Enantioselective Catalysts for the Synthesis of α-Substituted Allylboronates - An Accelerated Approach towards Isomerically Pure Homoallylic Alcohols

Article abstract of DOI:10.1002/anie.201509198

The use of a convenient protecting group for boronates allows a selective, catalyzed S_N2′ reaction to generate allylboronates which are applied for the synthesis of enantiomerically pure homoallylic alcohols. Depending on the configuration of both catalyst and the protecting group any of the four possible stereoisomers can be formed. The rationale behind the selective addition is supported by density functional theory calculations.

Full text of DOI:10.1002/anie.201509198

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201509198
German Edition: DOI: 10.1002/ange.201509198
Asymmetric Synthesis
Enantioselective Catalysts for the Synthesis of a-Substituted
Allylboronates—An Accelerated Approach towards Isomerically Pure
Homoallylic Alcohols
Marcus Brauns, FrØdØric Muller, Daniel Gülden, Dietrich Bçse, Wolfgang Frey, Martin Breugst,*
and Jçrg Pietruszka*
Dedicated to Professor Steven V. Ley on the occasion of his 70th birthday
Abstract: The use of a convenient protecting group for
boronates allows a selective, catalyzed SN2’ reaction to generate
allylboronates which are applied for the synthesis of enantio-
merically pure homoallylic alcohols. Depending on the con-
figuration of both catalyst and the protecting group any of the
four possible stereoisomers can be formed. The rationale
behind the selective addition is supported by density functional
theory calculations.
Z
-configured homoallylic alcohols are often found as
structural motifs in natural products, for example in oxy-
lipines such as solandelactone A + B (1) and neohalicholac-
tone (2) and related compounds of marine origin
(Scheme 1).[1,2] Many syntheses rely on the same key reaction,
the Nozaki–Hiyama–Kishi reaction[3] starting with vinyl
iodides 3 (3a[4]/3b[5]). The present study led to a general
approach towards intermediate 4.[4b, 5b] Central to the success
was the fast and selective allylation of aldehyde 5 with
a readily available allylboronate 6.[6]
Scheme 1. Allylboronates in the synthesis of marine oxylipin side
chains.
The starting point of the endeavor was the known eight-
step synthesis of diol 7 based boronates 6c/6d from prop-
argylic alcohol (8) via mesylate 9 (Scheme 2).[4f, 7] The
stereogenic center in a-position to the boron moiety stems
from a (3,3)-sigmatropic rearrangement.[8] While the route
was applied in a number of syntheses, the overall yields are
only moderate and a shorter alternative would be more
[*] M. Sc. M. Brauns, Dr. D. Bçse, Prof. Dr. J. Pietruszka
Institut für Bioorganische Chemie der Heinrich-Heine-Universität
Düsseldorf im Forschungszentrum Jülich
Scheme 2. Comparison of approaches towards boronates 6c/6d.
Stetternicher Forst, Geb. 15.8, 52426 Jülich (Germany)
E-mail: j.pietruszka@fz-juelich.de
Homepage: http://www.iboc.uni-duesseldorf.de
attractive, especially if it would circumvent the sophisticated
separation of diastereoisomers. Since boronic acid 10 is
commercially available, boronate 11 is readily accessible in
one step in high yield upon condensation (96%). It is an ideal
substrate for SN2’ reactions directly leading to allylboronates
6c/6d.
F. Muller, Dr. M. Breugst
Department für Chemie, Universität zu Kçln
Greinstrasse 4, 50939 Kçln (Germany)
E-mail: mbreugst@uni-koeln.de
Homepage: http://physorg.uni-koeln.de
D. Gülden, Prof. Dr. J. Pietruszka
Suitable parameters have been screened following various
references (see the Supporting Information), but best results
were obtained using the conditions promoted by Carosi and
Hall:[9] The combination of copper thiophene carbonate
(CuTc) and phosphoramidite ligand 12 led to high yields
(82–92%) and good diastereo- (> 95:5) and regioselectivities
(> 20:1); formation of a-product is negligible. The general
Institut für Bio- und Geowissenschaften (IBG-1: Biotechnologie)
Forschungszentrum Jülich, 52428 Jülich (Germany)
Dr. W. Frey
Institut für Organische Chemie, Universität Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201509198.
