Stereoselective alkene isomerization over one position

Article abstract of DOI:10.1021/ja3036477

Although controlling both the position of the double bond and E:Z selectivity in alkene isomerization is difficult, 1 is a very efficient catalyst for selective mono-isomerization of a variety of multifunctional alkenes to afford >99.5% E-products. Many reactions are complete within 10 min at room temperature. Even sensitive enols and enamides susceptible to further reaction can be generated. Catalyst loadings in the 0.01-0.1 mol% range can be employed. E-to-Z isomerization of the product from diallyl ether was only <10 ^-6 times as fast as its formation, showing the extremely high kinetic selectivity of 1.

Full text of DOI:10.1021/ja3036477

Communication
pubs.acs.org/JACS
Stereoselective Alkene Isomerization over One Position
Casey R. Larsen and Douglas B. Grotjahn*
Department of Chemistry and Biochemistry, San Diego State University, San Diego, California 92182-1030, United States
S
* Supporting Information
ABSTRACT: Although controlling both the position of
the double bond and E:Z selectivity in alkene isomerization
is difficult, 1 is a very efficient catalyst for selective mono-
isomerization of a variety of multifunctional alkenes to
afford >99.5% E-products. Many reactions are complete
within 10 min at room temperature. Even sensitive enols
and enamides susceptible to further reaction can be
generated. Catalyst loadings in the 0.01−0.1 mol% range
can be employed. E-to-Z isomerization of the product from
diallyl ether was only <10−6 times as fast as its formation,
showing the extremely high kinetic selectivity of 1.
Figure 1. Alkene isomerization induced by bifunctional catalyst 1.
selective alkene migrations over only one bond (mono-
isomerization), while still maintaining high stereocontrol for
E-isomeric products over a wide substrate scope. Our results
here show the success of this approach, including the synthesis
of sensitive enol and enamide derivatives, even in cases where
further tautomerization to a carbonyl or imine function is possible.
A second general goal was to lower catalyst loading from
≥2 mol%, as we previously reported. Indeed, in many cases we
find that loadings 20−200 times lower than 2 mol% can be used.
Finally, we quantitate the extremely high kinetic selectivity of
1 by determining in one case that the rate of forming internal
(E)-alkene from terminal isomer is over 106 times faster than
the rate of E-to-Z isomerization.
Table 1 shows results for substrates which can only undergo
a single isomerization, whereas Table 2 focuses on those com-
pounds which can in principle be isomerized further to another
alkene isomer. In all cases, the rapid generation of exclusively
(E)-alkenes was observed. Both diallyl ether (2) and sulfonamide
3 gave (E,E)-dipropenyl products (Table 1, entries 1 and 2).
Compound 3a was previously unknown, whereas 2a had been
made but only as a mixture with its geometrical isomers.18 The
sulfonamide case may be slower because the larger N-Ts unit
hinders the catalyst; in support of this notion, buildup of the
intermediate (allyl)(propenyl) isomer was seen before reaction
completion. Other allyl ethers (entries 3−5) are all rapidly
transformed into the corresponding (E)-propenyl ethers as
the sole product, even with as little as 0.05 mol% 1 (entry 4).
Allyl-substituted cyclohexanols 8 and 9 (entries 7 and 8) show
the facile production of exclusively E-isomer. Sensitive control
using substrate sterics is shown by much slower reaction of 10
(entry 9). Presumably, 1,3-diaxial interactions as 1 acts on the
allyl unit account for the observed differences between 10
and 9. In short, a wide variety of (E)-propenyl-substituted com-
pounds can be generated by catalyst 1 with very high selectivity.
Alkenyl aromatics such as trans-anethole and isoeugenol are
important in the pharmaceutical and fragrance industries and
are extracted from natural sources or generated from allyl aro-
matic isomers by the use of excess base and heat, thus creating
large amounts of caustic waste (using aqueous NaOH or KOH,
atalyzed isomerization of alkenes is an example of atom-
C
economical chemistry.1a,e Particularly useful alkene iso-
merizations include those of allyl alcohols into the correspond-
ing carbonyl compounds,1 allyl ethers into their enol ethers,2
and allylamines into their enamines.3 Thus, alkene isomer-
ization has attracted synthetic interest.1e However, the draw-
back of many known transition metal isomerization catalysts is
the lack of E:Z selectivity of olefinic product formation.
