Stereoselective transformations of α-trifluoromethylated ketoximes to optically active amines by enzyme-nanometal cocatalysis: Synthesis of (s)-inhibitor of phenylethanolamine n-methyltransferase

Article abstract of DOI:10.1002/cctc.201402114

One-pot cascade synthesis of optically active α-trifluoromethylated amines directly from ketoximes was accomplished with the use of Candida antarctica lipase B and catalysts prepared by atomic layer deposition (ALD). Compared to the commercial palladium catalyst, the ALD-prepared catalysts showed much higher activity and afforded various α-trifluoromethylated amides in good yields and with high enantioselectivity. One of the enantiopure amides was further hydrolyzed into the corresponding amine, which was treated as a crucial starting material for total synthesis of (S)-inhibitor of phenylethanolamine N-methyltransferase without loss of chiral information.

Full text of DOI:10.1002/cctc.201402114

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201402114
Stereoselective Transformations of a-Trifluoromethylated
Ketoximes to Optically Active Amines by Enzyme–
Nanometal Cocatalysis: Synthesis of (S)-Inhibitor of
Phenylethanolamine N-Methyltransferase
Guilin Cheng,[a] Qi Wu,[a] Zeyu Shang,[b] Xinhua Liang,*[b] and Xianfu Lin*[a]
One-pot cascade synthesis of optically active a-trifluoromethy-
lated amines directly from ketoximes was accomplished with
the use of Candida antarctica lipase B and catalysts prepared
by atomic layer deposition (ALD). Compared to the commercial
palladium catalyst, the ALD-prepared catalysts showed much
higher activity and afforded various a-trifluoromethylated
amides in good yields and with high enantioselectivity. One of
the enantiopure amides was further hydrolyzed into the corre-
sponding amine, which was treated as a crucial starting materi-
al for total synthesis of (S)-inhibitor of phenylethanolamine
N-methyltransferase without loss of chiral information.
Introduction
Trifluoromethyl-containing compounds have attracted growing
attention in various fields because the introduction of the tri-
fluoromethyl group having powerful electron-withdrawing
ability often leads to special physical, chemical, and biological
properties.[1] a-Trifluoromethylated amines, especially the opti-
cally active ones, are very important building blocks in the syn-
thesis of fluorinated pharmaceuticals, particularly in the design
of bioactive products.[2] For example, inhibitors of phenyletha-
nolamine N-methyltransferase (PNMT) exhibit interesting bio-
logical activity owing to the privileged chiral synthons of a-tri-
fluoromethylated amines, and the synthetic routes of PNMT in-
hibitors have been reported; however, their enantiomers were
separated only by chiral HPLC.[3]
amines through a one-pot sequential process of KR/DKR/KR
with palladium and lipase as the catalysts.[4] Although moder-
ate conversions were obtained, the reaction time was very
long (10 days). In addition, the methods related to DKR re-
quired a start from racemic amines, which are difficult to be
synthesized and purified. Kim and co-workers reported that
optically active amines could be transformed from ketoximes,[5]
however, these reports did not involve trifluoromethylated
amines and there was no consideration of hydrogen pressure
in the reaction, which is the most critical reaction factor.
Herein, we envisioned that optically active a-trifluoromethylat-
ed amines could be synthesized directly from ketoximes.
This work was focused on the design of an efficient catalyst
for both hydrogenation of ketoximes and racemization of the
following amines with the a-trifluoromethyl group. Atomic
layer deposition (ALD) was used to prepare highly dispersed
metal nanoparticles (NPs) for this study. ALD is a layer-by-layer
process[6] and has received increasing attention for the prepa-
ration of metal NPs, such as Pt and Pd.[7] Owing to their high
cohesive energy, metal NPs tend to form through an island
growth mechanism (Volmer–Weber mechanism) during the ini-
tial stages of ALD processes, and metal NPs can be highly dis-
persed on porous catalyst substrates. Thermostable Candida
antarctica lipase B (CALB; trade name, Novozym 435) was
chosen as the catalyst for resolution (Scheme 1).
