Mercury-free preparation and selective reactions of propargyl (and propargylic) grignard reagents

Article abstract of DOI:10.1021/ol071397f

ZnBr₂ was found to catalyze formation of propargyl and propargylic Grignard reagents, and thus put an end to the standard method using a mercury catalyst. The Grignard reagents were submitted to addition reaction with carbonyl compounds and allylation with the cyclic monoacetate to afford the propargyl-type products selectively. Furthermore, the product from the monoacetate was transformed to an acetylene analogue of 2-(5,6-epoxyisoprostane A2)phosphorylcholines.

Full text of DOI:10.1021/ol071397f

ORGANIC
LETTERS
2
007
Vol. 9, No. 18
535-3538
Mercury-Free Preparation and Selective
Reactions of Propargyl (and
Propargylic) Grignard Reagents
3
†
Hukum P. Acharya, Kei Miyoshi, and Yuichi Kobayashi*
Department of Biomolecular Engineering, Tokyo Institute of Technology,
4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8501, Japan
ykobayas@bio.titech.ac.jp
Received June 13, 2007
ABSTRACT
2
ZnBr was found to catalyze formation of propargyl and propargylic Grignard reagents, and thus put an end to the standard method using a
mercury catalyst. The Grignard reagents were submitted to addition reaction with carbonyl compounds and allylation with the cyclic monoacetate
to afford the propargyl-type products selectively. Furthermore, the product from the monoacetate was transformed to an acetylene analogue
2
of 2-(5,6-epoxyisoprostane A )phosphorylcholines.
1
As illustrated in Scheme 1, reaction of propargyl (HCt
CCH -) and propargylic (RCtCCH -) Grignard reagents
the reaction course in the transition state. While the high
2
2
level of native selectivity has found many applications in
organic synthesis, study to improve the low selectivity was
scarcely reported probably due to the use of a poisonous
mercury catalyst such as HgCl
2
for the preparation of
Scheme 1. Standard Preparation of the Propargyl (and
Propargylic) Grignard Reagents 2 and the Reaction Products
with Electrophiles
2-4
reagents 2 from halides 1 and magnesium in Et
2
O.
1
In the preparation of 2a (R ) H), the mercury catalyst is
understood to speed up the formation of 2a over the
unwanted consumption of 2a via abstraction of the acetylenic
(
1) (a) Yanagisawa, A. Sci. Synth. 2004, 7, 541-547. (b) Marshall, J.
A. Chem. ReV. 2000, 100, 3163-3185. (c) Brandsma, L. PreparatiVe
Acetylenic Chemistry, 2nd ed.; Elsevier: Amsterdam, The Netherlands,
1
988; pp 35-36. (d) Gaudemar, M. Ann. Chim. 1956, 1, 161-213.
2) Early examples of the preparation of 2a: (a) Stadler, P. A.; Nechvatal,
A.; Frey A. J.; Eschenmoser, A. HelV. Chim. Acta 1957, 40, 1373-1409.
b) Sondheimer, F.; Wolovsky, R.; Ben-Efraim, D. A. J. Am. Chem. Soc.
(
(
1
961, 83, 1686-1691. (c) Sondheimer, F.; Amiel, Y.; Gaoni, Y. J. Am.
Chem. Soc. 1962, 84, 270-274. (d) Viola, A.; MacMillan, J. H. J. Am.
Chem. Soc. 1968, 90, 6141-6145. (e) Hopf, H.; Bohm, I.; Kleinschroth, J.
Organic Syntheses; Wiley: New York, 1990; Collect. Vol. VII, pp 485-
4
90.
(
3) Preparation of 2b: (a) Mesnard, D.; Miginiac, L. J. Organomet.
Chem. 1990, 397, 127-137. (b) Eckenberg, P.; Groth, U.; K o¨ hler, T. Liebigs
+
2
with electrophiles (E ) has produced 4 and/or 5 with low
Ann. Chem. 1994, 673-677. (c) Hernandez, E.; Soderquist, J. A. Org. Lett.
2
005, 7, 5397-5400. (d) Zhang, L.; Kozmin, S. J. Am. Chem. Soc. 2004,
to high selectivity depending mainly on the steric factor and
126, 10204-10205.
1
(
4) Preparation of 2 (R * H): Yanagisawa, A.; Habaue, S.; Yamamoto,
†
Dedicated to the late Professor Yoshihiko Ito, Kyoto University.
