Synlett 2016; 27(13): 1963-1968
DOI: 10.1055/s-0035-1562344
letter
© Georg Thieme Verlag Stuttgart · New York

Benzylic Ammonium Ylide Mediated Epoxidations

Lukas Roiser
a   Institute of Organic Chemistry, Johannes Kepler University Linz, Altenbergerstraße 69, 4040 Linz, Austria   Telefon: +43(732)24688747   eMail: mario.waser@jku.at
,
Raphaël Robiette*
b   Institute of Condensed Matter and Nanosciences, Université catholique de Louvain, Place Louis Pasteur 1 box L4.01.02, 1348 Louvain-la-Neuve, Belgium   eMail: raphael.robiette@uclouvain.be
,
Mario Waser*
a   Institute of Organic Chemistry, Johannes Kepler University Linz, Altenbergerstraße 69, 4040 Linz, Austria   Telefon: +43(732)24688747   eMail: mario.waser@jku.at
› Institutsangaben
Weitere Informationen

Publikationsverlauf

Received: 31. März 2016

Accepted after revision: 22. Mai 2016

Publikationsdatum:
15. Juni 2016 (online)

 


Abstract

A high yielding synthesis of stilbene oxides using ammonium ylides has been developed. It turned out that the amine leaving group plays a crucial role as trimethylamine gives higher yields than DABCO or quinuclidine. The amine group also influences the diastereoselectivity, and detailed DFT calculations to understand the key parameters of these reactions have been carried out.


#

Onium ylides are versatile reagents for (dia)stereoselective epoxide, aziridine, and cyclopropane syntheses.[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] In contrast to sulfonium ylides, which are frequently used for a variety of three-ring forming reactions,[3] ammonium ylides have been less routinely used in the past.[4–8] This is mainly because of the weaker leaving group ability of the amine group as compared to their sulfur analogues.[10]

We have recently succeeded in developing a high yielding and highly trans-selective epoxidation reaction using amide-stabilised ammonium salts 1 as ylide precursors (Scheme [1]).[7] Key to success in this transformation was the use of trimethylamine as the amine leaving group of choice, which was clearly superior to other tertiary amines such as DABCO, quinuclidine, or triethylamine. Expanding the application scope of this operationally simple method we became interested in the synthesis of stilbene oxides 5, starting from benzylic ammonium salts 4 (Scheme [1]). This transformation has been reported in the past using DABCO and quinuclidine as the amino group.[5] However, it was found that the nature of the leaving group, as well as the electronic properties of the aryl group of the benzylic ammonium salt 4 play a crucial role hereby. While electron-rich or phenyl-based ammonium salts 4 gave epoxides 5 in up to 40% yields only (with a cis/trans ratio depending on the leaving group ability), more stabilised ylides (with electron-withdrawing aryl substituents) allowed for higher yields and higher trans selectivities.[5] Based on computational results, Aggarwal and co-workers explain the observed generally low yields by the high barrier to elimination, which is a consequence of the poor leaving group ability of the amine group.[5c] They also rationalize the observed variations in diastereoselectivity according to the nature of the substrate by the degree of reversibility in betaine formation: reaction of stabilized ylides (substituted by electron-poor aryl groups) lead to reversibility of betaine formation and high trans selectivity, whilst groups that destabilize the ylide (electron-rich aryl groups) or reduce the barrier to ring closure (better amine leaving group) lead to reduced reversibility and lower diastereoselectivity.[11]

Zoom Image
Scheme 1 Recent results in the field of ammonium ylide mediated epoxidations and targeted (high yielding) synthesis of stilbene oxides 5 with trimethylamine-based ammonium salts 6.

Based on these observations and our excellent previous experience with trimethylamine-based ammonium salts,[7] we became interested in investigating if the beneficial effect of this superior leaving group also holds for the synthesis of stilbene oxides 5 and evaluating the effect of Me3N on the diastereoselectivity of this reaction.

