Stereoselective Nucleophilic Additions to
Aldehydes and Development of New
Methodology in Organic Synthesis
Tessie Borg
Doctoral Thesis
Stockholm 2013
Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan i Stockholm framlägges till offentlig granskning för avläggande av teknologie doktorsexamen i kemi med inriktning mot organisk kemi onsdagen den 8 maj kl 10.00 i kollegiesalen, KTH, Brinellvägen 8, Stockholm. Avhandlingen försvaras på engelska. Opponent är Professor Siméon Arseniyadis, Centre de Recherches de Gif, France.
ISBN 978-91-7501-690-0
ISSN 1654-1081
TRITA-CHE-Report 2013:8
© Tessie Borg, 2013
Tessie Borg 2013: “Stereoselective Nucleophilic Additions to Aldehydes and Development of New Methodology in Organic Synthesis”, KTH Chemical Science and Engineering, Royal Institute of Technology, SE-100 44 Stockholm, Sweden.
Abstract
This thesis is divided into four separate parts with nucleophilic addition to aldehydes as the common feature in three of them.
The first part deals with the investigation of the stereochemical induction and elucidation of the factors that dictate the π-facial selectivities in Mukaiyama aldol addition to α- and α,β-heteroatom substituted aldehydes. An explanation for the unexpected shift from 1,2-anti to 1,2-syn selectivity seen in the reaction when applying nucleophiles of different sizes in the addition to α-chloro aldehydes is offered.
The next two parts describes the addition of 1,3-bis(silyl)propenes and C3 substituted 1,3-bis(silyl)propenes to aldehydes and the development of two highly stereoselective new methodologies for the construction of 1,3-dienes and 2,3,4,5-tetrasubstituted tetrahydrofuranes, respectively.
The last part describes the attempts made towards the total synthesis of (±)-aspidophylline A, where the intention was to apply a domino carbopalladation-carbonylation reaction as the key step in the synthetic route.
Keywords: Stereoselective synthesis, Mukaiyama aldol reaction, Induction
models, Polar Felkin-Anh, Cornforth-Evans, Sakurai allylation, Allylsilane, 1,3-Diene, Tetrahydrofuran, Natural product, (±)-aspidophylline A.
Abbreviations
Ald. Aldehyde 9-BBN 9-borabicyclo[3.3.1]nonane Bn benzyl cHex cyclohexyl dba dibenzylideneaceton DIPEA N,N-diisopropylethylamine DMAc dimethylacetamide DMF dimethylformamide DMPU N,N'-dimethyl-N,N'-trimethyleneurea Dppb 1,4-bis(diphenylphosphino)butane FA Felkin-Anh iPr isopropyl LA Lewis acidLDA Lithium diisopropylamide
Me methyl
MS molecular sieves
NaHMDS Sodium bis(trimethylsilyl)amide
NMO 4-methylmorpholine-N-oxide
Nu nucleophile
PFA polar Felkin-Anh
Pg protecting group
Ph phenyl
PhMe toluene
PPTs pyridinium p-toluenesulfonate
TBAF tetra-n-butylammonium fluoride
TBDPS tertbutyldiphenylsilyl TBS tertbutyldimethylsilyl tBu tertbutyl Tf triflate THF tetrahydrofuran TIPS triisopropylsilyl TMS trimethylsilyl Ts p-toluenesulfonyl TS transition structure
List of Publications
This thesis is based on the following papers, referred to in the text by their Roman numerals I-IV:
I. Mukaiyama Aldol Addition to α-Chloro substituted aldehydes.
Origin of Unexpected Syn Selectivity
Tessie Borg, Jakob Danielsson and Peter Somfai Chem. Commun. 2010, 46, 1281-1283.
II. Diastereoselective Nucleophilic Addition to Aldehydes with Polar
α- and α,β-Substituents
Tessie Borg, Jakob Danielsson, Maziar Mohiti, Per Restorp and Peter Somfai
Adv. Synth. Catal. 2011, 353, 2022-2036.
III. Lewis Acid-Promoted Addition of 1,3-Bis(silyl)propenes to Aldehydes: A Route to 1,3-Dienes
Tessie Borg, Pavel Tuzina and Peter Somfai J. Org. Chem. 2011, 76, 8070-8075.
IV. Diastereoselective formation of 2,3,4,5-Tetrasubstituted
Tetrahydrofuranes by a Lewis Acid-Promoted Addition of C3-Substituted 1,3-bis(silyl)propenes to Aldehydes
Tessie Borg, Brian Timmer and Peter Somfai Submitted manuscript.
Table of Contents
Abstract Abbreviations List of Publications
1. Introduction ... 1
2. Diastereoselectivity in the Nucleophilic Addition to Aldehydes with Polar α− or α,β−Substituents ... 5
2.1.
Introduction ... 5
2.2.
Asymmetric induction models ... 6
2.2.1.
1,2-induction models: The polar Felkin-Anh model ... 7
2.2.2.
1,2-induction models: The Cornforth-Evans model ... 8
2.2.3.
1,3-induction models ... 9
2.2.4.
Merged 1,2 and 1,3-induction models ... 10
2.2.5.
Aim of the study ... 12
2.3.
Diastereoselectivity in Mukaiyama aldol additions to α-heteroatom substituted aldehydes. ... 12
2.3.1.
Determination of stereochemistry ... 17
2.3.2.
Diastereoselectivity in Mukaiyama aldol additions to polar α,β-disubstituted aldehydes. ... 18
2.3.3.
Mukaiyama aldol additions to polar α,β-disubstituted aldehydes ... 18
2.3.4.
Sakurai allylation reactions to polar α,β-disubstituted aldehydes ... 22
2.3.5.
Synthesis of α,β-bisalkoxy and α-chloro-β-silyloxy aldehydes ... 24
2.3.6.
Determination of stereochemistry ... 24
2.4.
Conclusion and outlook ... 26
3. Addition of 1,3-Bis(silyl)propenes to Aldehydes: Synthesis of 1,3-(E)-Dienes ... 27
3.1.
Introduction ... 27
3.1.1.
Generation of dienes ... 28
3.1.2.
Allylic silanes ... 29
3.1.3.
Aim of the study and synthetic strategy ... 31
3.2.
Addition of 1,3-bis(silyl)propenes to aldehydes ... 32
3.3.
Mechanistic aspects ... 36
3.4.
Conclusions and outlook ... 39
4. Addition of 1,3-Bis(silyl)propenes to Aldehydes: Synthesis of 2,3,4,5-Tetrasubstituted Tetrahydrofurans ... 41
4.1.
Introduction ... 41
4.1.1.
Generation of furans ... 41
4.1.2.
Allylic silanes as dipole equivalents ... 42
4.2.
Addition of substituted 1,3-bis(silyl)propenes to aldehydes ... 43
4.3.
Mechanistic aspects of the furan formation ... 46
4.4.
Conclusions and outlook ... 47
5. Attempts Towards the Total Synthesis of (±)-Aspidophylline A ... 49
5.1.
Introduction ... 49
5.2.
Retrosyntethic analysis of (±)-aspidophylline A. ... 49
5.3.
Synthetic efforts towards (±)-aspidophylline A. ... 50
5.4.
Conclusions and outlook ... 55
6. Concluding Remarks ... 57
Acknowledgments Appendices
1. Introduction
At a time when more and more resources are invested in applied research to meet our needs in a society under constant development, it is easy to forget the importance of basic research. However, the applied research heavily relies on the progress made in basic research. It is the fundamental research that gives us a better understanding of how everything works and how we can use our resources in the best way to live in balance with the environment and develop a sustainable future society.