H. Tetrahedron 1992, 48, 1969-1980.
1
0.1021/ol071397f CCC: $37.00
© 2007 American Chemical Society
Published on Web 08/09/2007

1
proton from 1a (R ) H) to produce HCtCMe, which is
these anions with 1a to afford MgBr
activation of Mg with Br(CH Br in THF followed by
reaction with 1a produced similar precipitates. On the other
hand, attempted preparation in Et O with and without the
Br) resulted in the
complete recovery of Mg.
To identify effective catalysts, the preparation of 2a in
2
as precipitates. Pre-
1
ultimately changed to BrMgCtCMe (6). On the other hand,
the role of the catalyst in the preparation of 2 (R * H) is
not clearly stated to the best of our knowledge. To avoid
the problem associated with recovery of the mercury catalyst,
2 2
)
1
2
pre-activation of Mg (with Br(CH
2 2
)
1
Rieke Mg was once used for the preparation of 2 (R )
5
TMS). Use of zinc and lithium reagents similar to 2 (R )
,6,7
8
Si(alkyl)
3
, alkyl)4 and the dianion, LiCtCCH
2
Li, has been
Et O was examined with a number of metal salts (2-5 mol
2
an alternative choice. These anions have been synthesized
by reaction of 1 with Zn, by lithiation of 2-alkynes with
t-BuLi,6 and by lithiation of propyne and allene (CH
%) from a stock room. While most of the common metal
salts placed in group 1 of Table 1 were ineffective,
4
,7
2
dCd
8
CH
2
) with nBuLi, respectively. On the other hand, several
metals have been shown to assist generation of 2 in Barbier-
type reactions. These methods, however, seem less attractive
9
Table 1. Results of the Metal Salts Attempted for Preparation
of 2a in Et O
2
a
because of the complicated operation to prepare Rieke Mg,
inconvenience in handling gaseous propyne and allene for
the preparation of 2-alkynes as well as for the direct lithiation,
low reactivity and product selectivity, or the narrow range
of electrophiles for Barbier-type reactions.
consumption
of Mgb
concn
of 2a (M)c
group
metal salts
1
CuCl2
NiCl2
HfCl4
ZrCl4
no
no
CoCl2
CeCl3
FeCl3
Bu3SnCl
Zn(OAc)2
Zn(acac) 2
no
no
no
In contrast to the above method, two recent reports have
1
described the preparation of 2a (R ) H) in the usual way
without a mercury catalyst.10 Due to the significance, we
reinvestigated the preparation and confirmed that the mercury
catalyst is indeed necessary. Instead, we found an environ-
Cp2TiCl2
TiCl4
Zn(OTf)2
no
2
Zn(OTs)2
ZnEt2
yes
yes
yes
0
0
1
mentally acceptable catalyst for the preparation of 2 (R )
3
ZnX (X ) Cl, Br, I)
0-0.52
2
H, TMS, alkyls). Herein, we describe the preparation of 2
and applications including the synthesis of 5,6-epoxyiso-
prostane phosphorylcholine and its acetylene derivative.
After confirmation of the successful preparation of 2a in
a
Preparation was examined with 1a (1.0 mL, 13 mmol), Mg (650 mg,
27 mg-atom), and a catalyst (2-5 mol %) in Et O (18 mL) at 0-5 °C for
2
b
1
-2 h. Mg turnings from Nacalai Tesque, Japan, were used for the
c
investigation. Determined by titration with methyl orange.
2
Et O (0.38 M by titration with methyl orange, 54% yield
based on 0.70 M for 100% conversion) by adding propargyl
bromide (1a) (1 mL scale) to Mg turnings (2 equiv) in the
consumption of Mg was observed with TiCl
though concentration of 2a by titration was almost 0 M. To
our delight, production of 2a was brought about with ZnX
group 3, Table 1). Further investigation with other zinc salts
was, however, unsuccessful (see the zinc salts listed in groups
4
(group 2),
presence of HgCl
solvent and the catalyst was briefly studied. In THF at 0-5
C (ca. 2 h), Mg was consumed smoothly with formation of
gray precipitates. Surprisingly, the concentration was almost
M by titration, which was consistent with the result that
an attempted reaction of the mixture (supernatant and
precipitates) with Ph(CH CHO (7) did not afford any
2
(0.2 mol %), the requirement of the
2
(
°
1
and 2).