1548
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1548 –1552

Angewandte
Communications
Chemie
applicability of this short and efficient method has been
demonstrated for various a-alkyl-substituted allylboronates
with different chain lengths (albeit not R1 = Me; see the
Supporting Information).
Despite the dramatic improvement of the sequence, there
are still drawbacks that need addressing: a) The synthesis of
diol 7 is step intensive;[6f] b) The reaction rates for allyl
additions are relatively low and in some cases reaction times
> 7 days are necessary; c) Both g-products 6 and dia-6 form
predominantly the Z-homoallylic alcohols, while the corre-
sponding E compounds cannot be obtained. Tartrate 13/14
based bisboronate 15 might be an alternative (Scheme 3).[10]
Scheme 4. Selective synthesis of all stereoisomeric homoallyl alcohols.
reagents provided products within minutes. Surprisingly, even
at low temperatures down to À508C complete conversion was
observed within hours. All products 18/19 were obtained in
essentially enantiomerically pure form; the configuration of
the stereogenic center was dictated by the tetraol configu-
ration. Furthermore, we were surprised to observe that
reagents 16 and ent-16 provided in high selectivity (E/Z =
14:1) the E-configured homoallylic alcohols 18 and ent-18,
and not the expected Z compounds, in analogy to boronates
17 and ent-17 forming 19 and ent-19, respectively.[12] Appa-
rently, the relative configuration—side chain to auxiliary—is
the dominant factor for the double-bond configuration. The
reaction is not limited to benzaldehyde, but could be extended
also giving comparable yields (79–91%) and selectivity for E-
(20–23) and Z-products (24–27).
Two questions arise based on these findings: a) How can
the relative acceleration between the two related types of
reagents (dia-6c vs. ent-17a) be explained, and b) what is the
rationale behind the unexpected E-selectivity for 16a? One
aspect could be the relatively close distance (3.1 Š) between
the boron atom and the secondary oxygen atom which might
activate the allylboronate functionality. The hypothesis is
supported by findings of Morken as well as Hall and co-
workers, who demonstrated the possibility of enhancing
reactivity by intra- and intermolecular Lewis acid activa-
tion.[13]
Scheme 3. Tetraol-based synthesis of allylboronates 16 and 17 in three
steps.
Zhou and Shan[10a] reported on the selective condensation
of tetraol 14 and phenylboronic acid and we demonstrated
that the corresponding reaction of tetraol 14 with boronic acid
10 is feasible in high yield (97%), and in particular that the
established substitution conditions are also applicable (R1 =
Et, n-Pe): High yields (16/17: 89–93%) and selectivities were
observed for all eight cases despite the fact that two boron
moieties were present! Depending on the configuration of
ligand 12 vs. ent-12 and the tetraol 14 vs. ent-14 any of the
symmetric reagents are obtainable with diastereomeric ratios
> 95:5 for 16 and ent-16 and > 91:9 for 17 and ent-17. The
stereochemical outcome of this three-step sequence from
methyl tartrate was established by X-ray structure analyses of
bisboronates 16a and 17a.[11]
Having all four stereoisomeric allylboronates 16a and 17a
(and their enantiomers) in our hands, allyl additions were
tested utilizing benzaldehyde as the electrophile (Scheme 4).
While diol-based boronate 6c requires long reaction times of
several days, it was noted that even in dilute solution the new
To answer these questions, we analyzed the reaction as
depicted in Scheme 5 by using DFT calculations [M06-2X-
Angew. Chem. Int. Ed. 2016, 55, 1548 –1552
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1549

Angewandte
Communications
Chemie
Scheme 5. Transition states and intermediates of allylations of benzal-
dehyde.
D3/def2-TZVPP/IEFPCM(CH2Cl2)//M06-L-D3/6-31+G-
(d,p)].[14] As similar results were obtained for the reactions of
ent-17a in CH2Cl2 and in n-pentane solutions, we will discuss
only the values obtained in dichloromethane for the sake of
clarity (see the Supporting Information for details). The free
energies for the transformations are highly exergonic
(À31.3 < DG < À26.9 kcalmolÀ1). Independent of the
employed allylboronate, they all fall within a similar range,
with the tetraol-based systems 16a and ent-17a being slightly
more exergonic. Based on the calculated high exergonicities
and the low reaction temperatures we have to conclude that
the selectivities arise from differences in activation energies
(i.e., kinetic control) in line with previous investigations.[14]
Consequently, we have calculated transition states—that are
commonly depicted as TSA and TSB—for the formation of all
intermediates IA and IB (and their enantiomers) obtained
from benzaldehyde and allylboronates dia-6c, 16a, and ent-
17a. The calculated activation free energies are summarized
in Table 1 and selected transition-state structures are depicted
in Figure 1 (for others, see the Supporting Information). In all
cases, chair-like transition states are significantly lower in
energy than their corresponding boat-like structures, in line
with the accepted mechanistic model.[6c,d,15]
Figure 1. Calculated transition-state structures TS2 and TS3 with
selected bond lengths (in Š) and side views highlighting the dispersive
stabilization of TS3. For clarity, the hydrogens on the phenyl rings of
the boronates are not shown.
