High-pressure organic chemistry. 19. High-pressure-promoted, silica gel-catalyzed reaction of epoxides with nitrogen heterocycles

Article abstract of DOI:10.1021/jo951106t

The reaction of epoxides with nitrogen heterocycles such as indoles, pyrroles, imidazoles, and pyrazoles was studied under both high-pressure and silica gel-catalyzed conditions. Whereas it has been reported that the treatment of indole with styrene oxide

Full text of DOI:10.1021/jo951106t

984
J . Org. Chem. 1996, 61, 984-990
High -P r essu r e Or ga n ic Ch em istr y. 19. High -P r essu r e-P r om oted ,
Silica Gel-Ca ta lyzed Rea ction of Ep oxid es w ith Nitr ogen
Heter ocycles1,†
Hiyoshizo Kotsuki,* Katsunori Hayashida, Tomoyasu Shimanouchi, and Hitoshi Nishizawa
Department of Chemistry, Faculty of Science, Kochi University, Akebono-cho, Kochi 780, J apan
Received J une 19, 1995X
The reaction of epoxides with nitrogen heterocycles such as indoles, pyrroles, imidazoles, and
pyrazoles was studied under both high-pressure and silica gel-catalyzed conditions. Whereas it
has been reported that the treatment of indole with styrene oxide at 10 kbar and 42 °C for 24 h
gave 2-(3-indolyl)-2-phenylethanol in 56% yield, the same compound was obtained in 88% yield
when the reaction was conducted on silica gel at rt for 1 week. Similarly, efficient reaction of
epoxides with pyrroles, imidazoles, and pyrazoles was achieved. In terms of stereochemical features,
the epoxide ring opening reaction of (R)-(+)-styrene oxide with indole was found to proceed
stereoselectively in an SN2 fashion at the benzyl carbon, in either case.
In tr od u ction
Epoxides are versatile intermediates in organic syn-
thesis because they are both susceptible to attack by
several nucleophiles2 and readily accessible in optically
pure form.3 The epoxide opening reaction with certain
nucleophiles is generally performed with acid or base
catalysis, and in the absence of such a catalyst the
reaction is moderately slow.4 These methods have cer-
tain disadvantages in that they require large excesses
of reagents, long reaction times, and drastic conditions
and entail undesirable rearrangement and polymeriza-
tion side reactions. In our recent papers, we have
demonstrated a simple, effective, uncatalyzed way to
carry out nucleophilic ring opening reactions of epoxides
by applying a high-pressure technique.5
For example, indoles reacted with vinyl epoxides at 10
kbar to provide tryptophol derivatives in good yields (eq
1).6 We also found that aminolysis of epoxides could be
achieved efficiently at 10 kbar of pressure (eq 2)1 and
documented the interesting behavior of silica gel as a
mild solid catalyst. The synthetic versatility of silica gel
for some related epoxide opening reactions was disclosed
by others.7 In addition to these examples, the develop-
ment of new reactions taking advantage of the catalytic
activity of silica gel surfaces is receiving considerable
attention from synthetic chemists.8 Thus, we set out to
determine the potential of epoxide opening reactions with
indoles or their analogs and to characterize the relation-
ship between high-pressure and silica gel-catalyzed reac-
tions.
Herein, we describe our detailed study of epoxide
opening reactions with a variety of nitrogen heteroaro-
matic nucleophiles: indoles, pyrroles, imidazoles, and
pyrazoles.
Resu lts a n d Discu ssion
Rea ction of In d oles. As reported, indoles have
adequate nucleophilicity toward several vinyl epoxides
at high pressure even in the absence of any catalyst.6 We
further found that aminolysis is successful under either
high-pressure or silica gel-catalyzed conditions.1 On the
basis of these observations, we first examined the nu-
cleophilic epoxide-opening reaction with indoles by ad-
sorption on silica gel.
* To whom correspondence should be addressed. Tel: +81-888-44-
8298. Fax: +81-888-44-8360. E-mail: kotsuki@cc.kochi-u.ac.jp.
† Dedicated to Prof. T. Tokoroyama upon his retirement from Osaka
City University.
When a mixture of indole (2, 1 mmol) and styrene oxide
(1, 1 mmol) was adsorbed on silica gel (500 mg) and
allowed to react at rt for 1 week, 2-(3-indolyl)-2-phenyl-
ethanol (3) was obtained in 88% yield (Table 1, run 1b).
Compared to the result obtained at high pressure (Table
1, run 1a, 10 kbar, 42 °C, 24 h in acetonitrile; 56% yield),6
the silica gel-catalyzed method was usually slow but it
involves an appealingly simple procedure. Not unexpect-
edly, conducting the same reaction in dichloromethane
solution resulted, for the most part, in a recovery of the
starting materials. This means that the supporting
environment on a silica gel surface plays an essential role
in facilitating the desired epoxide-opening reaction.
X Abstract published in Advance ACS Abstracts, J anuary 1, 1996.
(1) Part 18: Kotsuki, H.; Shimanouchi, T.; Teraguchi, M.; Kataoka,
M.; Tatsukawa, A.; Nishizawa, H. Chem. Lett. 1994, 2159.
(2) Parker, R. E.; Isaacs, N. S. Chem. Rev. 1959, 59, 737. Buchanan,
J . G.; Sable, H. Z. In Selective Organic Transformations; Thyagarajan,
B. S., Ed.; Wiley: New York, 1972; Vol. 2, p 1. Rao, A. S.; Paknikar,
S. K.; Kirtane, J . G. Tetrahedron 1983, 39, 2323. Smith, J . G. Synthesis
1984, 629. Bonini, C.; Righi, G. Ibid. 1994, 225.
(3) Pfenninger, A. Synthesis 1986, 89. Hanson, R. M. Chem. Rev.