0
2
Concentrations of 2a prepared with ZnX of group 3 are
summarized in Table 2. For the two catalyst quantities (2
2 2
)
alcoholic products. These results indicate a pathway to
quench anions 2a and/or 6. One possibility is a coupling of
Table 2. Concentration of 2a vs Quantity of ZnX
2
(
5) Paquette, L. A.; Han, Y.-K. J. Am. Chem. Soc. 1981, 103, 1831-
mol %
of ZnX2
1
1
835.
(
6) Daniels, R. G.; Paquette, L. A. Tetrahedron Lett. 1981, 22, 1579-
ZnCl2
ZnBr2
ZnI2
582.
(
7) (a) Corey, E. J.; Kirst, H. A. Tetrahedron Lett. 1968, 5041-5043.
b) Corey, E. J.; R u¨ cker, C. Tetrahedron Lett. 1982, 23, 719-726. (c)
Commercon, A.; Normant, J.; Villieras, J. J. Organomet. Chem. 1975, 93,
4
4
2
1
2
3
4
5
0.39
0.41
0.51
0.52
(
0.31
0.29
0
15-421. (d) Pearson, N. R.; Hahn, G.; Zweifel, G. J. Org. Chem. 1982,
7, 3364-3366. (e) Ma, S.; Zhang, A.; Yu, Y.; Xia, W. J. Org. Chem.
000, 65, 2287-2291.
0.35
a
0.49
(8) (a) Hooz, J.; Calzada, J. G.; McMaster, D. Tetrahedron Lett. 1985,
a
Mg turnings from Aldrich gave a similar result (0.48 M).
2
6, 271-274. (b) Hooz, J.; Cabezas, J.; Musmanni, S.; Calzada, J. Org.
Synth. 1990, 69, 120-127. (c) Cabezas, J.; Alvarez, L. X. Tetrahedron
Lett. 1998, 39, 3935-3958. (d) Lipshutz, B. H.; Lower, A.; Berl, V.; Schein,
K.; Wetterich, F. Org. Lett. 2005, 7, 4095-4097.
(9) For example: (a) Masuyama, Y.; Ito, A.; Fukuzawa, M.; Terada,
and 4 mol %), the highest concentrations (0.41 and 0.52 M,
respectively) were recorded with ZnBr and were slightly
better than that obtained above with HgCl (0.38 M). A
similar result was attained with Mg from a different company
see footnote a of Table 2). A somewhat lower concentration
K.; Kurusu, Y. Chem. Commun. 1998, 2025-2026. (b) Alcaide, B.;
Almendros, P.; Aragoncillo, C. Org. Lett. 2000, 2, 1411-1414. (c) Nair,
V.; Jayan, C. N.; Ros, S. Tetrahedron 2001, 57, 9453-9459. (d) Inoue,
M.; Nakada, M. Angew. Chem., Int. Ed. 2006, 45, 252-255.
2
2
(
10) (a) Oppolzer, W.; Flaskamp, E.; Bieber, L. W. HelV. Chim. Acta
001, 84, 141-145. (b) Schrock, R. R.; Duval-Lungulescu, M.; Tsang, W.
C. P.; Hoveyda, A. H. J. Am. Chem. Soc. 2004, 126, 1948-1949.
(
2
was recorded with 1 mol %. On the basis of these results,
3536
Org. Lett., Vol. 9, No. 18, 2007

we recommend the use 2-5 mol % of ZnBr
successful preparation of 2a. Note that an attempted prepara-
tion of 2a in THF with ZnBr (4 mol %) produced the
precipitates and a supernatant of 0 M concentration.
The ZnBr -assisted preparation established above was
applied to 1b (R ) TMS) successfully with 2 mol % of
ZnBr . The yield of 2b (75-86%, 0.49-0.56 M, 3 runs; cf.
.65 M for 100% production) was a little higher than that
obtained with HgCl (65-77%, 3 runs). The preparation of
b with ZnBr attempted in THF was also successful (0.43
M). The concentration was, however, substantially dropped
to 0.21 M after 12 h at 0 °C, suggesting a similar process to
5 8
quench 2b. Bromides 1c and 1d (R ) C H11, C H17) also
produced 2c and 2d in 85% and 83% yields, respectively.
With the nontoxic preparation of the Grignard reagents 2
in hand, we turned our attention to reactions delineated in
eqs 1-3. Addition of 2a and 2b to cyclohexanone 8 produced
2
in Et
2
O for
Previously, lithium anion 12 was prepared by lithiation of
TMSCtCMe with n-BuLi for reaction with 11 (eq 2), and
product 13 was transformed into clavulone II and its
analogues. Now, preparation of TMSCtCMe and its
lithiation to 12 are no longer necessary.
