Catalytic wittig reactions of semi- and nonstabilized ylides enabled by ylide tuning

Article abstract of DOI:10.1002/anie.201406103

The first examples of catalytic Wittig reactions with semistabilized and nonstabilized ylides are reported. These reactions were enabled by utilization of a masked base, sodium tert-butyl carbonate, and/or ylide tuning. The acidity of the ylide-forming proton was tuned by varying the electron density at the phosphorus center in the precatalyst, thus facilitating the use of relatively mild bases. Steric modification of the precatalyst structure resulted in significant enhancement of E selectivity up to >95:5, E/Z. Time for a tune up: Catalytic Wittig reactions with semi- and nonstabilized ylides were enabled by use of a masked base (NaOCO₂tBu) and/or ylide tuning. The acidity of the ylide-forming proton was tuned by varying the electron density at the P center in the precatalyst, thus facilitating the use of relatively mild bases. Steric modification of the precatalyst structure resulted in significant enhancement of E selectivity.

Full text of DOI:10.1002/anie.201406103

Angewandte
Chemie
DOI: 10.1002/anie.201406103
Olefination
Catalytic Wittig Reactions of Semi- and Nonstabilized Ylides Enabled
by Ylide Tuning**
Emma E. Coyle, Bryan J. Doonan, Andrew J. Holohan, Killian A. Walsh, Florie Lavigne,
Elizabeth H. Krenske,* and Christopher J. OꢀBrien*
In memory of Leonard Mostyn OꢀBrien and Clifford Henry Spence
Abstract: The first examples of catalytic Wittig reactions with
semistabilized and nonstabilized ylides are reported. These
reactions were enabled by utilization of a masked base, sodium
tert-butyl carbonate, and/or ylide tuning. The acidity of the
ylide-forming proton was tuned by varying the electron density
at the phosphorus center in the precatalyst, thus facilitating the
use of relatively mild bases. Steric modification of the
precatalyst structure resulted in significant enhancement of
E selectivity up to > 95:5, E/Z.
T
he Wittig reaction is one of the most powerful methods for
the construction of a carbon–carbon double bond.[1] In the
60 years since its discovery, this reaction has been extensively
investigated, both in practical application[2] and mechanism.[3]
However, the traditional Wittig reaction suffers from several
limitations; in particular the removal of the phosphine oxide
by-product can be challenging. In an effort to address this
concern, in 2009 we developed the first catalytic Wittig
reaction (CWR, catalytic in phosphine), which relied on an
organosilane to chemoselectively reduce a cyclic phosphine
oxide of type 1 (Figure 1).[4a] Subsequently, we have devel-
oped modified protocols to allow for use of a soluble base,
acyclic phosphine oxides and lowering of the reaction
temperature.[4] Although these protocols were confined to
stabilized ylides,[4d] they represented the start of a process to
develop a sustainable, robust, selective, user-friendly olefina-
Figure 1. Key concepts in the proposed CWR. EDG=electron-donating
group, EWG=electron-withdrawing group, HMDS=1,1,1,3,3,3-hexam-
ethyldisilazane, DIPEA=N,N-diisopropylethylamine.
[*] Dr. E. E. Coyle,[+] B. J. Doonan,[+] A. J. Holohan, K. A. Walsh,
Dr. F. Lavigne, Dr. C. J. O’Brien
tion methodology. Indeed, a recent life cycle analysis by
Huijbregts and co-workers has shown that the CWR offers
a clear reduction in both cumulative energy demand and
greenhouse gas emissions compared to the stoichiometric
Wittig reaction.[5] In addition, other research groups have
applied our reduction strategy to the Appel, aza-Wittig, and
Staudinger reactions.[6c,e,g,h] Moreover, related organophos-
phorus catalysis has also seen significant advancement.[6] The
next major challenge was the design of a catalytic protocol
that would enable CWRs with semistabilized and nonstabi-
lized ylides. Herein, we demonstrate that this is now possible.
The key design features include utilization of a masked base
and ylide tuning.