Chemistry is all around us, not only as something unhealthy and toxic that you should avoid but rather something that exists in our everyday life. From the moment you get up in the morning until you go to bed you will encounter chemicals and chemical reactions. In your first cup of coffee a molecule is responsible for the aroma of coffee, in the shower it is the surfactants in soap and shampoo that get you clean. Even when you are in love or at sleep dreaming it is chemical reactions that are responsible for those experiences.1
What is even more important is that without the knowledge of molecules and chemical reactions, we will not be able to solve the global problems we face in areas such as health, environment, climate, energy and renewable raw materials.2
From the beginning the term “organic chemistry” was invented for the study of compounds derived from biological sources.3 At that time it was believed that
the synthesis of organic compounds only could occur within a living organism while inorganic compounds was synthesized from non-living material. The turning point came when Friedrich Wöhler in 1828 showed that ammonium cyanate could be used to synthesize urea.4 Today organic chemistry is defined
as the part of science of chemistry that deals with organic compounds5 and can
be found not only in the field of chemical science but also in the disciplines of biology, life science, material science and physics. One carbon can bind up to four carbons or other atoms and the connectivity together with the three-dimensional structure will be responsible for the properties of the molecule. The world around us is chiral6 and so are the most complex natural products
1 Ellervik, U. in Ond Kemi, Fri tanke förlag, Stockholm, 2011.
2 Kemi -den gränslösa vetenskapen, Andreas Nilsson Ed, KVA and IVA, Stockholm, 2011. 3 Yeh. B. J.; Lim, W. A. Nature Chemical Biology 2007, 3, 521 – 525.
4 Wöhler, F. Ann. Phys. Chem. 1828, 12, 253-256.
5 Most carbon-containing compounds are organic compounds, and almost all organic compounds
contain at least a C-H bond or a C-C bond.
and the interaction with other chiral compounds varies with the enantiomer.7
The development of new methodology in the field of asymmetric organic synthesis deals with the invention of new reliable and high yielding reactions that produce one enantiomer in preference of the other. There are three different ways of applying stereocontrol on the reaction and introduce asymmetry in a molecule:8
Reagent-control:
The formation of a new stereogenic center is governed by a catalyst or reagent not covalently bound to the substrate.
Auxiliary-control:
The formation of a new stereogenic center is controlled by stoichiometric amount of chiral auxiliary covalently bound to the substrate but not part of the final structure.
Substrate-control:
The formation of a new stereogenic center is controlled by the chirality already present in the substrate.
There are an almost infinite number of molecules, natural or unnatural, more or less complex for a synthetic organic chemist to construct and today the possibilities are endless.9 With that said it still remains indubitable that there is
a further need for development of new efficient, high yielding synthetic methods with high levels of stereocontrol as several of the methods used today still are far from satisfying. The aim of this doctoral thesis was to get a greater understanding of the factors influencing the stereochemical preferences in stereoselective additions of nucleophiles to aldehydes substituted with a heteroatom in α- or in α- and β-position and develop new methodology for stereoselective construction of carbon-carbon bonds in organic synthesis. Chapter 2 describes our investigation of the factors governing the stereoinduction in Lewis acid mediated nucleophilic addition to α- and α,β-heteroatom substituted aldehydes.
7 For examples of compounds where the two enantiomers show different effects when interacting
with a living environment see: Lin, G.-Q.; Li, Y.-M.; Chan, A. S. C. Principles and applications of
asymmetric synthesis; Wiley-Interscience: New York, 2001, pp. 4-7.
8 For further discussion, see: a) Nógrádi, M. Stereoselective synthesis, VCH Verlagsgesellschaft mbH:
Weinheim, 1995; b) Koskinen, A. M. P. Asymmetric Synthesis of Natural Products, John Wiley & Sons, Ltd: West Sussex, 2nd ed., 2012.
9 For examples of contemporary organic synthesis of complex organic compound see: Nicolaou, K.
C.; Sorensen, E. J. Classics in Total Synthesis; VCH: New York, 1996; Nicolaou, K. C.; Snyder, S. A.
Classics in Total Synthesis II; Wiley-VCH: Weinheim, 2003; Nicolaou, K. C.; Chen, S. J. Classics in Total Synthesis III; Wiley-VCH: Weinheim, 2011;
Chapter 3 deals with the development of a mild, straightforward and highly stereoselective synthesis of (E)-1,3-dienes, by a Lewis acid mediated addition of 1,3-bis(silyl)propens to aldehydes.
Chapter 4 deals with the development of a novel, highly stereoselective method for the construction of 2,3,4,5-tetrasubstituted tetrahydrofurans by addition of C3-substituted 1,3-bis(silyl)propens to aldehydes.
Chapter 5 describes our attempts to perform a total synthesis of (±)-aspidophylline A.
2. Diastereoselectivity in the Nucleophilic
Addition to Aldehydes with Polar α− or
α,β−Substituents
(Papers I and II)
2.1 Introduction
Nucleophilic additions to carbonyl compounds are an important synthetic tool for selective C-C bond formation10 and among these reactions the aldol
reaction is one of the most central.11 Even in the simplest case the reaction
combines two simple molecules into a more complex β-hydroxy carbonyl compound, and simultaneously two new stereogenic centers are formed. The aldol reaction was discovered as early as 1872,12 and until now it has been
widely used in the construction of a wide array of natural products,13 where
epothilone14 and spongistatin 215 are just two examples (Figure 1).
10 a) Comprehensive Organic Synthesis; Trost, B. M.; Fleming, I. Eds.; Pergamon: New York, 1991; Vol. 2;
b) Houben-Weyl: Methods of Organic Chemistry; Helmchen, G.; Hoffmann, R. W.; Mulzer, J.; Schaumann, E. Eds.; Thieme: Stuttgart, 1995; Vol. E 21 b, Chapter 1.3; c) Heathcock, C. H. In
Asymmetric Synthesis; Morrison, J. D. Ed.; Academic: Orlando, 1984; Vol. 3, p 111-212; d). Evans, D.
A.; Nelson, J. V.; Taber. T. R. Top. Stereochem. 1982, 13, 1-115; e) Mukaiyama, T. Org. React. 1982, 28, 203-331.
11 a) Palomo, C.; Oiarbide, M.; García, J. M. Chem. Eur. J. 2002, 8, 36-44; b) Modern Aldol Reactions,
Mahrwald, R., Eds, Wiley-VCH Verlag GmbH & C KGaA, Weinheim, 2004.
12 The aldol reaction was discovered independently by Charles Adolphe Wurtz in Germany and
Aleksandr Borodin in Russia. Wurtz, C. A. Bull. Soc. Chim. 1872, 17, 436–442.
13 a) Mahrwald, R.; Schetter, B. Angew. Chem. Int. Ed. 2006, 45, 7506–7525; b) Mukaiyama, T.; Shiina,
I.; Iwadare, H.; Saitoh, M.; Nishimura, T.; Ohkawa, N.; Sakoh, H.; Nishimura, K.; Tani, Y.; Hasegawa, M.; Yamada, K.; Saitoh, K. Chem. Eur. J. 1999, 5, 121–161. c) Paterson, I.; Cowden, C. J.; Wallace, D. J. in Modern Carbonyl Chemistry, Wiley-VCH Verlag GmbH, Weinheim, 2000, pp 249-297.
14 a) Nicolaou, K. C.; He, Y.; Vourloumis, D.; Vallberg, H.; Yang, Z. Angew. Chem. Int. Ed. 1996, 35,
2399-2401; b) Schinzer, D.; Limberg, A.; Böhm, O. M. Chem. Eur. J. 1996, 2, 1477-1482.
15 Evans, D. A.; Trotter, B. W.; Coleman, P. J.; Cote, B.; Dias, L. C.; Rajapakse, H. A. Tetrahedron 1999,
Figure 1. Natural products synthesized by using an aldol approach.
2.2 Asymmetric induction models
The power of this type of transformation lies in the ability to predict and control the stereochemical outcome of the reaction. Much effort has been devoted over the years to elucidate the factors that influence the carbonyl π-facial selectivity of these nucleophilic addition reactions and to define models that accurately predict their stereochemical outcome.16 The first model was
presented in middle of the 20th century by Cram17 and in the years that
followed Cornforth,18 Karabatsos,19 Felkin,20 Anh-Eisenstein21 and Evans22 all
presented models with the intention of predicting the stereofacial selectivity. Currently, the models used to rationalize and predict the π-facial selectivity in nucleophilic addition to an aldehyde with a proximate stereocenter under non-chelating reaction conditions are divided into 1,2- or 1,3-asymmetric induction models depending on if the stereocenter affecting the stereochemical outcome is placed α or β to the carbonyl. Furthermore, there is also a combined 1,2- and
16 a) Mengel, A.; Reiser, O. Chem. Rev. 1999, 99, 1191-1223. b) Carreira, E. M. In Comprehensive
Asymmetric Catalasys; Jacobsen, E. N.; Pflatz A.; Yamamoto, H. Eds.; Springer-Verlag, Heidelberg, 1999; Vol 3, chap. 29; c) Swamura, M.; Ito, Y. In Catalytic Asymmetric Synthesis; Ojima, I. Ed.