1991, 91, 437. Schurig, V.; Betschinger, F. Chem. Rev. 1992, 92, 873.
Besse, P.; Veschambre, H. Tetrahedron 1994, 50, 8885.
(4) March, J . Advanced Organic Chemistry, 4th ed.; Wiley: New
York, 1992; p 376.
(5) For a general review of our work in this area, see: Kotsuki, H.
Kagaku to Kogyo (Osaka) 1994, 68, 265.
(6) Kotsuki, H.; Nishiuchi, M.; Kobayashi, S.; Nishizawa, H. J . Org.
Chem. 1990, 55, 2969.
(7) Bennett, F.; Patel, N. M.; Girijavallabhan, V. M.; Ganguly, A.
K. Synlett 1993, 703. Raubo, P.; Wicha, J . Synlett 1993, 25.
(8) McKillop, A.; Young, D. W. Synthesis 1979, 401, 481. Hojo, M.
Yuki Gosei Kagaku Kyokaishi 1984, 42, 635. Cornelis, A.; Laszlo, P.
Synthesis 1985, 909. Isobe, K.; Yagasaki, A. Acc. Chem. Res. 1993,
26, 524. Corma, A. Chem. Rev. 1995, 95, 559.
0022-3263/96/1961-0984$12.00/0 © 1996 American Chemical Society

High-Pressure Organic Chemistry
J . Org. Chem., Vol. 61, No. 3, 1996 985
Ta ble 1. Rea ction of In d oles w ith Ep oxid esa
Sch em e 1
substituent effect of an electron-donating group at the
5-position of indole was briefly examined, but no striking
effect was observed (Table 1, runs 4 and 5). The only
significant difference in reactivity was observed for
1-methylindole (6) (Table 1, runs 3a and 3b) for which
the silica gel-catalyzed method was superior to the high-
pressure reaction. These results can be reasonably
explained by the following mechanistic rationale. At high
pressure, the driving force for the reaction is a proton
transfer from indole to epoxide as pointed out earlier,6
while on silica gel surfaces protonation of an epoxide ring
occurs readily as depicted in Scheme 1.
In accordance with our previous observations, indoles
were only reactive toward vinyl epoxides. However, the
extension of the above methodology into aliphatic systems
would be valuable. We therefore tried the reaction of
phenyl glycidyl ether (14) with indole (2) and found that
the combination of the above two methods was most
successful. Thus, when the reaction of 14 (1 mmol) and
2 (1 mmol) containing 1000 mg of silica gel (LC-5H) was
conducted in 1.5 mL of acetonitrile at 10 kbar, the yield
of 15 increased to 29% (Table 1, run 7e). Although the
generality of the procedure has yet to be documented,
the increased activity of silica gel under high pressure
suggests the potential discrimination that might be
available with high-pressure techniques.1,10
Rea ction of P yr r oles. Pyrrole derivatives are im-
portant intermediates not only for the synthesis of drugs,
pigments, and pharmaceuticals but also for the develop-
ment of organic functional materials.11 In general,
pyrroles have rather weak nucleophilicity compared with
indoles and also show a lability to acidic conditions.
Much worse is their enhanced fragility as a result of
introduction of alkyl groups on the pyrrole nucleus. It
might therefore be necessary to perform the epoxide-
based alkylation of pyrroles under carefully controlled
conditions.12 The application of high-pressure or silica
gel-catalyzed reaction is certainly feasible here.
After numerous experiments, it was found that syn-
thetically useful amounts of products were obtained when
relatively reactive vinyl epoxides were employed as the
substrates (Table 2).
Thus, subjecting a 1:1 mixture of pyrrole (16) and
styrene oxide (1) in acetonitrile to 10 kbar of pressure
(9) Although the reason is unclear, in a series of indole reactions
the use of CH2Cl2 as the solvent gave decreased yields.
(10) Dauben, W. G.; Hendricks, R. T. Tetrahedron Lett. 1992, 33,
603.
(11) J ones, R. A.; Bean, G. P. The Chemistry of Pyrroles; Academic
Press: London, 1977. Lipshutz, B. H. Chem. Rev. 1986, 86, 795. For
a general review see: Katritzky, A., Rees, C. W., Eds. Comprehensive
Heterocyclic Chemistry; Pergamon Press: Oxford, 1984.
(12) For example, see: Tanis, S. P.; Raggon, J . W. J . Org. Chem.
1987, 52, 819.
The generality and scope of this method are sum-
marized in Table 1, where the yields obtained at high
pressure are also listed for comparison.9 As can be seen
from these data, the high pressure and the silica gel-
catalyzed method exhibit comparable efficiency. The

986 J . Org. Chem., Vol. 61, No. 3, 1996
Ta ble 2. Rea ction of P yr r oles w ith Vin yl Ep oxid esa
Kotsuki et al.
followed by purification by preparative TLC at 0 °C13
produced the 2-substituted compound 17 and the 3-sub-
stituted 18 in 22% and 8% yields, respectively (Table 2,
run 8a). Replacing the solvent with dichloromethane
caused only a slight difference in product distribution
(Table 2, run 8b). The structures of 17 and 18 were
readily confirmed by 1H and 13C NMR spectroscopy.