Fundamentally, the key barrier to the utilization of
semistabilized and nonstabilized ylide classes in the CWR is
the selective deprotonation of the phosphonium salt requisite
for ylide generation (Figure 1, III). The success of this critical
deprotonation hinges on the choice of base, which must be of
sufficient power to remove the ylide-forming proton of the
National Centre for Sensor Research (NCSR), Dublin City University
Glasnevin, Dublin 9 (Ireland)
E-mail: christopher.obrien@dcu.ie
Dr. E. H. Krenske
School of Chemistry & Molecular Biosciences
The University of Queensland
Brisbane, QLD 4072 (Australia)
E-mail: e.krenske@uq.edu.au
[+] These authors contributed equally to this work.
[**] We thank Drs Martin Feeney and Gary Hessman, Trinity College
Dublin, for high-resolution mass spectrometry and Audrey Larkin
for the graphical abstract. Financial support for this work was
received from Dublin City University (Career Start), Enterprise
Ireland (CF/2011/1029), and the Australian Research Council
(FT120100632). Computational resources were provided by the
National Computational Infrastructure National Facility (Australia)
and the University of Queensland Research Computing Centre.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201406103.
Angew. Chem. Int. Ed. 2014, 53, 12907 –12911
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12907

.
Angewandte
Communications
phosphonium salt (pKa (in DMSO = dimethylsulfoxide,
DMSO) 17–18 for semistabilized, 22–25 for nonstabilized),[7]
but mild enough to be compatible with the wider CWR. A
further challenge for nonstabilized ylides will be to ensure
a viable rate of phosphonium salt formation (Figure 1, II).[3b]
We envisaged two possible approaches to achieving
efficient deprotonation. The first was identification of a suit-
able base. A literature search showed that NaHMDS, nBuLi,
NaOH, and NaOtBu are typically used.[3g] However, these
bases would be incompatible with the overall CWR, because
of their likely reaction with the silane and possibly organo-
halide. Furthermore, the bases used in our previous inves-
tigations, Na2CO3 and DIPEA, would be insufficient to
completely effect the desired deprotonation (pKa of the
conjugate acid in H2O is less than 12). To balance the need for
a stronger base while avoiding unwanted side reactions, we
hypothesized that a masked base, such as carbonate 2, could
be used to slowly release NaOtBu into the solution in situ
(Scheme 1).[8]
Scheme 1. Proposed in situ generation of NaOtBu.
As the pKa of the ylide-forming proton for nonstabilized
ylides is higher than 20, it is unlikely that 2 alone would
achieve a viable rate of deprotonation necessary to employ
this ylide class in the CWR. Therefore we would incorporate
a second approach to lower the pKa of the ylide-forming
proton sufficiently in order to facilitate the use of 2. Central to
this strategy is the concept that the introduction of electron-
withdrawing groups on the phenyl ring of the precatalyst
would lead to a withdrawal of electron density from the
phosphorus center, hence lowering the pKa of the ylide-
forming proton (Figure 1, approach 2).
The removal of electron density from the phosphorus
center may come at a cost: lower nucleophilicity of the
phosphine will have an impact upon 1) the rate of phospho-
nium salt formation (Figure 1, II), and 2) the rate of
phosphine oxide reduction (Figure 1, I).[9] Consequently, to
compensate for the retarded nucleophilicity of the phosphine
and lower reduction rate, we envisioned that the reaction
temperature would need to be increased to rebalance the
catalytic cycle. Thus, the success of ylide tuning relies on the
identification of a precatalyst structure that achieves the
desired electron-withdrawing effect, while maintaining ease
of reduction and a viable rate of phosphonium salt formation.
To this end, an initial series of phosphine oxides 1a–d
(Figure 2) were prepared, where the electron density at the
phosphorus center was varied by the introduction of electron-
withdrawing or electron-donating substituents. Notably, the-
oretical calculations supported the idea that the pKa of the
ylide-forming proton may be tuned by modifying the sub-
stituent at the phosphorus center. Density functional theory
(DFT) calculations (M06-2X/6-311 + G(d,p)//M06-2X/6-
31G(d)) gave predictions of the pKas of the P-benzyl-
Figure 2. Precatalyst library, computed pKas (DMSO), and values of
DG for proton transfer from phosphonium ions to DIPEA (toluene).