Wiley-VCH, Weinhem, 2nd ed, 2000; chap 8B1; d) Carreira, E. M.; Fettes, A.; Marti, C. Org. React., 2006,
67, 1-216.
17 Cram, D. J.; Elhafez, F. A. A. J. Am. Chem. Soc. 1952, 74, 5828-5835. 18 Cornforth, J. W.; Cornforth, R. H.; Mathew, K. K. J. Chem. Soc. 1959, 112-117. 19 Karabatsos, G. J. J. Am. Chem. Soc. 1967, 89, 1367-1371.
20 Cherest, M.; Felkin, H.; Prudent, N. Tetrahedron Lett. 1968, 2199-2204.
21 a) Anh, N. T.; Eisenstein, O. Nouv. J. Chim., 1977, 1, 61-70; b) Anh, N. T. Top. Curr. Chem., 1980, 88,
145-162;. c) Anh, N. T.; Eisenstein, O.; Lefour, J-M.; Dâu, M-E. J. Am. Chem. Soc. 1973, 95, 6146-6147.
22 Evans, D. A.; Siska, S. J.; Cee, V. J. Angew. Chem., Int. Ed. 2003, 42, 1761-1765. O O OH R OH O O S N epothilone A (R=H) epothilone B (R=Me) O O O OH O O O H AcO OH OH HO OH HO O O OAc OMe HO H O spongistatin 2 with the aldol disconnections made in one of the total synthesis
completed marked O O OH O OH 9 5 1
C1-C9 segment of the epothilones with three possible aldol
disconnections marked
1 5
1,3-asymmetric model. For α-heteroatom substituted aldehydes under non-chelating reaction conditions the models used for prediction are the polar Felkin-Anh model (PFA) or the Cornforth-Evans model.
2.2.1 1,2-induction models: The polar Felkin-Anh model
For additions to α-substituted aldehydes under non-chelation controlled conditions the Felkin-Anh model has been used with great success.23 In thismodel the torsional strain and hyperconjugation are important control elements governing the conformation of the transition state. The best vicinal acceptor (σC-X*) is oriented perpendicular to the C=O moiety (Figure 2). This
orientation will create the best overlap between the σC-X* and the π-orbital and
the delocalization of electron density will stabilize this conformation.24
Moreover, the perpendicular alignment of the polar substituent also means that the orbital interaction between the incoming nucleophile and the σC-X* is
optimized. The nucleophilic attack, then proceeds along the Bürgi-Dunitz angle25 via TS1 to avoid steric interactions between the α-(L)-substituent and
the nucleophile, thus affording the 1,2-anti isomer and this is in accord with experimental observations.26
Figure 2. Polar Felkin-Anh model, X= polar heteroatom, L= large group, S=
small group.
The PFA model has been the accepted model used to predict facial selectivities in nucleophilic additions to aldehydes with a α-heteroatom substituent for a long time.
23 a) See ref 20; For computational studies, see: b) Wu, Y.-D.; Houk, K. N. J. Am. Chem. Soc. 1987,
109, 908-910; c) Houk, K. N.; Paddon-Row, M. N.; Rondan, N. G.; Wu, Y.-D.; Brown, F. K.;
Spellmeyer, D. C.; Metz, J. T.; Loncharich, R. J. Science 1986, 231, 1108-1117; d) Wong, S. S.; Paddon-Row, M. N. Aust. J. Chem. 1991, 44, 765-770; e) Frenking, G.; Köhler, K. F.; Reetz, M. T.
Tetrahedron 1991, 43, 9005-9018.
24 Alabugin, I. V.; Zeidan, T. J. Am. Chem. Soc. 2002, 124, 3175-3185.
25 a) Bürgi, H. B.; Dunitz, J. D.; Shefter, E. J. Am. Chem. Soc. 1973, 95, 5065-5067; b) Bürgi, H. B.;
Dunitz, J. D.; Lehn, J. M.; Wipff, G. Tetrahedron 1974, 30, 1563-1572.
26 See, for example: Nakai, K.; Kaneko, M.; Loh, T.-P.; Terada, M. Nakai, T. Tetrahedron Lett. 1990, 31,
3909-3912. L O H S X Nu L OH Nu S X S L X O H Nu TS1 1,2-anti
2.2.2 1,2-induction models: The Cornforth-Evans model
Recently, Evans questioned the accuracy of the PFA-model. Instead he proposed resurrecting the Cornforth model and applying some modifications.27The original Cornforth model18 is based on the assumption that the
minimization of the dipole moment between the C=O framework and the polar α-substituent is the most important factor for the determination of the transition state energies in nucleophilic additions to α-heteroatom substituted aldehydes. Evans modified this model by incorporating the Bürgi-Dunitz trajectory and adjusting the dihedral angle to achieve a staggered conformation in the transition state (Figure 3, TS2).28
Figure 3. Cornforth-Evans model.
Both the PFA and the Cornforth model predict the same 1,2-anti isomer and to provide clarity and differentiate between the two models an elegant experiment was designed by Evans and coworkers. By combining the Zimmerman-Traxler model29 with either the PFA or the Cornforth-Evans model in the addition of
E(O)- and Z(O)-boron enolates to α-alkoxy substituted aldehydes they were able to distinguish which of the two models was most accurate (Figure 4).
27 a) See ref 22. The same observations has been made by other groups, see for example: b)
Diaz-Oltra, S.; Murga, J.; Falomir, E.; Carda, M.; Peris, G.; Marco, J. A. J. Org. Chem. 2005, 70, 8130– 8139.
28 For a discussion on staggered transition states see: Paddon-Row, M. N.; Rondan, N. G.; Houk, K.
N. J. Am. Chem. Soc. 1982, 104, 7162-7166.
29 Zimmerman, H. E.; Traxler, M. D. J. Am. Chem. Soc. 1957, 79, 1920-1923. L O H S X Nu L OH Nu S X X S L O H Nu TS2 1,2-anti
Figure 4. Cornforth-Evans and PFA transition state models for addition of
boron enolates to α-heteroatom substituted aldehydes. X=OPg.
The experimental outcome provided the expected 2,3-anti aldol adduct in the addition of E(O)-enolate and the 2,3-syn adduct in the addition of the Z(O)-enolate. More importantly, in the addition of the Z(O)-enolate the diastereoselectivity in favor for 3,4-anti diastereomer was excellent whereas in contrast the E(O)-enolate provided diminished diastereofacial selectivity and in some cases the 3,4-syn becames the major observed isomer. This outcome is not consistent with the PFA model, as this model predicts that the addition of the Z(O)-enolate would be hampered by severe syn-pentane interaction (shown in TS6), while addition of E(O)enolate should proceed without any destabilizing interactions (TS4, Figure 4). These predictions are opposite to the observed experimental outcomes. Instead the experimental outcome from this investigation is in accordance with the Cornforth-Evans model, in which the syn-pentane interaction seen in TS3 would give diminished 3,4-selectivities for reactions with the E(O)-enolate, while no such destabilizing interaction is predicte with the Z(O)-enolate (see TS5). A computational study further supports these findings and lead to the conclusion that the Cornforth-Evans model was valid in addition of enol boranes to α-heteroatom substituted aldehydes having electronegative substituents such as X= F, OMe, Cl, while the PFA model was accurate for those substrates with less electronegative substituents (X= NMe2, SMe, PMe2).30
2.2.3 1,3-induction models
In contrast to the 1,2-induction no general model for 1,3-induction has evolved for the prediction of nucleophilic addition to β-substituted aldehydes. For the
30 Cee, V. J.; Cramer, C. J.; Evans, D. A. J. Am. Chem. Soc. 2006, 128, 2920–2930.
R O H X R OH X 2,3-anti O iPr (cHex)2B O iPr (9-BBN) E(O)-enolate Z(O)-enolate O BR'2 O H H iPr Me X H R O BR'2 O H H iPr Me H R X O BR'2 O Me H iPr H X H R O BR'2 O Me H iPr H H R X Cornforth-Evans PFA TS3 TS4 TS5 TS6 iPr O Me 2 3 4 R OH X 2,3-syn iPr O Me 2 3 4 3.4-anti:3:4-syn 21:79-67:33 3.4-anti:3:4-syn 89:11-98:2
particular case when the β-substituent is a polar heteroatom it is proposed by Evans that TS7 (Figure 5) in which both steric and electrostatic effects are minimized, accounts for the selectivities seen in BF3·OEt2 mediated addition of
enol silanes to these types of aldehydes.31
Figure 5. 1,3-asymmetric induction model.