the overall dipole moment and to avoid unfavorable steric
interactions with the dioxaborolane system. The findings do
correlate very well with the experimental results (ent-19 was
formed from dia-6c in 83% yield and > 99% ee[7]).
A similar orientation (phenyl equatorial, ethyl axial) was
also found in the lowest-energy transition state TS3 for
boronate ent-17a (Figure 1) resulting also in the formation of
ent-19. In contrast, the lowest-energy transition state TS2 for
the reaction of the diastereomeric boronate 16a with
benzaldehyde features both the phenyl and the ethyl sub-
stituent in an equatorial position (Figure 1) yielding 18 as the
kinetically preferred product. For both cases 16a and ent-17a,
the computed E/Z selectivity is in very good qualitative
agreement with the experimental observations (Scheme 4).
Furthermore, the calculations also correctly predict the
tetraol-based systems to be significantly more reactive than
the diol-based analogue (20 min at room temperature in
dilute solution for ent-17a versus 2 days in concentrated
solution for dia-6c[7]).
For the diol-based boronate dia-6c, the lowest-energy
transition state TS1 leads (after hydrolysis of IA) to the
product ent-19, while transition states for the other stereo-
isomers are significantly higher in energy (Table 1). In this
transition-state structure, the ethyl substituent is located in an
axial position (see the Supporting Information) to minimize
Table 1: Calculated activation free energies (in kcalmolÀ1) for the
reactions of benzaldehyde and allylboronates dia-6c, 16a, and ent-17a.
To rationalize the observed reactivities and stereoselec-
tivities, we initially focused on putative boron–oxygen
stabilization in the bisboronate series as outlined above. In
Boronate
DG°(R,E)
DG°(S,E)
DG°(R,Z)
DG°(S,Z)
À
fact, the transannular B O bond lengths (green dashes in
Figure 1) are significantly shorter than the sum of the van der
Waals radii (3.41 Š)[16] in both TS2 and TS3, indicating an
attractive interaction between these atoms. Based on natural
bond orbital analyses,[17] we found a binding interaction (ca.
2 kcalmolÀ1) between the “uninvolved” boron and the oxygen
atom next to the reaction center for TS2. However, this
interaction was not observed for TS3. Instead, an interaction
(ca. 5 kcalmolÀ1) between the carbonyl oxygen of benzalde-
hyde and the “uninvolved” boron atom was found, which is
+19.6
(TS1)
+22.8
+29.3
+26.1
+13.8
(TS2)
+20.3
+19.7
+23.2
+20.6
+15.1
À
also reflected in an even shorter B O bond. While these
additional B–O interactions rationalize the higher reactivities
of 16a/ent-17a over dia-6c, they are not responsible for the
change in selectivity observed for the bisboronates, as these
interactions are present in most transition state conformers of
+15.3
(TS3)
+15.8
1550
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1548 –1552

Angewandte
Communications
Chemie
TS2 and TS3. A closer examination of both structures
[1] a) Y. Seo, K. W. Cho, J.-R. Rho, J. Shin, Tetrahedron 1996, 52,
(Figure 1, right) suggests that the different selectivity of ent-
17a is rather caused by a different orientation of the
bisboronate. While there are no significant additional inter-
actions between the boronate 16a and the aldehyde in TS2,
the nonreacting pentenyl substituent of ent-17a interacts with
the aldehyde in TS3 by additional dispersive effects (e.g., C–
H–p interactions).[18] These stabilizing effects are also verified
by an NCIPLOTanalysis (see the Supporting Information for
details).[19]
