Metalation of oxazole-borane complexes: A practical solution to the problem of electrocyclic ring opening of 2-lithiooxazoles

Full text of DOI:10.1021/jo960813z

5192
J . Org. Chem. 1996, 61, 5192-5193
Communications
Meta la tion of Oxa zole-Bor a n e Com p lexes:
A P r a ctica l Solu tion to th e P r oblem of
Electr ocyclic Rin g Op en in g of
2-Lith iooxa zoles
Edwin Vedejs* and Sean D. Monahan
Chemistry Department, University of Wisconsin,
Madison, Wisconsin 53705
Received May 3, 1996 (Revised Manuscript Received J une 14,
1996 )
Attempts to trap 2-lithiooxazoles with electrophiles
must contend with the complications due to the facile
equilibrium between valence bond tautomers 1 and 2.1-3
Using benzaldehyde as the electrophile, Whitney and
Rickborn reported variable yields of 3, increasing from
35% to 75% as the time for reaction of 1/2 (R4 ) Ph; R5
) H) with PhCHO was increased from 0.25 to 24 h (rt)
prior to workup.3a According to these authors, the long
reaction time allows anionic adducts derived from the
acyclic tautomer 2 to equilibrate with the precursor(s)
of cyclic 3. In a related example, Hodges et al. found that
metalation of the parent oxazole with butyllithium fol-
lowed by reaction with benzaldehyde (-78 °C to rt, 24
h) affords only 2% of the “normal” product 4, 34% of
benzyl alcohol, and 20% of the 4-substituted oxazole 6.3b
The latter product is presumably derived from conversion
of 2 into 5 followed by cyclization. An added complication
is that 2 can react with oxophilic electrophiles at enolate
oxygen as well as carbon,1,3a,b,c and direct metalation at
C4 is yet another possibility if C2 and C5 are blocked.3a
The recently reported 2-chlorozinc analogs behave more
predictably and can be intercepted at oxazole C2 using
acyl halides in the presence of CuI or Pd(0) catalysts.4,5
However, no prior study has reported practical yields in
the alkylation of 2-metalloxazoles at C2.3b,f,g
of borane complexation as described below, resulting in
a practical method for the functionalization of oxazoles
at C2.
Commercial THF-borane was combined with 5-phen-
yloxaxole7 7a , and the solution was concentrated to yield
the crystalline borane complex 8. Despite initial concerns
that the oxazole ring might be reduced by borane,8
8
proved to be surprisingly stable and could be stored for
several weeks at room temperature. Complex formation
was at least 98% complete according to NMR assay, but
8 did not crystallize efficiently. Fortunately, isolation of
8 was not necessary, and good overall yields were
obtained using oxazole borane complexes generated in
situ. The first metalation experiments followed the
2-lithiooxazole precedent.3a Thus, 8 was generated in
THF at room temperature (30 min) and was then treated
with LiTMP/THF at -78 °C. The resulting solution of 9
was quenched with benzaldehyde, followed by decom-
plexation using 5% acetic acid in ethanol to give 10
(>90% yield, Table 1). Attempted deprotonation with
LDA was not successful, but n-BuLi or s-BuLi (1.1 equiv,
30 min at -78 °C in THF) gave good results and were
more convenient than LiTMP.9 No complications due to
nucleophilic addition to the iminium subunit were de-
tected.10 Reaction times as short as 30 min or as long as
16 h at -78 °C gave similar results, in contrast to the
strongly time dependent aldehyde additions of the equili-
brating tautomers 1/2 (R4 ) Ph; R5 ) H).3a Quenching
of 9 with hydrocinnamaldehyde also worked well and
produced the alcohol 11.
It occurred to us that the undesired electrocyclic ring
opening process from 1 to 2 might be prevented if the
crucial electron pair at oxazole nitrogen could be locked
in place by complexation with a suitable Lewis acid.
Complexation was also expected to activate the C2-H
bond for metalation.6 Suppression of the electrocyclic
pathway has now been achieved by the simple expedient
(1) Schro¨der, R.; Scho¨llkopf, U.; Blume, E.; Hoppe, I. Liebigs Ann.
Chem. 1975, 533-46.
(2) Reviews: (a) Iddon, B. Heterocycles 1994, 37, 1321-46. (b)
Gilchrist, T. L. Adv. Heterocycl. Chem. 1987, 41, 41.
(3) (a) Whitney, S. E.; Rickborn, B. J . Org. Chem. 1991, 56, 3058.
(b) Hodges, J . C.; Patt, W. C.; Connolly, C. J . J . Org. Chem. 1991, 56,
449. (c) Dondoni, A.; Fantin, G.; Fogagnolo, M.; Medici, A.; Pedrini, P.
