Asymmetric hydroalkoxylation of non-activated alkenes: Titanium-catalyzed cycloisomerization of allylphenols at high temperatures

Article abstract of DOI:10.1002/anie.201409252

The asymmetric catalytic addition of alcohols (phenols) to non-activated alkenes has been realized through the cycloisomerization of 2-allylphenols to 2-methyl-2,3-dihydrobenzofurans (2-methylcoumarans). The reaction was catalyzed by a chiral titanium-carboxylate complex at uncommonly high temperatures for asymmetric catalytic reactions. The catalyst was generated by mixing titanium isopropoxide, the chiral ligand (aS)-1-(2-methoxy-1-naphthyl)-2-naphthoic acid or its derivatives, and a co-catalytic amount of water in a ratio of 1:1:1 (5 mol % each). This homogeneous thermal catalysis (HOT-CAT) gave various (S)-2-methylcoumarans with yields of up to 90 and in up to 85 e at 240°C, and in 87 ee at 220 °C.

Full text of DOI:10.1002/anie.201409252

Angewandte
Chemie
DOI: 10.1002/anie.201409252
Asymmetric Catalysis
Asymmetric Hydroalkoxylation of Non-Activated Alkenes: Titanium-
Catalyzed Cycloisomerization of Allylphenols at High Temperatures**
Johannes Schlꢀter, Max Blazejak, Florian Boeck, and Lukas Hintermann*
Dedicated to Professor Keisuke Suzuki on the occasion of his 60th birthday
Abstract: The asymmetric catalytic addition of alcohols
(phenols) to non-activated alkenes has been realized through
the cycloisomerization of 2-allylphenols to 2-methyl-2,3-dihy-
drobenzofurans (2-methylcoumarans). The reaction was cata-
lyzed by a chiral titanium–carboxylate complex at uncom-
monly high temperatures for asymmetric catalytic reactions.
The catalyst was generated by mixing titanium isopropoxide,
the chiral ligand (aS)-1-(2-methoxy-1-naphthyl)-2-naphthoic
acid or its derivatives, and a co-catalytic amount of water in
a ratio of 1:1:1 (5 mol% each). This homogeneous thermal
catalysis (HOT-CAT) gave various (S)-2-methylcoumarans
with yields of up to 90% and in up to 85% ee at 2408C, and in
87% ee at 2208C.
Table 1). We have shown that aluminum isopropoxide is
a catalyst that gives high yields of 2 in short reaction times at
2508C (microwave heating; 10 bar).[12] The basic alkoxide
ligands in this system render competing hidden catalysis by
Table 1: Screening for asymmetric hydroalkoxylation with titanium
catalysts.[a]
Entry
Ligand
Solvent
T [8C]
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
7
8
9
-
PhCl
PhCl
PhMe
PhCl
PhMe
PhMe
PhMe
PhMe
PhMe
250
250
250
250
250
180
200
240
280
–
–
10
16
62
–
4
32
67
–
–
0
13
64
–
68
70
48
L2
L2
L1
L1
L1
L1
L1
L1
H
ydroalkoxylation is defined as the addition of alcohols (or
phenols) to alkenes in order to produce ethers.[1–3] For non-
activated alkenes, this synthetically important hydrofunction-
alization[2] is catalyzed by strong mineral acids, with limited
prospect for stereoselective catalysis. Contrary to the related
hydroamination of alkenes, for which several metal-catalyzed
asymmetric versions have been reported,[4] asymmetric
hydroalkoxylations of non-activated alkenes, as opposed to
activated allenes,[5] are hardly known. Hydroalkoxylations
that are catalyzed by strong Lewis acids and lead to
Markovnikov selectivity have been described,[1b,6] but they
suffer from competing hidden catalysis by Brønsted acids,[7]
and asymmetric versions are still unknown.[1b] Hartwig has
reported singular examples of IrI-catalyzed hydroalkoxyla-
tions with moderate induction.[8] Alternative approaches to
catalytic, stereo- or regioselective hydroalkoxylations rely on
photocatalysis[9] or oxidation–reduction processes.[10,11] We
now report the redox-neutral, asymmetric hydroalkoxylation
of a non-activated alkene that is catalyzed by a chiral
titanium–carboxylate complex under thermally forcing con-
ditions (220–2408C).