To the best of our knowledge, rare reports have referred to
chemoenzymatic synthesis of chiral a-trifluoromethylated
amines especially by dynamic kinetic resolution (DKR). As the
racemization rate of a-trifluoromethylated amines is very slow
owing to the formation of a hydrogen bond between CF3 and
NH2, which leads to difficult achievement of equilibrium be-
tween amine and imine. Recently, we have developed an ele-
gant chemoenzymatic resolution of a-trifluoromethylated
[a] G. Cheng, Dr. Q. Wu, Prof. X. Lin
Department of Chemistry
Zhejiang University
Hangzhou 310027 (P. R. China)
Tel : (+86)571-8795-3001
E-mail: llc123@zju.edu.cn
Results and Discussion
[b] Z. Shang, Prof. X. Liang
Department of Chemical and Biochemical Engineering
Missouri University of Science and Technology
Rolla, MO 65409 (USA)
Metal nanocatalysts synthesized by ALD
With ten cycles of Pd ALD on r-alumina, the Pd content is
0.95 wt% based on inductively coupled plasma analysis com-
bined with mass spectrometry (ICP-MS). The coated particles
were visualized with a JEOL 2010F 200 kV Schottky field emis-
Tel : (+1)573-341-7632
E-mail: liangxin@mst.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201402114.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2129 – 2133 2129

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Table 1. Dynamic kinetic resolution of a-trifluoromethylated amine 1a.
Entry[a] Catalyst
T [8C] t [d] Conv. [%][b] eeamide [%][c]
1
2
3
4
5
5
6
commercial Pd/Al2O3 70
7
5
7
7
5
7
5
70
88
91
84
87
92
75
99
99
99
99
99
99
99
ALD Pd/Al2O3
ALD Pd/Al2O3
ALD Pd/Al2O3
ALD Ni/Al2O3
ALD Ni/Al2O3
ALD Ni/silica gel
70
70
60
70
70
70
Scheme 1. Lipase–nanometal cocatalyzed transformations of a-trifluorome-
thylated ketoximes to optically active amides.
[a] Reaction conditions: substrate rac-1a (0.1 mmol), IPAC (0.5 mmol), TEA
(0.2 mmol), CAL-B (50 mg), and nanocatalyst (0.002 mmol) in dry toluene
(1 mL) in the presence of 4 ꢁ molecular sieve (100 mg). [b] Determined
by GC using AT·SE-30 column. [c] Determined by GC using chiral column.
sion transmission electron microscope equipped with an
Oxford detector unit for elemental analysis while imaging. In
Figure 1a one STEM image of cross-sectioned surface of r-alu-
mina deposited with Pd NPs is shown . Pd NPs are clearly visi-
reason for the high reactivity of ALD catalysts could be as-
cribed to the highly dispersed metal NPs and uniform particle
size. This set of experiments indicates that both palladium and
nickel on Al2O3 as the basic support could be suitable for DKR
of rac-1a. However, nickel on acid support such as silica gel
gave only low conversion (entry 6), which could be caused by
the deactivation of the enzymatic resolution catalyst or hydrol-
ysis of the acyl donor. The influence of the catalyst substrates
was in accordance with what was discovered by Parvulescu
and co-workers.[8]
Figure 1. a) Cross-sectioned STEM image of Pd ALD-deposited Al2O3 particles
and b) high-resolution TEM image of Ni ALD-deposited SiO2 particles.
Hydrogenation of a-trifluoromethylated ketoximes
A hydrogenation study on ketoxime 2a was then conducted in
dry toluene with different catalysts. According to the data re-
ported by Kim and co-workers,[5b] asymmetric reductive acyla-
ble as white dots and are highly dispersed on the inside
porous structure as well as on the outside surface of the parti-
cle substrate. The Pd NPs produced exhibit a narrow particle
size distribution with an average particle size of approximately
3 nm diameter. Similar highly dispersed metal NPs were ob-
tained by Ni ALD. After one cycle of Ni ALD, the content of Ni
on r-alumina and silica gel is 1.59 wt% and 0.73 wt%, respec-
tively. In Figure 1b, one HRTEM image of Ni NPs dispersed on
20–30 nm silica particles is shown. The average Ni NP size is
approximately 2.5 nm.
Table 2. Hydrogenation of a-trifluoromethylated ketoxime 2a.