2
1
3
14
2
1
Next investigated was allylic substitution of 14 with copper
15
2
reagents derived from 2b (eq 3), in which the regioselection
at the R or γ position of the allylic moiety should be
controlled as well. According to the guideline established
0
2
1
6
2
2
by us for alkyl and aryl reagents, a 2:1 complex of 2b and
CuCN in THF was examined first, but produced the diol
derivative of 14. After unsuccessful attempts, we found that
a 1:1 complex formulated as 15 (4 equiv) at 0 °C for 4 h
afforded the propargyl type of the products, but with a low
regioselectivity on the substrate 14 (Table 3, entry 1).
Table 3. Allylic Substitution of 14 with Propargylic Anion 15
Derived from 2b and CuCNa
method for
preparation of 2b
additive
(equiv)
ratio of
16:17b
combined
yield (%)
c,d
entry
1
2
3
4
5
HgCl2-catalyzed
HgCl2-catalyzed
HgCl2-catalyzed
ZnBr2-catalyzed
ZnBr2-catalyzed
68:32
85:15
94:6
94:6
87:13
83
81
76
77
79
MgCl2 (4)
MgCl2 (8)
MgCl2 (8)
LiCl (8)
a
Reactions were carried out with 2b (4 equiv) and CuCN (4.2 equiv) in
b
1
c
THF at 0 °C for 3-6 h. Determined by H NMR spectroscopy. Isolated
yield. d The products 16 and 17 were separable by chromatography on silica
gel.
Fortunately, higher than the native selectivity was attained
with 8 equiv of MgCl
examined in entries 1-3 was prepared by using the HgCl
catalyzed method, we also executed the reaction with 2b
prepared with ZnBr to confirm almost the same selectivity
2
(entry 3; cf. entry 2). Since 2b
2
-
2
and yield was obtained (entry 4), thus showing no interfer-
ence to the allylation by the mercury and zinc residues
remaining in the solutions of 2b.
Finally, the allylation established above (eq 3) was applied
to the synthesis of the acetylene analogue of 2-(5,6-
1
7,18
epoxyisoprostane A
2
)phosphorylcholine
(18b in Figure
(
13) Tanaka, H.; Hasegawa, T.; Iwashima, M.; Iguchi, K.; Takahashi,
T. Org. Lett. 2004, 6, 1103-1106.
14) While the authors of ref 13 did not mention the preparation, TMSCt
CMe has been synthesized by the methods: (a) Corey, E. J.; Kirst, H. A.;
Katzenellenbogen, J. A. J. Am. Chem. Soc. 1970, 92, 6314-6319. (b) Wang,
J.; Gurevich, Y.; Botoshansky, M.; Eisen, M. S. J. Am. Chem. Soc. 2006,
(
the homopropargylic alcohols in good yields with a high
product selectivity over the allenes (eq 1). The high selectivi-
ties observed are consistent with those reported for cyclic
1
28, 9350-9351. (c) See ref 6.
ketones3
HgCl
yield and selectivity of the reaction. Likewise, 4-hydroxy-
-cyclopentenone (11) underwent reaction with 2b to afford
alcohol 13 with good yield and high selectivity (eq 2).
a,7b,11,12
with 2a and 2b that were prepared with
(15) Reactions with the propargylic copper reagents: (a) Lam, H. W.;
Pattenden, G. Angew. Chem., Int. Ed. 2002, 41, 508-511. (b) Bednarski,
P. J.; Nelson, S. D. J. Med. Chem. 1989, 32, 203-213. (c) Wender, P. A.;
Hilinski, M. K.; Soldermann, N.; Mooberry, S. L. Org. Lett. 2006, 8, 1507-
1510. (d) See ref 7c.
2
, thus concluding no influence of the zinc residue on
2
(
16) (a) Kobayashi, Y.; Nakata, K.; Ainai, T. Org. Lett. 2005, 7, 183-
1
86. (b) Ito, M.; Matsuumi, M.; Murugesh, M. G.; Kobayashi, Y. J. Org.
Chem. 2001, 66, 5881-5889.
(
11) Quayle, P.; Rahman, S.; Ward, E. L. M. Tetrahedron Lett. 1994,
5, 3801-3804.
12) A similar selectivity for the 2-bromo derivative: Hirama, M.;
Fujiwara, K.; Shigematu, K.; Fukazawa, Y. J. Am. Chem. Soc. 1989, 111,
120-4122.
