectroscopic data are given in the
Supporting Information.
(h) Diacetate from 17: 95% yield. mp 220-221 °C. HRMS:
m/z calcd for C33H28O6 520.1886, found 520.1879.
3,4-Dim eth ylp h en ol. The reaction of 1.37 g (11.2 mmol)
of 4,4-dimethyl-2,5-cyclohexadienone (prepared according to
the published procedure22), 7.42 g (67.4 mmol) of resorcinol,
and 4.0 mL of 10% aqueous HCl (11 mmol) gave a solid
product. This was subjected to silica gel chromatography
(EtOAc: hexane 1:3) and vacuum distillation, which afforded
1.268 g (92%) of 3,4-dimethylphenol, mp 66-68 °C (lit.23 mp
65.11 ( 0.01 °C). MS (EI): m/z (rel intens) 122 (M+, 68), 121
(41), 107 (100), 91 (24), 77 (38). 1H NMR (CDCl3, 250 MHz)
δ: 7.005 (d, J ) 8.0 Hz, 1H), 6.666 (d, J ) 2.5 Hz, 1H), 6.000
(dd, J ) 8.0, 2.8 Hz, 1H), 2.232 (s, 3H), 2.204 (s, 3H). 13C NMR
(CDCl3, 63 MHz): 153.42, 137.95, 130.47, 128.64, 116.58,
112.34, 19.82, 18.72.
(a) Triacetate from 1: 83% yield, mp 129-130 °C. Analyti-
cal and spectroscopic data as well as X-ray crystal structure
were reported previously.2
(b) Diacetate from 4: 95% yield, mp 194-5 °C. Analytical
and spectroscopic data were reported previously.2
Ack n ow led gm en t. We thank Prof. Vincenzo Cam-
marata for assistance with the fluorescence studies and
Dr. K. P. Sarathy of the Auburn University NMR Center
for a generous allocation of instrument time. Also, we
are indebted to Mr. Steven Kizito for providing the
steroids.
(c) Triacetate from 10: 92% yield, mp 135-136 °C. Anal.
Calcd for C23H24O7: C, 66.98; H, 5.87. Found: C, 66.89; H,
5.90. HRMS: m/z calcd for C23H24O7 412.1522, found 412.1526.
(d) Triacetate from 11 (major isomer): mp 190-192 °C.
HRMS: m/z calcd for C28H26O7 474.1678, found 474.1681.
(e) Triacetate from 12: 92% yield. HRMS: m/z calcd for
C
26H26O7 450.1678, found 450.1675.
(f) Triacetate from 13a : 91% yield, mp 145-146 °C.
Su p p or tin g In for m a tion Ava ila ble: Copies of NMR
spectra for 10-12, 17, and the acetate derivatives of 11, 12,
13b and 17, listings of mass spectral data for 9-12 and 13b-
17, 13C NMR data for 13b, and listings of mass spectral, 1H
NMR, and 13C NMR data for the acetate derivatives (26 pages).
This material is contained in libraries on microfiche, im-
mediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
Analytical and spectroscopic data were reported previously.2
(g) Triacetate from 13b: 94% yield, mp 88-92 °C. HRMS:
m/z calcd for C26H28O7 452.1835, found 452.1831.
(22) (a) Zimmerman, H. E.; Binkley, R. W.; McCullough, J . J .;
Zimmerman, G. A. J . Am. Chem. Soc. 1967, 89, 6589. (b) Zimmerman,
H. E.; Hackett, P.; J uers, D. F.; McCall, J . M.; Schroder, B. J . Am.
Chem. Soc. 1971, 93, 3653.
(23) Andon, R. J . L.; Biddiscombe, D. P.; Cox, J . D.; Handley, R.;
Harrop, D.; Herington, E. F. G.; Martin, J . F. J . Chem. Soc. 1960,
5246-5254.
J O9715478