Table 1 Syntheses of Epoxides 5 Using Trimethyl Amine Based Ammonium Salts 6

Entry

Ar

R

5

t-BuOK (equiv)

Temp (°C)

trans:cis a

Yield (%)b

 1

Ph

Ph

5a

2

25

 61:39

42

 2

Ph

Ph

5a

3

25

 61:39

69

 3

Ph

Ph

5a

4

25

 61:39

36

 4

Ph

Ph

5a

3

40

 63:37

37

 5

Ph

Ph

5a

4

40

 66:34

93

 6

Ph/DABCO salt

Ph

5b

4

40

 80:20

41

 7

Ph/Et3N salt

Ph

5b

4

40

 89:11

31

 8

4-O2NC6H4

Ph

5b

4

40

  n.d.c

 0d

 9

4-O2NC6H4

Ph

5b

3

25

 95:5

19d

10

4-F3CC6H4

Ph

5c

4

40

> 99:1

97

11

4-BrC6H4

Ph

5d

4

40

 91:9

89

12

4-FC6H4

Ph

5e

4

40

 55:45

94

13

3-MeOC6H4

Ph

5f

4

40

 73:27

97

14

4-MeC6H4

Ph

5g

4

40

 47:53

52e

15

Ph

4-MeOC6H4

5h

4

40

 68:32

88

16

Ph

2-MeOC6H4

5i

4

40

 41:59

83

17

Ph

4-Me2NC6H4

5j

4

40

 86:14

99

18

Ph

4-MeC6H4

5g

4

40

 60:40

16

19

Ph

4-MeC6H4

5g

3

25

 64:36

66

20

Ph

4-BrC6H4

5d

4

40

 60:40

90

21

Ph

4-ClC6H4

5k

4

40

 64:36

81

22

Ph

4-NCC6H4

5l

4

40

 50:50

26

23

3-MeOC6H4

4-BrC6H4

5m

4

40

 66:34

84

24

4-F3CC6H4

4-BrC6H4

5n

4

40

 97:3

55

25

Ph

Cy

5o

4

40

 62:38

 8

a Determined by 1H NMR on the crude product.

b n.d. = not determined.

c Isolated yield of the mixture of both diastereomers.

d Epoxide 5b decomposition under the basic conditions.

e Partial oligomerization of 6g under the basic conditions.[12]

Initial experiments to identify the best suited conditions for the reaction of the parent ammonium salt 6a (Ar = Ph) with benzaldehyde (2a) showed that a combination of t-BuOK as the base and THF as the solvent clearly outperforms other base/solvent systems. With these basic conditions set, we then further fine-tuned the reaction conditions and elucidated the scope of this transformation. Table [1] gives an overview of the most significant results obtained in a rather detailed screening of different reaction conditions and substrates. As shown in Table [1], entries 1–5, the amount of base as well as the reaction temperature play a significant role. For room temperature reactions it was found best to use three equivalents of t-BuOK which allowed for 69% isolated yield of a 61:39 trans:cis mixture of 5a (Table [1], entry 2), whereas a slightly higher temperature of 40 °C allowed us to obtain 5a in an excellent 93% yield and a slightly higher trans selectivity (66:34) when using four equivalents of base (Table [1], entry 5). Noteworthy, larger or lower amounts of base reduced the yields significantly, either due to decomposition of the product or because of insufficient reactivity when using less base. In addition, longer reaction times usually had no positive effect, even in those cases where conversion was slow.

To illustrate the different leaving group ability of trimethylamine compared to DABCO and triethylamine, we also used the analogous ammonium salts under the optimized conditions (Table [1], entries 6 and 7). These experiments clearly demonstrate the significantly higher yield using trimethylamine but also showed that with this more reactive system the diastereoselectivity slightly decreases, giving increasing amounts of the cis epoxides.

Having identified simple conditions that allowed for high yields within a short reaction time, we next tested a variety of differently substituted ammonium salts 6 (Table [1], entries 8–14) and various aromatic aldehydes 2 (Table [1], entries 15–24). Except in the case of electron-poor aryl groups (Table [1], entries 8 and 9) and/or electron-poor aryl aldehydes (Table [1], entry 22), for which 2 and 5 are prone to decomposition under the basic conditions, a variety of differently substituted aryl groups are well accepted in this reaction. Yields for epoxides 5 starting from electron-rich and electron-neutral substrates were equally high, apart for 5g synthesized from ammonium salt 6g (Table [1], entry 14), which tends to oligomerize under the basic reaction conditions.[12] As expected, the diastereoselectivity strongly depends on the electronic properties of the ammonium salt, with more stabilized ylides (obtained from 6 with electron-poorer aryl groups) gave higher trans selectivities. The opposite (but less pronounced) trend was observed by changing the nature of the aldehyde 2. Hereby, less reactive electron-rich aryl groups (i.e., Table [1], entries 15 and 17) give a higher trans selectivity than electron poorer ones (Table [1], entry 20). Unfortunately, using cyclohexane carbaldehyde as the acceptor mainly resulted in decomposition, giving only minute amounts of the corresponding epoxide 5o (Table [1], entry 25).