Though many metal catalysts (e.g., Ru,4−9 Ir,10 Fe,7 Ni,11,12
Pd,11,13 Pt,14 and Rh7,15) are capable of performing alkene
isomerization, challenges in the field include achieving high
positional and stereochemical selectivity, substrate generality,
and simplicity of catalyst use. For example, there are many
O-allyl2 and N-allyl3 susbtrates for which isomerization over
only one position is possible. Here the challenge is to obtain
only the E- or Z-isomer. The preparation of enamines from allyl
amines with high E-selectivity has been catalyzed by rhodium
complexes.15 In the oxygen case, the generation of predomi-
nately trans-enol ethers from allylic ethers occurs using an
activated iridium catalyst,10 whereas a Ru hydride complex
generates a mixture featuring the energetically more stable
cis-enol ether (Z > 55−68%).6 These systems vary in their cis:trans
selectivity, which is reported to be substrate dependent. Metathesis
catalysts occasionally isomerize olefins as a side reaction;16
subsequent modification and optimization, for example by Wipf
and Donohoe, has expanded this methodology to use in natural
product synthesis.5,8,11 However, many of these systems require
the use of additives, are tolerant of alkenes of only one
functionality, and most importantly have difficulty controlling
E:Z selectivity in a general way for all substrate classes.
We have previously disclosed complex 1 as an “alkene
zipper” for isomerization of alkenes with a wide variety of
functional groups under mild conditions (Figure 1).4,17 The
ability of 1 to isomerize an alkene up to 30 positions along an
alkyl chain highlights its activity. In contrast, here our goal
was to see if the activity of 1 could be controlled to perform
Received: April 16, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja3036477 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
Table 2. Mono-isomerization of Linear Alkene Substrates
Table 1. Mono-isomerization of Linear Allyl Substrates
a
b
c
Acetone-d6 solvent, RT, 2 mol% 1, NMR yield. Isolated. 5 mol% 1;
11% 17 remaining (ratios of products).
only 2% of conjugated isomer 15b20 is formed, where this
conversion may be catalyzed by 1 or by acid- or base-catalyzed
enolization. Linear pent-4-en-1-yl systems 16 and 17 build up
synthetically useful amounts of the trans-mono-isomerized
species shown. The substituents on 16 may hinder catalyst
approach to the H at C-2, as required for further isomerization;
however, a similar result in the case of 17 suggests that other
factors (e.g., substituent electronics or coordination) may play
a role, which would be a subject of further investigation. In all
cases shown, high selectivity for the (E)-2-alkene prevails.
The effects of branching presented by 1,1′-disubstituted
alkenes were explored (Table 3). Of mechanistic more than
Table 3. Mono-isomerization of 1,1′-Disubstituted Alkenes
to Trisubstituted Alkenes
a
b
c
Acetone-d6 solvent, RT, NMR yield. Isolated. 70 °C; produces
d
∼5% of (Z)-7a. No reaction (estimated limit of detection ∼2%).
e
Neat reaction; product (4.59 g) isolated by distillation.
56% conversion, E:Z = 82:18 in 12 h at 200 °C).12 Kannan
catalyzed the isomerization of eugenol to isoeugenol with
hydrotalcites at 200 °C (<77%, E:Z selectivity only 84:16).12b To
overcome such drawbacks, Sharma12a screened RuCl2(PPh3)3 and
RuCl3(AsPh3)2 in a variety of solvents over 3 h at 85 °C. The
conversion (>99%) and selectivity for each of the valuable alkenyl
aromatics were good (<95.6% trans); however, about 4% cis-isomer
remained even after extensively optimizing reaction conditions.12a
The most impressive result to date appears to be 99% trans-
anethole (80 °C, 15 min), in a variety of solvents, using 1 mol%
[RuCl2(η6-PhOCH2CH2OH)(P(OMe)3)].19 In contrast, using
much lower loadings of 1 without heating, both (E)-isoeugenol (11a) and
(E)-anethole (12a) were afforded in >99% yield in <13 min (0.1 mol%
1) and 40 min (0.05 mol% 1), respectively, where in each case no
trace of the Z-isomer is detected (detection limit ≈ 0.1%).