2.2.4 Merged 1,2 and 1,3-induction models
To make matters more complicated, as soon as the aldehyde contains an α− and a β−substituent it becomes difficult to define a single model to rationalize the observed stereochemical outcome in nucleophilic additions to aldehydes. In the simplest case of a nucleophilic addition to an α,β-substituted aldehyde it could be expected that the relative stereochemistry of the major product could be predicted by analyzing the individual effects from the α- and the β-stereocenter respectively. One diastereomer of the aldehyde should then have a matched combination of the influence from α- and the β-stereocenter, whereas the other diastereomer should have a mismatch combination. This is indeed the case for α-methyl-β-alkoxy aldehydes, for which TS8 (Figure 6), where both stereodirecting elements promote the addition to the same C=O π-face, has been proposed to explain the high selectivities seen in the addition to the anti-diastereomer. The addition to the corresponding mismatched syn-substituted aldehyde proceeded with varying diastereoselectivities.32
Figure 6. Merged FA-polar 1,3-asymmetric induction model for addition to
anti-α-methyl-β-alkoxy aldehyde.
31 a) Evans, D. A.; Duffy, J. L.; Dart, M. J. Tetrahedron Lett. 1994, 35, 8537–8540; b) Evans, D. A.;
Dart, M. J.; Duffy, J. L.; Yang, M. G. J. Am. Chem. Soc. 1996, 118, 4322–4343; c) Reetz, M. T.; Kessler, K.; Jung, A. Tetrahedron Lett. 1984. 25, 729–732.
32 a) See ref 31b; b) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Yang, M. G.; Livingstone, A. B. J. Am.
Chem. Soc. 1995, 117, 6619–6620; c) Evans, D. A.; Allison, B. D.; Yang, M. G.; Masse, C. E. J. Am. Chem. Soc. 2001, 123, 10840–10852; O H Nu OH Nu Hb Ha O H Nu TS7 1,3-anti R X X H R R X O H Nu OH Nu Hb Me O H Nu TS8 1,3-anti R OP PO H R R OP Me Me
Evans and coworkers investigated the nucleophilic addition to syn- and anti-substituted α,β-bisalkoxy aldehydes.33 Experimentally it was found that the
Mukaiyama aldol reactions with anti-α,β-bisalkoxy aldehydes in most cases gave good diastereoselectivities whereas the corresponding syn-isomer resulted in moderated selectivities. By analyzing both the PFA and the Cornforth-Evans transition states, and by considering the inherent and the developing syn-pentane interactions,34 it was established that the Cornforth-Evans model was
the most accurate for the prediction of the stereochemical outcome (Figure 7,
TS9 and TS10). In the most favorable Cornforth-Evans TS (TS10) for the
addition to syn substituted aldehydes it was found that the positioning of the β-alkyl substituent (R) is causing a destabilization syn-pentane interaction with the aldehyde hydrogen whereas in the addition to anti substituted aldehydes this destabilizing interaction was absent (TS9). This finding concluded that the Cornforth-Evans model accurately predicted the stereochemical outcome in additions to α,β-alkoxy aldehydes. The same analysis was performed by invoking the PFA transition state models and it was revealed that the syn substituted aldehyde should give higher diastereoselectivity compared to the anti substituted aldehyde, however this is inconsistent with the stereochemical trends seen experimentally and strengthen the validity of the Cornforth-Evans model for predicting the stereochemical outcome in nucleophilic addition to α,β-substituted aldehydes.
Figure 7. Cornforth-Evans transition state structures for addition to anti- and
syn-α,β-bisalkoxy aldehyde (TS9 and TS10).
33 Evans, D. A.; Cee, V. J.; Siska, S. J. J. Am. Chem. Soc. 2006, 128, 9433–9441.
34 For discussions of the role of syn-pentane interactions in aldol reactions, see: a) Roush, W. R. J. Org.
Chem. 1991, 56, 4151–4157; b) Patel, D. V.; Van Middlesworth, F.; Donaubauer, J.; Gannett, P.; Sib,
C. J. J. Am. Chem. Soc. 1986, 108, 4603–4614.
R OP anti,anti O OP HNu H H O TS9 Nu OH OP R OP syn,anti Nu OH OP R P R OP anti H O OP Nu O OP HNu H R O TS10 H P R OP syn H O OP Nu
2.2.5 Aim of the study
The understanding of the interactions that dictates the stereochemical outcome of nucleophilic additions to carbonyl compounds is significant for the planning and the synthesis of complex molecules. The different biases that influence the transition state energies can in one compound be reinforcing and in another work against each other. The aim with this study was to investigate what influence the α- and β-substituents will have on the stereochemical outcome in nucleophilic additions to aldehydes and gain a greater understanding of the transition states for these kinds of reactions.
2.3 Diastereoselectivity in Mukaiyama aldol additions to
α-heteroatom substituted aldehydes.
Initial focus was directed towards the investigation of the stereochemical outcome in Mukaiyama aldol additions to aldehydes substituted with an α-heteroatom. A set of aldehydes was chosen with α-heteroatoms of different electronegative character, from the less electronegative α-NTsBn moiety to substituents with more prominent electronegativity, namely chloro, α-silyloxy and α-fluoro. The aldehydes were subjected to Mukaiyama aldol conditions applying silyl enol ethers of different steric bulk (Table 1).
Interestingly, the Mukaiyama addition to α-chloro aldehyde 2.1 and 2.2 (Table 1, entries 1 and 4) furnished the syn isomer 2.7a and 2.10a in good diastereoselectivity and excellent yields. Neither the Cornforth-Evans nor the PFA model can account for this anti-PFA selectivity, as both of these models predict the formation of the anti isomer. Decreasing the bulk of the silyl enol ether 2.6 lead to a stereochemical drift and reversal of the diastereoselectivity (entries 2-3 and 5-6). This observation is not in alignment with the PFA model. In the PFA model an increase in size of the nucleophile should instead result in higher anti selectivities. The same trend, if not as pronounced, was observed in the addition to α-silyloxy and α-fluoro aldehydes 2.3 and 2.4 (entries 7-12). In these additions there were poor anti selectivities and an increase of steric bulk of the nucleophile lead to even lower diastereoselectivities. In contrast, the addition to α-sulfonamide 2.5 afforded the anti adduct as the major isomer with increased selectivity with more sterically hindered nucleophiles, and hence follows the trend expected from a aldehyde reacting through the PFA manifold.
Table 1. Addition of silyl enol ethers to α-substituted aldehydes.a
Entry Aldehyde X R 2.6 Yieldb Products syn:antic
1 2.1 Cl iPr a 99 2.7a:2.7b 84:16 2 2.1 Cl iPr b 92 2.8a:2.8b 35:65 3 2.1 Cl iPr c 94 2.9a:2.9b 40:60 4 2.2 Cl CH2Ph a 99 2.10a:2.10b 78:22 5 2.2 Cl CH2Ph b 97 2.11a:2.11b 29:71 6 2.2 Cl CH2Ph c 94 2.12a:2.12b 40:60
7 2.3 OTBS iPr a 66 2.13a:2.13b 50:50
8 2.3 OTBS iPr b 69 2.14a:2.14b 25:75
9 2.3 OTBS iPr c 66 2.15a:2.15b 18:82
10 2.4 F CH2Ph a 43 2.16a:2.16b 43:57 11 2.4 F CH2Ph b 35 2.17a:2.17b 27:73 12 2.4 F CH2Ph c 63 2.18a:2.18b 36:64 13 2.5 NTsBn iPr a 85 2.19a:2.19b > 2:98 14 2.5 NTsBn iPr b 94 2.20a:2.20b 7:93 15 2.5 NTsBn iPr c 60 2.21a:2.21b 22:78
a) Reaction conditions: To a solution of 2.1-2.5 in CH
2Cl2 at -60 °C was added BF3·OEt2 (3 equiv) and 2.6 (2 equiv) and the mixture was stirred for 18h. b) Isolated yield. c) Determined by 1H NMR spectroscopy of the crude reaction mixture.