J . Org. Chem. 1987, 52, 3413. (d) Dondoni, A.; Dall’Occo, T.; Fantin,
Gi.; Fogagnolo, M.; Medici, A.; Pedrini, P. J . Chem. Soc., Chem.
Commun. 1984, 258. (e) Kozikowski, A. P.; Ames, A. J . Org. Chem.
1980, 45, 2548. (f) Wasserman, H. H.; McCarthy, K. E.; Prowse, K. S.
Chem.Rev. 1986, 86, 845. (g) J acobi, P. A.; Ueng, S.; Carr, D. J . Org.
Chem. 1979, 44, 2042. (i) Howe, R. K.; Lee, L. F. Eur. Pat. Appl. 27020;
Chem. Abstr. 1981, 95, 80933.
(4) Crowe, E.; Hossner, F.; Hughes, M. J . Tetrahedron 1995, 51,
8889.
(5) Harn, N. K.; Gramer, C. J .; Anderson, B. A. Tetrahedron Lett.
1995, 36, 9456.
(6) (a) Kessar, S. V.; Singh, P.; Vohra, R.; Kaur, N. P.; Singh, K. N.
J . Chem. Soc., Chem. Commun. 1991, 568. (b) Kessar, S. V.; Singh,
P.; Singh, K. N.; Dutt, M. J . Chem. Soc., Chem. Commun. 1991, 570.
(c) Kessar, S. V.; Vohra, R.; Kaur, N. P.; Singh, K. N.; Singh, P. J .
Chem. Soc., Chem. Commun. 1994, 1327. (d) Kessar, S. V.; Singh, P.;
Singh, K. N.; Kaul, V. K.; Kumar, G. Tetrahedron Lett. 1995, 36, 8481.
(e) Ebden, M. R.; Simpkins, N. S. Tetrahedron Lett. 1995, 36, 8697.
(7) Van Leusen, A. M.; Hoogenboom, B. E.; Siderius, H. Tetrahedron
Lett. 1972, 2369-72.
(8) Knapp, K. K.; Keller, P. C.; Rund, J . V. J . Chem. Soc., Chem.
Commun. 1978, 971.
(9) To a solution of 7b7 (1.00 g, 6.89 mmol) in 34 mL of THF at rt
under nitrogen was added BH3-THF (6.9 mL, 1 M in THF, Aldrich).
After 30 min, the solution was cooled to -78 °C, and n-BuLi (4.44 mL,
1.63 M in hexane, Aldrich) was added dropwise. After 30 min,
benzaldehyde (0.77 mL, 7.58 mmol) was added. The mixture was
stirred for 30 min, and 34 mL of 5% HOAc in ethanol was added. The
cooling bath was removed, and the mixture was stirred for 18 h at rt
to cleave the borane complex. Conventional workup (supporting
information) gave 10, isolated by crystallization from ether, two crops
(1.4 g, 81%).
(10) (a) Turchi, I. J . In Oxazoles; Turchi, I. J ., Ed.; Wiley: New York,
1986; Chapter 1. (b) Vedejs, E.; Grissom, J . W. J . Org. Chem. 1988,
53, 1876.
(11) Vedejs, E. Tucci, F. C. Unpublished results.
S0022-3263(96)00813-4 CCC: $12.00 © 1996 American Chemical Society

Communications
J . Org. Chem., Vol. 61, No. 16, 1996 5193
Ta ble 1. Rea ction of 2-Lith iooxa zole-Bor a n e Com p lexes
Ta ble 2. Rea ction s of 2-Lith io-5-p h en yloxa zole-Bor a n e
w ith Ald eh yd esa
Com p lex 9 w ith Electr op h ilesa
entry oxazole
aldehyde
PhCHO
PhCHO
PhCHO
PhCH2CH2CHO n-BuLi
PhCHO
PhCHO
base
product yield (%)
entry
electrophile
Me3SiCl
C2Cl6(2 equiv)
C2Cl6(0.5 equiv)
CH3I
T (°C)
product
yield (%)
1
2
3
4
5
6
7a
7a
7a
7a
7b
15
LiTMP
n-BuLi
s-BuLi
10
10
10
11
14
16
94b
1
2
3
4
5
-20
-78
-78
-20
-20
17
19
20
21
22
78b
86
88;c 81d
90b
79
84e
70
84
74c
65d
s-BuLi
n-BuLi
PhCH2CH2OTf
a
Reactions were performed in THF using 1.05 equiv of n-BuLi
to generate the anion at -78 °C, followed by addition of the
electrophile at the indicated temperature. After 16 h, the reactions
were allowed to warm and were treated with 5% acetic acid/ethanol
a
Reactions were performed in THF at -78 °C using 1.05 equiv
of base and 1.1 equiv of aldehyde. After reaction times of 30 min
or longer (see notes), the resulting mixtures were quenched and
b
b
decomplexed with 5% acetic acid in ethanol (18 h, rt). Yield of
to decomplex the borane. Yield of borane complex prior to acetic
major fraction after chromatography and crystallization; 16 h
acid/ethanol treatment; 13% of unreacted 8 was also isolated.
reaction time with PhCHO. c Yield of major fraction after chro-
c Yield of sublimed product. Ca. 5% recovered 7a was also
d
d
matography; 2 h reaction time with PhCHO. Yield of product
present.
crystallized after aqueous workup; 30 min reaction time with
PhCHO. e Yield of crystallized product; 20 min reaction time with
RCHO.
equiv of hexachloroethane, then the bis-oxazole 201,13 was
obtained in good yield (Table 2, entry 3). This product
is presumably formed by reaction of the initial product
18 with 9.
Alkylation reactions of 9 were also studied (Table 2,
entries 4 and 5). Treating the anion with iodomethane
at -20 °C (16 h) afforded 2-methyl-5-phenyloxazole (21)14
(74% yield). Attempts to perform a similar alkylation
with phenethyl bromide or iodide failed to give significant
conversion of 9, but the corresponding triflate, Ph(CH2)2-
OSO2CF3,15 was sufficiently reactive at -20 °C (18 h) and
afforded 2216 (65% yield). Similarly, oxazole (7b) was
alkylated to give 23 (76%).17 Prior literature attempts
to alkylate oxazoles at C2 resulted in yields of 30% or
less.3b,f,g The reaction of 9 with benzoyl chloride was also
explored briefly. However, a mixture of products was
obtained including ca. 15% of the benzoate ester of alcohol
10, apparently derived from reduction of the expected
2-benzoyl-5-phenyloxazole. The benzoylation was not
explored further in view of the successful organozinc
alternative.4,5
In summary, the 2-lithiooxazole complexes 9 and 13
have been shown to undergo facile reaction with repre-
sentative electrophiles. The troublesome anionic elec-
trocyclic ring opening reaction is completely suppressed
using the borane complexation procedure.
A similar complexation-metalation procedure proved
effective for the parent oxazole 7b.3b,4 Thus, 12 was
prepared in THF solution, and 13 was generated by
treatment with butyllithium as before. Addition of
benzaldehyde afforded 14 (70% isolated; Table 1, entry
5). No benzyl alcohol was observed (<3%), and no ring-
opened products or C4-substituted oxazoles were detected,
in contrast to the results using 2-lithiooxazole 1/2 (R4 )
R5 ) H).3b In the same way, the 5-alkyloxazole 1511 was
converted into 16 (84%; Table 1, entry 6).
To further define the scope of this methodology, the
anion 9 derived from 5-phenyloxazole-borane complex
(8) was trapped with a variety of electrophiles (Table 2).
Trimethylchlorosilane afforded a 6:1 mixture of 17:8
according to NMR assay (Table 2, entry 1), but attempted
decomplexation using protic conditions (5% acetic acid
in ethanol) resulted only in the desilylated oxazole 7a .
Anion trapping with excess (2 equiv) hexachloroethane
resulted in the formation of 1912 (Table 2, entry 2). On
the other hand, if quenching was performed using 0.5
Ack n ow led gm en t. This work was supported by a
grant from the National Institutes of Health (CA17918).
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
1
cedures and H NMR spectra for 8, 10, 11, 14, 16, and 19-23
(17 pages).
J O960813Z
(12) (a) Harrison, R. G.; J amieson, W. B.; Ross, W. J .; Saunders, J .
C. Ger. Offen. Chem. Abstr. 1976, 86, 155658. (b) Harrison, R. G. Ger.
Offen. 2,459,380; Chem. Abstr. 1976, 86, 121320. (c) Maquestiau, A.;
Ben A., Fouad B.; Flammang, R. Bull. Soc. Chim. Belg. 1990, 99, 89-
101.
(13) Hayes, F. N.; Rogers, B. S.; Ott, D. G. J . Am. Chem.Soc. 1955,
77, 1850.
(14) Kondrat’eva, G. Y.; Chzhi-ken, K. Zh. Obsch. Khim. 1962, 32,
2348.
(15) Lee, C. C.; Unger, D. Can. J . Chem. 1973, 51, 1494.
(16) Kashima, C.; Arao, H.; Okada, R. Heterocycles 1990, 30, 487.
(17) Note Ad d ed in P r oof: The same methods could be used for
BH3 activation, lithiation, and PhCH2CH2OTf alkylation of thiazole
(85% isolated yield).