Mild and efficient palladium(II)-catalyzed racemization of allenes

Article abstract of DOI:10.1039/b316482a

Allenes undergo racemization in the presence of catalytic amounts of Pd(OAc)₂/LiBr under mild conditions; the reaction proceeds via a bromopalladation-debromopalladation sequence and tolerates various functional groups.

Full text of DOI:10.1039/b316482a

Mild and efficient palladium(II)-catalyzed racemization of allenes†
Attila Horváth and Jan-E. Bäckvall*
Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, SE-106 91 Stockholm,
Sweden. E-mail: jeb@organ.su.se
Received (in Cambridge, UK) 17th December 2003, Accepted 3rd February 2004
First published as an Advance Article on the web 9th March 2004
Allenes undergo racemization in the presence of catalytic
amounts of Pd(OAc)2/LiBr under mild conditions; the reaction
proceeds via a bromopalladation–debromopalladation sequence
and tolerates various functional groups.
propargylic alcohols and organocopper reagents according to
literature procedures. The acetates 4a,b and amides 6a–c were
obtained from the corresponding alcohols 3a,b and amines 5a,b,
respectively, by use of standard methods. Finally, b-allenyl ester 8
was prepared via an orthoester Claisen rearrangement according to
a literature procedure.17
Our first attempt to racemize allenyl carbamate 6a resulted in a
clean but slow reaction (Table 1, entry 1). Increasing the
temperature to 50 °C accelerated the reaction by one order of
magnitude (entry 2). A further increase of rate was obtained by
adding water to the reaction mixture (entries 3–6). This, however,
turned out to be applicable only to carbamates 6a and 6b.
To find a general procedure we turned our attention to a-allenyl
alcohol 3c. The reaction at 50 °C was fast but proceeded to only
about 75–80% racemization even when reaction times were
prolonged. This inhibition is probably due to deactivation of the
catalyst via dimer formations (see the mechanisms in Scheme 3 and
4). To avoid this deactivation we first screened several solvents
such as ethanol, dioxane, toluene, acetone, acetic acid, dichloro-
methane and acetonitrile. Of these solvents acetonitrile gave the
best result. We next decreased the concentration of allene to slow
down the dimer formation. This simple dilution resulted in a clean
but slower reaction. Finally, changing the catalyst to Pd(OAc)2 and
LiBr speeded up the reaction and afforded more than 90%
racemization after 1.5 h (Table 1, entry 9).
To determine the scope of the reaction, racemization of several
other optically active allenes was studied using the catalytic system
Pd(OAc)2/LiBr in acetonitrile. Our objective with this racemization
is to finally couple it with a kinetic resolution process to obtain a
DKR process. Since most of the published resolutions deal with
allenes bearing a functional group in the a-position12,18 our primary
goal was to racemize such substrates. Reaction of allenic ester 2
with 5 mol% of Pd(OAc)2 and 10 mol% of LiBr in acetonitrile at 50
°C resulted in a fast racemization with a half-time (t1/2) of less than
1 min and with > 90% racemization within 7 min (Table 2, entry 1).
The rate of racemization of the different allenes tested varied from
having t1/2 from 1 min to several hours. In general, allenes having
a phenyl substituent racemized rapidly (entries 1, 3 and 9). On the
Racemization is an important tool in asymmetric synthesis and has
attracted considerable attention in the past few years.1–3 Many large
scale syntheses include a racemization step because the most
common way in industry to obtain enantiomerically pure com-
pounds is still via resolution of racemic mixtures. A major
disadvantage of resolution is that only a maximum yield of 50% can
be obtained; the other 50% has to be discarded or recycled.
Recycling usually involves racemization of the unwanted enantio-
mer, often under rather harsh conditions.
Racemization also plays an important role in dynamic kinetic
resolution (DKR).3 The Pd(0)-catalyzed racemization of allylic
compounds via p-allyl intermediates has been investigated4 and
utilized in dynamic kinetic asymmetric transformations5 and
DKR.6 The Pd(II)-catalyzed racemization of allylic acetates via a
1,3-acetate shift7 has also been utilized in DKR.8
During our work on the stereoselective synthesis of 2-bromo-
1,3-dienes from a-acetoxy allenes we found that the starting allene
epimerized during the reaction (Scheme 1).9 Partial racemization of
allenes has previously been observed in reactions with organo-
copper or Grignard reagents (2–5 equiv.).10 We have now studied
the palladium(II)-catalyzed racemization of allenes as a first step
toward dynamic kinetic resolution of chiral allenes.
Optically active allenyl acid 1 (Chart 1) was obtained from the
corresponding racemic ethyl ester11 that was hydrolyzed and
resolved with cinchonidine according to a literature method.12
Methylation of 1 afforded 2. Allenyl alcohols 3a–c,9,13,14 amines
5a,b 15 and allene 7 16 were prepared from the appropriate
Table 1 Optimization of the racemization processa
Scheme 1 Epimerization during the palladium-catalyzed SN2A reaction.
b
c
Entry Substrate
Solvent
CH3CN
t1/2
t90%
1d
6a
23 h
n.d.
n.d.
2
3
4
5
6
6a
6a
6a
6a
6a
CH3CN
CH3CN/H2O 99/1
CH3CN/H2O 95/5
110 min
75 min
65 min
n.d.
e
e
e
CH3CN/H2O 80/20 12 min
CH3CN/H2O 50/50
7 min
e
7
3c
CH3CN
6 min
8f
9f,g
3c
3c
CH3CN
CH3CN
45 min
15 min
130 min
90 min
a Unless otherwise noted the reactions were carried out in 0.05 M solution
at 50 °C employing 5% PdBr2(PhCN)2. b t1/2 is the time taken to reach 50%
racemization. c Time for 90% racemization. d Reaction was carried out at 20
°C. e Reaction stopped after about 76% racemization. f 0.005 M solution
was used. g 5% Pd(OAc)2 and 10% LiBr was used.
Chart 1
† Electronic supplementary information (ESI) available: details of synthesis
and characterization of all new compounds. See http://www.rsc.org/
suppdata/cc/b3/b316482a/
964
C h e m . C o m m u n . , 2 0 0 4 , 9 6 4 – 9 6 5
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