To be able understand the stereochemical outcome obtained in the nucleophilic additions to α-chloro aldehydes an examination of the transition state of the reaction was done. In a recent theoretical study of α-heteroatom substituted aldehydes it was revealed that the preferred transition state geometry was dependent on the heteroatom. Less electronegative heteroatoms (P, S and N) favor the PFA manifold while highly electronegative elements (F, O, Cl) favor the Cornforth-Evans model.30 In order to examine the preference of the
α-chloro aldehydes for either model we performed a similar reaction set up (Scheme 1). Addition of E(O)-boron enolate 2.2235 to 2.1 gave 2.23b in modest
yield and selectivity while the addition of Z(O)-boron enolate 2.2435 to 2.1
gave adduct 2.25b in excellent yield and diastereoselectivity. The results from
35 Brown, H. C.; Ganesan, K.; Dhar, R. K. J. Org. Chem. 1993, 58, 147–153. H O R' R X OTMS OH R X R' O OH R X R' O
2.1-2.5 2.6a, R'=tBu 2.7a-2.21a 2.7b-2.21b
2.6b, R'=iPr 2.6c, R'=Me
3,4-syn 3,4-anti BF3.OEt2
these experiments indicate that the α-chloro aldehyde also preferentially reacts through a Cornforth-Evans manifold in boron-enolate mediated aldol reactions and we assumed that the same is true for the Mukaiyama aldol reaction.
Scheme 1. Addition of boron enolate 2.22 and 2.24 to α-chloro aldehyde 2.1. Now that the preferential orientation of the aldehyde in the addition reaction was established, it was still not obvious why both α-chloro and α-alkoxy afforded poor PFA or even anti-PFA selectivities in the Mukaiyama aldol reaction (entries 1-9, Table 1). In order to rationalize the stereochemical outcome the possible transition state structures were examined. It is known that the Mukaiyama reaction in general proceeds through an open transition state,22,36 even though exceptions are known.37 The addition could either take
place in an antiperiplanar or synclinal TS conformation with the Lewis acid placed syn to the aldehyde hydrogen, resulting in four possible TS structures (Figure 8, TS11-TS14).
36 a) Denmark, S. E.; Lee, W. J. Org. Chem. 1994, 59, 707–709; b) Murata, S. Suzuki, M.; Noyori, R. J.
Am. Chem. Soc. 1980, 102, 3248–3249; c) Yamamoto, Y.; Yatagai, H.; Naruta, Y.; Maruyama, K. J. Am. Chem. Soc. 1980, 102, 7107–7109.
37 a) Myers, A. G.; Widdowson, K. L. J. Am. Chem. Soc. 1990, 112, 9672–9674; b) Myers, A. G.;
Widdowson, K. L.; Kukkola, P. J. J. Am. Chem. Soc. 1992, 114, 2765–2767.
O H Cl OH Cl O iPr (cHex)2B O iPr (9-BBN) 50% 99% iPr O Me 2 3 4 OH Cl iPr O Me 2 3 4 OH Cl iPr O Me 2 3 4 OH Cl iPr O Me 2 3 4 2.1 2.22 2.24 2.23a 2.23b 2.25a 2.25b dr 13:87 dr 6:94
Figure 8. Transition state structures in nucleophilic addition to α-chloro aldehydes
Transition state structures TS11 and TS12 correspond to Cornforth-Evans structures in which the silyl enol ether has a synclinal or antiperiplanar orientation. In the anti-Cornforth structures TS13 and TS1438 the dihedral
angle between the α-chloro and the carbonyl group in the aldehyde is relatively small resulting in destabilization due to dipole repulsion,39 however the
diastereotopic C=O face is comparatively more exposed for attack. In TS12 and TS14 the major steric interaction is between the enol silane R’(OTMS) moieties and the substituent on the α-carbon,31b,40 while the destabilizing
interaction in the synclinal structures TS11 and TS13 is between the nucleophile and the Lewis acid coordinated to the carbonyl oxygen.31b,40,41 By
applying a bulky Lewis acid and varying the size of the nucleophile, we were aiming to differentiate between the four possible transition structures. α-Chloro aldehyde 2.1 was subjected to enol silane 2.6a and 2.6c in the presence of the steric demanding Lewis acid trityl tetrafluoroborate (Scheme 2).42 A
sterically demanding Lewis acid, such as trityl tetrafluoroborate, is expected to
38 a)For steric and/or electronic reasons all other synclinal structures has been excluded. For a
discussion, see ref. 31b, footnote 31. b) The aldehyde rotamer having the α-chloro substituent perpendicular to the C=O moiety (anti-PFA TS) is judged less likely since it would result in a destabilizing steric interaction between R and one substituent on the enol silane (R’ or OTMS).
39 For 2-chloropropanal the corresponding conformation having a 30° angle between the C=O and
C-Cl bonds is about 1.9 kcal mol-1 higher in energy than the most stable conformer. Notable is that
this energy is dramatically reduced when the angle is decreased. See ref. 30.
40 Heathcock, C. H.; Davidsen, S. H.; Hug, K. T.; Flippin, L. A. J. Org. Chem. 1986, 51, 3027–3037. 41 Heathcock, C. H.; Flippin, L. J. Am. Chem. Soc. 1983, 105, 1667–1668.
42 Mukaiyama, T.; Kobayashi, S.; Murakami, M. Chem. Lett. 1985, 14, 447–450. R O H Cl R OH Cl OTMS R' R' O R OH Cl R' O Cl R H O H LAH H OTMS R' Cl R H O H LAH H OTMS(R') R'(OTMS) Cl R H O H LA H H TMSO R' Cl R H O H LA H H (R')TMSO (TMSO)R' TS11 TS12 TS13 TS14 Cornforth-Evans
synclinal TS antiperiplanar TSCornforth-Evans
anti-Cornforth
synclinal TS antiperiplanar TSanti-Cornforth
or
or
anti
disfavor a reaction proceeding through the synclinal transition state mode (TS11 and TS13) and this effect is expected to be increased when changing the silyl enol ether R’-substituent from Me (2.6c) to tBu (2.6a). However, the use of Ph3CBF4 as Lewis acid in the reaction did not change the
diastereoselectivity to any significant extent (compare with Table 1, entries 1 and 3). These results imply that the antiperiplanar transition state structures
TS12 and TS14 are responsible for the stereochemical outcome in the
Mukaiyama additions to α-chloro aldehydes.
Scheme 2. Nucleophilic addition of silyl enol ether to α-chloro aldehydes and the effect of using a sterically more demanding Lewis acid.
To distinguish between the antiplanar TS12 and TS14 a closer look at the BF3·OEt2 mediated addition of silyl enol ethers to the α-chloro aldehydes is
necessary. To avoid steric interaction between the α-chloro substituent and pinacolone enol silane 2.6a we argue that the addition proceeds through TS14 (Table 1, entries 1 and 4). As the size of the nucleophile is reduced this steric interaction will be less pronounced, while the magnitude of the dipole interaction in TS14 remains unchanged. This would explain the increased formation of the anti diastereomer seen (Table 1, entries 2, 3, 5 and 6). The similar trend is observed for the α-silyloxy aldehyde 2.321 when the size of the
nucleophile is changed (Table 1, entries 7-9, R= tBu→Me, syn:anti 50:50→18:82) and we assume that the same factors as those outlined in Figure 8 also determines the stereochemical outcome for the α-alkoxy aldehydes.43
The higher anti selectivities seen in the reaction with α-silyloxy and α-fluoro aldehydes compared with α-chloro aldehydes can be explained by assuming an increased dipole repulsion for more electronegative elements (F and O), which will further destabilize the anti-Cornforth structure TS14. This means that the higher electronegativity in the α-fluoro and the α-silyloxy substituent will make the preference for the Cornforth-Evans reaction pathway more difficult to override. Furthermore, the fluorine substituent is significantly smaller than the chlorine atom, and as such the steric interaction in TS12 should be minimized.
43 TS structure similar to TS14 has previously been invoked to rationalize the stereochemical outcome
in the addition to an α-silyloxy aldehyde, see: Smith III, A. B.; Condon, S. M.; McCauley, J. A.; Leazer Jr, J. L.; Leahy, J. W.; Maleczka Jr, R. E. J. Am. Chem. Soc. 1997, 119, 947–961.