Entry[a]
Catalyst
H2 [MPa]
T [8C]
t [h]
Conv. [%][b]
Dynamic kinetic resolution of a-trifluoromethylated amines
1
2
3
4
5
6
7
8
commercial Pd/Al2O3
ALD Pd/Al2O3
ALD Pd/Al2O3
ALD Pd/Al2O3
ALD Pd/Al2O3
ALD Pd/Al2O3
ALD Ni/Al2O3
2.00
0.10
0.10
0.10
0.10
0.05
0.10
0.10
100
100
70
70
70
70
70
70
48
48
48
48
36
48
36
36
96
99
99
99
98
87
98
95
The dynamic kinetic resolution of a-trifluoromethylated amine
(rac-1a) was studied as a model substrate with the ALD NP
catalysts and the commercially available Pd/Al2O3 that was
studied in our previous report.[4] The conversion of a-trifluoro-
methylated amine 1a and the selectivity to the corresponding
amide for different catalysts are listed in Table 1. The results in-
dicate that almost all ALD catalysts have much higher activity
than the conventional Pd/Al2O3 catalyst even at low tempera-
ture (entry 4) or during short reaction time (entry 2). The
ALD Ni/silica gel
[a] Reaction conditions: substrate ketoxime 2a (0.1 mmol) and nanocata-
lyst (0.001 mmol) in dry toluene (1 mL) under hydrogen pressure. [b] De-
termined by GC using AT·SE-30 column.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2129 – 2133 2130

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
tion of aromatic ketoximes was performed at 1 atm of hydro-
gen or lower hydrogen pressure. However, at the referred hy-
drogen pressure (0.1 MPa),[5] no reduction of ketoxime 2a was
observed for the commercial Pd/Al2O3 even after 24 h. The in-
hibitive role for trifluoromethyl group could be attributed to
the high negative inductive effect that decreased the electron
density on C=N double bond of the ketoximes. Inspired by the
conclusion made by Niedermann and co-workers,[9] the hydro-
genation pressure was enhanced to 2.0 MPa and the reaction
was observed (Table 2, entry 1). In contrast, using ALD cata-
lysts, the hydrogenation reaction was completed within a very
short time at 1008C under 2.0 MPa. To achieve one-pot cas-
cade chemoenzymatic reactions, which are hydrogenation of
a-trifluoromethylated ketoximes followed by DKR of the inter-
mediate amines, hydrogen pressure was lowered to 0.1 MPa
and complete conversion was obtained (entry 2). The reaction
time and temperature were fur-
corresponding enantiopure amide (S)-3a in high yield and ex-
cellent ee (Table 3, entries 2 and 3). For economic benefit, ALD
Ni/Al2O3 was chosen and the optimal reaction conditions were
also applied to other substrates; the corresponding amides
were obtained with good yield (70–95%) and high enantio-
meric excess (90–99%, entries 5–7). However, at a hydrogen
pressure of 0.1 MPa, lower catalytic activity was observed with
relatively poor yield for the substrate 2b. Therefore, higher
pressure (0.15 MPa) and higher temperature (1008C) were ap-
plied to the hydrogenation step (entry 4). With regards to the
substrate 1 f, significant side reactions took place in hydroge-
nation of the ketoxime; therefore, lower pressure and tempera-
ture were used (entry 8). The results from Table 3 indicate that
both aromatic and aliphatic a-trifluoromethylated ketoximes
are good substrates toward the asymmetric reductive acylation
by the combination of ALD catalysts and lipase. Notably, the
ther reduced to match the con-
Table 3. Asymmetric transformations of a-trifluoromethylated ketoximes to chiral amides under optimized
conditions.
ditions required for DKR reac-
tions, and the results were also
satisfying. The results are consis-
tent with those reported by Par-
vulescu and co-workers for race-
mization of chiral amines[8] and
our group’s previous work for
DKR reactions.[4] However, if we
further brought down the hydro-
gen pressure, the reaction was
not complete with a significant
amount of substrate 2a left
(entry 6). Similar results were ob-
tained for the ALD-deposited
nickel NPs on Al2O3 or silica gel.