Thus, for compound 17 the protons on the pyrrole ring
appear at 6.00 (m), 6.16 (dd, J ) 5.3, 2.4 Hz), and 6.69
(dd, J ) 3.7, 2.4 Hz) ppm and the carbons at 105.7 (d),
108.1 (d), 117.1 (d), and 131.8 (s) ppm, indicative of a
2-substituted pyrrole derivative. The methylene proton
signal at 4.10 ppm and carbon signal at 66.4 ppm is in
good agreement with the proposed primary alcohol
structure. In compound 18, except for the pyrrole ring
fragment, quite similar data were obtained: 6.13 (dd, J
) 4.1, 2.6 Hz), 6.61 (dd, J ) 4.1, 2.0 Hz), and 6.74 (dd, J
) 4.8, 2.6 Hz) ppm for protons and 108.0 (d), 115.9 (d),
118.3 (d), and 123.6 (s) ppm for carbons. The exclusive
formation of these isomers for compounds 17 and 18
clearly shows that the pyrrole attack on styrene oxide
(1) occurs at the benzyl position with the expected
regioselectivity, as is true of indoles.
On the other hand, the silica gel-catalyzed reaction of
16 with 1 produced 17 (29%) and 18 (16%) along with
the 2,5-disubstituted derivative 19 (7%) (Table 2, run 8c).
J udging from 13C NMR measurements, this adduct 19
was deemed to be a diastereomeric mixture that was
homogeneous on TLC. Interestingly, probably due to
some decomposition of impurities on the silica gel, the
silica gel-catalyzed method led to much easier purifica-
tion than the high-pressure method. In any event, these
results demonstrate the power of these two techniques.
Similar results were obtained for N-methylpyrrole (20)
(Table 2, runs 9a-c) and also for the reaction of indene
oxide (12) with pyrrole (16) (Table 2, runs 11a-c). Under
similar conditions, 2,5-dimethylpyrrole (24) efficiently
reacted with styrene oxide (1) to afford the highly labile
compound 25 (Table 2, runs 10a-c).
As expected, all of these products were rather unstable
even in a refrigerator, and hence, all attempts to increase
the product yields were fruitless. For example, prolonged
exposure of the substrates to silica gel or longer times at
high pressure led only to low yields. In spite of this
drawback, the above results offer a valuable tool for
preparing pyrrole derivatives, which are not easily ac-
cessible by other means.
(13) Due to the instability of pyrrole adducts, separation by TLC
was best achieved at 0 °C.
Rea ction of Dia zoles. We subsequently investigated

High-Pressure Organic Chemistry
J . Org. Chem., Vol. 61, No. 3, 1996 987
Ta ble 3. Rea ction of Ep oxid es w ith Im id a zoles a n d
P yr a zolesa
Sch em e 2
1
mined on the basis of its H and 13C NMR spectral data.
Following are the characteristic signals of the aliphatic
portion of this compound: 2.18 (1H, br, OH), 4.11 (2H,
d, J ) 5.7 Hz, CH(OH)CH2N), and 4.89 (1H, t, J ) 5.7
Hz, CH(OH)CH2) ppm for protons and 52.1 (CH2N) and
70.3 (CHOH) ppm for carbons, which are in good agree-
ment with those reported for related amino alcohol
systems.1,15 This opening reaction occurs exclusively at
the less hindered side of the epoxide ring. This is
consistent with an SN2-type attack of an imidazole
nitrogen atom due to its hard nucleophilicity. In ac-
cordance with this interpretation, imidazole (29) reacted
smoothly even with aliphatic epoxides 31, 33, and 14 to
afford the corresponding adducts 32, 34, and 35 (Table
3, runs 13-15).16 Similar results were obtained for
2-methylimidazole (36) and pyrazole (38) (Table 3, runs
16 and 17).
Since the presently developed methods can be used
under very mild conditions, we believe they provide
tremendous utility for preparing a variety of imidazole
and pyrazole derivatives, as well as other analogs.17
Ster eoch em ica l F ea tu r e of Ep oxid e-Op en in g Re-
a ction w ith In d ole. We previously suggested that some
kind of ionization process might be involved in the
reaction of indole (2) with styrene oxide (1),6 since it is
clear that at high pressure this process is considerably
accelerated.18 However, at high pressure the stereo-
chemical course of our epoxide ring openings was unclear.
To clarify this and to determine the mechanistic similari-
ties between the high-pressure and silica gel-catalyzed
reactions, we turned to (R)-(+)-styrene oxide ((+)-1) as a
suitable substrate.
The results are summarized in Scheme 2. The enan-
tiomeric purity of the product was determined by HPLC
analysis using DAICEL CHIRALECEL OD column (Fig-
ure 1). Thus, the high pressure reaction of (+)-1 with 2
produced (-)-3 with 92% ee, which was subsequently
recrystallized from hexane-dichloromethane to furnish
a nearly pure sample (>99% ee by HPLC), mp 82.0-82.5
the epoxide-opening reaction using diazoles as the sub-
strates. The importance of this kind of transformation
is well established, particularly in the field of medicinal
chemistry.14 In general, the epoxide-based alkylation of
these heterocycles requires rather drastic conditions, i.e.,
prolonged heating at high temperature with the concomi-
tant use of a strong base. In light of the observations
described earlier, we expected that the application of our
strategy to these systems would produce the desired
adduct.
The results are summarized in Table 3. Imidazole (29)
was efficiently reacted with styrene oxide (1) under high-
pressure or silica gel-catalyzed conditions to afford 30 in
59% and 47% yields, respectively, as the sole product
(Table 3, runs 12a,b). The structure of 30 was deter-
°C, [R]21 -0.435 (c 0.46, CHCl3). From the silica gel-
D
catalyzed reaction, high enantiomeric excess (88% ee) of
the same sign of specific rotation was observed.
The absolute configuration at the newly formed asym-
metric center in (-)-3 was confirmed by independent
synthesis from commercially available (R)-(-)-styrenediol
(40) as outlined in Scheme 3. The corresponding TBDMS
(15) Harris, C. E.; Fisher, G. B.; Beardsley, D.; Lee, L.; Goralski, C.
T.; Nicholson, L. W.; Singaram, B. J . Org. Chem. 1994, 59, 7746 and
references cited therein.
