Enhancing regiocontrol in carboaluminations of terminal alkynes. Application to the one-pot synthesis of coenzyme Q10

Article abstract of DOI:10.1021/ol701469e

Two new "generations" of methodological advances are reported for the Negishi carboalumination of terminal alkynes. Use of simple, inexpensive additives that alter the Al-Zr complex formed between Me₃Al and Cp₂ZrCl₂ give rise to an especially effective reagent mix that results in virtually complete control of regiochemistry upon carboalumination of 1-alkynes. One timely application to coenzyme Q₁₀ is highlighted. Regioisomers from subsequent coupling, which would otherwise be very difficult to separate, are avoided.

Full text of DOI:10.1021/ol701469e

ORGANIC
LETTERS
2
007
Vol. 9, No. 19
737-3740
Enhancing Regiocontrol in
3
Carboaluminations of Terminal Alkynes.
Application to the One-Pot Synthesis of
Coenzyme Q10
Bruce H. Lipshutz,* Tom Butler, Asher Lower, and Jeff Servesko
Department of Chemistry and Biochemistry, UniVersity of California,
Santa Barbara, California 93106
lipshutz@chem.ucsb.edu
Received June 22, 2007
ABSTRACT
Two new “generations” of methodological advances are reported for the Negishi carboalumination of terminal alkynes. Use of simple, inexpensive
additives that alter the Al Zr complex formed between Me Al and Cp ZrCl give rise to an especially effective reagent mix that results in
−
3
2
2
virtually complete control of regiochemistry upon carboalumination of 1-alkynes. One timely application to coenzyme Q10 is highlighted.
Regioisomers from subsequent coupling, which would otherwise be very difficult to separate, are avoided.
The now classical Negishi carboalumination (CA) reaction
applied to terminal alkynes is routinely effected by prior
as the regioisomeric product. Should this happen on scale-
up, the loss would also likely present major economic issues,
since separation depending upon the target can be nontrivial.
mixing of Me
of Cp ZrCl
3
Al (trimethylaluminum; TMA) and g25 mol
(zirconocene dichloride; Figure 1) in 1,2-
3
%
2
2
One solution has recently been put forth, where the Cp
1
dichloroethane (DCE). While the stereochemistry of addition
is essentially all cis, the regiochemistry is usually at best
95:5, and thus can present separation issues especially for
reactions run on a larger scale. An improvement may occur
by using the impressive Wipf modification based on addition
2
of either water or MAO (Scheme 1), although ratios on
simple substrates up to only 97:3 have occasionally been
seen. For this modest enhancement, the CA needs to be
perfomed at -70 °C. Thus, even a 90% yielding reaction is
at a disadvantage, since an additional 5-10% will be lost
(
1) (a) Van Horn, D. E.; Negishi, E. J. Am. Chem. Soc. 1978, 100, 2252.
b) Negishi, E.; Baba, S. J. Chem Soc., Chem. Commun. 1976, 596.
2) Wipf, P.; Lim, S. Angew. Chem., Int. Ed. 1993, 32, 1068.
(
Figure 1. Zirconocene catalysts.
(
1
0.1021/ol701469e CCC: $37.00
© 2007 American Chemical Society
Published on Web 08/18/2007