Entry[a] Substrates
Catalyst
H2 pressure [MPa]
Reduction[b] Acylation[c]
Yield [%][d] eeamide [%][e]
1
2
3
2a commercial Pd/Al2O3 2.00[f]
0.01
0.01
0.01
62
93
92
99
99
99
2a ALD Pd/Al2O3
2a ALD Ni/Al2O3
0.10
0.10
4
5
2b ALD Ni/Al2O3
2c ALD Ni/Al2O3
0.15[g]
0.10
0.01
0.01
78
82
98
98
One-pot chemoenzymatic syn-
thesis of optically active a-tri-
fluoromethylated amines
With the results from the sepa-
rate investigations on the DKR
and hydrogenation in hand, the
combination in a one-pot cas-
cade chemoenzymatic reaction
is feasible. For commercial cata-
lyst Pd/Al2O3, hydrogen pressure
should be turned down manual-
ly for acylation after reductive
process. To our satisfaction, with
ALD prepared catalysts, hydro-
gen was consumed during re-
duction reaction to the appropri-
ate pressure for the following
acylation.
6
7
2d ALD Ni/Al2O3
2e ALD Ni/Al2O3
0.10
0.10
0.01
0.01
81
70
98
99
8
9
2 f ALD Ni/Al2O3
2g ALD Ni/Al2O3
0.08[h]
0.10
0.01
0.01
85
95
99
90
[a] Reaction conditions: substrate ketoxime (0.1 mmol), IPAC (0.5 mmol), TEA (0.2 mmol), CAL-B (50 mg) and
nanocatalyst (0.001 mmol for reduction, 0.002 mmol for acylation) in dry toluene (1 mL) in the presence of 4 ꢁ
molecular sieve (100 mg) at 708C. [b] For 36 hr. [c] For 6 days. [d] Determined by GC using AT·SE-30 column.
[e] Determined by GC using chiral column. [f] Reduction for 48 hr. [g] Hydrogenation at 1008C. [h] Hydrogena-
tion at 608C.
With respect to the substrate
2a, different catalysts were used.
The ALD catalysts under opti-
mized conditions afforded the
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2129 – 2133 2131

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
nature of the substituent on the benzene ring affects the yield.
Although a clear rationale is not available to the best of our
knowledge, this could be attributable to electronic effect and
steric hindrance.
was demonstrated. Highly dispersed metal nanoparticles were
prepared by atomic-layer deposition (ALD) method. Good
yields and high enantiomeric excesses were obtained for the
ALD-prepared catalysts combined with Candida Antarctica
lipase B.
The amides could be readily hydrolyzed to optically active
amines, which have been used as a crucial chiral synthon to
accomplish total synthesis of (S)-3-trifluoromethyl-tetrahydroi-
soquinoline as an important (S)-inhibitor of phenylethanola-
mine N-methyltransferase . Interestingly, (R)-inhibitor of phenyl-
ethanolamine N-methyltransferase could be obtained in the
same way by careful selection of lipase, and alkaline protease
as an example is in progress to switch the enantioselectivity of
amines.
Synthetic applications
The amide products can be readily hydrolyzed to the corre-
sponding optically active a-trifluoromethylated amines, which
are extremely useful as chiral building blocks. As a representa-
tive example, optically active amide 3b was hydrolyzed under
acidic condition (6n HCl) to obtain (S)-amine 1b quantitatively
without loss of enantiopurity (Scheme 2). Amine 1b was treat-
ed with ethyl chloroformate to form chiral carbamate 5, which
Experimental Section
Metal nanocatalysts synthe-
sized by ALD
The deposition of Pd and Ni nano-
particles by ALD was performed in
a vibrating fluidized-bed reactor,
which was similar to the one de-
scribed in detail previously.[10] Pd
ALD was performed at 2008C by
using alternating exposures to PdII
hexafluoroacetylacetonate
[Pd-
(hfac)2] and formalin (a 37% solu-
tion of formaldehyde in water con-
taining 10–15% methanol), which
were purchased from Sigma–Al-
drich and were used as received.