In order to investigate the origin of the observed better yield and lower diastereoselectivity using the trimethylamine-based ylide, as compared to DABCO, quinuclidine, and Et3N derivatives, we carried out DFT calculations. We have computed the free energy profile of the reaction of PhCHNMe3 ylide with benzaldehyde at the B3LYP-D3/6-311+G**(DCM)//B3LYP/6-31G*(DCM) level of theory (Table [2] and Figure [1]).[13] For purpose of comparison with Aggarwal’s results, we also considered DABCO and quinuclidine derivatives.[5c]

Zoom Image
Figure 1 Calculated free energy profile (kcal/mol) for the epoxidation reaction between PhCHNMe3 ylide and benzaldehyde. See Supporting Information for the estimation of free energy barrier to addition.

Our calculations are in good agreement with the previously predicted (partial) reversibility of betaine formation in reaction of quinuclidine-based ylide and the less or non-reversible addition step in the case of Me3N and DABCO derivatives.[14] While for trimethylamine-based ylides this reduced reversibility in betaine formation can be accounted for by a better leaving group ability (see free energy barriers from anti-betaine in brackets in Table [2]), for DABCO derivatives the main reason is in fact a more exothermic betaine formation.

Accordingly, the generally low diastereoselectivity observed in reaction of trimethylamine-based ylides can be accounted for by a less reversible addition step which is therefore the selectivity determining step. This latter involving no significant barrier,[15] it is expected to occur with no (or very low) selectivity. Conversely, for quinuclidine-based ylides the addition is predicted to be (partially) reversible, meaning that selectivity must be determined at the elimination step, which explains the high preference for trans-epoxides in this case.[5]

The observed increase in trans diastereoselectivity with electron-withdrawing character of ylide substituents (see Table [1], entries 7–9) can be explained by their impact on the degree of reversibility of addition, and hence on the origin of the selectivity-determining step. Indeed, stabilization of the ylide by EWG increases addition reversibility because of a less exothermic betaine formation and an increased free energy barrier to elimination[16] (Figure [2]). In the case of EWG substitution, selectivity is thus determined at the elimination step, and the predicted relative higher stability of trans-TSelim over its cis isomer (by 1.9 kcal/mol for 4-F3CC6H4CHNMe3 ylide) accounts for the observed high selectivity.

Table 2Computed Relative Free Energies (kcal/mol)

R3N =

Me3N

quinuclidine

DABCO

trans

cis

trans

cis

trans

cis

TSadd

  –a

 –a

 –a

 –a

 –a

 –a

syn betaine

  –7.7

 –7.9

 –8.8

 –9.1

–15.8

–15.9

TSrot

  1.6

 –0.3

 n.d.b

 n.d.b

 n.d.b

 n.d.b

anti betaine

  –2.1

 –1.2

 –3.6

 –2.6

–10.3

–9.1

TSelim

  1.8 (3.9)c

 3.1 (4.3)c

 2.8 (6.4)c

 5.5 (8.2)c

–4.2 (6.1)c

–1.4 (7.7)c

Epoxide + R3N

–26.3

–23.1

–22.8

–19.6

–31.5

–28.3

a Betaine formation occurs without enthalpic barrier.[15]

b n.d. = not determined.
c Free energy barrier from anti betaine reported in brackets.

Zoom Image
Figure 2 Effect of EWG substitution. Calculated free energy profile (kcal/mol) for the epoxidation reaction between 4-F3CC6H4CHNMe3 ylide and benzaldehyde.

