Regio- and enantioselective iridium-catalyzed allylic alkylation with in situ activated P,C-chelate complexes

Article abstract of DOI:10.1002/anie.200460016

Come together: Successful cooperation of iridium and copper yielded a new catalyst system for allylic alkylation. Branched alkylation products can be obtained rapidly and with high enantioselectivity from simple linear precursors (see scheme; cod = cycloo

Full text of DOI:10.1002/anie.200460016

Angewandte
Chemie
Allylic Substitution
Regio- and Enantioselective Iridium-Catalyzed
Allylic Alkylation with In Situ Activated P,C-
Chelate Complexes**
Gunter Lipowsky, Nicole Miller, and Günter Helmchen*
Dedicated to Professor Armin de Meijere
on the occasion of his 65th birthday
nation with ligand L1 according to Scheme 2.[7c] For example,
in the case of 4c the optimum value was found to be 5c/6c =
95:5. However, the enantioselectivity of 86% ee was still
The asymmetric allylic substitution plays an important role in
organic synthesis. However, this reaction is still largely limited
to symmetrically substituted substrates and palladium cata-
lysts.[1] With the exception of a few examples,[2] the partic-
ularly easily accessible monosubstituted allylic substrates 1
(Scheme 1) are not suitable substrates, because in intermo-
Scheme 2. Ir-catalyzed allylic substitution with monosubstituted allylic
carbonates. M=Na,Li.
Scheme 1. Allylic substitution (X: leaving group).
below the critical value of 90% ee. Moreover, reaction times
of typically several days at room temperature were unsat-
isfactory; a typical example is the reaction of substrate 4a
(Scheme 2) according to entry 1, Table 1. A slight, but not
sufficient improvement resulted with LiCl as additive
(Table 1, entry 2).
We have now developed a new catalyst system that led to
very substantial improvement with respect to selectivity and
activity for a broad range of substrates. The starting point of
the new development was the observation that complexes
lecular reactions using Pd catalysts they yield achiral, linear
substitution products 3. Remarkably, complexes of most of
the other transition metals, e.g. Ir, Mo, and Ru, preferentially
give rise to the branched, chiral products.[3–5] While Mo and
Ru complexes furnished good results only in special cases, Ir
catalysts are applicable to reactions with C[6] as well as N[7]
and O[8] nucleophiles.[9]
In the case of Ir catalysis, results for alkylations have not
been satisfactory so far. Thus, using standard reaction
[Ir(cod)LCl],
formed
spontaneously
upon
mixing
conditions[6b,c]
(phosphorus
amidites
as
ligands,
[{Ir(cod)Cl}2] with a phosphane L, do not react with allylic
compounds 1 (X = Cl, OAc). An active catalyst is generated
[{Ir(cod)Cl}2]/ligand = 1:2, Nu= NaCH(COOCH3)2, THF,
658C; cod = cyclooctadiene), we achieved following results:
a) with phosphorus amidite[10] L2 as the ligand regioselectiv-
ities (2/3) of > 90:10 were obtained for a broad range of
acetates (1, X = OAc); however, the ee values of the products
were low (4–86%).[11] b) In contrast, with phosphorus amidite
L1 ee values ranged from 56 to 94%, while regioselectivities
covered the range from 30:70 up to 91:9.[6c] An improvement
of regioselectivities resulted by using carbonates 4 in combi-
À
by C H activation only after addition of the typically strongly
basic nucleophile (e.g. malonate or amine). Thus, in the case
of the achiral ligand P(OPh)3 we identified the complex K1 as
[*] G. Lipowsky,N. Miller,Prof. Dr. G. Helmchen
Organisch-Chemisches Institut
Universität Heidelberg
Im Neuenheimer Feld 270,69120 Heidelberg (Germany)
Fax: (+49)6221-544-205
E-mail: g.helmchen@urz.uni-heidelberg.de
À
the active species, which is formed by ortho C H activation
[**] This work was supported by the Deutsche Forschungsgemeinschaft
(SFB 623) and by the Fonds der Chemischen Industrie. We thank
Markus Krauter for competent experimental assistance,Dr. F.
Rominger for the X-ray crystal structure,and Degussa AG for
iridium salts.
and elimination of HCl.[6b] Similarly, Hartwig et al. found that
complex K2, besides other compounds, is formed from L1
with amines as bases.[7b] The amine, e.g. pyrrolidine, acts as
both nucleophile and base.
Angew. Chem. Int. Ed. 2004, 43, 4595 –4597
DOI: 10.1002/anie.200460016
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4595

Communications
Table 1: Ir-catalyzed allylic alkylation of carbonate 4a according to Scheme 2.[a]
dominant in the 31P NMR spectrum (see
above). Addition of only THT did not affect
the activity of the catalyst (cf. entries 7 and
3). Similarly, addition of CuI did not show
an effect without additional THT (cf.