In summary, computational investigations confirm the
experimentally observed stereoselectivities for the reactions
of the allyl boronates dia-6c, 16a, and ent-17a with benzal-
dehyde. The higher reactivities of the bisboronates can be
attributed to favorable B–O interactions. The high selectiv-
ities of these systems primarily arise from steric interactions
with the phenyl groups of the adjacent dioxaborinane (see the
Supporting Information), while the change in selectivity (16a
vs. ent-17a) is caused by additional dispersive interactions in
a tweezer-like geometry in the case of the latter. Finally, after
having a) established a short and convenient approach
towards the new reagents 16 and 17 and b) provided a rational
for the unexpected high reactivity, we applied the method in
the syntheses of the side chains of the marine oxylipins 1 and
2. Again, the homoallylic alcohols (here 4a from 17a and 4b
from 17b) were obtained in good yield and selectivity, thus
considerable shortening the sequence towards the natural
products (Scheme 6). The flexibility of the approach should
10583 – 10596; b) H. Niwa, K. Wakamatsu, K. Yamada, Tetrahe-
dron Lett. 1989, 30, 4543 – 4546; c) P. J. Proteau, J. V. Rossi,
W. H. Gerwick, J. Nat. Prod. 1994, 57, 1717 – 1719.
[2] Review: J. D. White, J. Yang, Synlett 2009, 1713 – 1729.
[3] A. Fürstner, Chem. Rev. 1999, 99, 991 – 1045, and references
therein.
[4] a) D. J. Critcher, S. Connolly, M. Wills, Tetrahedron Lett. 1995,
36, 3763 – 3766; b) M. Bischop, V. Doum, A. C. M. Nordschild
nØe Rieche, J. Pietruszka, D. Sandkuhl, Synthesis 2010, 527 – 537.
[5] a) J. D. White, M. S. Jensen, J. Am. Chem. Soc. 1995, 117, 6224 –
6233; b) C. Barloy-Da Silva, A. Benkouider, P. Pale, Tetrahedron
Lett. 2000, 41, 3077 – 3081; c) J. Yu, J.-Y. Lai, J. Ye, N. Balu, L. M.
Reddy, W. Duan, E. R. Fogel, J. H. Capdevila, J. R. Falck,
Tetrahedron Lett. 2002, 43, 3939 – 3941; d) J. Pietruszka, T.
Wilhelm, Synlett 2003, 1698 – 1700; e) J. D. White, W. H. C.
Martin, C. Lincoln, J. Yang, Org. Lett. 2007, 9, 3481 – 3483; f) J.
Pietruszka, A. C. M. Rieche, Adv. Synth. Catal. 2008, 350, 1407 –
1412; g) J. D. White, C. M. Lincoln, J. Yang, W. H. C. Martin,
D. B. Chan, J. Org. Chem. 2008, 73, 4139 – 4150; h) N. C.
Eichenauer, A. C. M. Nordschild, M. Bischop, D. Schumacher,
M. K. W. Mackwitz, R. Tschersich, T. Wilhelm, J. Pietruszka,
Eur. J. Org. Chem. 2015, 5620 – 5632.
[6] a) Modern Carbonyl Chemistry (Ed.: J. Otera), Wiley-VCH,
Weinheim, 2000; b) Y. Yamamoto, N. Asao, Chem. Rev. 1993, 93,
2207 – 2293; c) R. W. Hoffmann, Pure Appl. Chem. 1988, 60,
123 – 130; d) Boronic Acids (Ed.: D. G. Hall), Wiley-VCH,
Weinheim, 2005; e) H. Lachance, D. G. Hall, Allylboration of
Carbonyl Compounds. Organic Reactions (Ed.: S. E. Denmark),
Wiley, New York, 2008, pp. 1 – 573; f) C. A. Berg, N. C. Eiche-
nauer, J. Pietruszka, Pure Appl. Chem. 2012, 84, 2339 – 2416;
selected examples: g) M. Althaus, A. Mahmood, J. R. Suµrez,
S. P. Thomas, V. K. Aggarwal, J. Am. Chem. Soc. 2010, 132,
4025 – 4028; h) J. L. Y. Chen, H. K. Scott, M. J. Hesse, C. L.
Willis, V. K. Aggarwal, J. Am. Chem. Soc. 2013, 135, 5316 – 5319;
i) S. Touchet, A. Mace, T. Roisnel, F. Carreaux, A. Bouillon, B.
Carboni, Org. Lett. 2013, 15, 2712 – 2715; j) T. S. N. Zhao, Y.