substituted phosphonium cations derived from 1a–d (1aa–
1da) and values of DG for deprotonation by DIPEA
(Figure 2).[11] The decrease in electron density at the phos-
phorus center across the series 1aa–da results in an 8 kcal
molÀ1 decrease in the value of DG for proton transfer in
toluene, a 104-fold increase in the deprotonation equilibrium
constant (Figure 1, III) at 1108C.[12]
With the strategy set, we began our CWR studies by
examining the synthesis of stilbene (4) employing precatalysts
1a–d (Table 1). Base 2 in combination with 1b resulted in the
desired CWR, producing 4 in a yield of 81% and E/Z
selectivity of 66:34 (Table 1, entry 1). Importantly, the syn-
thesis of 4 could not be achieved in high yield with any of our
previous CWR conditions or by utilizing NaOtBu as the base
(Table 1, footnote [c].) Following the successful adoption of
masked base 2 into the CWR, we probed the effect of varying
the electron density at the phosphorus center by employing
the less basic DIPEA. Gratifying, ylide tuning was clearly
demonstrated by the increase in yield observed when
progressing from electron-rich 1a to electron-deficient 1d
(entries 6–9, Table 1), albeit at the anticipated elevated
temperature. To the best of our knowledge, this is the mildest
base used for Wittig reactions employing semistabilized
ylides.[13]
12908 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 12907 –12911

Angewandte
Chemie
Table 1: Optimization of CWR conditions for the synthesis of 4.[a]
Entry
Cond.
T [8C]
Conv. (Yield) [%]
E/Z[b]
=
P O
1[c]
2
3
4
5
6
7
8
9
10
11
12
13
1b
1a
3a
3b
3c
1a
1b
1c
1d
3a
3b
3c
3d
A
A
A
A
A
B
B
B
B
B
B
B
B
110
110
110
110
110
140
140
140
140
140
140
140
140
100 (81)
100 (80)
100 (82)
100 (80)
100 (82)
55 (37)
65 (43)
88 (61)
91 (72)
42
66:34
80:20
95:5
95:5
95:5
75:25
80:20
75:25
82:18
94:6
46
71
84 (79)
95:5
95:5
94:6
[a] Benzaldehyde (1.2 mmol), benzyl bromide (1.0 mmol), phosphine
oxide 1 or 3 (0.1 mmol), Ph2SiH2 (1.2 mmol), A) 2 (2.0 mmol), 1.0m in
toluene, or B) DIPEA (1.2 mmol), 3.0m in toluene. [b] E/Z ratio
determined by 1H NMR spectroscopy of the unpurified reaction mixture.
[c] Using Na2CO3 (1.5 mmol), the yield was 21%, and using NaOtBu
(1.5 mmol) or DIPEA (1.2 mmol), only a trace of product was observed.
Although the CWR had been extended fully to semi-
stabilized ylides using two protocols, the diastereoselectivity
effected by the 1 series of precatalysts was poor, typically
around 66:34, E/Z. Moreover, varying the acyclic substituent
on phosphorus provided only slight improvement in E/Z
selectivity (Table 1, entries 1 and 2). Further improvement of
diastereocontrol would require a new cyclic core structure of
the precatalyst. An analysis of the reports on Wittig reactions
showed that Vedejs and co-workers accomplished highly E-
selective Wittig reactions by employing a bicyclic phosphine,
reduced 3b.[14] Crucially, use of 3b would couple the required
five-membered cyclic structure, which is vital to ensure
a sufficient rate of phosphine oxide reduction, with a steric
shield that should provide high E selectivity.[4a,14] DFT
calculations also predicted that precatalysts of type 3 would
afford a higher E/Z selectivity than series 1. For example, the
computed E selectivity for the formation of stilbene (4) from
3ba is 3.0 kcalmolÀ1 (DDG°), compared to 1.7 kcalmolÀ1
from 1ba (see the Supporting Information).
With a new catalyst structure identified, another precata-
lyst library (Figure 2, 3a–d) was prepared and tested
(Table 1). All type-3 precatalysts produced 4 with high
E selectivity, while essentially mirroring the yields obtained
with series 1. Following optimization of both deprotonation
and diastereoselectivity, we embarked on a substrate study
employing both 2 and DIPEA as base (Scheme 2). The results
demonstrated that the CWR utilizing semistabilized ylides
could tolerate a variety of aryl, heteroaryl, and aliphatic
aldehydes and organohalides. Notable results are the syn-
thesis of resveratrol analogues 10 and 17, which have been
demonstrated to have more potent anticancer activity than
resveratrol.[15] The protocol performed well on a multigram
Scheme 2. Substrate study utilizing semistabilized ylides. For each
product the compound number, yield of isolated product, E/Z ratio
(determined by 1H NMR spectroscopy of the unpurified reaction
mixture), halide, and reaction conditions are given (A: 2 (2.0 mmol),
1108C, 1.0m in toluene; B: DIPEA (1.2 mmol), 1408C, 3.0m in
toluene). The reactions were performed in duplicate; see the Support-
ing Information for details. [a] On a 25 mmol scale, the yield was 77%
(5.8 g). Ts=4-toluenesulfonyl.