In terms of yield, the better results obtained for reaction of Me3N-based ylides, as compared to DABCO and quinuclidine derivatives, can be explained by the higher reactivity toward epoxidation of the former due to a better leaving group ability, which then decreases the occurrence of competitive degradation reactions.[17]

Zoom Image
Scheme 2 Syntheses of epoxides 5 using phenyldiazomethane 7 as the ammonium ylide precursor

One very versatile strategy to carry out ylide-mediated epoxidations without the need of strong basic conditions relies on the use of diazo compounds as ylide precursors. This strategy is commonly used for sulfonium ylide mediated epoxidations, where the diazo compound can be used to obtain ylides in situ upon reaction with a metal catalyst (usually Rh- or Cu-based) and a (catalytic amount of) sulfide as was impressively shown by Aggarwal’s group.[18] Interestingly, it was also shown that by changing from sulfonium to tellurium ylides,[19] the corresponding alkenes are obtained instead of the epoxides. Much to our surprise (and to the best of our knowledge) no analogous reports for ammonium ylides have been reported so far. We thus tested the feasibility of such a transformation by using phenyldiazomethane 7 in combination with Rh2(OAc)4 and DABCO (8) as the amine component (Scheme [2]). After thorough screening of different conditions, we found that such a reaction is possible in principle, albeit with low yield only. These low yields can mainly be attributed to rather pronounced side-reactions such as the dimerization of the diazo compound to stilbene derivatives or different carbene insertion products. Also, variations of the metal catalyst or the in situ generation of the diazo species under phase-transfer conditions (as reported previously[18d]) were not beneficial. Attempts to use the amine component in a catalytic fashion also were not very successful. Thus, it can be concluded that, although possible, the use of diazo compounds as ammonium ylide precursors for epoxide formations is not as effective as the comparable use for sulfonium ylide mediated reactions.

In conclusion, a high-yielding protocol for the synthesis of stilbene oxides using ammonium ylides has been developed.[19] [20] The nature of the amine leaving group plays a crucial role herein and it was found that trimethylamine gives significantly higher yields than DABCO or quinuclidine. The nature of the amine group also influences the diastereoselectivity in these reactions. A detailed DFT study allows understanding the different yields and selectivities when using different amine leaving groups and aryl substituents.


#

Acknowledgment

This work was supported by the Austrian Science Funds (FWF): Project No. P26387-N28. Computational resources have been provided by the supercomputing facilities of the Université catholique de Louvain (CISM/UCL) and the Consortium des Équipements de Calcul Intensif en Fédération Wallonie Bruxelles (CÉCI) funded by the Fond de la Recherche Scientifique de Belgique (F.R.S.-FNRS) under convention 2.5020.11. The NMR spectrometers used were acquired in collaboration with the University of South Bohemia (CZ) with financial support from the European Union through the EFRE INTERREG IV ETC-AT-CZ program (project M00146, ‘RERI-uasb’). RR is a Chercheur qualifié of the F.R.S.-FNRS.

Supporting Information

  • References and Notes


    • For selected reviews, see:
    • 1a Aggarwal VK In Comprehensive Asymmetric Catalysis . Jacobsen EN, Pfaltz A, Yamamoto H. Springer; New York: 1999. Vol. 2 679
    • 1b McGarrigle EM, Myers EL, Illa O, Shaw MA, Riches SL, Aggarwal VK. Chem. Rev. 2007; 107: 5841

      For seminal work on sulfur ylides, see:
    • 2a Corey EJ, Chaykovsky M. J. Am. Chem. Soc. 1962; 84: 867
    • 2b Corey EJ, Chaykovsky M. J. Am. Chem. Soc. 1965; 87: 1353
    • 2c Johnson AW, Lacount RB. Chem. Ind. 1958; 1440

      For examples highlighting the use of sulfur ylides in epoxidation reactions, see:
    • 3a Zhou Y.-G, Hou X.-L, Dai L.-X, Xia L.-J, Tang M.-H. J. Chem. Soc., Perkin Trans. 1 1999; 77
    • 3b Zanardi J, Leriverend C, Aubert D, Julienne K, Metzner P. J. Org. Chem. 2001; 66: 5620
    • 3c Aggarwal VK, Hynd G, Picoul W, Vasse J.-L. J. Am. Chem. Soc. 2002; 124: 9964
    • 3d Aggarwal VK, Richardson J. Chem. Commun. 2003; 2644
    • 3e Davoust M, Briere J.-F, Jaffres P.-A, Metzner P. J. Org. Chem. 2005; 70: 4166
    • 3f Aggarwal VK, Charmant JP. H, Fuentes D, Harvey JN, Hynd G, Ohara D, Picoul W, Robiette R, Smith C, Vasse J.-L, Winn CL. J. Am. Chem. Soc. 2006; 128: 2105
    • 3g Illa O, Arshad M, Ros A, McGarrigle EM, Aggarwal VK. J. Am. Chem. Soc. 2010; 132: 1828
    • 3h Illa O, Namutebi M, Saha C, Ostovar M, Chen CC, Haddow MF, Nocquet-Thibault S, Lusi M, McGarrigle EM, Aggarwal VK. J. Am. Chem. Soc. 2013; 135: 11951

      For ammonium ylide mediated cyclopropanations, see:
    • 4a Gaunt MJ, Johansson CC. C. Chem. Rev. 2007; 107: 5596
    • 4b Johansson CC. C, Bremeyer N, Ley SV, Owen DR, Smith SC, Gaunt MJ. Angew. Chem. Int. Ed. 2006; 45: 6024
    • 4c Papageorgiou CD, Cubillo de Dios MA, Ley SV, Gaunt MJ. Angew. Chem. Int. Ed. 2004; 43: 4641
    • 4d Bremeyer N, Smith SC, Ley SV, Gaunt MJ. Angew. Chem. Int. Ed. 2004; 43: 2681
    • 4e Papageorgiou CD, Ley SV, Gaunt MJ. Angew. Chem. Int. Ed. 2003; 42: 828

      For benzylic ammonium ylide mediated epoxidations, see:
    • 5a Kimachi T, Kinoshita H, Kusaka K, Takeuchi Y, Aoe M, Ju-ichi M. Synlett 2005; 842
    • 5b Kinoshita H, Ihoriya A, Ju-ichi M, Kimachi T. Synlett 2010; 2330
    • 5c Robiette R, Conza M, Aggarwal VK. Org. Biomol. Chem. 2006; 4: 621
    • 5d Xiao X, Lin D, Tong S, Mo H. Synlett 2011; 2823

      For cyano-stabilized ammonium ylide mediated epoxidations, see:
    • 6a Jonczyk A, Konarska A. Synlett 1999; 1085
    • 6b Kowalkowska A, Sucholbiak D, Jonczyk A. Eur. J. Org. Chem. 2005; 925
    • 6c Alex A, Larmanjat B, Marrot J, Couty F, David O. Chem. Commun. 2007; 2500

      For amide-stabilized ammonium ylide mediated epoxidations, see:
    • 7a Waser M, Herchl R, Müller N. Chem. Commun. 2011; 47: 2170
    • 7b Herchl R, Stiftinger M, Waser M. Org. Biomol. Chem. 2011; 9: 7023
    • 7c Pichler M, Novacek J, Robiette R, Poscher V, Himmelsbach M, Monkowius U, Müller N, Waser M. Org. Biomol. Chem. 2015; 13: 2092
    • 7d Novacek, J.; Roiser, L.; Zielke, K.; Robiette, R.; Waser, M. Chem. Eur. J. 2016, DOI: 10.1002/chem.201602052.

      For ammonium ylide mediated aziridinations, see:
    • 8a Yadav LD. S, Kapoor R Garima Synlett 2009; 3123
    • 8b Aichhorn S, Gururaja GN, Reisinger M, Waser M. RSC Adv. 2013; 3: 4552
  • 9 Aggarwal VK, Harvey JN, Robiette R. Angew. Chem. Int. Ed. 2005; 44: 5468
  • 10 For a recent report on selenium ylide mediated epoxidations, see: Banach A, Scianowski J, Uzarewicz-Baig M, Wojtczak A. Eur. J. Org. Chem. 2015; 3477
  • 11 Ylide addition onto the aldehyde occurs without enthalpic barrier and thus no significant selectivity is expected for this step.
  • 12 Winberg HE, Fawcett FS, Mochel WE, Theobald CW. J. Am. Chem. Soc. 1960; 82: 1428
  • 13 Computations have been carried out at the B3LYP-D3/6-311+G(d,p)//B3LYP/6-31G* level of theory, including a continuum description of dichloromethane solvent for both the geometry optimization and the single-point energy calculation using Jaguar, version 8.5; Schrodinger, Inc.: New York, 2014. Thermal and entropic contributions to free energy were computed by performing frequency calculations at the B3LYP/6-31G (d) level of theory using the fine DFT grid within Jaguar. See ESI for full computational details and data.
  • 14 In the earlier computational study (ref. 5c) only electronic energies were obtained, but no entropic and thermal contributions were included.
  • 15 Betaine formation occurs without enthalpic barrier. A set of constrained geometry optimization at successively smaller values of the C–C distance showed that the interaction between reactants is uniformly attractive. The free-energy barrier of a diffusion controlled reaction can, however, be estimated at 3.7 kcal/mol in THF at 40 °C (see Supporting Information for details).
  • 16 Robiette R, Trieu-Van T, Aggarwal VK, Harvey JN. J. Am. Chem. Soc. 2016; 138: 734
  • 17 In contrast to DABCO, quinuclidine, or triethylammonium salts the herein used trimethylammonium salts cannot undergo Hofmann elimination reactions. A possible side reaction that sometimes occurs with all these ylides is the Stevens rearrangement: Maeda Y, Sato Y. J. Chem. Soc., Perkin Trans. 1 1997; 1491

    • For seminal studies, see:
    • 18a Aggarwal VA, Abdel-Rahman H, Jones RV. H, Lee HY, Reid BD. J. Am. Chem. Soc. 1994; 116: 5973
    • 18b Aggarwal VA, Abdel-Rahman H, Jones RV. H, Standen MC. H. Tetrahedron Lett. 1995; 36: 1731
    • 18c Aggarwal VA, Abdel-Rahman H, Fan L, Jones RV. H, Standen MC. H. Chem. Eur. J. 1996; 2: 1024
    • 18d Aggarwal VK, Alonso E, Bae I, Hynd G, Lydon KM, Palmer MJ, Patel M, Porcelloni M, Richardson J, Stenson RA, Studley JR, Vasse J.-L, Winn CL. J. Am. Chem. Soc. 2003; 125: 10926
  • 19 Zhu S, Xing C, Pang W, Zhu S. Tetrahedron Lett. 2006; 47: 5897
  • 20 General Epoxidation Procedure Ammonium salt 6 (1 equiv) was suspended in dry THF (0.05 M) and stirred at 40 °C. t-BuOK (4 equiv) was added, and the mixture was stirred vigorously. After 10 min, 2 equiv of aldehyde 2 were added, and the mixture was stirred for 3 h at 40 °C. The reaction was then quenched by addition of a half-saturated NaCl solution. After phase separation, the aqueous phase was extracted three times with CH2Cl2, and the combined organic phases were dried with Na2SO4 and evaporated to dryness. Purification by column chromatography (gradient of heptanes and EtOAc) gave the corresponding epoxides in the reported yields as a mixture of diastereomers. Compound 5a: Obtained in 93% (trans/cis = 66:34) on a 0.5 mmol scale as a white solid. Analytical data match those reported previously.5,21 Selected Data 1H NMR (300 MHz, δ, CDCl3, 298 K):trans isomer: 3.94 (s, 2 H), 7.36–7.49 (m, 5 H);cis isomer: 4.42 (s, 2 H), 7.18–7.28 (m, 5 H) ppm. ESI-HRMS: m/zcalcd for C14H12O: 197.0961 [M + H]+; found: 197.0961.
  • 21 Robinson MW. C, Davies AM, Buckle R, Mabbett I, Taylor SH, Graham AE. Org. Biomol. Chem. 2009; 7: 2559

  • References and Notes


    • For selected reviews, see:
    • 1a Aggarwal VK In Comprehensive Asymmetric Catalysis . Jacobsen EN, Pfaltz A, Yamamoto H. Springer; New York: 1999. Vol. 2 679
    • 1b McGarrigle EM, Myers EL, Illa O, Shaw MA, Riches SL, Aggarwal VK. Chem. Rev. 2007; 107: 5841

      For seminal work on sulfur ylides, see:
    • 2a Corey EJ, Chaykovsky M. J. Am. Chem. Soc. 1962; 84: 867
    • 2b Corey EJ, Chaykovsky M. J. Am. Chem. Soc. 1965; 87: 1353
    • 2c Johnson AW, Lacount RB. Chem. Ind. 1958; 1440