(16) The use of CH2Cl2 as the solvent for high-pressure reactions
gave generally complex mixtures of products.
(14) For example, see: Hu¨mmer, W.; Gracza, T.; J a¨ger, V. Tetra-
hedron Lett. 1989, 30, 1517. Murabayashi, A.; Makisumi, Y. Hetero-
cycles 1990, 31, 537. Gala, D.; DiBenedetto, D. J . Tetrahedron Lett.
1994, 35, 8299. Bartroli, J .; Turmo, E.; Belloc, J .; Forn, J . J . Org.
Chem. 1995, 60, 3000.
(17) The benzene analogs such as benzimidazole were unreactive
under these conditions.
(18) le Noble, W. J . Prog. Phys. Org. Chem. 1967, 5, 207. Asano,
T.; le Noble, W. J . Chem. Rev. 1978, 78, 407. le Noble, W. J .; Asano,
T.; van Eldik, R. Ibid. 1989, 89, 549.

988 J . Org. Chem., Vol. 61, No. 3, 1996
Kotsuki et al.
high-pressure method is apparent from its reliability in
conducting the reaction in an almost neutral medium.
At the same time, we have explored the novel aspects of
silica gel as a mild acid catalyst. The combination of
these two techniques provides a new and powerful tool
for less reactive substrates. Finally, the stereochemistry
in the epoxide-opening step was determined using opti-
cally pure styrene oxide, and the reaction occurs highly
stereoselectively in an SN2 inversion manner at the
benzyl carbon, by either method. These two methods
offer broad utility in organic synthesis, and their ap-
plication to natural product synthesis and to the design
of novel heterocyclic ligands22 and functionalized materi-
als is now in progress in our laboratory.
Exp er im en ta l Section
All high-pressure reactions were conducted with a Hikari-
Koatsu HR-15-B3 apparatus, and for silica gel-catalyzed
reactions Wakogel C-300 was employed unless otherwise
noted. HPLC analyses were carried out with a UV (λ ) 254
nm) detector and DAICEL CHIRALCEL OD column with
hexane/i-PrOH as eluent. For general experimental informa-
tion see our previous paper.23
F igu r e 1. Analytical HPLC data: (A) racemic 3; (B) (-)-3
obtained at high pressure. Column: DAICEL CHIRALCEL
OD. Eluent: hexane/2-propanol ) 80:20. Flow rate: 0.5 mL/
min.
Sch em e 3
Typ ica l P r oced u r e for th e Rea ction of Ep oxid es w ith
Nitr ogen Heter ocycles. High -P r essu r e Con d ition s.
A
mixture of epoxide (1.5 mmol) and nitrogen heterocycles (1.5
mmol) in acetonitrile (1.5 mL) was placed in a Teflon reaction
vessel and allowed to react at 10 kbar under the conditions
indicated in Tables 1-3. After evaporation of the solvent, the
residue was purified by preparative TLC or by silica gel column
chromatography.
Silica Gel-Ca ta lyzed Con d ition s. To a mixture of ep-
oxide (1.5 mmol) and nitrogen heterocycles (1.5 mmol) in a
few milliliters of CH2Cl2 was added silica gel (500 mg). After
evaporation of most of the organic solvent, the residue was
allowed to stand at rt for the period indicated in Tables 1-3.
After completion of the reaction, silica gel was removed by
filtration and rinsed well with AcOEt. The organic phase was
treated as above.
2-(5-Meth yl-3-in d olyl)-2-p h en yleth a n ol (9): Rf (hexane/
AcOEt, 1:1) 0.41; mp 137.5-138.0 °C (from CH2Cl2/hexane);
ether 4119 was treated with MsCl/Et3N followed by
reaction with indolylmagnesium iodide, prepared in situ
from indole (2) and MeMgI,20 to afford 43, which was
finally desilylated by careful treatment with aqueous HF
to afford 44 with 70% ee. Although the SN2 inversion at
the stage of introduction of an indole function onto 42
was incomplete, the major isomer of 44 was identified
as (-)-3. Hence, the absolute configuration of (-)-3 was
finally deduced to be the R configuration, suggesting that
the epoxide ring opening occurred predominantly in an
SN2 inversion manner at the benzyl carbon.21 This result
can be explained by the SN2 type addition of indole to a
protonated epoxide intermeidate via a concerted mech-
anism (vide supra, Scheme 1).
1
FTIR (KBr) ν 3513, 3260, 1485, 1454 cm-1; H NMR (CDCl3)
δ 1.65 (1H br), 2.38 (3H, s), 3.9-4.6 (3H, m), 7.03 (1H, s), 6.9-
7.4 (8H, m), 7.94 (1H, br); 13C NMR (CDCl3) δ 21.5, 45.6, 66.4,
110.8, 115.3, 118.8, 122.1, 123.8, 126.5, 127.2, 128.2 (×2), 128.5
(×2), 128.6, 134.7, 141.7. Anal. Calcd for C17H17NO: C, 81.24;
H, 6.82; N, 5.57. Found: C, 81.34; H, 6.64; N, 5.41.
2-(5-Meth oxy-3-in d olyl)-2-p h en yleth a n ol (11): Rf (hex-
ane/AcOEt, 1:1) 0.34; mp 89.5-90.5 °C (from CH2Cl2/hexane);
FTIR (KBr) ν 3409, 1483, 1451 cm-1; 1H NMR (CDCl3) δ 1.73
(1H, br), 3.73 (3H, s), 4.0-4.5 (3H, m), 6.85 (1H, s), 6.7-7.4
(8H, m), 7.96 (1H, br); 13C NMR (CDCl3) δ 45.6, 55.8, 66.2,
101.4, 111.8, 112.1, 115.4, 122.7, 126.5, 127.3, 128.2 (×2), 128.4
(×2), 131.6, 141.6, 153.7. Anal. Calcd for C17H17NO2: C,
76.38; H, 6.41; N, 5.24. Found: C, 76.50; H, 6.42; N, 5.41.