ligand exchange to a mixed cluster 2 (R ) i-Bu; Scheme 2).
The newly positioned isobutyl group could exert the desired
influence on the observed regioselectivity favoring, at least
initially, 3 (R ) i-Bu). It was eventually discovered that by
Scheme 1. Carboalumination with Catalyst 1
equilibrating only 5% Cp
2 2
ZrCl , 10% of inexpensive IBAO,
and Me Al, all in toluene at room temperature for only 15
3
min, a reagent was formed that effects highly selective
carboalumination. After the usual removal of solvent under
vacuum and redissolution in THF, introduction of an
electrophilic coupling partner affords the desired E-alkenyl
product. Scheme 3 shows a few representative examples, all
ligands on zirconium are replaced by the racemic ethylene
2
bridged bis-indenyl analogue, (ebi)ZrCl (i.e., the Brintzinger
4
catalyst, 1). Together with catalytic quantities of MAO, a
very selective CA takes place in toluene at rt (Scheme 1).
While straightforward procedurally, catalyst 1 is quite
Scheme 3. Representative Examples for CA with IBAO
2 2
expensive relative to the cost of Cp ZrCl . From an academic
perspective, given the requirement that both Al and Zr be
5
present within the catalytically active species, there might
be a simpler alternative that relies on changes at aluminum
rather than at the ligand(s) on zirconium. In this Letter we
disclose our latest efforts in this area leading to simplified
and economically attractive new procedures for gaining
essentially total control of regiochemistry in Negishi car-
boalumination reactions.
Fundamental mechanistic studies on carboalumination by
5
Negishi and Takahashi suggest that an Al p-orbital-alkyne-π
interaction directs a bimetallic aggregate 2 to the sterically
and stereochemically favored terminal position (Scheme 2).
Scheme 2. Potential Interaction of Catalyst with Allkyne
of which reflect >99% control of regiochemistry in the
carboalumination step. In the specific case of neutraceutical
7
coenzyme Q10, one immediate application of this new
3
technology, the 48-carbon side chain precursor 5 was
8
prepared from solanesol (4, Scheme 4), a waste product from
Scheme 4. Preparation of the CoQ10 Side Chain Precursor
Thus, a bulkier residue R on aluminum in 2 should enhance
regioselectivity. This would necessitate a “mixed” alane (i.e.,
2, R * Me) which, on the other hand, could also sterically
retard rates of carboalumination to an intolerable degree, as
already seen with variations of ligands on zirconium (e.g.,
3
2 2
Cp* ZrCl ).
To generate such a mixed alane in situ, advantage could
be taken of the rapid exchange of ligands on aluminum at
room temperature. Given the nontransferable nature of
tobacco plants. Treatment of 5 under these modified car-
boalumination conditions, which required 0.25 equiv of
5
isobutyl groups on Al (e.g., as in DIBAL), MAO’s isobutyl
6
analogue, isobutylaluminoxane (IBAO), might undergo
(6) Negishi, E. Organometallics in Organic Synthesis; Wiley, New York,
1
980; Vol. 1.
(
3) Lipshutz, B. H.; Butler, T.; Lower, A. J. Am. Chem. Soc. 2006, 128,
5396.
4) Wild, W. P.; Zsolnai, L.; Huttner, G.; Brintzinger, H. H. J.
(7) Littaru, G. P. The Fourth Conference of the International Coenzyme
1
Q10 Association; IOS Press: Amsterdam, The Netherlands, 2006. Littarru,
G. P. Energy and Defense. Facts and PerspectiVes on CoQ10 in Biology
and Medicine; Casa Editrice Scientifica Internazionale, 1994; p 1.
(8) (a) Rowland, R. L.; Latimer, P. H.; Giles, J. A. J. Am. Chem. Soc.
1956, 78, 4680. (b) Sheen, S. J.; Davis, D. L.; DeJong, D. W.; Chaolin, S.
F. J. Agric. Food Chem. 1978, 26, 259.
(
Organomet. Chem. 1982, 232, 233. Wild, F. W. R. P.; Wasiucionek, M.;
Huttner, G.; Brintzinger, H. H. J. Organomet. Chem. 1982, 288, 63.
(5) Negishi, E.; Kondakov, D. Y.; Choueiry, D.; Kasai, K.; Takahashi,
T. J. Am. Chem. Soc. 1996, 118, 9577.
3738
Org. Lett., Vol. 9, No. 19, 2007