Prior to the Pd ALD, an alumina
film with eight cycles of alumina
ALD (ꢀ1 nm) was deposited as
a buffer or seed layer. The alumina
ALD was performed using alternat-
ing reactions of trimethylaluminum
(TMA, 97%, Sigma–Aldrich) and
deionized water at 1778C. Ni ALD
was performed at 3008C by using
alternating exposures to nickelo-
cene and H2. Porous alumina parti-
cles, with a particle size of approxi-
Scheme 2. Synthesis of (S)-inhibitor of PNMT starting from the a-trifluoromethylated ketoxime 2b.
mately 50 mm and a surface area of 103.5 m2 gÀ1, were used as the
substrates for Pd and Ni ALD. Silica particles with a particle size of
approximately 20–30 nm, and silica gel particles with a particle size
of 30–75 mm, an average pore size of 15 nm, and a BET surface
area of 240 m2 gÀ1 were also used as the substrates for Ni ALD. Ten
cycles of Pd ALD and one cycle of Ni ALD were applied. The de-
tailed procedures for the Pd[11] and Ni ALD[12] can be found else-
where.
was cyclized with polyphosphoric acid to give lactam (S)-6. Re-
duction of the lactam with BH3 THF produced 3-trifluorometh-
.
yl-tetrahydroisoquinoline (S)-7 with 72% yield and 98% ee.[3]
Notably, the ketoxime was starting from trifluoromethyl
ketone, which was formed by the reaction of a-trifluoromethy-
lated Weinreb amide with benzyl Grignard reagent. To our
relief, the enantiomeric excess could be maintained at a high
level (98% ee) during five synthetic steps and the target mole-
cule could be easily separated and purified in the form of hy-
drochloride salt.
General procedure for chemoenzymatic transformations of
a-trifluoromethylated ketoximes
A suspension containing substrate 2a (0.1 mmol), isopropyl acetate
(IPAC, 0.5 mmol), triethylamine (TEA, 0.2 mmol), CAL-B (50 mg),
1.59 wt% ALD Ni/Al2O3 (0.001 mmol) and 4 ꢁ molecular sieve
(100 mg) in dry toluene (1 mL) was stirred at 708C and 0.1 MPa hy-
Conclusions
Asymmetrically conversion of prochiral a-trifluoromethylated
ketoximes to chiral amides by enzyme–nanometal cocatalysts
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2129 – 2133 2132

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[2] a) B. Tçrçk, G. K. S. Prakash, Adv. Synth. Catal. 2003, 345, 165–168; b) K.
Kato, Y. Gong, T. Saito, Y. Yokogawa, J. Mol. Catal. B 2004, 30, 61–68;
c) F. Gosselin, P. D. O’Shea, S. Roy, R. A. Reamer, C. Chen, R. P. Volante,
Org. Lett. 2005, 7, 355–358; d) J. Xu, Z. J. Liu, X. J. Yang, L. M. Wang,
G. L. Chen, J. T. Liu, Tetrahedron 2010, 66, 8933–8937; e) Y. W. Wu, L.
Deng, J. Am. Chem. Soc. 2012, 134, 14334–14337.
[3] a) G. L. Grunewald, T. M. Caldwell, Q. F. Li, K. R. Criscione, J. Med. Chem.
1999, 42, 3315–3323; b) G. L. Grunewald, J. Lu, K. R. Criscione, C. O.
Okoro, Bioorg. Med. Chem. Lett. 2005, 15, 5319–5323.
[4] G. L. Cheng, B. Xia, Q. Wu, X. F. Lin, RSC Adv. 2013, 3, 9820–9828.
[5] a) Y. K. Choi, M. J. Kim, Y. Ahn, M.-J. Kim, Org. Lett. 2001, 3, 4099–4101;
b) K. Han, J. Park, M.-J. Kim, J. Org. Chem. 2008, 73, 4302–4304; c) K.
Han, Y. Kim, J. Park, M.-J. Kim, Tetrahedron Lett. 2010, 51, 3536–3537.
[6] a) T. Suntola, Thin Solid Films 1992, 216, 84–89; b) R. L. Puurunen, J.
Appl. Phys. 2005, 97, 121301.
drogen pressure in a 50 mL autoclave. After 36 h, another part of
1.59 wt% ALD Ni/Al2O3 (0.002 mmol) was added to the autoclave
and the hydrogen pressure was declined to 0.01 MPa. Under the
optimal conditions, the reaction mixture was stirred for 6 d, and
then cooled to RT and filtered. The filtrate was concentrated and
analyzed by GC to indicate that all of the substrate was consumed
and the yield was 92%. The filtrate was then subjected to GC by
using chiral column to confirm the enantiomeric excess of the
product (S)-3a (99% ee).