Yang, T. Lessing, K. J. Szabó, J. Am. Chem. Soc. 2014, 136, 7563 –
7566; k) D. Bçse, P. Niesobski, M. Lübcke, J. Pietruszka, J. Org.
Chem. 2014, 79, 4699 – 4703.
Scheme 6. Synthesis of key intermediates 4a/4b for marine oxylipins.
[7] J. Pietruszka, N. Schçne, Synthesis 2006, 24 – 30.
[8] a) J. Pietruszka, N. Schçne, Angew. Chem. Int. Ed. 2003, 42,
5638 – 5641; Angew. Chem. 2003, 115, 5796 – 5799; b) J. Pie-
truszka, N. Schçne, W. Frey, L. Grundl, Chem. Eur. J. 2008, 14,
5178 – 5197.
[9] L. Carosi, D. G. Hall, Angew. Chem. Int. Ed. 2007, 46, 5913 –
5915; Angew. Chem. 2007, 119, 6017 – 6019.
[10] a) Y. Zhou, Z. Shan, Tetrahedron Lett. 2007, 48, 3531 – 3534;
review on TADDOLs: b) D. Seebach, A. K. Beck, A. Heckel,
Angew. Chem. Int. Ed. 2001, 40, 92 – 138; Angew. Chem. 2001,
113, 96 – 142.
[11] CCDC 1428548 and 1428549 contain the supplementary crys-
tallographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre.
[12] The configuration was confirmed by comparison of optical
rotation with literature values.[7,8b,9] The enantio- and diastereo-
selectivity was determined by HPLC or GC analysis (see the
Supporting Information). The E preference is independent of
the temperature and solvent used.
render it versatile for a considerable number of further
applications. A general method providing access to all four
possible isomeric homoallylic alcohols has been presented. It
is worth noting that the allylation reagents were synthesized
in only three steps from dimethyl tartrate.
Acknowledgments
We thank the Deutsche Forschungsgemeinschaft DFG and
the Fonds der Chemischen Industrie (Liebig-Scholarship to
M.Bre. and scholarship for D.B.) for their generous support of
our projects. This work used the Cologne High Efficiency
Operating Platform for Sciences.
[13] a) S. N. Mlynarski, C. H. Schuster, J. P. Morken, Nature 2014,
505, 386 – 390; b) V. Rauniyar, D. G. Hall, J. Am. Chem. Soc.
2004, 126, 4518 – 4519; c) H. Lachance, X. Lu, M. Gravel, D. G.
Hall, J. Am. Chem. Soc. 2003, 125, 10160 – 10161.
[14] a) Y. Zhao, D. G. Truhlar, J. Chem. Phys. 2006, 125, 194101; b) Y.
Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215 – 241; c) F.
Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297 –
Keywords: allylation · asymmetric catalysis ·
asymmetric synthesis · boron · computational chemistry
How to cite: Angew. Chem. Int. Ed. 2016, 55, 1548–1552
Angew. Chem. 2016, 128, 1574–1578
Angew. Chem. Int. Ed. 2016, 55, 1548 –1552
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1551

Angewandte
Communications
Chemie
3305; d) E. Canc›s, B. Mennucci, J. Tomasi, J. Chem. Phys. 1997,
107, 3032 – 3041.
[18] J. P. Wagner, P. R. Schreiner, Angew. Chem. Int. Ed. 2015, 54,
12274 – 12296; Angew. Chem. 2015, 127, 12446 – 12471.
[15] a) S. E. Denmark, E. J. Weber, Helv. Chim. Acta 1983, 66, 1655 –
1660; b) B. W. Gung, X. Xue, W. R. Roush, J. Am. Chem. Soc.
2002, 124, 10692 – 10697.
[19] E. R. Johnson, S. Keinan, P. Mori-Sµnchez, J. Contreras-García,
A. J. Cohen, W. Yang, J. Am. Chem. Soc. 2010, 132, 6498 – 6506.
[16] S. Alvarez, Dalton Trans. 2013, 42, 8617 – 8636.
[17] J. E. D. Glendening, K. Badenhoop, A. E. Reed, J. E. Carpenter,
J. A. Bohmann, C. M. Morales, C. R. Landis, F. Weinhold,
NBO6.0, Theoretical Chemistry Institute, University of Wis-
consin, Madison, 2013.
Received: October 2, 2015
Published online: December 14, 2015
1552
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1548 –1552