H O R Cl OTMS OH Cl R O OH Cl R O 2.1 2.6a, R=tBu 2.6c, R=Me Ph3CBF4 CH2Cl2 -60 °C 2.7a:2.7b, R=tBu, dr 84:16 2.9a:2.9b, R=Me, dr 37:63
The effects discussed above should be exclusive for substrates that preferentially react through the Cornforth-Evans transition state structures. Aldehydes that follow the PFA reaction manifold are expected to be less affected by changing the size of the nucleophile.44 This is indeed the case for
the Mukaiyama aldol addition to Ν−Ts-N-Bn protected valinal 2.5, where the reaction proceeded with excellent stereoselectivity regardless of the size of the nucleophile (Table 1, entries 13-15).45
2.3.1 Determination of stereochemistry
The relative stereochemistry of aldol adducts 2.7-2.9, 2.16-2.18 was determined by analyzing the coupling constants of corresponding epoxide, and for the aldol adducts 2.10-2.12 where the Ha and Hb protons overlapped the
stereochemistry was determined by NOESY measurements of the epoxide. In those cases a NOE cross peak was observed between the Ha and Hb proton in
the cis-isomer whereas no such cross peak was found for the trans-isomer. The stereochemical determination was realized by an anti-1,3-reduction of the aldol adducts followed by an epoxide formation (Scheme 3).
Scheme 3. Stereochemical determination of the aldol adducts 2.7-2.12, 2.16-2.18.
The relative stereochemistry of aldol adducts 2.19-2.21 was determined by analyzing the JHa-Hb coupling constant of the corresponding oxazolidinone
(Scheme 4).46
Scheme 4. Stereochemical determination of the aldol adducts 2.19-2.21.
44Alvarez-Ibarra, C.; Arjona, O.; Pérez-Ossorio, R.; Pérez-Rubalcaba, A.; Quiroga, M. L.; Santesmases,
M. J. J. Chem. Soc. Perkin Trans. 2 1983, 1645–1648.
45
Theoretical investigation supports this finding and suggests that α-amino substituted aldehydes react through PFA transition state structures in boron aldol additions, see ref 30.
46 Mikami, K.; Kaneko, M.; Loh, T. P.; Terada, M.; Nakai, T. Tetrahedron Lett. 1990, 31, 3909-3912. OH R X R' OH OH R X R' O 2.7-2.12, X=Cl 2.16-2.18, X= F Me4NHB(OAc)3 MeCN:AcOH -30 °C R K2CO3 EtOH 70 °C O Ha Hb R' OH 1 3 4 1 3 4 73-99% 3,4-cis, JHa-Hb= 4.3 Hz 3,4-trans, JHa-Hb= 2.3-2.4 Hz 3 4 85-99% X=Cl, F OH NHBn R' OH OH NTsBn R' O 1. Me4NBH(OAc)3, MeCN:AcOH Ha triphosgene, DIPEA iPr Hb three steps 82-95% 3J Ha-Hb= 7.5-7.6 Hz 2. Na+C10H8-, THF or SmI2, pyrrolidine, H2O:THF R OH O BnN O CH2Cl2 2.19-2.21
The relative stereochemistry of boron enolate addition products 2.23-2.25 was determined as outlined in Scheme 5, by analyzing the coupling constants of corresponding epoxides.
Scheme 5. Stereochemical determination of the boron enolate addition
products 2.23-2.25.
2.3.2 Diastereoselectivity in Mukaiyama aldol additions to
polar α,β-disubstituted aldehydes.
In a continuation of the investigation discussed above the Mukaiyama aldol addition to aldehydes having both a polar α- and β-substituent was examined. The aldehydes investigated were selected with different electronic and steric properties of the α-substituent, while the β-substituent, an β-alkoxy moiety, was held constant: α,β-bisalkoxy aldehydes 2.26-2.29, α-amino-β-silyloxy aldehydes 2.30-2.33 and α-chloro-β-silyloxy aldehydes 2.34-2.35. These substrates were subjected to Mukaiyama aldol conditions and applied in the Sakurai allylation with allyltrimethylsilane (2.46) with the aim to unravel any stereochemical trends and examine the generality in Evans merged α,β-stereoinduction model.
2.3.3 Mukaiyama aldol additions to polar α,β-disubstituted
aldehydes
The α,β-disubstituted aldehydes 2.26-2.35 were applied in Mukaiyama aldol addition with silyl enol ether 2.6a (Table 2).
OH X OH OH X O Me4NHB(OAc)3 MeCN:AcOH -30 °C R K2CO3 EtOH 70 °C O Ha Hb R' OH 1 3 4 1 3 4 26-84% 3,4-cis, JHa-Hb= 4.2 Hz 3,4-trans, JHa-Hb= 2.3-2.5 Hz 3 4 92-99% 2 2 2.23-2.25
Table 2. Addition of silyl enol ethers to α,β-disubstituted aldehydes.a
Entry Aldehyde α,β X P Yieldb Products Ratioc
1 2.26 anti OBn Bn 99 2.36a:2.36b 14:86
2 2.27 anti OBn Bn 93 2.37a:2.37b 16:84
3 2.28 syn OBn Bn 90 2.38a:2.38b 49:51
4 2.29 syn OBn Bn 92 2.39a:2.39b 45:55
5 2.30 anti NTsBn TBS 92 2.40a:2.40b > 2:98 6 2.31 anti NTsBn TBS 81 2.41a:2.41b > 2:98 7 2.32 syn NTsBn TBS 91 2.42a:2.42b 53:47 8 2.33 syn NTsBn TBS 49 2.43a:2.43b 56:44 9 2.34 anti Cl TBS 99 2.44a:2.44b 95:5 10 2.35 syn Cl TBS 94 2.45a:2.45b 91:9
a) Reaction conditions: To a solution of 2.26-2.35 in CH
2Cl2 at -60 °C was added BF3·OEt2 (3 equiv) and 2.6a (2 equiv) and the mixture was stirred for 18h.
b) Isolated yield. c) Determined by 1H NMR spectroscopy of the crude reaction mixture.
Addition of TMS-enol ether 2.6a to anti-aldehydes 2.26, 2.27 proceeded in good yields and afforded the corresponding 3,4-anti isomers in good diastereoselectivities (Table 2, entries 1, 2). In contrast, additions to syn-aldehydes 2.28, 2.29 delivered triols 2.38 and 2.39 in equally good yields but with poor diastereoselectivities (entries 3, 4). Similar observations have been made previously and were rationalized by invoking Cornforth-Evans TS structures (e. g. see TS9 and TS10, Figure 7). In this scenario the lower diastereoselectivities obtained for the syn-isomers 2.28 and 2.29 was traced to an unfavorable syn-pentane interaction between R and the CHO proton in
TS10, which is absent in structure TS9.
H O tBu X OTMS OH X tBu O 2.26, R=(CH2)2Ph 2.27, R=n-C6H13 2.28, R=(CH2)2Ph 2.29, R=n-C6H13 2.30, R=(CH2)2Ph 2.31, R=c-C6H11 2.32, R=(CH2)2Ph 2.33, R=c-C6H11 2.34, R=(CH2)2Ph 2.35, R=(CH2)2Ph 2.36a-2.45a 2.26b-2.45b 2.6a 3,4-syn 3,4-anti BF3.OEt2 CH2Cl2 R PO α β R PO 3 5 4 OH X tBu O R PO 3 5 4
According to the analysis of the transition states Evans did for the addition of nucleophiles to α,β-bisalkoxy aldehydes, the PFA model predicts higher selectivity for the addition to the syn isomer. By applying this conclusion to α-amino-β-silyloxy aldehydes 2.30-2.33, it is expected that the stereocenters in the syn-substituted aldehyde should be mutually reinforcing. However the experimental data in Table 2 (entries 5-8), where the anti-isomers react with excellent diastereoselectivity while the syn-aldehydes show low π-facial discrimination, revealed the opposite trend to what would be expected for substrate that follows the PFA model.33 It is clear that the diastereofacial
discrimination for substrates 2.30-2.23 cannot be analyzed simply by applying Evan’s merged α,β-stereoinduction model or in terms of contribution from the individual stereocenters.
The π-facial selectivities obtained in the Mukaiyama aldol reactions with α-chloro-β-silyloxy aldehydes 2.34 and 2.35 (Table 2, entries 9-10) afforded the 3,4-syn products 2.44a and 2.45a, respectively, in excellent yield and with high diastereoselectivity. Interestingly, both of the additions of silyl enol ether 2.6a to 2.34 and 2.35 afforded higher selectivities than the additions to the corresponding α-chloro substituted aldehydes 2.1 and 2.2. It appears that the β-substituent has a reinforcing effect on the stereochemical outcome in the addition to aldehydes 2.34 and 2.35.
In the Mukayiama addition to α-chloro substituted aldehydes the reaction proceeds either through a Cornforth-Evans or an anti-Cornforth TS depending on the size of the nucleophile (see TS12 and TS14 in Figure 8). Assuming this is also true for α-chloro-β-silyloxy aldehydes six different transition structures are depicted, three anti-Cornforth and three Cornforth-Evans for each set of aldehydes, resulting from a rotation around the Cα-Cβ bond (Figure 9). Analyzing the factors governing in each set, anti-Ca330 seems to be the energetically most favored TS for the addition to anti substituted aldehydes, with an optimal position of the dipole Cl↔OP and no developing syn-pentane interaction within the molecule except for OP↔H. The Cornforth-Evans transition state structures Ca1-Ca3 suffer from unfavorable interactions; gauche interaction between Cl↔R and syn-pentane interaction between OP↔C=O in
Ca1. Gauche interaction and an unfavorable dipole alignment between Cl↔OP and syn-pentane interaction R↔H exist in Ca2, and developing syn-pentane interactions within the molecule both between OP↔H and C=O↔R are present in Ca3. From this analysis it is concluded that the diastereofacial bias exerted by the α- and the β-stereocenter in the addition to anti-α-chloro-β-silyloxy aldehydes are a matched combination when the reaction proceeds through an anti-Cornforth manifold. Indeed, this is in alignment with the experimental result, where the addition of 2.6a to 2.34 gave excellent diastereoselectivity in favor for the anti-Cornforth adduct.
Figure 9. Anti-Cornforth and Cornforth-Evans transition state structures for
nucleophilic addition to anti- and syn-aldehydes 2.34-2.35. P= TBS.
By performing the same analysis for the syn substituted aldehyde 2.35, it is then concluded that there is developing syn-pentane interactions in all three rotamers of the anti-Cornforth TS, between OP↔C=O in anti-Cs1, between R↔H in anti-Cs2 and between both OP↔H and C=O↔R in anti-Cs3. The dipole Cl↔OP is minimized by an antiparallel alignment in anti-Cs3 and OP↔C=O in anti-Cs2. Depending on which interaction is most costly, the dipole alignment or the syn-pentane interaction, the addition can take place either through anti-Cs2 or anti-Cs3. The most favorable Cornforth-Evans TS in addition to the α,β-syn aldehyde is Cs1, as Cs2 and Cs3 both suffer from gauche interaction (Cl↔OP and Cl↔R respectively), syn-pentane interactions (C=O↔R or OP↔H) and unfavorable dipole alignment of Cl↔OP. However, the Cs1 is far from optimal as there are syn-pentane interactions between OP↔C=O and R↔H together with the unfavorable dipole alignment of OP↔C=O.
This evaluation of the different transition states, interestingly, leads to the conclusion that even in this case, for the addition to α-chloro-β-silyloxy aldehydes, the diastereofacial discrimination cannot be analyzed simply by
R PO 3,4-syn O OP ClNu H R H Nu OH Cl H R PO α,β-anti H O Cl Nu α β O R ClNu H H H PO O H ClNu H OP HR 5 4 3
anti-Ca1 anti-Ca2 anti-Ca3
R PO 3,4-anti O OP HNu H H Cl Nu OH Cl R O H HNu H R ClPO O R HNu H OP ClH 3 4 5 Ca1 Ca2 Ca3 R PO 3,4-syn O OP ClNu H H H Nu OH Cl R R PO α,β-syn H O Cl Nu α β O H ClNu H R H PO O R ClNu H OP HH 5 4 3
anti-Cs1 anti-Cs2 anti-Cs3
R PO 3,4-anti O OP HNu H R Cl Nu OH Cl H O R HNu H H ClPO O H HNu H OP ClR 5 4 3 Cs1 Cs2 Cs3
applying Evans merged α,β-stereoinduction model or in terms of contribution from the individual stereocenters. It seems that the α-stereocenter is dictating the stereochemical outcome even in additions to α-chloro aldehydes bearing an additional heteroatom on β-position.
2.3.4 Sakurai allylation reactions to polar α,β-disubstituted
aldehydes
In order to further investigate the nucleophilic addition to these substrates, Sakurai allylations47 of aldehydes 2.26-2.30, 2.32, 2.34-2.35 were performed.
The π-facial selectivities in these additions are presented in Table 3. For aldehydes reacting through a Cornforth-Evans manifold, provided that the α-substituent is small enough, the diastereofacial bias should be unaffected regardless of the size of the nucleophiles for additions to anti aldehydes (see
TS9, Figure 7).48 Indeed, this is the case for the addition to
anti-α,β-bisbenzyloxy aldehyde 2.26 and 2.27 in entries 1 and 2 (compare with Table 2, entries 1 and 2). In the additions to syn aldehydes 2.28 and 2.29 (entries 3 and 4) the selectivities are slightly increased as compared with the Mukaiyama aldol additions (Table 2, entries 3 and 4).49
47 Hosomi, A.; Sakurai, H. Tetrahedron Lett. 1976, 1295-1298. 48
This is true for benzylprotected α,β-bisalkoxyaldehydes, however changing the protecting group from a benzyl to a TBS-group will effect the sterical interference between the α-substituent and the nucleophile. See ref 33, Table 3.
49
This trend has previously been observed in the additions of small nucleophiles to α,β-bisalkoxy aldehydes, see ref 33.
Table 3. Sakurai allylation of α,β-disubstituted aldehydes.a
Entry Aldehyde α,β X P R’ Yieldb Products Ratioc
1 2.26 anti OBn Bn (CH2)2Ph 95 2.47a:2.47b 13:87
2 2.27 anti OBn Bn n-C6H13 93 2.48a:2.48b 16:84
3 2.28 syn OBn Bn (CH2)2Ph 83 2.49a:2.49b 21:79
4 2.29 syn OBn Bn n-C6H13 87 2.50a:2.50b 33:67
5 2.30 anti NTsBn TBS (CH2)2Ph 67 2.51a:2.51b 13:87
6 2.32 syn NTsBn TBS (CH2)2Ph <5 2.52a:2.52b 45:55
7 2.34 anti Cl TBS (CH2)2Ph 88 2.53a:2.53b 60:40
8 2.35 syn Cl TBS (CH2)2Ph 94 2.54a:2.54b 59:41
a) Reaction conditions: To a solution of 2.26-2.35 in CH
2Cl2 at -60 °C was added BF3·OEt2 (3 equiv) and 2.46 (2 equiv) and the mixture was stirred for 18h.
b) Isolated yield. c) Determined by 1H NMR spectroscopy of the crude reaction mixture.
Drawing on our conclusions from the Mukaiyama aldol addition, substrates following the PFA model are expected to show diminished levels of anti selectivity as the size of the nucleophile is decreased for the matched case. Indeed, the Sakurai allylations to anti-α-amino-β-alkoxy aldehyde 2.30 (entry 5) gave a diminished selectivity compared with the addition of enolsilane 2.6c (Table 2, entry 5). In the addition to the mismatched syn substituted amino aldehyde 2.32, there was almost no reaction at all and little selectivity (Table 3, entry 6). This is most likely due to the nucleophilicity of the allylsilane, which is a weaker nucleophile as compared to the enol silylether 2.6. Addition of allylsilane 2.46 to α-chloro-β-alkoxy aldehydes 2.34 and 2.35 furnished the allylated products 2.53 and 2.54 in very poor selectivities (entries 7 and 8). As reveled in the addition to α-chloro substrates the reaction is expected to proceed preferably via anti-Cornforth-Evans manifold when applying bulky nucleophiles however as the size of of the nucleophile is reduced the steric interaction in TS14 (Figure 8) will be less pronounced and so will the preference for reacting trough anti-Evans compared to Cornforth-Evans (TS12, Figure 8). It is then expected that Sakurai allylation of both anti- and syn-α-chloro-β-alkoxy aldehydes should give decreased selectivities compared with the Mukaiyama addition of enolsilane 2.6a to α,β-chloro aldehydes 2.34 and 2.35 (Table 2, entries 9, 10) and to α-chloro aldehydes 1
H O TMS X OH X 2.26-2.30, 2.32, 2.34-2.35 2.47a-2.54a 2.47b-2.54b 2.46 3,4-syn 3,4-anti BF3.OEt2 CH2Cl2 R PO α β R PO 3 5 4 OH X R PO 3 5 4
and 2 (Table 1, entries 1 and 4). Indeed, the experimental results are in agreement with previous discussion where the addition to anti α-chloro-β-alkoxy aldehyde 2.34 and the syn aldehyde 2.35 gave only a slight excess of the 3,4-syn isomer (Table 3, entries 7 and 8). Thus, the results support the conclusions drawn from the Mukaiyama reaction.
2.3.5 Synthesis of α,β-bisalkoxy and α-chloro-β-silyloxy
aldehydes
The α,β-bisalkoxy aldehydes 2.26-2.29 were prepared from corresponding allylic alcohol using standard transformations (Scheme 6).
Scheme 6. Preparation of α,β-bisalkoxy substituted aldehydes 2.26-2.29. The α-chloro-β-silyloxy aldehydes 2.34-2.35 were prepared from syn- and anti-α-halohydrin50 using standard transformations (Scheme 7).
Scheme 7. Preparation of α-chloro-β-silyloxy substituted aldehydes.
2.3.6 Determination of stereochemistry
All attempts to determine the relative stereochemistry of aldol adducts 2.36b failed, instead we determined the relative stereochemistry of allylation product
2.47b which were realized by a deprotection followed by an acetal formation
and an analyzing the coupling constants (Scheme 8). The relative stereochemistry of compound 2.36b, 2.37b and 2.48b was assigned in analogy with that of 2.47b.
50 Julia, M.; Verpeaux, J.-N.; Zahneisen, T. Bull. Soc. Chim. Fr. 1994, 131, 539-554. 1. PMBCl, TBAI, NaH 2. OsO4, NMO 3. BnBr, NaH 52-72% (three steps) 1. DDQ 2. DMSO,(COCl)2, Et3N or IBX 37-73% (two steps) 2.26-2.29 R OH R OPMB OH OH R H OBn OBn O 1. TBDMSOTf, 2,6-lutidine 2. OsO4, NMO 51-87% (two steps) NaIO4 or Pb(OAc)4 94-98% 2.34-2.35 R R TBSO Cl R H TBSO Cl O Cl OH OH OH
Scheme 8. Stereochemical determination of the aldol adducts 2.47b.
The relative stereochemistry of 2.44a was determined by analyzing the coupling constants of 2.56 and was realized by a deprotection of the aldol adduct followed by a cyclization (Scheme 9).
Scheme 9. Stereochemical determination of the aldol adducts 2.44a.
The same strategy was tried without success for the determination of the relative stereochemistry of compound 2.45a. Instead, an anti-selective 1,3-reduction followed by an epoxidation made the analysis of coupling constant possible (Scheme 10).
Scheme 10. Stereochemical determination of the aldol adducts 2.45a.
The relative stereochemistry of aldol adduct 2.40b was determined by analyzing the J coupling constants of the corresponding oxazolidinone (Scheme 11) and compound 2.41b was assigned in analogy.
Scheme 11. Stereochemical determination of the aldol adducts 2.40b.
OBn OH
three steps 51% 3JHa-Hb= 9.4 Hz (ax-ax)
OBn Ph O O OAc Ph Ph 1. Pd/C, H2, EtOH 2. PhCH(OCH3)2 , pTsOH, CH2Cl2 3. (CH3CO)2O, pyridine, CH2Cl2 O O R' Hb Ha OAc R Hb Ph 2.47b 2.55 TBSO OH two steps 34% 3JHa-Hb= 11.1 Hz (ax-ax) 3J Hb-Hc= 10.0 Hz (ax-ax) Cl Ph tBu 1. TBAF, THF 2. CH(OMe)3 , PPTS, MeOH 2.44a 2.56 O O OMe tBu OH Ph Cl O Ha OH Hb Cl Hc H tBu OMe H Ph 2.45a 2.57 OH Cl tBu O Me4NHB(OAc)3 MeCN:AcOH -0 °C K2CO3 EtOH 70 °C O Ha Hb tBu OH 67% 3JHa-Hb= 4.1 Hz 39% TBSO Ph OH Cl tBu OH TBSO Ph TBSO Ph 2.58 2.40b 2.59 OH NTsBn tBu O 1. Me4NBH(OAc)3, MeCN:AcOH 3. triphosgene, DIPEA, CH2Cl2 three steps 82% 3J ox= 8.5 Hz (cis) TBSO 2. Na+C 10H8-, THF Ph tBu OH TBSO Ph BnN O O
2.4 Conclusion and outlook
In conclusion, this investigation has shown that the size of the nucleophile in the Mukaiyama addition to α-chloro substituted aldehydes will affect the stereochemical outcome. Small enol silanes will preferentially react through Cornforth-Evans TS structures, while sterically more hindered nucleophiles will react through anti-Cornforth-Evans TS. We have shown that there is a similar relationship between the size of the nucleophile and the stereochemical outcome for additions to α-alkoxy and α-fluoro substituted aldehydes, and we propose that similar factors govern the diastereofacial selectivity with this class of substrates. We have also shown that the opposite is true for α-sulfonamide substituted aldehydes. These substrates will give a higher diastereoselectivity with increased size of the nucleophile in favor for the anti diastereomer. Moreover, we have highlighted the fact that the stereochemical outcome in the Mukaiyama aldol addition and Sakurai allylation of α,β-bisheteroatom substituted aldehydes cannot be predicted or rationalized with current stereoinduction models. The more complex the molecules get, the more difficult it is to take in account all the biases affecting the stereochemical outcome. At some level of complexity the point will be reached where it has to be questioned whether it still is possible to perform an accurate prediction in terms of the stereochemical outcome on these flexible systems, or if perhaps another strategy is preferable, especially when planning a total synthesis of complex molecules.
3 Addition of 1,3-Bis(silyl)propenes to
Aldehydes: Synthesis of 1,3-(E)-Dienes
(Paper III)
3.1 Introduction
Dienes are important motifs in organic chemistry and highly suitable for further synthetic elaborations (Figure 10).
Figure 10. Possible pathways for further synthetic elaborations of dienes.
One of the most powerful reactions dienes participate in is the Diels Alder reaction. Via an inter- or intra molecular fashion dienes can be transformed into either six membered carba- or heterocycles depending on the choice of dienophile.51 In addition dienes can also undergo transformations to form
peroxides by reacting with molecular oxygen,52 mono-53 or dicyclopropanes54
by 1,2-addition of carbenes and they can also form β-lactams55 by reacting
with activated isocyanates. The broad ranges of transformation possible for dienes make them useful as intermediates in the construction of natural
51 a) For Intermolecular Diels–Alder see: Oppolzer, W. In Comprehensive Organic Synthesis; Trost, B. M.,
Flemming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 5, pp 315-328, 339-377; b) For Intramolecular Diels–Alder see: Roush, W. R. In Comprehensive Organic Synthesis; Trost, B. M., Flemming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 5, pp 513-546; c) For Hetero Diels–Alder see: Weinreb, S. M. In Comprehensive Organic Synthesis; Trost, B. M., Flemming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 5, pp 401-444.
52 For a review on photooxidation of 1,3-dienes: Clennan, E. L. Tetrahedron 1991, 47, 1343-1382. 53 Woodworth, R. C.; Skell, P. S. J. Am. Chem. Soc. 1957, 79, 2542-2544.
54 Orchin, M.; Herrick, E. C. J. Org. Chem. 1959, 24, 139-140. 55 Moriconi, E. J.; Meyer, W. C. J. Org. Chem. 1971, 36, 2841-2849.
R X R R O O R R NR R O Diels Alder Hetero Diels Alder O2, hv Carbene Isocyanate β-Lactams Cyclopropanes Peroxides Heterocycles X= O, N Carbacycles