Entry L1
[mol%] ([mol%])
Additive A[b] TBD
Additive B[c] t[d] [h] Yield
Ratio
ee(5a) [%][g]
[mol%] ([mol%])
5a+6a [%][e] 5a/6a[f] (Abs. config.)
1
4
4
4
8
4
4
4
4
4
4
4
0.4
–
–
–
8
8
8
8
8
8
8
–
–
–
–
120
22
4
12
140
2
77
87
89
80
60
87
83
89:11
94:6
97:3
97:3
92:8
97:3
98:2
98:2
98:2
98:2
99:1
98:2
88 (+)(S)
90 (+)(S)
95 (+)(S)
92 (+)(S)
75 (+)(S)
95 (+)(S)
95 (+)(S)
95 (+)(S)
96 (+)(S)
96 (+)(S)
96 (+)(S)
94 (+)(S)
2[h]
3
LiCl (100)
–
–
entries 8,
7 and 3). Increased activity
resulted only upon addition of both addi-
tives (entry 9). In this case, the signals of K2
had completely disappeared in the 31P NMR
spectrum, and in addition to the signal at
98 ppm there was a broad signal at 130 ppm.
The activity of the catalyst could be notice-
ably increased by using an excess of THT as
well as CuI (entries 10, 11; cf. General
Procedure). The high activity of the new
catalyst allowed Ir-catalyzed allylic alkyla-
tions for the first time to be carried out with
less than 1 mol% of the iridium catalyst
with very good results (entry 12).
4
5
6
7
8
9
10
11
12[i]
PPh3 (4)
PPh3 (4)
THT (4)
–
THT (4)
THT (20)
THT (80)
THT (5)
–
CuI (4)
–
6
CuI (24)
CuI (4)
CuI (20)
CuI (100)
CuI (5)
5.5 74
1.5 83
1.5 92
<1
24
12
8
1.2
84
77
[a] Reaction conditions: 1 mmol 4a, L1/[{Ir(cod)Cl}2]=2:1,2 equiv NaCH(COOMe) ,0.25 m in THF,
2
RT. [b] Addition before base and substrate. [c] Addition after substrate. [d] Reaction time. [e] Yield of
1
isolated product. [f] Determined from the H NMR spectrum of the crude product. [g] Determined by
HPLC on a chiral column (Daicel Chiralcel OJ-H,250”4.6 mm,5 mm with guard cartridge OJ-H,10”
4 mm,5 mm,0.5 mLmin À1): 5a (n-hexane/isopropanol 90:10,20 8C,254 nm): tR[(+)(S)-5a]=26 min,
tR[(À)(R)-5a]=29 min. [h] Reaction temperature: 508C. [i] 0.2 mol% [{Ir(cod)Cl}2],0.4 mol% L1.
The optimal reaction conditions (cf.
General Procedure) were applied to further
These results suggested that we generate the catalyst
before the allylic substitution using a base other than the
nucleophile. In our studies on the intramolecular allylic
amination,[7d] this concept indeed led to a very active catalyst.
This catalyst was prepared by treatment of [Ir(cod)(L1)Cl]
with 1,5,7-triazabicyclo[4.4.0]undec-5-ene (TBD). In this
reaction a species of unknown structure displaying a singlet
at 98 ppm (31P NMR) was formed initially. After approx-
imately 2h the concentration of this compound reached a
maximum, and after prolonged reaction time (5–10 h) it was
transformed mainly into K2 (31P NMR signals at 128 and
153 ppm). On the basis of these results it appeared possible to
avoid the formation of K2, which displays low catalytic
activity (Table 1, entry 1) because one coordination site is
blocked by L1. Indeed, use of the system prepared over a
period of only 2h, largely devoid of K2, resulted in very rapid
and highly selective alkylation (Table 1, entry 3). For a
crosscheck, we increased the amount of L1 (L1:Ir= 2:1); as
anticipated, this led to a slow reaction (entry 4).
substrates, including the problematic alkyl- and alkoxyalkyl-
substituted compounds, in order to investigate the substrate
scope of the new catalyst (Table 2). For all examples both very
high activity and enantioselectivity were found.
The absolute configuration of dienylic esters such as 5a
and 5b was not determined previously. We have now
prepared the derivative 7 from 5a (Scheme 3); its absolute
configuration could be determined by X-ray crystal structure
analysis on account of the two heavy atoms. (Note that the
change of the descriptors in Table 2is only a consequence of
the different CIP priorities of the substituents R in 5.)