      For examples highlighting the use of sulfur ylides in epoxidation reactions, see:
    • 3a Zhou Y.-G, Hou X.-L, Dai L.-X, Xia L.-J, Tang M.-H. J. Chem. Soc., Perkin Trans. 1 1999; 77
    • 3b Zanardi J, Leriverend C, Aubert D, Julienne K, Metzner P. J. Org. Chem. 2001; 66: 5620
    • 3c Aggarwal VK, Hynd G, Picoul W, Vasse J.-L. J. Am. Chem. Soc. 2002; 124: 9964
    • 3d Aggarwal VK, Richardson J. Chem. Commun. 2003; 2644
    • 3e Davoust M, Briere J.-F, Jaffres P.-A, Metzner P. J. Org. Chem. 2005; 70: 4166
    • 3f Aggarwal VK, Charmant JP. H, Fuentes D, Harvey JN, Hynd G, Ohara D, Picoul W, Robiette R, Smith C, Vasse J.-L, Winn CL. J. Am. Chem. Soc. 2006; 128: 2105
    • 3g Illa O, Arshad M, Ros A, McGarrigle EM, Aggarwal VK. J. Am. Chem. Soc. 2010; 132: 1828
    • 3h Illa O, Namutebi M, Saha C, Ostovar M, Chen CC, Haddow MF, Nocquet-Thibault S, Lusi M, McGarrigle EM, Aggarwal VK. J. Am. Chem. Soc. 2013; 135: 11951

      For ammonium ylide mediated cyclopropanations, see:
    • 4a Gaunt MJ, Johansson CC. C. Chem. Rev. 2007; 107: 5596
    • 4b Johansson CC. C, Bremeyer N, Ley SV, Owen DR, Smith SC, Gaunt MJ. Angew. Chem. Int. Ed. 2006; 45: 6024
    • 4c Papageorgiou CD, Cubillo de Dios MA, Ley SV, Gaunt MJ. Angew. Chem. Int. Ed. 2004; 43: 4641
    • 4d Bremeyer N, Smith SC, Ley SV, Gaunt MJ. Angew. Chem. Int. Ed. 2004; 43: 2681
    • 4e Papageorgiou CD, Ley SV, Gaunt MJ. Angew. Chem. Int. Ed. 2003; 42: 828

      For benzylic ammonium ylide mediated epoxidations, see:
    • 5a Kimachi T, Kinoshita H, Kusaka K, Takeuchi Y, Aoe M, Ju-ichi M. Synlett 2005; 842
    • 5b Kinoshita H, Ihoriya A, Ju-ichi M, Kimachi T. Synlett 2010; 2330
    • 5c Robiette R, Conza M, Aggarwal VK. Org. Biomol. Chem. 2006; 4: 621
    • 5d Xiao X, Lin D, Tong S, Mo H. Synlett 2011; 2823

      For cyano-stabilized ammonium ylide mediated epoxidations, see:
    • 6a Jonczyk A, Konarska A. Synlett 1999; 1085
    • 6b Kowalkowska A, Sucholbiak D, Jonczyk A. Eur. J. Org. Chem. 2005; 925
    • 6c Alex A, Larmanjat B, Marrot J, Couty F, David O. Chem. Commun. 2007; 2500

      For amide-stabilized ammonium ylide mediated epoxidations, see:
    • 7a Waser M, Herchl R, Müller N. Chem. Commun. 2011; 47: 2170
    • 7b Herchl R, Stiftinger M, Waser M. Org. Biomol. Chem. 2011; 9: 7023
    • 7c Pichler M, Novacek J, Robiette R, Poscher V, Himmelsbach M, Monkowius U, Müller N, Waser M. Org. Biomol. Chem. 2015; 13: 2092
    • 7d Novacek, J.; Roiser, L.; Zielke, K.; Robiette, R.; Waser, M. Chem. Eur. J. 2016, DOI: 10.1002/chem.201602052.

      For ammonium ylide mediated aziridinations, see:
    • 8a Yadav LD. S, Kapoor R Garima Synlett 2009; 3123
    • 8b Aichhorn S, Gururaja GN, Reisinger M, Waser M. RSC Adv. 2013; 3: 4552
  • 9 Aggarwal VK, Harvey JN, Robiette R. Angew. Chem. Int. Ed. 2005; 44: 5468
  • 10 For a recent report on selenium ylide mediated epoxidations, see: Banach A, Scianowski J, Uzarewicz-Baig M, Wojtczak A. Eur. J. Org. Chem. 2015; 3477
  • 11 Ylide addition onto the aldehyde occurs without enthalpic barrier and thus no significant selectivity is expected for this step.
  • 12 Winberg HE, Fawcett FS, Mochel WE, Theobald CW. J. Am. Chem. Soc. 1960; 82: 1428
  • 13 Computations have been carried out at the B3LYP-D3/6-311+G(d,p)//B3LYP/6-31G* level of theory, including a continuum description of dichloromethane solvent for both the geometry optimization and the single-point energy calculation using Jaguar, version 8.5; Schrodinger, Inc.: New York, 2014. Thermal and entropic contributions to free energy were computed by performing frequency calculations at the B3LYP/6-31G (d) level of theory using the fine DFT grid within Jaguar. See ESI for full computational details and data.
  • 14 In the earlier computational study (ref. 5c) only electronic energies were obtained, but no entropic and thermal contributions were included.
  • 15 Betaine formation occurs without enthalpic barrier. A set of constrained geometry optimization at successively smaller values of the C–C distance showed that the interaction between reactants is uniformly attractive. The free-energy barrier of a diffusion controlled reaction can, however, be estimated at 3.7 kcal/mol in THF at 40 °C (see Supporting Information for details).
  • 16 Robiette R, Trieu-Van T, Aggarwal VK, Harvey JN. J. Am. Chem. Soc. 2016; 138: 734
  • 17 In contrast to DABCO, quinuclidine, or triethylammonium salts the herein used trimethylammonium salts cannot undergo Hofmann elimination reactions. A possible side reaction that sometimes occurs with all these ylides is the Stevens rearrangement: Maeda Y, Sato Y. J. Chem. Soc., Perkin Trans. 1 1997; 1491

    • For seminal studies, see:
    • 18a Aggarwal VA, Abdel-Rahman H, Jones RV. H, Lee HY, Reid BD. J. Am. Chem. Soc. 1994; 116: 5973
    • 18b Aggarwal VA, Abdel-Rahman H, Jones RV. H, Standen MC. H. Tetrahedron Lett. 1995; 36: 1731
    • 18c Aggarwal VA, Abdel-Rahman H, Fan L, Jones RV. H, Standen MC. H. Chem. Eur. J. 1996; 2: 1024
    • 18d Aggarwal VK, Alonso E, Bae I, Hynd G, Lydon KM, Palmer MJ, Patel M, Porcelloni M, Richardson J, Stenson RA, Studley JR, Vasse J.-L, Winn CL. J. Am. Chem. Soc. 2003; 125: 10926
  • 19 Zhu S, Xing C, Pang W, Zhu S. Tetrahedron Lett. 2006; 47: 5897
  • 20 General Epoxidation Procedure Ammonium salt 6 (1 equiv) was suspended in dry THF (0.05 M) and stirred at 40 °C. t-BuOK (4 equiv) was added, and the mixture was stirred vigorously. After 10 min, 2 equiv of aldehyde 2 were added, and the mixture was stirred for 3 h at 40 °C. The reaction was then quenched by addition of a half-saturated NaCl solution. After phase separation, the aqueous phase was extracted three times with CH2Cl2, and the combined organic phases were dried with Na2SO4 and evaporated to dryness. Purification by column chromatography (gradient of heptanes and EtOAc) gave the corresponding epoxides in the reported yields as a mixture of diastereomers. Compound 5a: Obtained in 93% (trans/cis = 66:34) on a 0.5 mmol scale as a white solid. Analytical data match those reported previously.5,21 Selected Data 1H NMR (300 MHz, δ, CDCl3, 298 K):trans isomer: 3.94 (s, 2 H), 7.36–7.49 (m, 5 H);cis isomer: 4.42 (s, 2 H), 7.18–7.28 (m, 5 H) ppm. ESI-HRMS: m/zcalcd for C14H12O: 197.0961 [M + H]+; found: 197.0961.
  • 21 Robinson MW. C, Davies AM, Buckle R, Mabbett I, Taylor SH, Graham AE. Org. Biomol. Chem. 2009; 7: 2559

Zoom Image
Scheme 1 Recent results in the field of ammonium ylide mediated epoxidations and targeted (high yielding) synthesis of stilbene oxides 5 with trimethylamine-based ammonium salts 6.
Zoom Image
Figure 1 Calculated free energy profile (kcal/mol) for the epoxidation reaction between PhCHNMe3 ylide and benzaldehyde. See Supporting Information for the estimation of free energy barrier to addition.
Zoom Image
Figure 2 Effect of EWG substitution. Calculated free energy profile (kcal/mol) for the epoxidation reaction between 4-F3CC6H4CHNMe3 ylide and benzaldehyde.
Zoom Image
Scheme 2 Syntheses of epoxides 5 using phenyldiazomethane 7 as the ammonium ylide precursor