[a] Reaction conditions: 1a (1.1 mmol); Ti(OiPr)4 (5 mol%), Ln
(5 mol%); PhMe (3 mL). [b] Determined by quantitative 1H NMR
(qNMR) analysis using an internal standard. [c] Determined by HPLC on
a chiral stationary phase.
Brønsted acids[7] unlikely. We chose this process as a platform
to search for an asymmetric catalytic hydroalkoxylation
reaction by combining aluminum isopropoxide with chiral
steering ligands such as those shown in Figure 1.
While experiments with aluminum-based catalysts were
not successful, combinations of Ti(OiPr)4, which itself is not
catalytically active (Table 1, entry 1), with some ligands were
The cyclization of 2-allylphenols (1) to 2-methylcoumar-
ans (2-metyl-2,3-dihydrobenzofurans; 2) served as the test
reaction to screen for hydroalkoxylation catalysts (see
[*] M. Sc. J. Schlꢀter, M. Sc. M. Blazejak, Dr. F. Boeck,
Prof. Dr. L. Hintermann
Figure 1. Chiral, chelating oxygen donor ligands used in this study.
Department Chemie, Technische Universitꢁt Mꢀnchen
Lichtenbergstr. 4, 85748 Garching bei Mꢀnchen (Germany)
E-mail: lukas.hintermann@tum.de
productive. Catalysts with BINOL (L2) showed low activity
and did not result in stereoselectivity (entries 2 and 3), but the
chiral carboxylic acid MeO-BINA-Cox[13] (L1; Figure 1) led
to the in situ formation of a catalyst that produced 2-
methylcoumaran (2a) with distinct enantiomeric excess at
2508C (Table 1, entries 4 and 5).
[**] The reported research received funding by the Deutsche For-
schungsgemeinschaft (HI 854/5-1).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201409252.
Angew. Chem. Int. Ed. 2015, 54, 1 – 5
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 3: Ligand effects on the catalytic cyclization of 1a to 2a.[a]
The influence of the solvent (entries 4,5), the reaction
temperature (entries 5–9),[14] and other factors, including the
ligand/metal ratio and the concentration, were elucidated
with (aS)-L1 as the steering ligand at reaction times of
15 min.[15,16] The product was not formed in catalytic reactions
performed below 2008C (entry 6). The remarkable enantio-
selectivity of 70% ee that was obtained at 2408C (entry 8)
provides an example for asymmetric catalysis at uncommonly
high temperatures (HOT-CAT).[17,18] Curiously, the repetition
of experiments from Table 1 over time gave variable results
(e.g. 8–32% yield for entry 8). The cause of these variations
was the presence of trace amounts of water in either the
starting material 1a or the solvent. Catalytic reactions
performed with dried substrate and solvent gave a low yield
(Table 2, entry 1), whereas reactions with co-catalytic
amounts of water showed the highest catalytic activity and
selectivity (entries 2 and 3), which again eroded upon
addition of additional water (entries 4 and 5).
Entry
Ligand
Yield [%][b]
ee [%]
1a
1b
L1
56
48
75
75
L1+tBu2Py
(5 mol%)
L1+tBu2Py
(20 mol%)
1c
41
73
2
3
4
L2
L3
L4
10
10
80
0
Table 2: Influence of water on the asymmetric catalytic hydroalkoxyla-
36
57
tion.[a]
Entry
H2O [ppm][b]
H2O [mol%][c]
Yield [%][d]
ee [%]
1
2
3
4
5
10
0.1
5.1
7.8
15.4
31.3
19
55
59
32
10
60
70
60
48
7
330
500
1000
2050
5
6
L5
L6
38
71
80
[a] Reaction conditions: 1a (1.1 mmol); PhMe (3 mL). [b] Sum of added
water and water content of the solvent (10 ppm). [c] Relative to 1a.
[d] Determined by qNMR analysis using an internal standard.