(17) (a) Watson, A. D.; Subbanagounder, G.; Welsbie, D. S.; Faull, K.
F.; Navab, M.; Jung, M. E.; Fogelman, A. M.; Berliner, J. A. J. Biol. Chem.
1999, 274, 24787-24798. (b) Subbanagounder, G.; Wong, J. W.; Lee, H.;
Faull, K. F.; Miller, E.; Witztum, J. L.; Berliner, J. A. J. Biol. Chem. 2002,
277, 7271-7281.
3
(
4
Org. Lett., Vol. 9, No. 18, 2007
3537

2
Scheme 2. Synthesis of the Acetylene Derivative and Natural 2-(5,6-Epoxyisoprostane A )phosphorylcholines 18b and 18a
). Monoacetate (1R)-1419 of >99% ee by HPLC was
21
1
active epoxy aldehyde 23 gave aldol adduct 24, which was
transformed to dienone 25 stereoselectively at the newly
20
converted into pivaloate 19 (Scheme 2). The reaction was
formed olefin through mesylation and Al
nation of the mesyloxy group. Hydrolysis of methyl ester
5 by using PPL afforded acid 26, which upon condensation
2 3
O -assisted elimi-
2
with lyso-PC (27) using Yamaguchi reagent furnished 18b.
To elucidate a step for tritium labeling, early intermediate
2
1, ester 25, and acid 26 were submitted to hydrogenation
instead of T , Pd/BaSO , quinoline). The former two
afforded cis olefins (81% from 21, 78% of 28 from 25
(H
2
2
4
(
Scheme 3)), while 26 gave a mixture of products. Previ-
18a
ously, 28 was transformed to 18a by us in two steps.
Scheme 3. Hydrogenation of Acetylene 25
Figure 1. The 2-(5,6-epoxyisoprostane A
family (18a-c).
2
)phosphorylcholine
somewhat slow with 4 equiv of 15 and hence an excess of
the reagent was used to produce, after 2 h at 0 °C, a 93:7
mixture of (1R)-16 and the regioisomer in good yield. After
separation of the mixture by chromatography, (1R)-16
obtained in 80% yield was converted to enone 22 through
alkylation at the acetylenic carbon and subsequent oxidation
of the hydroxyl group. Aldol reaction of 22 with the optically
In summary, a new method for mercury-free preparation
of 2 (R ) H, TMS, C H11, C H17) was established for the
5 8
first time, facilitating investigation of reactions with 2. We
also studied reactions of 2 shown in eqs 1-3, which proceed
regio- and/or stereoselectively. Furthermore, the reaction of
eq 3 was applied to the synthesis of 2-(5,6-epoxyisoprostane
1
(
18) (a) Acharya, H. P.; Kobayashi, Y. Angew. Chem., Int. Ed. 2005,
4, 3481-3484. (b) Jung, M. E.; Berliner, J. A.; Angst, D.; Yui, D.;
Koroniak, L.; Watson, A. D.; Li, R. Org. Lett. 2005, 7, 3933-3935.
19) (a) Sugai, T.; Mori, K. Synthesis 1988, 19-22. (b) Laumen, K.;
Schneider, M. P. J. Chem. Soc., Chem. Commun. 1986, 1298-1299.
20) Although (1S)-14 is the direct substrate to give (1R)-16, synthesis
4
A
2
)phosphorylcholine 18a and its acetylenic analogue 18b.
(
Acknowledgment. This work was supported by a Grant-
in-Aid for Scientific Research from the Ministry of Educa-
tion, Science, Sports, and Culture, Japan.
(
of (1S)-14 by the asymmetric hydrolysis of the diacetate by PLE of several
lots was unsuccessful in our hand. The hydrolysis with PLE: (a) Wang,
Y.-F.; Chen, C.-S.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1984, 106,
Supporting Information Available: Experimental pro-
cedures and spectral data of compounds described herein.
This material is available free of charge via the Internet at
http://pubs.acs.org.
3
695-3696. (b) Laumen, K.; Schneider, M. Tetrahedron Lett. 1984, 25,
5
875-5878. Synthesis of 19: (c) Curran, T. T.; Hay, D. A.; Koegel, C. P.
Tetrahedron 1997, 53, 1983-2004. (d) Myers, A. G.; Hammond, M.; Wu,
Y. Tetrahedron Lett. 1996, 37, 3083-3086.
(21) Rossiter, B. E.; Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc.
1
981, 103, 464-465.
OL071397F
3538
Org. Lett., Vol. 9, No. 18, 2007