IBAO in chlorobenzene in this case, followed by Ni-
catalyzed cross-coupling with the 6-chloromethylated 2,3-
dimethoxy-5-methyl-p-benzoquinone 6 (DMMCQ) afforded
CoQ10 directly in good yield and with excellent regioselec-
tivity (Scheme 5). Most noteworthy is the ratio of regio-
bearing the alkoxy moiety (MeOH). Success with isobutanol,
however, necessitated a switch from toluene to dichlo-
romethane (DCM), both class 2 solvents, to achieve reason-
able rates of carboalumination. Under these new conditions
the observed ratio of CoQ10:isomer 7 was >99:1, although
the isolated yield was somewhat lower than expected
3
(Scheme 7).
Scheme 5. Carboalumination/Coupling with IBAO
Scheme 7. i-BuOH-Modified Carboalumination/Coupling to
CoQ10
The generality of this latest route to highly stereo- and
regio-defined alkenes is suggested by the examples illustrated
in Scheme 8. Unlike the case above leading to CoQ10, more
isomeric products: that is, the desired quinone (CoQ10) to
its regioisomer 7 associated with this one-pot process as
determined by both NMR and HPLC analyses (>99%).
Routinely, 7-10% of isomer 7 is formed following the
traditional approach to carboalumination.1
Scheme 8. Representative Examples of Carboalumination with
i-BuOH
,6
The rationale of building bulk around Al in a Zr/Al
complex to gain regioselectivity in Negishi carboaluminations
of 1-alkynes, while presumably manifested by ligand ex-
change of isobutyl (from IBAO) for methyl, could potentially
be solved in several ways. For example, using residues on
aluminum with steric demands in between those of the two
alkyl moieties studied (Me or i-Bu) might be enough to
encourage the Al/Zr complex to migrate in the desired
direction. Alternatively, ligands on aluminum other than alkyl
were also envisioned, the simplest and most expedient being
oxygen based (i.e., an Al-OR bond). Thus, with alcohol as
an additive rather than MAO or IBAO, the catalyst should
be altered by insertion of an alkoxy rather than alkyl group.
Of several alcohols examined with the 48-carbon alkyne 5
as the benchmark substrate, an isobutanol-modified catalyst
8
(Scheme 6) led to complete carboalumination at ambient
typical 1-alkynes could be carboaluminated in toluene in the
presence of only 10% i-BuOH. Each is representative of
standard reactions of vinylalanes (e.g., [1] a copper(I)-
Scheme 6. Presumed Influence of i-BuOH on Catalyst
9
catalyzed conjugate addition; [2] simple proton quenching
of the vinylalane; and [3] a Ni-catalyzed cross-coupling to
1
an aryl iodide). The presumed alkoxy-substituted vinylalane
intermediates are noticeably of lower reactivity than those
formed by using the MAO- or IBAO-derived (alkyl-
substituted) reagents. Nonetheless, carboalumination progresses
at quite reasonable rates (2-4 h) at substrate concentrations
of ca. 1.3 M in toluene, and subsequent bond formations
temperature. Other alcohols were unproductive, either in
terms of conversion in the carboalumination step (t-BuOH,
PhOH), or subsequent coupling of the derived vinylalane
(9) Lipshutz, B. H.; Dimock, S. H. J. Org. Chem. 1991, 56, 5761.
Org. Lett., Vol. 9, No. 19, 2007
3739

occur as usual in THF in good yields and excellent
regioselectivities based on GC analyses.
addition of Me
controlled.
3
Al to a 1-alkyne can be essentially fully
Acknowledgment. We warmly acknowledge generous
financial support provided by Zymes, LLC, and the BASF.
Consultation by several scientists (Drs. Volker Berl, Karin
Schein, and Frank Wetterich) at the BASF is acknowledged
with thanks. We also are indebted to Boulder Scientific for
In summary, the evolution of carboalumination chemistry
of terminal alkynes, dating back to the seminal contribution
of Negishi, has now seen several “generations” of develop-
1
ment. As testimony to the insight of its creator, the sequence,
in a sense, has come full circle. That is, all of the reagents
2 2
supplying the Cp ZrCl used in this work, and the Albemarle
Corporation for providing TMA in chlorobenzene.
1
as originally put forth in 1978, albeit in modified amounts,
2 2
are to be found in the latest process: Cp ZrCl (reduced from
Supporting Information Available: Representative pro-
cedures for carboalumination, along with full spectral data
for all new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
2
5 to 5 mol %), excess Me Al (reduced from 2 to 1.5 equiv),
3
and DCM or toluene as solvent (environmentally preferred
alternatives to DCE). Through the expediency of an additive
(catalytic IBAO), or introduction of a simple alcohol
(isobutanol) to the Negishi recipe, regiochemistry of net cis-
OL701469E
3740
Org. Lett., Vol. 9, No. 19, 2007