scale, as 5.8 g of 10 (77% yield) was prepared with 25 mmol of
the organohalide using 1d and DIPEA. However, no differ-
ence in diastereoselectivity was observed when switching
from type-1 to type-3 precatalysts for secondary halides
(Scheme 2, compound 9).
To further extend the scope of the CWR, the use of
nonstabilized ylides was investigated. Anticipating a slow rate
of phosphonium salt formation, we used organoiodides rather
than organobromides. Pleasingly, the combination of masked
base 2 and ylide tuning, 3a–d, allowed the CWR to be applied
to this ylide class with moderate E selectivity (Table 2). As
a result of the slow rate of phosphonium salt formation, the
reaction time was extended to 48 h for maximum yield.
Interestingly, 3c and not 3d turned out to be the optimum
precatalyst (Table 2). This highlights the importance of
Angew. Chem. Int. Ed. 2014, 53, 12907 –12911
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12909

.
Angewandte
Communications
Table 2: Optimization of CWR conditions for the synthesis of 22.[a]
the full employment of semistabilized ylides. Notably, cata-
lytic use of phosphorus species offers a strategic advantage,
the ability to tailor the electronic and steric structure of the
catalyst. This was highlighted with the concept of ylide tuning;
addition of EWGs to the precatalyst allowed semistabilized
ylides to be formed using the mild base DIPEA. A
combination of ylide tuning and 2 allowed nonstabilized
ylides to be employed in the CWR for the first time. An
increase in diastereoselectivity was achieved through modi-
fication of the precatalyst core structure, attaining an E/Z
selectivity of up to > 95:5 for semistabilized, and 75:25 for
nonstabilized ylide types.
E/Z[b]
=
P O
Entry
t [h]
Conv. (Yield) [%]
1
2
3a
3b
3c
3c
3d
48
48
48
24
48
100 (64)
88 (70)
100 (95)
88 (74)
75:25
75:25
75:25
75:25
75:25
3[c]
4
5
100 (83)
[a] 4-Chlorobenzaldehyde (1.2 mmol), (2-iodoethyl)benzene (1.0 mmol),
phosphine oxide 1 or 3 (0.2 mmol), Ph2SiH2 (1.2 mmol), 2 (2.0 mmol
added at 0 h, additional 1.5 mmol added at 24 h for 48 h reactions),
Received: June 11, 2014
Published online: September 22, 2014
1
0.7m in toluene, 1408C. [b] E/Z ratio determined by H NMR spectros-
copy of the unpurified reaction mixture. [c] Using (2-bromoethyl)ben-
zene, the yield was 76%.
Keywords: homogeneous catalysis · olefination · phosphorus ·
Wittig reaction · ylide
.
finding a balance between lowering the pKa of the phospho-
nium salt, while maintaining a sufficient rate of phosphonium
salt formation. Though organoiodides proved optimal, bro-
mides could be employed with a 20% drop in yield (Table 2,
entry 3). A substrate study (Scheme 3) showed that aryl,
heteroaryl, and aliphatic aldehydes could be employed with
a variety of organohalides. Significant results were the
synthesis of 25, 26, and 27 that highlight both the diversity
of potential products and the relatively mild nature of the
protocol as the TBS group was retained. Furthermore,
terminal olefination was also viable, shown by the synthesis
of 29, 30, and 31.
[1] a) G. Wittig, G. Geissler, Justus Liebigs Ann. Chem. 1953, 580,
44 – 57; b) G. Wittig, U. Schollkopf, Chem. Ber. 1954, 87, 1318 –
1330.