3-(3-In d olyl)-1-p h en oxy-2-p r op a n ol (15): Rf (hexane/
Et2O, 1:1) 0.11; mp 82.0-83.5 °C (from CH2Cl2/hexane); FTIR
(KBr) ν 3418, 1595, 1495, 1454 cm-1; 1H NMR (CDCl3) δ 2.24
(1H, br), 3.08 (2H, d, J ) 6.4 Hz), 3.92 (1H, dd, J ) 13.5, 6.2
Hz), 3.98 (1H, dd, J ) 13.5, 4.0 Hz), 4.29 (1H, m), 6.6-7.7 (10H,
m), 8.03 (1H, br); 13C NMR (CDCl3) δ 29.5, 70.1, 71.2, 111.2,
111.3, 114.6 (×2), 118.8, 119.5, 121.0, 122.1, 122.9, 127.6, 129.4
(×2), 136.3, 158.6. Anal. Calcd for C17H17NO2: C, 76.38; H,
6.41; N, 5.24. Found: C, 76.38; H, 6.57; N, 5.29.
Con clu sion s
The high-pressure-promoted and silica gel-catalyzed
reactions of epoxides with several nitrogen heterocycles
have been demonstrated to be effective for the alkylation
of these heterocycles. The synthetic superiority of the
(19) [R]20D -38.46 (c 1.04, CHCl3). Prepared from diol 40 according
to the literature procedure: Kim, S.; Chang, H. Bull. Chem. Soc. J pn.
1985, 58, 3669. Mai, K.; Patil, G. J . Org. Chem. 1986, 51, 3545.
Yamamoto, K.; Takemae, M. Bull. Chem. Soc. J pn. 1989, 62, 2111.
Leigh, D. A.; Martin, R. P.; Smart, J . P.; Truscello, A. M. J . Chem.
Soc., Chem. Commun. 1994, 1373.
(20) Heath-Brown, B.; Philpott, P. G. J . Chem. Soc. 1965, 7165.
(21) For a similar discussion, see: Biggs, J .; Chapman, N. B.; Wray,
V. J . Chem. Soc. B 1971, 71. Lin, B.; Whalen, D. L. J . Org. Chem.
1994, 59, 1638. Boaz, N. W. Tetrahedron: Asymmetry 1995, 6, 15.
2-(2-P yr r olyl)-2-p h en yleth a n ol (17): pale brown oil; Rf
(hexane/AcOEt, 1:1) 0.41; FTIR (neat) ν 3385, 1601, 1562,
1
1493, 1453 cm-1; H NMR (CDCl3) δ 1.75 (1H, br), 4.10 (3H,
(22) For a successful application to catalytic asymmetric synthesis,
see: Kotsuki, H.; Hayakawa, H.; Wakao, M.; Shimanouchi, T.; Ochi,
M. Tetrahedron: Asymmetry 1995, 6, 2665.
(23) Kotsuki, H.; Nishikawa, H.; Mori, Y.; Ochi, M. J . Org. Chem.
1992, 57, 5036.

High-Pressure Organic Chemistry
J . Org. Chem., Vol. 61, No. 3, 1996 989
m), 6.00 (1H, m), 6.16 (1H, dd, J ) 5.3, 2.4 Hz), 6.69 (1H, dd,
J ) 3.7, 2.4 Hz), 7.26 (5H, m), 8.30 (1H, br); 13C NMR (CDCl3)
δ 47.0, 66.4, 105.7, 108.1, 117.1, 127.0, 128.2 (×2), 128.6 (×2),
131.8, 140.4; MS m/ z (rel intensity) 187 (M+, 13), 156 (100),
129 (10), 128 (10), 103 (6), 91 (6), 77 (7), 51 (3), 32 (13); HRMS
calcd for C12H13NO 187.0997, found 187.0979.
1H NMR (CDCl3) δ 2.34 (1H, br s), 2.88 (1H, dd, J ) 15.4, 7.0
Hz), 3.24 (1H, dd, J ) 15.4, 6.6 Hz), 4.08 (1H, d, J ) 6.6 Hz),
4.43 (1H, q, J ) 6.8 Hz), 6.05 (1H, m), 6.59 (1H, m), 6.73 (1H,
dd, J ) 4.6, 2.4 Hz), 7.13 (4H, m), 8.10 (1H, br); 13C NMR
(CDCl3) δ 39.8, 52.6, 81.7, 107.8, 116.0, 118.4, 123.7, 124.5,
124.9, 126.7, 126.8, 139.9, 144.4; MS m/ z (rel intensity) 199
(M+, 100), 181 (47), 180 (48), 171 (84), 156 (43), 141 (17), 128
(25), 115 (26), 104 (93), 80 (40), 77 (20), 63 (8), 51 (8), 39 (7);
HRMS calcd for C13H13NO 199.0997, found 199.0986.
2-(3-P yr r olyl)-2-p h en yleth a n ol (18): pale brown oil; Rf
(hexane/AcOEt, 1:1) 0.33; FTIR (neat) ν 3380, 1492, 1453 cm-1
;
1H NMR (CDCl3) δ 1.63 (1H, br), 4.04 (3H, m), 6.13 (1H, dd, J
) 4.1, 2.6 Hz), 6.61 (1H, dd, J ) 4.1, 2.0 Hz), 6.74 (1H, dd, J
) 4.8, 2.6Hz), 7.28 (5H, m), 8.10 (1H, br); 13C NMR (CDCl3) δ
47.0, 67.1, 108.0, 115.9, 118.3, 123.6, 126.5, 128.2 (×2), 128.4
(×2), 142.5; MS m/ z (rel intensity) 187 (M+, 12), 156 (100),
3-Hyd r oxy-2-[5-(3-h yd r oxy-2-ben zocyclop en tyl)p yr r o-
lyl]ben zocyclop en ta n e (28): Rf (hexane/AcOEt, 1:1) 0.12; mp
73.0-74.5 °C (from CH2Cl2/hexane); FTIR (KBr) ν 3312, 1588,
1
1460 cm-1; H NMR (CDCl3-acetone-d6) δ 2.82 (2H, dd, J )
129 (14), 128 (13), 77 (4), 51 (2), 32 (17); HRMS calcd for C12H13
NO 187.0997, found 187.0978.