Table 2 Racemization of optically active allenesa
Time for 90%
racemization
b
Entry
Substrate
t1/2
1
2
< 1 min
3.3 h
7 min
n.d.
3
15 min
40 min
21 min
30 mind
1.3 hd
2.25 h
6 min
25 min
1.5 h
2.3 h
1.7 h
n.d.
Scheme 3
4c
5
6
Scheme 4
7
n.d.
insertion of the second allene into the palladium–carbon bond of
intermediate 12.
8
n.d.
If an internal nucleophile is present in the molecule a vinylpalla-
dium species can be formed via an intramolecular nucleophilic
attack (Scheme 4). This vinyl complex (14) can also insert an allene
to give a dimer (15) that will tie up palladium and decrease the
catalytic activity. Formation of dimers via vinylpalladium com-
plexes such as 12 and 14 is common, and is involved in several
catalytic transformations.19 To avoid the dimerization according to
Scheme 4, nucleophiles in the a-position can be protected.
In conclusion, a simple, efficient and mild method for the
racemization of allenes has been developed. Various allenes were
rapidly racemized in the presence of 5 mol% of palladium acetate
and 10 mol% lithium bromide except when a nucleophile was
present a to the cumulated double bonds. The mechanism proposed
involves a bromopalladation–debromopalladation sequence. The
combination of the racemization with kinetic resolution for DKR is
currently being investigated in our laboratory.
9
30 min
10
75 min
a Unless otherwise noted the reactions were carried out in CH3CN at 50 °C
(0.005 M) employing 5% Pd(OAc)2 and 10% LiBr. b t1/2 is the time taken
to reach 50% racemization. c A single diastereomer epimerized during the
reaction to afford an approximately 1:1 mixture of the diastereomers.
d Acetonitrile:water 8:2 was used as solvent.
other hand, alcohol 3b and tosyl derivative 6c having a methyl
group in the terminal position of the allene racemized slowly and
the t1/2 values obtained were 3.3 and 2.25 h, respectively (entries 2
and 8). The decreased reaction rate is probably due to the proposed
dimer formation (in Scheme 4). Attempts to racemize amine 5a and
acid 1 under these reaction conditions resulted in complicated
mixtures of products.
Scheme 2 shows a likely mechanism for the racemization. After
anti-bromopalladation of one of the double bonds of (S)-3c the s-
allyl complex 9 formed rearranges to the other possible s-allyl
intermediate 11 via p-allyl complex 10. anti-Elimination of
bromide ion and palladium(II) from 11 gives the enantiomeric
allene (R)-3c.
Another, less likely, pathway would involve anti-bromopallada-
tion/syn-debromopalladation at the same double bond (or syn-
addition/anti-elimination). Syn–anti interconversion of the four,
diastereomeric p-allyl complexes does not play a role in the
racemization.
Reversible formation of a vinylpalladium intermediate 12 via
attack by bromide on the terminal allene carbon (Scheme 3, path B)
followed by irreversible insertion of a second allenic unit gives
dimeric complex 13. This pathway will tie up the catalyst in dimer
13 and decrease the rate of racemization. The deactivation can be
suppressed by dilution. Although dilution does not change the ratio
between the p-allyl complex 10 (path A) and vinyl complex 12
(path B) it inhibits the dimer formation by slowing down the
Notes and references
1 F. F. Huerta, A. B. E. Minidis and J. E. Bäckvall, Chem. Soc. Rev., 2001,
30, 321.
2 E. J. Ebbers, G. J. A. Ariaans, J. P. M. Houbiers, A. Bruggink and B.
Zwanenburg, Tetrahedron, 1997, 53, 9417.
3 O. Pàmies and J. E. Bäckvall, Chem. Rev., 2003, 103, 3247.
4 K. L. Granberg and J. E. Bäckvall, J. Am. Chem. Soc., 1992, 114,
6858.
5 B. M. Trost and D. Toste, J. Am. Chem. Soc., 1999, 121, 3543.
6 Y. K. Choi, J. H. Suh, D. Lee, I. T. Lim, J. Y. Jung and M.-J. Kim, J.
Org. Chem., 1999, 64, 8423.
7 L. E. Overman, Angew. Chem., Int. Ed. Engl., 1984, 23, 579.
8 J. V. Allen and J. M. J. Williams, Tetrahedron Lett., 1996, 37, 1859.
9 A. Horváth and J. E. Bäckvall, J. Org. Chem., 2001, 66, 8120.
10 A. Claesson and L. I. Olsson, J. Chem. Soc., Chem. Commun., 1979, 12,
524.
11 J. A. Marshall, M. A. Wolf and E. M. Wallace, J. Org. Chem., 1997, 62,
367.
12 G. Kresze, W. Runge and E. Ruch, Justus Liebigs Ann. Chem., 1972,
756, 112.
13 J. S. Cowie, P. D. Landor and S. R. Landor, J. Chem. Soc., Perkin
Trans., 1973, 7, 720.
14 R. K. Duke and R. W. Rickards, J. Org. Chem., 1984, 49, 1898.
15 A. Claesson and C. Sahlberg, Tetrahedron, 1982, 38, 363.
16 C. J. Elsevier and P. Vermeer, J. Org. Chem., 1989, 54, 3726.
17 K. Mori, T. Nukada and T. Ebata, Tetrahedron, 1981, 37, 1343.
18 M. Pietzsch, O. Vielhauer, D. Pamperin, B. Ohse and H. Hopf, J. Mol.
Catal. B: Enzymatic, 1999, 6, 51; A. Cipiciani and F. Bellezza, J. Mol.
Catal. B: Enzymatic, 2002, 17, 261; G. Gil, E. Ferre, A. Meou, J. Le
Petit and C. Triantaphylides, Tetrahedron Lett., 1987, 28, 1647; S.
Ramaswamy, R. A. H. F. Hui and J. B. Jones, J. Chem. Soc., Chem.
Commun., 1986, 20, 1545.
19 L. S. Hegedus, N. Kambe, Y. Ishii and A. Mori, J. Org. Chem., 1985, 50,
2240; S. Ma and Z. Yu, Angew. Chem., Int. Ed., 2002, 41, 1775; S. Ma
and Z. Yu, Org. Lett., 2003, 5, 1507.
Scheme 2 Proposed mechanism for the racemization of allenes.
C h e m . C o m m u n . , 2 0 0 4 , 9 6 4 – 9 6 5
965