[2] a) D. Edmonds, A. Abell in Modern Carbonyl Olefination (Ed.:
T. Takeda), Wiley-VCH, Weinheim, 2004, pp. 1 – 17; b) A. Abell,
D. M. K. Edmonds in Organophosphorus Reagents (Ed.: P. J.
Murphy), Oxford University Press, Oxford, 2004, pp. 99 – 127;
c) O. I. Kolodiazhnyi, Phosphorus Ylides: Chemistry and Appli-
cation in Organic Chemistry, Wiley-VCH, New York, 1999;
d) K. C. Nicolaou, M. W. Hꢀrter, J. L. Gunzner, A. Nadin,
Liebigs Ann. Recl. 1997, 1283 – 1301; e) N. J. Lawrence in
Preparation of Alkenes: A Practical Approach (Ed.: J. M. J.
Williams), Oxford University Press, Oxford, 1995, pp. 55 – 64;
f) B. E. Maryanoff, A. B. Reitz, Chem. Rev. 1989, 89, 863 – 927;
g) for a review of asymmetric Wittig-type olefination reactions,
see: T. Rein, T. M. Pedersen, Synthesis 2002, 579 – 584.
In conclusion, the first catalytic Wittig reactions using
a wide range of semistabilized and nonstabilized ylides have
been shown. The use of 2, a masked form of NaOtBu, enabled
[3] For discussions on the mechanisms and selectivity, see: a) P. A.
Byrne, D. G. Gilheany, Chem. Soc. Rev. 2013, 42, 6670 – 6696;
b) P. A. Byrne, D. G. Gilheany, J. Am. Chem. Soc. 2012, 134,
9225 – 9239; c) R. Robiette, J. Richardson, V. K. Aggarwal, J. N.
Harvey, J. Am. Chem. Soc. 2006, 128, 2394 – 2409; d) R.
Robiette, J. Richardson, V. K. Aggarwal, J. N. Harvey, J. Am.
Chem. Soc. 2005, 127, 13468 – 13469; e) V. K. Aggarwal, J. R.
Fulton, C. G. Sheldon, J. de Vicente, J. Am. Chem. Soc. 2003,
125, 6034 – 6035; f) E. Vedejs, M. J. Peterson in Advances in
Carbanion Chemistry, Vol. 2 (Ed.: V. Snieckus), JAI, Greenwich,
1996, pp. 1 – 85; g) E. Vedejs, M. J. Peterson, Top. Stereochem.
1994, 21, 1 – 157.
[4] a) C. J. OꢁBrien, J. L. Tellez, Z. S. Nixon, L. J. Kang, A. L.
Carter, S. R. Kunkel, K. C. Przeworski, G. A. Chass, Angew.
Chem. Int. Ed. 2009, 48, 6836 – 6839; Angew. Chem. 2009, 121,
6968 – 6971; b) C. J. OꢁBrien, Z. S. Nixon, A. J. Holohan, S. R.
Kunkel, J. L. Tellez, B. J. Doonan, E. E. Coyle, F. Lavigne, L. J.
Kang, K. C. Przeworski, Chem. Eur. J. 2013, 19, 15281 – 15289;
c) C. J. OꢁBrien, F. Lavigne, E. E. Coyle, A. J. Holohan, B. J.
Doonan, Chem. Eur. J. 2013, 19, 5854 – 5858. d) While two
stilbene examples were reported in Ref. [4], the benzyl bromides
used were substituted with EWGs. The pKa of benzyl phospho-
nium salts substituted with EWGs are significantly lower than
the nonsubstituted analogue, Ref. [7]. Thus, selective deproto-
nation of benzyl phosphonium salts containing EWGs is notably
easier to achieve than with the actual benzyl system.
Scheme 3. Substrate study using nonstabilized ylides. For each prod-
uct the compound number, yield of isolated product, E/Z ratio
(determined by 1H NMR spectroscopy of the unpurified reaction
mixture), phosphine oxide, halide, and reaction time are given. The
reactions were performed in duplicate; see the Supporting Information
for details. N/A=not applicable, TBS=tert-butyldimethylsilyl.
[5] H. A. van Kalkeren, A. L. Blom, F. P. J. T. Rutjes, M. A. J.
Huijbregts, Green Chem. 2013, 15, 1255 – 1263.
12910 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 12907 –12911

Angewandte
Chemie
[6] a) X. Tang, C. Chapman, M. Whiting, R. M. Denton, Chem.
Commun. 2014, 50, 7340 – 7343; b) X. Tang, J. An, R. M. Denton,
Tetrahedron Lett. 2014, 55, 799 – 802; c) H. A. van Kalkeren, C.
te Grotenhuis, F. S. Haasjes, C. A. Hommersom, F. P. J. T.
Rutjes, F. L. van Delft, Eur. J. Org. Chem. 2013, 7059 – 7066;
d) J. An, X. Tang, J. Moore, W. Lewis, R. M. Denton, Tetrahe-
dron 2013, 69, 8769 – 8776; e) A. D. Kosal, E. E. Wilson, B. L.
Ashfeld, Angew. Chem. Int. Ed. 2012, 51, 12036 – 12040; Angew.
Chem. 2012, 124, 12202 – 12206; f) N. L. Dunn, M. Ha, A. T.
Radosevich, J. Am. Chem. Soc. 2012, 134, 11330 – 11333;
g) H. A. van Kalkeren, J. J. Bruins, F. P. J. T. Rutjes, F. L. van
Delft, Adv. Synth. Catal. 2012, 354, 1417 – 1421; h) H. A.
van Kalkeren, S. H. A. M. Leenders, C. R. A. Hommerson,
F. P. J. T. Rutjes, F. L. van Delft, Chem. Eur. J. 2011, 17, 11290 –
11295; i) R. M. Denton, J. An, B. Adeniran, A. J. Blake, W.
Lewis, A. M. Poulton, J. Org. Chem. 2011, 76, 6749 – 6767; j) J. R.
Harris, M. T. Haynes, A. M. Thomas, K. A. Woerpel, J. Org.
Chem. 2010, 75, 5083 – 5091; k) S. P. Marsden, A. E. McGonagle,
B. McKever-Abbas, Org. Lett. 2008, 10, 2589 – 2591; l) for
a review on catalytic phosphorus(V)-mediated reactions, see:
S. P. Marsden, in Sustainable Catalysis: Challenges and Practices
for the Pharmaceutical and Fine Chemical Industries (Eds.: P. J.
Dunn, K. K. Hii, M. J. Krische, M. T. Williams), Wiley, New
York, 2013, pp. 339 – 362.
[9] We found electron-rich phosphine oxides reduced faster than
electron-poor precatalysts; these results will be reported in due
course.
[10] J.-P. Cheng, B. Liu, Y. Zhao, Y. Sun, X.-M. Zhang, Y. Lu, J. Org.
Chem. 1999, 64, 604 – 610.
[11] Computations were performed in Gaussian09: M. J. Frisch,
et al., Gaussian09, RevisionD.01, Gaussian, Inc., Wallingford
CT, 2013. Solvation energies in DMSO and toluene were
computed with the SMD model. The nOct group of 1a was
modeled as Me in 1aa. Full details and references are provided
in the Supporting Information.
[12] Ion-pairing effects significantly promoted deprotonation. For
example, the computed DG for deprotonation of 1ba decreases
by 11 kcalmolÀ1 when a BrÀ counterion is included.
[13] To our knowledge, the mildest base used for traditional Wittig
reactions of semistabilized and nonstabilized ylides is 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU; pKa (in H2O) ca. 12, (in
acetonitrile) 24), see: K. Okuma, O. Sakai, K. Shioji, Bull. Chem.
Soc. Jpn. 2003, 76, 1675 – 1676. However, DBU reacts with
organohalides (methyl iodide and benzyl bromide) and is
incompatible with the CWR. For use of guanidine bases (pKa
(acetonitrile) 22 – 26), see: D. Simoni, M. Rossi, R. Rondanin, A.
Mazzali, R. Baruchello, M. Malagutti, F. P. Invidiata, Org. Lett.
2000, 2, 3765 – 3768.
[7] Y. Fu, H.-J. Wang, S.-S. Chong, Q.-X. Guo, L. J. Liu, J. Org.
Chem. 2009, 74, 810 – 819.
[8] For a similar approach, see: S. Xu, S. Zhu, J. Shang, J. Zhang, Y.
Tang, J. Dou, J. Org. Chem. 2014, 79, 3696 – 3703.
[14] a) E. Vedejs, M. J. Peterson, J. Org. Chem. 1983, 48, 6509 – 6512;
b) E. Vedejs, J. Cabaj, M. J. Peterson, J. Org. Chem. 1983, 48,
1985 – 1986.
[15] S. Fulda, Drug Discovery Today 2010, 15, 757 – 765.
Angew. Chem. Int. Ed. 2014, 53, 12907 –12911
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12911