-
15.6, 6.2 Hz), 2.85 (2H, br s), 3.20 (2H, dd, J ) 15.6, 6.6 Hz),
4.16 (2H, br d, J ) 6.2 Hz), 4.54 (2H, br dt, J ) 6.8, 6.2 Hz),
5.71 (2H, t, J ) 2.3 Hz), 7.11 (8H, m); 13C NMR (CDCl3/acetone-
d6) δ 39.4 (×2), 52.4 (×2), 79.1 (×2), 104.4, 104.6, 123.5 (×2),
124.0 (×2), 125.5 (×2), 125.9 (×2), 130.8 (×2), 139.6 (×2), 142.3
(7), 142.4 (3); MS m/ z (rel intensity) 332 (M+ + 1, 100), 331
(M+, 88), 314 (97), 303 (7), 275 (5), 228 (7), 200 (17), 198 (20),
182 (9), 133 (23), 105 (2), 91 (2), 57 (2), 41 (24); HRMS calcd
for C22H21NO2 + H 332.1651, found 332.1638.
2-[5-(2-H yd r oxy-1-p h en ylet h yl)-2-p yr r olyl]-2-p h en yl-
eth a n ol (19): brown oil; Rf (hexane/AcOEt, 1:1) 0.20; FTIR
(KBr) ν 3297, 1489, 1456, 1414 cm-1; 1H NMR (CDCl3) δ 1.80
(2H, br), 3.7-4.3 (6H, m), 5.90 (1H, dd, J ) 2.6, 1.5 Hz), 7.23
(10H, m), 8.40 (1H, br); 13C NMR (CDCl3, major diastereomer)
δ 47.1 (×2), 66.3 (×2), 105.5 (×2), 126.7 (×2), 128.1 (×4), 128.3
(×4), 131.4 (×2), 140.5, 140.6; MS m/ z (rel intensity) 307 (M+,
15), 276 (100), 258 (9), 245 (39), 156 (8), 107 (5), 91 (5), 77 (3),
40 (2), 32 (28); HRMS calcd for C20H21NO2 307.1572, found
307.1580.
2-(1-Im id a zolyl)-1-p h en yleth a n ol (30): Rf (CHCl3/MeOH,
9:1) 0.38; mp 145.0-145.5 °C (from CH2Cl2/hexane) (lit.24 mp
1
151-152 °C); FTIR (KBr) ν 3119, 1595, 1512, 1451 cm-1; H
NMR (CDCl3/CD3OD) δ 2.18 (1H, br), 4.11 (2H, d, J ) 5.7 Hz),
4.89 (1H, t, J ) 5.7 Hz), 6.7-7.5 (8H, m); 13C NMR (CDCl3/
DMSO-d6) δ 52.1, 70.3, 118.2, 124.1 (×2), 125.4, 125.7, 126.1
(×2), 135.7, 140.5. Anal. Calcd for C11H12N2O: C, 70.18; H,
6.43; N, 14.88. Found: C, 69.96; H, 6.32; N, 14.95.
2-(1-Met h yl-2-p yr r olyl)-2-p h en ylet h a n ol (21): oil; Rf
1
(hexane/AcOEt, 1:1) 0.43; FTIR (neat) ν 3410, 1491 cm-1; H
NMR (CDCl3) δ 1.57 (1H, br), 3.33 (3H, s), 3.8-4.3 (3H, m),
6.15 (2H, d, J ) 2.4 Hz), 6.58 (1H, t, J ) 2.4 Hz), 7.25 (5H,
m); 13C NMR (CDCl3) δ 33.8, 46.5, 66.4, 105.6, 106.8, 122.3,
126.9, 128.2 (×2), 128.4, 128.7 (×2), 140.2; MS m/ z (rel
intensity) 201 (M+, 12), 183 (2), 170 (100), 149 (6), 128 (8),
115 (3), 105 (5), 91 (4), 77 (4), 57 (5), 42 (7); HRMS calcd for
C13H15NO 201.1154, found 201.1139.
1-(1-Im id a zolyl)-2-d eca n ol (32): Rf (CHCl3/MeOH, 9:1)
0.27; mp 56.0-58.0 °C (from hexane) (lit.25 mp 59.5-61.0 °C);
FTIR (KBr) ν 3229, 1514, 1460 cm-1; 1H NMR (CDCl3) δ 0.88
(3H, m), 1.1-1.7 (14H, m), 3.86 (3H, m), 4.56 (1H, br), 6.88
(2H, br s), 7.34 (1H, br s); 13C NMR (CDCl3) δ 14.0, 22.5, 25.5,
29.1, 29.4, 29.5, 31.8, 34.6, 53.4, 70.3, 119.7, 127.9, 137.4.
Anal. Calcd for C13H24N2O: C, 69.60; H, 10.78; N, 12.49.
Found: C, 69.66; H, 10.91; N, 12.41.