Significantly for green chemistry, 1 (0.1 mol%) acts on neat 11
within 10 min, and pure 11a can be distilled on a multigram scale.
Moreover, even days after completion of isomerization,
(Z)-alkenes are undetectable under our conditions.
Table 2 shows a variety of systems where a selectively mono-
isomerized product can be obtained in high yield, before further
alkene movement occurs. Hydrocarbon 14a resisted further
isomerization even after 48 h, forming only 9% of trisubstituted
isomer, which is consistent with catalyst approach to the
isopropyl C−H being hindered by branching. The ring of
cyclohexanone derivative 15 also presents branching, and 15a is
obtained exclusively as its E-isomer within 45 min. The
dramatic kinetic selectivity of 1 is shown by the fact that, after
77 h (100 times longer than the time required to obtain 15a),
a
b
Acetone-d6, RT, 2 mol% 1, NMR yield. Equilibrium ratios also
c
reached from reverse reaction. 70 °C.
practical interest, a polysulfone catalyst has shown to catalyze
olefin migration in similar substrates at room temperature (RT)
in high yields.21 RajanBabu11 recently optimized a multicomponent
alkene isomerization system made from 5 mol% of [(allyl)-
PdCl]2 or [(allyl)NiBr]2, phosphine ligand, AgOTf, and an addi-
tive. The ratio of E:Z components within the product mixture
was suggested to be based on the additive and also appeared to
B
dx.doi.org/10.1021/ja3036477 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
vary with substrate. In contrast, using single-component catalyst
1, only the E-isomer of the product is formed, though in some
cases in equilibrium with starting material. Branched alkene
ether 18 is completely transformed into its enol ether 18a, and
branched hydrocarbons 19 and 20 are fully converted to
trisubstituted alkenes 19a and 20a, respectively, all in less than
2 h. Alkenes illustrated in entries 4−7 reach an equilibrium
mixture of terminal and internal alkenyl isomers. Disubstituted
alkene 21, studied previously,4 affords an equilibrium mixture
(30:70) of the starting disubstituted alkene 21 and product
trisubstituted alkene 21a. In each case, ∼80% conversion to the
corresponding E-trisubstituted product is achieved. After
isomerization of 21, 22, and 23 to 21a, 22a, and 23a, products
were purified and isolated. When 21a, 22a, and 23a were
subjected to 1 for isomerization, each disubstituted alkene, 21,
22, and 23, was regenerated in the same ratio as in the forward
reaction. Trisubstituted alkenes 22a and 23a were subjected to
RajanBabu conditions in several separate attempts but did not
react, which leads to the conclusion the two catalyst systems
differ in kinetic selectivity.22 The RajanBabu catalyst system
polymerized 24, whereas entry 7 shows the mildness and
functional group tolerance of 1 in its ability to isomerize 24 to
(E)-24a. In summary, each disubstituted alkene is converted to
the more stable trisubstituted alkene, and where an E:Z mixture
is possible, only the E-analogue is formed.
completely tautomerized in 8 days.23a Amino acid allyl amides,
such as 27, were treated with Grubbs’ metathesis catalyst, and
isomerized products were significant components of the
mixture [20% (E)-27a and 30% (Z)-27a].24 In contrast, the
action of 1 on 27 within 5 min exclusively forms (E)-27a. The
success of 1 in making (E)-enol and enamide derivatives with
very high selectivity is especially notable, because the E and Z
geometrical isomers of such compounds are well-documented
to be of comparable thermodynamic stability, with the Z form
even predominating in some cases.23,25,26
Indeed, because E- and Z-isomers of many enols and
enamides are of comparable stability, compelling evidence for
unusually high kinetic selectivity using 1 comes from observing
mixtures from 2 and 1 long after complete formation of 2a.