Hydrolysis of the optically active amide 3b to the amine
(S)-1b
[7] a) T. Aaltonen, A. Rahtu, M. Ritala, M. Leskela, Electrochem. Solid-State
Lett. 2003, 6, C130–C133; b) W. M. M. Kessels, H. C. M. Knoops, S. A. F.
Dielissen, A. J. M. Mackus, M. C. M. van de Sanden, Appl. Phys. Lett.
2009, 95, 013114; c) J. W. Elam, A. Zinovev, C. Y. Han, H. H. Wang, U.
Welp, J. N. Hryn, M. J. Pellin, Thin Solid Films 2006, 515, 1664–1673;
d) X. H. Liang, L. B. Lyon, Y. B. Jiang, A. W. Weimer, J. Nanopart. Res.
2012, 14, 943.
[8] a) A. Parvulescu, D. De Vos, P. Jacobs, Chem. Commun. 2005, 5307–
5309; b) A. N. Parvulescu, P. A. Jacobs, D. E. De Vos, Chem. Eur. J. 2007,
13, 2034–2043; c) A. N. Parvulescu, P. A. Jacobs, D. E. De Vos, Adv. Synth.
Catal. 2008, 350, 113–121; d) A. N. Parvulescu, P. A. Jacobs, D. E. De Vos,
Appl. Catal. A 2009, 368, 9–16.
The chiral amide 3b (92.4 mg, 0.4 mmol), 6n HCl (1.5 mL), and
methanol (1.5 mL) were added to a round-bottomed flask with
a reflux condenser, and then agitated together at 808C for 15 h.
Once the reaction was complete, the mixture was concentrated to
remove methanol, and then neutralized with 4m NaOH, and there-
after extracted with dichloromethane. The organic layers were
combined and the solvent was then removed to give (S)-amine 1b
(75 mg, 0.396 mmol, 99% yield, 98% ee). Thereafter, the optically
pure amine 1b was acted as the starting material for the total syn-
thesis of 3-trifluoromethyl-tetrahydroisoquinoline (98% ee) as an
important (S)-inhibitor of PNMT.
[9] H.-P. Niedermann, Bubenheim, D. Gutheil, B. Kreuznach, PCT Int. Appl.
US 6307103 B1, 2001.
[10] a) X. H. Liang, L. F. Hakim, G. D. Zhan, J. A. McCormick, S. M. George,
A. W. Weimer, J. A. Spencer, K. J. Buechler, J. Blackson, C. J. Wood, J. R.
Dorgan, J. Am. Ceram. Soc. 2007, 90, 57–63; b) X. H. Liang, D. M. King,
P. Li, S. M. George, A. W. Weimer, AIChE J. 2009, 55, 1030–1039.
[11] X. H. Liang, L. B. Lyon, Y. B. Jiang, A. W. Weimer, J. Nanopart. Res. 2012,
14, 943.
Keywords: amines · enzyme catalysis · fluorine · kinetic
resolution · palladium
[1] a) G. K. S. Prakash, M. Mandal, G. A. Olah, Org. Lett. 2001, 3, 2847–2850;
b) Z. J. Liu, J. T. Liu, Chem. Commun. 2008, 5233–5235; c) J. Liu, J. B. Hu,
Future Med. Chem. 2009, 1, 875–888; d) H. Q. Xiao, Y. G. Huang, F. L.
Qing, Tetrahedron: Asymmetry 2010, 21, 2949–2955; e) A. Henseler, M.
Kato, K. Mori, T. Akiyama, Angew. Chem. Int. Ed. 2011, 50, 8180–8183;
Angew. Chem. 2011, 123, 8330–8333; f) M. Liu, J. Li, X. Xiao, Y. Xie, Y.
Shi, Chem. Commun. 2013, 49, 1404–1406.
[12] T. D. Gould, A. M. Lubers, B. T. Neltner, J. V. Carrier, A. W. Weimer, J. L.
Falconer, J. W. Medlin, J. Catal. 2013, 303, 9–15.
Received: March 10, 2014
Published online on May 30, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2129 – 2133 2133