2-(1-Met h yl-3-p yr r olyl)-2-p h en ylet h a n ol (22): oil; Rf
1
(hexane/AcOEt, 1:1) 0.38; FTIR (neat) ν 3422, 1491 cm-1; H
NMR (CDCl3) δ 1.58 (1H, br), 3.59 (3H, s), 4.03 (3H, m), 6.02
(1H, dd, J ) 2.4, 2.0 Hz), 6.42 (1H, dd, J ) 2.4, 2.0 Hz), 6.54
(1H, t, J ) 2.4 Hz), 7.28 (5H, m); 13C NMR (CDCl3) δ 36.1,
47.1, 67.1, 107.7, 119.9, 122.1, 123.7, 126.4, 128.1 (×2), 128.4
(×2), 142.6; MS m/ z (rel intensity) 201 (M+, 24), 183 (2), 170
(100), 154 (7), 149 (2), 128 (16), 115 (4), 102 (2), 77 (4), 65 (2),
42 (12); HRMS calcd for C13H15NO 201.1154, found 201.1160.
2-[1-Meth yl-5-(2-h yd r oxy-1-p h en yleth yl)-2-p yr r olyl]-2-
p h en yleth a n ol (23): Rf (hexane/AcOEt, 1:1) 0.36; mp 154.5-
156.0 °C (from CH2Cl2/hexane); FTIR (KBr) ν 3301, 1599, 1491,
1456 cm-1; 1H NMR (CDCl3) δ 1.80 (1H, br), 2.98 (3H, s), 3.7-
4.3 (6H, m), 6.18 (2H, s), 6.9-7.4 (10H, m); 13C NMR (CDCl3,
major diastereomer) δ 30.4, 46.9 (×2), 66.3 (×2), 104.5 (×2),
127.0 (×2), 128.2 (×4), 128.7 (×4), 132.1 (×2), 140.2 (×2); MS
m/ z (rel intensity) 322 (M+ + 1, 100), 304 (55), 290 (68), 259
(10), 244 (6), 202 (7), 184 (5), 168 (2), 121 (3), 41 (9); HRMS
calcd for C21H23NO2 + H 322.1807, found 322.1806.
tr a n s-2-(1-Im id a zolyl)cycloh exa n ol (34): Rf (CHCl3/
MeOH, 9:1) 0.24; mp 140.0-141.0 °C (from CH2Cl2/hexane)
(lit.26 mp 134-135 °C); FTIR (KBr) ν 3108, 1502, 1460 cm-1
;
1H NMR (CDCl3) δ 1.2-2.4 (8H, m), 3.3-3.9 (3H, m), 6.94 (2H,
br s), 7.47 (1H, br s); 13C NMR (CDCl3) δ 24.4, 25.2, 32.4, 34.3,
64.0, 73.0, 117.1, 128.3, 136.1. Anal. Calcd for C9H14N2O: C,
65.03; H, 8.49; N, 16.85. Found: C, 64.68; H, 8.36; N, 16.98.
3-(1-Im id a zolyl)-1-p h en oxy-2p r op a n ol (35): Rf (CHCl3/
MeOH, 9:1) 0.38; mp 112.5-114.0 °C (from CH2Cl2/hexane)
(lit.27 mp 108 °C); FTIR (KBr) ν 3123, 1597, 1588, 1512, 1495,
1
1466 cm-1; H NMR (CDCl3) δ 3.5-4.5 (5H, m), 6.00 (1H, br
s), 6.5-7.5 (8H, m); 13C MMR (CDCl3) δ 50.3, 68.4, 68.6, 114.4
(×2), 120.0, 121.1, 127.9, 129.4 (×2), 137.5, 158.0. Anal. Calcd
for C12H14N2O2: C, 66.03; H, 6.47; N, 12.84. Found: C, 66.04;
H, 6.59; N, 12.57.
2-(2-Met h yl-1-im id a zolyl)-1-p h en ylet h a n ol (37): Rf
(CHCl3/MeOH, 9:1) 0.37; mp 117.0-118.0 °C (from Et2O/
hexane) (lit.24 mp 117-118 °C); FTIR (KBr) ν 3111, 1534, 1503
2-(2,5-Dim et h yl-3-p yr r olyl)-2-p h en ylet h a n ol (25): Rf
(CH2Cl2/acetone, 4:1) 0.57; mp 70.5-71.5 °C (from CH2Cl2/
1
hexane); FTIR (KBr) ν 3400, 1527 cm-1; H NMR (CDCl3) δ
cm-1 1H NMR (CDCl3) δ 2.08 (3H, s), 3.98 (2H, d, J ) 6.0
;
1.73 (1H, br), 2.10, 2.19 (each 3H, s), 3.97 (3H, br s), 5.77 (1H,
d, J ) 2.4 Hz), 7.25 (5H, m), 7.45 (1H, br); 13C NMR (CDCl3)
δ 11.2, 13.1, 45.7, 66.8, 104.6, 118.0, 125.6, 126.2, 127.9 (×2),
128.1, 128.4 (×2), 142.8; MS m/ z (rel intensity) 216 (M+ + 1,
48), 215 (52, M+), 198 (100), 184 (81), 168 (2), 138 (11), 121
(11), 96 (22), 41 (8); HRMS calcd for C14H17NO + H 216.1388,
found 216.1396.
Hz), 3.9-4.2 (1H, br), 4.88 (1H, t, J ) 6.0 Hz), 6.69 (1H, d, J
) 1.3 Hz), 6.79 (1H, d, J ) 1.3 Hz), 7.27 (5H, s); 13C NMR
(CDCl3) δ 12.6, 53.7, 73.0, 120.0, 125.8 (×2), 126.0, 127.9, 128.5
(×2), 141.5, 144.9. Anal. Calcd for C12H14N2O: C, 71.26; H,
6.98; N, 13.85. Found: C, 71.16; H, 6.97; N, 13.75.