Strikingly, whereas 2 mol% of 1 isomerized 2 within 6 min,
even when 20 mol% of 1 was applied, after 25 d, only 10% of
(Z)-2a was seen (Scheme 1). Complete analysis of these and
Scheme 1. Alkene Positional Isomerization Is Much Faster
Than Geometrical Isomerization
Especially notable results (Table 4) using 1 are formation of
sensitive enols and enamides. Typically enols are short-lived
Table 4. Formation of Enols and Primary Enamide
Derivatives
similar reactions over several time points (see Supporting
Information) allowed us to conclude that the ratio of rates for
isomerization of 2 to (E)-2a and of further isomerization of
(E)-2a to (Z)-2a must be over 3,600,000:1, a remarkable result.
Scheme 2 shows further examples of chemo- and stereo-
selective isomerization of polyfunctional biologically active
Scheme 2. Alkene Isomerization of Biologically Active
Molecules at Low Catalytic Loading of 1
a
b
c
Acetone-d6, RT, 3 mol% 1. NMR yield. 0.1 mol% 1. Also in
d
e
mixture: 3% propanal and unreacted 25. 0.6 mol% 1. Isolated.
due to favorable tautomerization to the keto forms, yet enols23
and enamides24 are useful synthetic intermediates and subunits
in biologically active natural products, respectively. Bosnich23a
generated simple enols via double bond migration using
[Rh(diphos)(solvent)2]ClO4 activated by H2, where it is
thought that isomerization proceeds by alkene insertion into
a Rh−H bond. The cis:trans ratios of the enols and the rate of
subsequent tautomerization depended strongly on substrate.
In the Bosnich work, allyl alcohol 25 gave 89% of enol 25a as
an isomeric mixture (E:Z = 1:1), along with 11% aldehyde, in
14 min at RT, whereas 2-phenylallyl alcohol led to 47% of the
corresponding enol in all Z-configuration after 2 h. Table 4
illustrates the unique capability of 1 in giving (E)-enols and
(E)-enamides. Remarkably, after 5 min (E)-enol 25a is seen
(93%) along with 3% of the Z-isomer, where both subsequently
tautomerize to propanal. 2-Methyl-2-propenol (26) is
quantitatively converted to the corresponding enol 26a after
60 min with 0.1 mol% 1, whereas according to Bosnich the
keto analogue (4%) of 26a begins to appear at 16 min and is
compounds by 1. Carbohydrate 28 is a key starting material in
the generation of an irreversible inhibitor of yeast hexokinase,27
and both allyl and propenyl functional groups such as those in
28 and 28a are used as protecting groups.28,29 Compounds like
28a, but with varying E:Z ratios, are made from isomers like 28
under strongly basic conditions (t-BuOK in DMSO28).
Glucopyranoside 28 is converted by 1 with very low catalytic
loading to (E)-propenyl isomer 28a, isolated in 96% yield.
Bioallethrin (29) is a flexible pyrethroid pesticide30 with three
olefinic bonds. Pyrethroids demonstrate insecticidal actions
based on their ability to adopt a certain conformation.31 Novel
pyrethroid 29a is formed as a single alkene isomer, without
perturbing the allylic ester functionality or the other two alkene
units. Catalyst 1 enables selective structural alterations with the
potential of leading to novel bioactive compounds.
C
dx.doi.org/10.1021/ja3036477 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(10) Baudry, D.; Ephritikhine, M.; Felkin, H. J. Chem. Soc., Chem.
Commun. 1978, 16, 694. (b) Yamamoto, Y.; Miyairi, T.; Ohmura, T.;
Miyaura, N. J. Org. Chem. 1999, 64, 296. Ohmura, T.; Shirai, Y.;
Yamamoto, Y.; Miyaura, N. Chem. Commun. 1998, 1337. Ohmura, T.;
Yamamoto, Y.; Miyaura, N. Organometallics 1999, 18, 413.
(11) Lim, H. J.; Smith, C. R.; RajanBabu, T. V. J. Org. Chem. 2009,
74, 4565.
(12) (a) Sharma, S. K.; Srivastava, V. K.; Jasra, R. V. J. Mol. Catal. A:
Chemical 2006, 245, 200. (b) Jinesh, C. M.; Antonyrai, C. A.; Kannan,
S. Catal. Today 2009, 141, 176.
(13) Mirza-Aghayan, M.; Boukherroub, R.; Bolourtchian, M.;
Hoseini, M.; Tabar-Hydar, K. J. Organomet. Chem. 2003, 678, 1.