93[c]
The results imply that the in situ generation of the catalyst
requires a single equivalent of water per titanium atom.
Catalytic reactions were henceforth carried out with dry
reactants and solvents,[19] but with added water as the co-
catalyst.[20] To this effect, Ti(OR)4 (5 mol%), MeO-BINA-
Cox (5 mol%), and water (5 mol%) were preheated in
toluene at 608C for 10 min. After the addition of the
substrate, the reaction mixture was heated to 2408C for 20–
50 min. Both Ti(OiPr)4 and Ti(OtBu)4 gave similar results,[15]
but the heavier Group 4 metal alkoxides (Zr(OtBu)4: 21%
yield, 12% ee; Hf(OtBu)4: 25% yield, 34% ee) gave less
active and selective catalysts. The standard reaction with (aS)-
MeO-BINA-Cox (L1; Table 3, entry 1a) was not affected to
a great extent by the presence of 2,6-di-tert-butylpyridine
(entries 1b and c), excluding the interference of a strong
Brønsted acid in this catalysis. In the absence of a metal
precursor, carboxylic acid L1 did not show catalytic activity.[15]
Catalysis with (aR)-BINOL gave the racemic product in a low
yield (Table 3, entry 2). Dicarboxylic acid L3 displayed low
activity, but a distinct stereoselective induction (entry 3).
Derivatives of L1 were more successful: the methylated L4[21]
showed higher activity with lower enantioselectivity (entry 4).
[a] Reaction conditions: 1a (1.5 mmol); Ti(OiPr)4 (5 mol%), ligand
(5 mol%), H2O (5 mol%); PhMe (3 mL). [b] Determined by qNMR
analysis using an internal standard. [c] 50 min reaction time.
Auxiliary MNCB (L5)[22] displayed lower activity, but similar
selectivity like L1 (entry 5).
The screening pointed to the necessary presence of
a carboxylic acid[23] and a methoxy group in the ligand. It
also implied increased catalyst activity for ligands with
electron-donating groups (see entry 4). Indeed, the 6’-tert-
butylated ligand L6 showed both higher activity and enantio-
selectivity (entry 6). The optimized reaction conditions, but
with an extended reaction time of 50 min to increase
conversion, were applied to several 2-allylphenol substrates
(Table 4). Readily available L1 served as reference ligand,
while some reactions with L6 were performed to achieve
highest selectivities (Table 4).[24,25]
The yield of 2a was significantly higher at longer reaction
times, while the enantioselectivity remained unchanged
(entry 1a). The invariance of product ee value with reaction
time excludes any marked reversibility of the reaction at
2
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Table 4: Substrate scope of catalytic hydroalkoxylation.[a]
Heating of L1 (99.7% ee) in toluene to 2408C for 20 min left
its ee value unchanged.[15] From an actual catalytic reaction
(Table 3, entry 1a), L1 could be recovered (86%) with
unchanged enantiomeric excess (99.7%). The isopropyl-
(E1; 7%) and 2-allylphenyl esters of L1 (E2; 4%) were also
detected in the crude reaction mixture.[15] From a catalytic
reaction performed at 2658C, reisolated L1 (61%) had
a slightly lowered ee value of 99.4%, whereas from catalytic
reactions performed at 2908C, none of the ligand L1 could be
recovered.
Entry Substrate
Product
Yield
[%][b]
ee
[%]
1a
1b
1c
1d
L1: 84 75
L1: 77[c] 41[d,e]
L6: 92 80
L6: 82[c] 85[f]
1a
1b
2a
Results from preliminary studies of the in situ generated
titanium catalyst by NMR spectroscopy and FAB-MS spec-
trometry are consistent with the in situ generation of a species
of the type [Ti4(m-O)2(OH)4(MeO-BINA-CO2)4(OiPr)4] (m/
z = 1819.7, [MÀOH]+) with two-fold symmetry. Additional
studies are required to answer questions regarding the nature
of the catalyst and the reaction mechanism, which might be
similar to the one discussed for the reaction catalyzed by
aluminum isopropoxide.[12]
In conclusion, we have described the realization of an
asymmetric metal-catalyzed hydroalkoxylation of non-acti-
vated alkenes in a model system. This partial solution to
a long-standing problem takes the form of a cycloisomeriza-
tion of 2-allylphenols (1) to 2-methylcoumarans (2) and is
catalyzed by a new chiral catalyst based on a titanium–
carboxylate complex. The remarkably high temperature of
the process exceeds those earlier used in asymmetric catal-
ysis.[18] Studies of the scope and mechanism of the catalytic
reaction are ongoing.