2-(1-P yr a zolyl)-1-p h en yleth a n ol (39): Rf (hexane/AcOEt,
1:2) 0.43; mp 125.5-127.0 °C (from CH2Cl2/Et2O) (lit.28 mp
tr a n s-3-Hydr oxy-2-(2-pyr r olyl)ben zocyclopen tan e (26):
1
123-127 °C); FTIR (KBr) ν 3210, 1514, 1495, 1453 cm-1; H
oil; Rf (hexane/AcOEt, 1:1) 0.43; FTIR (neat) ν 3380, 1460 cm-1
;
1H NMR (CDCl3) δ 2.42 (1H, br s), 2.85 (1H, dd, J ) 15.8, 6.5
Hz), 3.21 (1H, dd, J ) 15.8, 6.8 Hz), 4.18 (1H, d, J ) 6.6 Hz),
4.40 (1H, dt, J ) 6.8, 6.6 Hz), 5.99 (1H, m), 6.13 (1H, dd, J )
5.5, 2.6 Hz), 6.65 (1H, m), 7.13 (4H, m), 8.08 (1H, br); 13C NMR
(CDCl3) δ 39.9, 53.0, 80.9, 106.2, 108.5, 117.2, 124.8, 125.0,
127.0, 127.5, 131.5, 140.1, 141.8; MS m/ z (rel intensity) 199
(M+, 100), 181 (37), 180 (54), 171 (49), 156 (34), 141 (11), 128
(19), 115 (21), 104 (86), 80 (35), 77 (15), 63 (7), 51 (7), 39 (7);
HRMS calcd for C13H13NO 199.0997, found 199.1008.
NMR (CDCl3/acetone-d6) δ 3.36 (1H, br), 4.26 (1H, dd, J ) 14.0,
7.0 Hz), 4.33 (1H, dd, J ) 14.0, 4.8 Hz), 5.08 (1H, dd, J ) 7.0,
4.8 Hz), 6.17 (1H, t, J ) 2.0 Hz), 7.1-7.6 (7H, m); 13C NMR
(24) Cooper, G.; Irwin, W. J . J . Chem. Soc., Perkin Trans. 1 1976,
545.
(25) Manley, P. W.; Lai, M. F. Eur. Pat. Appl. EP 105,575, 1984;
Chem. Abstr. 1984, 101, 72725s.
(26) Murabayashi, A.; Makisumi, Y. Heterocycles 1990, 31, 537.
(27) Banfi, A.; Benedini, F.; Sala, A. J . Heterocycl. Chem. 1991, 28,
401.
tr a n s-3-Hydr oxy-2-(3-pyr r olyl)ben zocyclopen tan e (27):
oil; Rf (hexane/AcOEt, 1:1) 0.35; FTIR (neat) ν 3387, 1476 cm-1
;
(28) Butler, D. E.; Alexander, S. M. J . Org. Chem. 1972, 37, 215.

990 J . Org. Chem., Vol. 61, No. 3, 1996
Kotsuki et al.
(CDCl3/acetone-d6) δ 58.1, 71.9, 103.8, 125.0 (×2), 126.5, 127.2
(×2), 129.5, 137.9, 141.0. Anal. Calcd for C11H12N2O: C,
70.18; H, 6.43; N, 14.88. Found: C, 70.21; H, 6.49; N, 14.87.
(-)-2-(3-In d olyl)-2-p h en yleth a n ol [(-)-3]: Rf (hexane/
EtO, 1:1) 0.15; mp for a recrystallized pure sample 82.0-82.5
then extracted thoroughly with AcOEt. Conventional workup
followed by purification by flash column chromatography
(elution with hexane/AcOEt ) 2:1) afforded 32 mg (27% from
41) of 44. This sample was found by HPLC to be 70% ee.
°C (from CH2Cl2/hexane, >99% ee); [R]21 -0.435 (c 0.46,
D
Ack n ow led gm en t. This work was partially sup-
ported by Scientific Research Grants (Nos. 05453063
and 07304047) from the Ministry of Education, Science
and Culture of J apan. We also thank the Asahi Glass
Foundation and the Naito Foundation for financial
support of this work. We are also grateful to Prof. Y.
Fukuyama of Tokushima Bunri University for HRMS/
MS measurements.
CHCl3).
P r ep a r a tion of 44. To a solution of 41 (140 mg, 0.5 mmol),
prepared from 40 according to the literature procedure,19 and
Et3N (0.21 mL, 1.5 mmol) in CH2Cl2 (3 mL) at 0 °C was added
methanesulfonyl chloride (114 mg, 1.0 mmol), and the mixture
was stirred for 0.5 h. Conventional workup producted the
crude mesylate 42. This sample was dissolved in dry Et2O (3
mL) and transferred via cannula into a cooled (0 °C) Et2O
solution (4 mL) of indolylmagnesium iodide, prepared from
iodole (176 mg, 1.5 mmol) and MeMgI (1.58 M in Et2O; 1 mL,
1.5 mmol),20 and the mixture was stirred at rt for 3 h under
Ar. After the reaction was quenched with water, the aqueous
phase was extracted with AcOEt and treated conventionally
to afford 43, which was directly subjected to the next step.
A solution of 43 in acetonitrile (3 mL) was placed in a Teflon
reaction vessel and treated dropwise with 35% aqueous HF
(3 mL). After being stirred for 3 h at rt, the mixture was
quenched with 1 M NaOH until the solution turned basic and
Su p p or tin g In for m a tion Ava ila ble: 1H and 13C NMR
spectra of 9, 11, 15, 17-19, 21-23, 25-28, 30, 32, 34, 35, 37,
39, (-)-3, and 41 (42 pages). This material is contained in
libraries on microfiche, immediately follows this article in the
microfilm version of the journal, and can be ordered from the
ACS; see any current masthead page for ordering information.
J O951106T