Gautheir, D.; Lindhardt, A. T.; Olsen, E. P. K.; Overgaard, J.;
Skrydstrup, T. J. Am. Chem. Soc. 2010, 132, 7998.
(14) Permin, A. B.; Petrosyan, V. S. Appl. Organomet. Chem. 1990, 4,
329. Scarso, A.; Colladon, M.; Sgarbossa, P.; Michelin, R. A.; Strukul,
G. Organometallics 2010, 29, 1487.
In summary, we show that very active catalyst 1 can be controlled
to provide mono-isomerized alkene products with very high (>106:1)
kinetic E-selectivity. These features are expected to make 1 an
attractive tool in synthesis,32 in part because the addition of an allylic
substituent (e.g., to a carbonyl) is frequently easier than the installa-
tion of a propenyl analogue. Using 1, allylic moieties can then be
readily and cleanly converted to propenyl units in a controlled
manner. Moreover, 1 allows facile access to the (E)-alkenyl aromatic
compounds useful in the pharmaceutical and fragrance industries.
Catalyst 1 achieves high E-selectivity in cases well-documented to be
of comparable thermodynamic stability as their Z-isomers, such as
enols, enol ethers, and enamides. In addition, the very mild, neutral
reaction conditions under which 1 operates allow formation of enols,
enamides, β,γ-unsaturated carbonyl compounds, and even multifunc-
tional bioactive molecules. We believe that the heterocyclic phosphine
ligand of 1 acts as a hemilabile base acting in cooperation with the
metal to promote these selective transformations. Mechanistic studies
are in progress to allow a greater understanding of the bifunctional
system and further alter and enhance its unique reactivity, in addition
to the possible applications and new chemistry catalyst 1 can provide.
(15) Tani, K.; Yamagata, T.; Akutagawa, S.; Kumobayashi, H.;
Taketomi, T.; Takaya, H.; Miyashita, A.; Noyori, R.; Otsuka, S. J. Am.
Chem. Soc. 1984, 106, 5208.
(16) Examples: Joe, D.; Overman, L. E. Tetrahedron Lett. 1997, 38,
8635. Furstner, A.; Theil, O. R.; Ackermann, L.; Schanz, H. J.; Nolan,
̈
N. P. J. Org. Chem. 2000, 65, 2204. Gurjar, M. K.; Yakambram, P.
Tetrahedron Lett. 2001, 42, 3633. Alcaide, B.; Almendros, P.; Alonso,
J. M.; Aly, M. F. Org. Lett. 2001, 3, 3781. Alcaide, B.; Almendros, P.;
Luna, A. Chem. Rev. 2009, 109, 3817.
(17) Erdogan, G.; Grotjahn, D. B. J. Am. Chem. Soc. 2009, 131, 10354.
(18) Taskinen, E.; Virtanen, R. J. Org. Chem. 1977, 42, 1443.
(19) Lastra-Barreira, B.; Crochet, P. Green Chem 2010, 12, 1311.
(20) Compound 15a can undergo further isomerization to β,γ-
unsaturated 15b (2%, 3 d, RT), increasing slowly with time (6%, 6.3 d).
ASSOCIATED CONTENT
■
S
* Supporting Information
Details of substrate preparation, characterization, and catalysis.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
grotjahn@sciences.sdsu.edu
Notes
The authors declare no competing financial interest.
(21) Markovic, D.; Vogel, P. Angew.Chem. Int. Ed. 2004, 43, 2928.
(22) Larsen, C. R.; Grotjahn, D. B. Top. Catal. 2010, 53, 1015.
(23) (a) Bergens, S. H.; Bosnich, B. J. Am. Chem. Soc. 1991, 113, 958.
Persistence of 26a: (b) Park, J.; Chin, C. S. J. Chem. Soc., Chem.
Commun. 1987, 1213. (c) Chin, C. S.; Lee, S. Y.; Park, J.; Kim, S.
J. Am. Chem. Soc. 1988, 110, 8244.
ACKNOWLEDGMENTS
■
NSF supported this work and upgrade of the departmental
NMR facility. Dr. LeRoy Lafferty assisted with NMR analysis,
and Prof. Andrew Cooksy assisted with kinetics analysis.