In conclusion, we have developed a highly active in situ
catalyst for the iridium-catalyzed enantioselective allylic
Table 2: Ir-catalyzed allylic alkylation of the carbonates 4a–4 f according
to Scheme 2 under optimal reaction conditions.[a]
Entry
Substrate
t[b] [h]
Yield
Ratio
ee(5) [%][e]
(Abs. config.)
(5+6) [%][c]
5:6[d]
To explore another possibility for avoiding formation of
K2, we added a soft, cheap auxiliary ligand L (Table 1,
additive A) prior to the base, and catalysis was induced with a
soft metal ion (Table 1, additive B).[12] PPh3 was the first
auxiliary ligand tested. Treatment of a 1:1:1 mixture of
[{Ir(cod)Cl}2], L1, and PPh3 with TBD gave complex K3
(31P NMR signals at 6 and 152ppm), [7b] which displayed low
catalytic activity (entry 5) as expected, because coordination
of PPh3 to IrI is strong. However, if after addition of 4a an
amount of CuI was added equivalent to the amount of Ir, a
very active catalyst was obtained (entry 6). In other words, the
selective removal of the auxiliary ligand PPh3 with the cheap
CuI as a scavenger is possible.[13]
The continued search for even more suitable auxiliary
ligands led to sulfides, in particular tetrahydrothiophene
(THT), as the best solution so far. Optimal results were
obtained by treatment of a 1:1:5 mixture of [{Ir(cod)Cl}2], L1,
and THT with TBD for only 2h, thus, avoiding the formation
of K2. At this time the species with a singlet at 98 ppm was
1
2
3
4
5
6
4a
4b
4c
4d
4e
4 f
1.5
1
1
<1
2
2
92
79
88
95
92
88
98:2
98:2
99:1
99:1
81:19
88:12
96 (+)(S)
96 (+)
96 (+)(R)
97 (+)(R)
96 (+)(R)
97 (+)
[a] M=Na,Table 1,entry 10 and General Procedure. [b] Reaction time.
[c] Yield of isolated product. [d] Determined from the 1H NMR spectrum
of the crude product. [e] Determined by HPLC on a chiral column; 5b by
determination as 4-bromobenzoate (Daicel Chiralcel OJ-H,250”
4.6 mm,5 mm with guard cartridge OJ-H,10”4 mm,5
mm,
0.5 mLminÀ1, n-hexane/isopropanol 99:1,28 8C,254 nm): tR[(+)-5b]=
96 min, tR[(À)-5b]=115 min; 5c (same HPLC setup as for 5b, n-hexane/
isopropanol 97:3,30 8C,220 nm): tR[(À)(S)-5c]=56 min, tR[(+)(R)-5c]=
62 min; 5d (Daicel Chiralcel OD-H,250”4.6 mm,5 mm with guard
cartridge OD-H,10”4 mm,5 mm,0.5 mLmin À1, n-hexane/isopropanol
90:10,20 8C,220 nm): tR[(+)(R)-5d]=12 min, tR[(À)(S)-5d]=14 min;
5e (same HPLC setup as for 5b, n-hexane/isopropanol 98:2,25 8C,
220 nm): tR[(+)(R)-5e]=42 min, tR[(À)(S)-5e]=54 min; 5 f (same
HPLC setup as for 5d, n-hexane/isopropanol 99:1,25 8C,220 nm):
tR[(+)-5 f]=12.3 min, tR[(À)-5 f]=14.9 min.
4596
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 4595 –4597

Angewandte
Chemie
J. Inorg. Chem. 2002, 2569 – 2586; c) B. Bartels, C. Garcia-Yebra,
G. Helmchen, Eur. J. Org. Chem. 2003, 1097 – 1103.
[7] a) T. Ohmura, J. F. Hartwig, J. Am. Chem. Soc. 2002, 124,
15164 – 15165; b) C. A. Kiener, C. Shu, C. Incarvito, J. F.
Hartwig, J. Am. Chem. Soc. 2003, 125, 14272 – 14273; c) G.
Lipowsky, G. Helmchen, Chem. Commun. 2004, 116 – 117; d) C.
Welter, O. Koch, G. Lipowsky, G. Helmchen, Chem. Commun.
2004, 895 – 896.
[8] a) F. Lopez, T. Ohmura, J. F. Hartwig, J. Am. Chem. Soc. 2003,
125, 3426 – 3427; b) Kinetic resolution: C. Fischer, C. Defieber,
T. Suzuki, E. M. Carreira, J. Am. Chem. Soc. 2004, 126, 1628 –
1629.
Scheme 3. Synthesis of the acetal (À)(S)-7; its absolute configuration
was determined by X-ray crystal structure analysis. a) Diisobutylalumi-
num hydride,THF, À788C,1 h,67%; b) 44, ’-dibromobenzophenone,
pTsOH (0.3 equiv),toluene,reflux,20 h,50%.
[9] Additionally it has to be noted that the price of iridium is far
below that of palladium. The prices per ounce (world market
prices, October 2003) are: rhodium $500, palladium $199,
iridium $90.
[10] B. L. Feringa, Acc. Chem. Res. 2000, 33, 346 – 353.
[11] B. Bartels, unpublished results.
[12] The reason for this demand is the selective coordination of
phosphorus in competition with the typically hard nucleophile.
[13] Realization of the obvious possibility of generating the catalyst
in the presence of the allylic substrate or another olefin in order
to saturate the free coordination site was not successful.
[14] Note added in proof: Use of a recently reported dimethoxy
derivative of ligand L1 (compound 3b in K. Tissot-Croset, D.
Polet, A. Alexakis, Angew. Chem. 2004, 43, 2426 – 2428) gave the
following improved results for the reaction of 4a according to
Table 2, entry 1: reaction time 45 min, yield 80%, regioselectiv-
ity 99:1, 98.5% ee; the regioselectivity of the reaction according
to entry 5 was improved to 5e/6e = 88:12.
alkylation, which by interplay of Ir and Cu allows reaction
times of less than an hour at room temperature and ee values
of 94–97% to be achieved for a broad spectrum of substrates.
The ratio of catalyst to substrate could be lowered down to
0.004.[14]
General procedure (corresponding to Table 1, entry 10): Under
an atmosphere of argon, a solution of [{Ir(cod)Cl}2] (13.4 mg,
0.02mmol) and L1 (21.6 mg, 0.04 mmol) in anhydrous THF
(0.5 mL) was treated with tetrahydrothiophene (18 mL, 0.20 mmol)
(Table 1, additive A) and TBD (17 mg, 0.12mmol), and the mixture
was stirred for 2h. Then substrate 4 (1.0 mmol) and subsequently CuI
(38 mg, 0.20 mmol) (additive B) as well as a solution of sodium
dimethylmalonate (2.0 mmol) in anhydrous THF (4 mL) were added.
The mixture was stirred for the time stated in Tables 1 and 2, and
conversion was monitored by thin-layer chromatography. When the
conversion was complete, Et2O (5 mL) and saturated NH4Cl solution
(5 mL) were added, and the aqueous phase was extracted with Et2O
(2” 20 mL). The organic phases were combined, washed with brine
(20 mL), dried over Na2SO4, filtered and concentrated in vacuo. The
ratio 5/6 of the crude product was determined by 1H NMR spectros-
copy. Then the crude product was purified by flash column
chromatography (silica, petroleum ether/ethyl acetate 15:1).
Received: March 16, 2004
Keywords: alkylation · allylic substitution · asymmetric catalysis ·
.
iridium · phosphorus amidites
[1] a) B. M. Trost, C. Lee in Catalytic Asymmetric Synthesis, 2nd Ed.
(Ed.: I. Ojima), Wiley, New York, 2000, p. 593 – 649; b) A. Pfaltz,
M. Lautens in Comprehensive Asymmetric Catalysis I–III (Eds.:
E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999,
p. 833 – 884; c) B. M. Trost, M. L. Crawley, Chem. Rev. 2003, 103,
2921 – 2943.
[2] a) S.-L. You, X.-Z. Zhu, Y.-M. Luo, X.-L. Hou, L.-X. Dai, J. Am.
Chem. Soc. 2001, 123, 7471 – 7472, and references therein;
b) Vinyl epoxides: B. M. Trost, C. Jiang, Org. Lett. 2003, 5,
1563 – 1565, and references therein.
[3] a) Ir: R. Takeuchi, Synlett 2002, 1954 – 1965; b) First enantiose-
lective reaction: J. P. Janssen, G. Helmchen, Tetrahedron Lett.
1997, 38, 8025 – 8026.
[4] Mo: a) B. M. Trost, I. Hachiya, J. Am. Chem. Soc. 1998, 120,
1104 – 1105 (arylallyl derivatives); b) B. M. Trost, S. Hildbrand,
K. Dogra, J. Am. Chem. Soc. 1999, 121, 10416 – 10417 (alkeny-
lallyl derivatives); c) F. Glorius, A. Pfaltz, Org. Lett. 1999, 1,
141 – 144 (alkylallyl derivatives).
[5] Ru: B. M. Trost, P. L. Fraisse, Z. T. Ball, Angew. Chem. 2002,
114, 1101 – 1103; Angew. Chem. Int. Ed. 2002, 41, 1059 – 1061.
[6] a) B. Bartels, G. Helmchen, Chem. Commun. 1999, 741 – 742;
b) B. Bartels, C. Garcia-Yebra, F. Rominger, G. Helmchen, Eur.
Angew. Chem. Int. Ed. 2004, 43, 4595 –4597
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4597