2a
2b
2c
L1: 88 78
2b L6: 93 84
L6: 88 87[f]
3a
3b
3c
L1: 71 72
2c L6: 79 75
L6: 78 81[f]
1c
1d
1e
1 f
1g
4a
4b
L1: 51 61[e]
2d
L6: 83 77[g]
5
2e L1: 56 72
6a
6b
6c
L1: 90 62
2 f L6: 93 69
L6: 88 73[f]
Received: September 18, 2014
Revised: November 6, 2014
Published online: && &&, &&&&
7a
7b
L1: 71 62
2g
L6: 86 73
Keywords: alkenes · asymmetric catalysis · heterocycles ·
.
microwave chemistry · titanium
8a
8b
L1: 88 67[e]
2h
1h
L1: 76 71[f]
[1] a) K. Tani, Y. Kataoka in Catalytic Heterofunctionalization
(Eds.: A. Togni, H. Grꢀtzmacher), Wiley-VCH, Weinheim,
2001; b) L. Hintermann, Top. Organomet. Chem. 2010, 31, 123 –
155; c) C. J. Weiss, T. J. Marks, Dalton Trans. 2010, 39, 6576 –
6588.
[a] Reaction conditions: 1 (1.5 mmol); Ti(OiPr)4 (5 mol%), L1 or L6
(5 mol%), H2O (5 mol%); PhMe (3 mL); internal pressure typically
14 bar. [b] Yield of isolated product. [c] Yield determined by qNMR
analysis using an internal standard. [d] At 3008C. [e] Reaction time
20 min. [f] At 2208C. [g] Catalyst loading: 8 mol% of each component.
[2] a) Hydrofunctionalization—Top. Organomet. Chem., Vol. 43
(Eds.: V. P. Ananikov, M. Tanaka), Springer, Berlin, 2013;
b) J. F. Hartwig, Nature 2008, 455, 314 – 322; c) Catalytic Hetero-
functionalization (Eds.: A. Togni, H. Grꢀtzmacher), Wiley-
VCH, Weinheim, 2001.
[3] Hydroalkoxylation denotes the addition of both aliphatic
alcohols and phenols to alkenes, in analogy to the general term
hydroamination. Phenol additions may be further specified as
hydroaryloxylations.
[4] a) J. Hannedouche, E. Schulz, Chem. Eur. J. 2013, 19, 4972 –
4985; b) T. E. Mꢀller, K. C. Hultzsch, M. Yus, F. Foubelo, M.
Tada, Chem. Rev. 2008, 108, 3795 – 3892.
[5] a) E. M. Barreiro, L. A. Adrio, K. K. Hii, J. J. Brazier, Eur. J.
Org. Chem. 2013, 1027 – 1039; b) R. L. LaLonde, Z. J. Wang, M.
Mba, A. D. Lackner, F. D. Toste, Angew. Chem. Int. Ed. 2010, 49,
598 – 601; Angew. Chem. 2010, 122, 608 – 611; c) G. L. Hamilton,
E. J. Kang, M. Mba, F. D. Toste, Science 2007, 317, 496 – 499;
d) Z. Zhang, R. A. Widenhoefer, Angew. Chem. Int. Ed. 2007,
46, 283 – 285; Angew. Chem. 2007, 119, 287 – 289.
2408C.[26] Even at 3008C, the reaction provided a notable
enantiomeric excess (entry 1b). Higher selectivity was
obtained with ligand L6 and at 2208C (entries 1c and d).
Core-alkylated 2-allylphenols were converted to 2-methyl-
coumarans in acceptable to high yields and enantioselectiv-
ities (entries 2–4), providing maximal values of 84% ee
(2408C) or 87% ee (2208C) for product 2b with ligand L6.
Core-halogenated 2-allylphenols were also tolerated
(entries 5–7). Substrate 2h with a geminal dimethyl unit
cyclized faster as a result of entropy effects (entry 8a);
lowering of the reaction temperature to 2208C also slightly
increased the enantioselectivity in this case (entry 8b).
The high reaction temperature posed questions regarding
the chemical and configurational stability of the catalyst.
Angew. Chem. Int. Ed. 2015, 54, 1 – 5
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[6] a) S. Antoniotti, S. Poulain-Martini, E. DuÇach, Synlett 2010,
2973 – 2988. For less hydrolyzable Lewis acidic metal catalysts,
see: b) H. Qian, X. Han, R. A. Widenhoefer, J. Am. Chem. Soc.
2004, 126, 9536 – 9537; c) A. Dzudza, T. J. Marks, Chem. Eur. J.
2010, 16, 3403 – 3422.
[7] a) T. T. Dang, F. Boeck, L. Hintermann, J. Org. Chem. 2011, 76,
9353 – 9361; b) D. C. Rosenfeld, S. Shekhar, A. Takemiya, M.
Utsunomiya, J. F. Hartwig, Org. Lett. 2006, 8, 4179 – 4182;
c) T. C. Wabnitz, J.-Q. Yu, J. B. Spencer, Chem. Eur. J. 2004,
10, 484 – 493.
[8] a) C. S. Sevov, J. F. Hartwig, J. Am. Chem. Soc. 2013, 135, 9303 –
9306. For a symmetric Ir-catalyzed hydroaryloxylation, see:
b) M. C. Haibach, C. Guan, D. Y. Wang, B. Li, N. Lease, A. M.
Steffens, K. Krogh-Jespersen, A. S. Goldman, J. Am. Chem. Soc.
2013, 135, 15062 – 15070. For a discussion of other claims of
asymmetric catalytic hydroalkoxylation, see Refs. [1b,7a].
[9] a) Y. Nishiyama, T. Wada, K. Kakiuchi, Y. J. Inoue, J. Org. Chem.
2012, 77, 5681 – 5686; b) Y. Nishiyama, T. Wada, S. Asaoka, T.
Mori, T. A. McCarty, N. D. Kraut, F. V. Bright, Y. Inoue, J. Am.
Chem. Soc. 2008, 130, 7526 – 7527; c) D. S. Hamilton, D. A.
Nicewicz, J. Am. Chem. Soc. 2012, 134, 18577 – 18580; d) A. J.
Perkowski, D. A. Nicewicz, J. Am. Chem. Soc. 2013, 135, 10334 –
10337.
[10] a) S. M. Podhajsky, M. S. Sigman, Organometallics 2007, 26,
5680 – 5686; b) Y. Zhang, M. S. Sigman, Org. Lett. 2006, 8, 5557 –
5560; c) K. M. Gligorich, M. J. Schultz, M. S. Sigman, J. Am.
Chem. Soc. 2006, 128, 2794 – 2797; d) H. Shigehisa, T. Aoki, S.
Yamaguchi, N. Shimizu, K. J. Hiroya, J. Am. Chem. Soc. 2013,
135, 10306 – 10309; e) D. Schuch, P. Fries, M. Dçnges, B.
Menꢁndez Pꢁrez, J. Hartung, J. Am. Chem. Soc. 2009, 131,
12918 – 12920; f) L. Li, S. B. Herzon, Nat. Chem. 2014, 6, 22 – 27;
g) L. Li, S. B. Herzon, J. Am. Chem. Soc. 2012, 134, 17376 –
17379.
[14] The temperature was measured externally with an IR probe and
calibrated against the internal temperature as determined by
a ruby thermometer. Errors are in the order of ꢀÆ 58C at the
highest values (3008C).
[15] Additional data is provided in the Supporting Information.
[16] Reaction times refer to hold times at the target temperature, not
including the heating and cooling phase. See the Supporting
Information for typical temperature and pressure progress
curves.
[17] We use the acronym HOT-CAT for “homogeneous thermal
catalysis” at high temperatures (> 2008C).
[18] For asymmetric catalysis at very high temperatures (> 1708C),
see: a) A. L. Reznichenko, H. N. Nguyen, K. C. Hultzsch,
Angew. Chem. Int. Ed. 2010, 49, 8984 – 8987; Angew. Chem.
2010, 122, 9168 – 9171; b) A. S. Tsai, R. M. Wilson, H. Harada,
R. G. Bergman, J. A. Ellman, Chem. Commun. 2009, 3910 –
3912; c) N.-F. K. Kaiser, U. Bremberg, M. Larhed, C. Moberg,
A. Hallberg, Angew. Chem. Int. Ed. 2000, 39, 3595 – 3598;
Angew. Chem. 2000, 112, 3741 – 3744. The last example features
short reaction times (5 – 7 min) with steep temperature gradi-
ents, where the peak temperatures do not correspond to average
reaction temperatures.
[19] The water content of toluene (2–10 mg H2O/mL) and 2-
allylphenol (distilled; 30 mg H2O/mL), both stored over molec-
ular sieves, was monitored by Karl Fischer titration.
[20] S. Ribe, P. Wipf, Chem. Commun. 2001, 299 – 307.
[21] B. Hungerhoff, P. Metz, Tetrahedron 1999, 55, 14941 – 14946.
[22] a) Y. Fukushi, C. Yajima, J. Mizutani, Tetrahedron Lett. 1994, 35,
599 – 602; b) Y. Fukushi, PhD Thesis, Hokkaido University,
Sapporo, 1996.
[23] No catalytic activity was observed with esters or amides of L1.
[24] Remaining starting material was detected in reactions that were
run for 20 min.
[11] For asymmetric catalytic diene–phenol cycloisomerizations, see:
a) A. Sakakura, M. Sakuma, K. Ishihara, Org. Lett. 2011, 13,
3130 – 3133; b) H. Nguyen, M. R. Gagnꢁ, ACS Catal. 2014, 4,
855 – 859.
[25] Yield (qNMR) and reaction time with 5 mol% of the in situ
generated catalyst with Ti(OiPr)4 at 2408C: 52% (15 min), 56%
(20 min), 67% (30 min), 87% (50 min). The ee value was stable
(73–75%) in all experiments.
[12] J. Schlꢀter, M. Blazejak, L. Hintermann, ChemCatChem 2013, 5,
3309 – 3315.
[13] MeO-BINA-Cox (L1) is 1-(2’-methoxy-1’-naphthyl)-2-naph-
thoic acid: T. Hattori, H. Hotta, T. Suzuki, S. Miyano, Bull.
Chem. Soc. Jpn. 1993, 66, 613 – 622.
[26] For a discussion of the characteristics of reversible asymmetric
catalyses, see: a) L. Hintermann, C. Dittmer, Eur. J. Org. Chem.
2012, 5573 – 5584; b) L. Hintermann in Asymmetric Synthesis—
More Methods and Applications (Eds.: M. Christmann, S. Brꢂse)
Wiley-VCH, Weinheim, 2012, pp. 109 – 115.
4
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Asymmetric Catalysis
J. Schlꢁter, M. Blazejak, F. Boeck,
L. Hintermann*
&&&& — &&&&
Asymmetric Hydroalkoxylation of Non-
Activated Alkenes: Titanium-Catalyzed
Cycloisomerization of Allylphenols at
High Temperatures
Best served hot and fast: The metal-
catalyzed asymmetric hydroalkoxylation
of non-activated alkenes was catalyzed by
a chiral titanium–carboxylate complex
that requires co-catalytic amounts of
water to reach full activity. This homoge-
neous thermal catalysis (HOT-CAT) pro-
ceeds at remarkably high temperatures
(above 2208C, typically 2408C) and pro-
duces 2-methylcoumarans from 2-allyl-
phenols in up to 90% yield and 87% ee.
Angew. Chem. Int. Ed. 2015, 54, 1 – 5
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!