(24) McNaughton, B. R.; Bucholz, K. M.; Camaano-Moure, A.;
̃
Miller, B. L. Org. Lett. 2005, 7, 733.
REFERENCES
(25) Epiotis, N. D.; Bjorkquist, D.; Bjorkquist, L.; Sarkanen, S. J. Am.
Chem. Soc. 1973, 95, 7558. Bingham, R. C. J. Am. Chem. Soc. 1976, 98,
535. Knorr, R. Chem. Ber. 1980, 113, 2441.
(26) Complementary Z-selectivity but much lower catalytic activity
was recently reported for isomerization of 6 by a Pt(II) catalyst (only
45% yield in 24 h at 50 °C, but 100% selectivity).14
(27) Bessell, E. M.; Westwood, J. H. Carbohydr. Res. 1972, 25, 11.
Yuasa, Y.; Yuasa, Y. Org. Proc. Rec. Devel. 2004, 8, 405.
(28) Cunningham, J.; Gigg, R.; Warren, C. D. Tetrahedron Lett. 1964,
1191.
(29) Mereyala, H. B.; Guntha, S. Tetrahedron Lett. 1993, 34, 6929.
(30) Nkazawa, H.; Horiba, M.; Yamamoto, S. Agric. Biol. Chem.
1980, 44, 1173. He, F. Toxicology 1994, 91, 43.
(31) Pulman, D. A. J. Agric. Food Chem. 2011, 59, 2770.
(32) Catalyst 1 is now available from Strem Chemicals. The work
described here was finished before that event.
■
(1) (a) Trost, B. M. Acc. Chem. Res. 2002, 35, 695. (b) Trost, B. M.;
Kulawiec, R. J. J. Am. Chem. Soc. 1993, 115, 2027. (c) Slugovc, C.; Ruba,
E.; Schmid, R.; Kirchner, K. Organometallics 1999, 18, 4230. (d) Bouziane,
A.; Carboni, B.; Bruneau, C.; Carreaux, F.; Renaud, J. Tetrahedron 2008,
́ ́
64, 11745. (e) Uma, R.; Crevisy, C.; Gree, R. Chem. Rev. 2003, 103, 27.
(2) Kuznik, N.; Krompiec, S. Coord. Chem. Rev. 2007, 251, 222.
(3) Krompiec, S.; Krompiec, M.; Penczek, R.; Ignasiak, H. Coord.
Chem. Rev. 2008, 252, 1819.
(4) Grotjahn, D. B.; Larsen, C. R.; Gustafson, J. L.; Nair, R.; Sharma,
A. J. Am. Chem. Soc. 2007, 129, 9592. Grotjahn, D. B.; Larsen, C. R.;
Erdogan, G.; Gustafson, J. L.; Sharma, A.; Nair, R. Catal. Org. React.
2009, 123, 379.
(5) Wipf, P.; Spencer, S. R. J. Am. Chem. Soc. 2005, 127, 225. Arisawa,
M.; Terada, Y.; Takahashi, K.; Nakagawa, M.; Nishida, A. J. Org. Chem.
2006, 71, 4255. Donohoe, T. J.; O’Riordan, T. J. C.; Rosa, C. P. Angew.
Chem., Int. Ed. 2009, 48, 1014.
(6) Suzuki, H.; Koyama, Y.; Moro-Oka, Y.; Ikawa, T. Tetrahedron
Lett. 1979, 20, 1415.
(7) (a) Stille, J. K.; Becker, Y. J. Org. Chem. 1980, 45, 2139.
(b) Jennerjahn, R.; Jackstell, R.; Piras, I.; Franke, R.; Jiao, H.; Bauer,
M.; Beller, M. ChemSusChem 2012, 5, 734.
(8) Hanessian, S.; Giroux, S.; Larrson, A. Org. Lett. 2006, 8, 5481.
(9) Sivaramakrishna, A.; Mushonga, P.; Rogers, I. L.; Zheng, F.;
Haines, R. J.; Nordlander, E.; Moss, J. R. Polyhedron 2008, 27, 1911.
D
dx.doi.org/10.1021/ja3036477 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX