Deprotonation–electrophile trapping of terminal epoxides

Abstract

Organolithium-induced deprotonation of terminal epoxides in the presence of appropriate diamine ligands allows trapping with a range of electrophiles, yielding functionalised di- and tri-substituted epoxides in good yields and with control of stereochemistry at the epoxide.

---

Introduction

In view of the many roles that epoxides play in organic synthesis, efficient (asymmetric) methods for the direct introduction of this important functional group would be highly valued.¹ At the outset of our studies the coupling of oxiranyl organometallics with electrophiles was one method for incorporation of epoxides that had yet to be fully developed, especially as an asymmetric transformation.² The first systematic studies of stabilised oxiranyl anions were reported in 1976 by Eisch and Galle.³ Their pioneering research demonstrated the apparent necessity of an anion-stabilising group attached to the epoxide ring to enhance the facility of metalation of epoxides and to increase the thermal stability of the oxiranyl anions produced. Metalation of cis-disubstituted trialkylsilyl epoxides and reaction of the resultant oxiranyl anions with simple electrophiles, such as Me₃SiCl, Me₃SnCl, CO₂, MeI, amides, and D₂O, led to high yields of the desired products, with complete configurational stability being observed. This early work has subsequently been extended in many directions, expanding the range of electrophiles and anion-stabilising groups, and even generating and trapping destabilised (alkyl-substituted) oxiranyl anions by ligand exchange (desulfinylation) of sulfinyloxiranes.⁴ Until the last few years it was still generally believed that, as stated in a 1996 review on the synthesis and utility of oxiranyl anions, the “metal–hydrogen exchange reaction is obviously not useful for generation of nonstabilised (H substituent) oxiranyl anions”.^2a

Arising from our studies concerning the lithiation–rearrangement of epoxides,⁵ we envisioned that an appropriate ligand might be able to serve the dual role of accelerating deprotonation and reducing the rate of oxiranyl anion decomposition, enabling the direct deprotonation–electrophile trapping of unfunctionalised terminal epoxides. Compared to lithiation–electrophile trapping using epoxides of medium-sized cycloalkenes, terminal epoxides (e.g.1a) represent a greater challenge due to comparatively reduced epoxide C–H acidity,⁶ combined with a higher propensity of both the starting and α-lithiated epoxides to undergo ring-opening reactions. In this paper we report a full account of our studies in this area.⁷

Results and discussion

We started our investigations with a study of various conditions for the deprotonation–deuteration of the simple terminal epoxide 1,2-epoxydodecane 1a (Scheme 1). The reaction was first carried out in Et₂O in the absence of a ligand. Whilst no reaction occurred at −90 °C, we performed the addition of a solution of 1a in Et₂O to ⁱPrLi (2.5 eq.) in Et₂O at −78 °C over a period of 10 min followed after 1.5 min by D₂O quench. This run led after chromatography to recovered 1a (35% yield, 0% D incorporation), trans-alkene 3a (R = ⁱPr, 34%) arising from reductive alkylation,⁸ and alcohol 4a (R = ⁱPr, 22%). The same reaction, but carried out in the presence of TMEDA (2.5 eq.) was found to accelerate the deprotonation step. In this case, in Et₂O at −78 °C or −90 °C the alkene dominated the reaction mixture (1a (0% D) : 3a = 20 : 80 and 50 : 50, respectively) with no 4a (R = ⁱPr) detected. On switching from Et₂O to THF, no reaction of 1a was observed at −90 °C with ⁱPrLi in the presence of TMEDA.

Scheme 1 Deprotonation–deuteration studies of 1,2-epoxydodecane 1a.

Significantly, replacing TMEDA with commercially available (−)-sparteine 5 (Fig. 1) as the ligand, we found that D incorporation was possible. When 1a was added to a mixture of 5 (2.5 eq.) and either ⁿBuLi, ^tBuLi or ⁱPrLi (2.5 eq.) in Et₂O at −90 °C, followed after 15 min by D₂O quench, 1a was recovered with 0% D, 10% D and 22% D incorporation, respectively. Increasing the reaction time to 1 h with ⁱPrLi in Et₂O before D₂O quench led to 1a in 50% yield with 45% D incorporation, along with 3a (R = ⁱPr, 18%) and 4a (R = ⁱPr, 11%). The reaction performed under the same conditions (ⁱPrLi–5 at −90 °C for 1 h) but with hexane as solvent allowed us to recover 1a in 60% yield with the same level of D incorporation but with significantly less 3a (R = ⁱPr, 6%) (Table 1, entry 1). Better results were obtained using ^sBuLi in hexane, which after only 15 min at −90 °C gave 1a in 70% yield with 75% D incorporation (Table 1, entry 3) with 9% of 3a (R = ^sBu) isolated. Longer reaction times resulted in increased formation of by-products and diminished recovery of 1a (Table 1, entries 2 and 4). Incorporation of deuterium occurred exclusively trans to the alkyl substituent on the oxirane ring.

Table 1 Effect of experimental conditions on the lithiation–deuteration of 1,2-epoxydodecane 1a in hexane at −90 °C

Entry^a	RLi	Ligand	Time/min	Amount of D^b in 1a (%)	Yield^c of 1a (%)
a Reactions were carried out by the addition over 10 min of a solution of 1a (1 eq.) in hexane to a mixture of RLi–ligand (2.5 eq. each) in hexane at −90 °C, followed by the addition of CD3OD after the time indicated. b Determined by ^1H NMR spectroscopy. c Yield following chromatography. d 4a (R = ^sBu) was only observed (5%) in this experiment.
1	^iPrLi	5	60	46	60
2	^iPrLi	5	180	50	50
3	^sBuLi	5	15	75	70
4^d	^sBuLi	5	60	90	40
5	^sBuLi	6	15	45	82
6	^sBuLi	7	15	50	85
7	^sBuLi	7	60	52	80
8	^sBuLi	8	15	63	91
9	^sBuLi	8	60	70	75

---

Fig. 1 Diamine ligands used in this study.

The above encouraging results with 5 as ligand prompted us to examine ligands such as 6–8⁹ that all possess the 3,7-diazabicyclo[3.3.1]nonane structural feature of 5 (Table 1, entries 5–8). For these ligands, it was found that the deprotonation step was slower than with 5. However, in these experiments little or no alkene was formed, which demonstrates the ability of these ligands to stabilise the intermediate oxiranyl anion 2a. The best results were obtained using ^sBuLi–8 in hexane at −90 °C (Table 1, entries 8 and 9). D₂O quenching after 1 h afforded 1a in 75% yield with 70% D incorporation along with 3a (R = ^sBu, 8%). Using ^sBuLi–8 in Et₂O was less effective and provided after 1 h a mixture of 1a and 3a (96 : 4 respectively), with only 48% D incorporation in 1a.

Attempts to trap the intermediate oxiranyl anion 2a with other external electrophiles (Me₃SiCl, ClCO₂Me, PhCHO, EtCHO) were found to be initially unsuccessful under the above conditions. However, reactions carried out with Me₃SiCl present during the generation of the oxiranyl anion led to α,β-epoxysilanes 9 in good yields (Scheme 2, Table 2). α,β-Epoxysilanes are interesting synthons for organic chemistry since, for example, they can be hydrolysed to give carbonyl compounds, undergo regio- and stereospecific α-opening by a range of nucleophiles to give diastereomerically pure β-hydroxysilanes and can serve as vinyl cation equivalents for the preparation of olefins of known stereochemistry.¹⁰

Table 2 Direct synthesis of α,β-epoxysilanes 9 from epoxides 1

Entry^a	1		Time/h	Yield of 9 (%)^b
a Reactions were carried out at −90 °C (−83 °C for entry 3) for the time indicated, followed by warming to −80 °C over 5 min (to −50 °C over 30 min for entries 1 and 7–10). b Yield following chromatography. c Ligand 5 used in place of ligand 8.
1		1a	2		9a
2		1b	2.5		9b
3		1c	4.5		9c
4		1d	2		9d
5		1e	3		9e
6		1f	4.5		9f
7		1g	3		9g
8^c		1a	0.25		(1S,2S)-9a
9^c		(S)-1a	0.25		(1S,2S)-9a
10^c		(R)-1a	0.25		(1R,2R)-9a

---

Scheme 2 Silylation of terminal epoxides 1.

As shown in Table 2, the process is compatible with a variety of functionalised epoxides¹¹ leading to stereochemically pure trans-α,β-epoxysilanes (entries 2–5).¹² The reaction can also be performed with 2,2-disubstituted epoxides affording trisubstituted epoxides (entries 6 and 7).¹³ For the unsymmetrical epoxide 1g (entry 7), silylation was found to proceed with a high degree of selectivity (97 : 3) trans to the phenyl substituent.^13b

Chiral non-racemic ligand 5 could in theory effect kinetic resolution on a racemic epoxide, and this was found to be the case to an extent (entry 8, ca. 30% ee for the enantiomer shown based on the specific rotation of the product). When employing 0.6 eq. each of ^sBuLi, 5 and Me₃SiCl, the product (1S,2S)-9a was returned in 18% yield and ca. 46% ee (based on its specific rotation), along with starting material 1a (63%) and alkene 3a (R = ^sBu) from reductive alkylation (5%). Importantly, it was demonstrated that ^sBuLi–5 could be used to silylate either enantiomer of epoxide 1a in high yield and with complete trans selectivity (entries 9 and 10). This is useful given the commercial availability of ligand 5.

A minor side-reaction of bis-silylation was observed in the silylation chemistry with sparteine 5 as the ligand. This led to the formation of an epoxydisilane 10, presumably via lithiation–silylation of 9 (Scheme 3, ca. 10% bis-silylation observed for R = C₁₀H₂₁).^3,10,13a,14

Scheme 3 Epoxydisilanes 10via bis-silylation of epoxides 1.

Methods to access epoxydisilanes 10 are of interest, as epoxydisilanes can be readily converted into acylsilanes and 1-halovinylsilanes.¹⁵ Acylsilanes in particular are finding increasing utility in synthesis.¹⁶ With this in mind it was decided to investigate the direct synthesis of epoxydisilanes 10 from readily available terminal epoxides 1.

Initially, the equivalents of Me₃SiCl and ^sBuLi–sparteine complex were increased from the original 1.3 eq. to 3 or 5 eq. This led to a considerable increase in the formation of the epoxydisilane 10a, however epoxysilane 9a and traces of the starting terminal epoxide 1a were still observed by crude ¹H NMR analysis (1a : 9a : 10a returned in ratios of 0.15 : 0.4 : 1 and 0.05 : 0.3 : 1 respectively). On the timescale of the reaction, and in the presence of excess reagents, competitive silylation of the ^sBuLi could contribute to incomplete disilylation. It was found that it is important to convert all of the epoxysilane 9 into the epoxydisilane 10, because of their similar mobility characteristics on column chromatography. Bearing this in mind, the reaction conditions were investigated in more detail.

To avoid observing epoxysilane 9 in the final product profile, it appeared advantageous to obtain complete conversion of the terminal epoxide 1 to the epoxysilane 9, before the addition of further reagents. To hopefully achieve this, the original reaction conditions (1.3 eq. of reagents at −90 °C, followed by warming to −50 °C over 1 h) were modified by then re-cooling back to −90 °C and adding a further 2 eq. of Me₃SiCl and ^sBuLi and leaving for 15 min at −90 °C, then warming to room temperature over 18 h before work-up. Under this protocol, it was found that, as with the monosilylation procedure, hexane proved a better solvent than Et₂O (1a : 9a : 10a returned in ratios of 0.3 : 1.3 : 1 and 2 : 3 : 1 respectively). To further improve this, instead of just ^sBuLi being added after warming to −50 °C and back to −90 °C, ^sBuLi–sparteine complex was added. This resulted in an encouraging improvement in the proportion of epoxydisilane 10a compared to the epoxysilane 9a (1a : 9a : 10a returned in ratio of 0 : 0.2 : 1 respectively). Indeed, warming to temperatures above −50 °C as part of a warm–cool cycle led to no epoxysilane 9a at all in the crude product; however, more noticeable alkenyl signals in the ¹H NMR spectra of the crude products suggested that ^sBuLi-induced reductive alkylation of epoxides 1a and 9a had become a significant side-reaction.¹⁷ A similar trend to the monosilylation chemistry was shown in the effect of ligand with ^sBuLi–sparteine complex giving superior results to ^sBuLi–TMEDA complex (1a : 9a : 10a returned in ratio of 0 : 0.2 : 1 and 0.4 : 0.5 : 1 respectively).

Further insight into the reaction pathway occurring during the above warm–cool protocol was achieved by quenching with CD₃OD at different stages, followed by GC-MS analysis to determine product ratios (after appropriate calibration graphs had been constructed) and D incorporations. The observation of D incorporation in 1 and/or 9 was expected to indicate the proportion of any oxiranyl anions 2 and 11 (Scheme 4) present at the time the reaction was quenched.

Scheme 4 In-depth studies of the bis-silylation reaction.

After only 15 min following the addition of Me₃SiCl and ^sBuLi–sparteine complex (all 1.3 eq.) to the epoxide 1a (R = C₁₀H₂₁) at −90 °C a ratio of 1a : 9a of 0.5 : 1 with 13% D incorporation in 1a was obtained. It was established that complete reaction of the terminal epoxide 1a and 2a was achieved if the reaction was warmed to −50 °C over 5 min and then left at −50 °C for 1 h. As expected at that point the major compound present was epoxysilane 9a (traces of alkene signals were also observed by ¹H NMR spectroscopy of the crude product). GC-MS analysis indicated a small amount of epoxydisilane 10a was present (9a : 10a ratio of 10 : 1), and a significant proportion of epoxysilane 9a was lithiated (29% D incorporation in 9a). The latter observations are likely to be due to the slight initial excess of ^sBuLi–sparteine complex employed. Following subsequent re-cooling to −90 °C and addition of further Me₃SiCl and ^sBuLi–sparteine complex (all 2 eq.) and stirring for 15 min, there was a modest change to the product profile (9a : 10a ratio of 5 : 1, 22% of 9a deuterated). Thus, from epoxysilane 9a the second deprotonation and silylation do not occur as readily as from the terminal epoxide 1a since, although the oxiranyl anion 11a is stabilised by the α-silyl group, the proton that has to be removed to generate 11a occupies a relatively more hindered site.¹² Compound 11a is also more hindered than 2a to subsequent electrophilic trapping. As the reaction warmed from −90 °C to 0 °C over 7 h, the proportion of epoxysilane 9a gradually fell relative to epoxydisilane 10a [after 2 h (−60 °C) 9a : 10a ratio of 2.5 : 1, with 46% of 9a deuterated; after 4 h (−30 °C) 9a : 10a ratio of 2 : 1, with 62% of 9a deuterated; after 7 h (0 °C) 9a : 10a ratio of 0.1 : 1, with 3% of 9a deuterated]. Therefore, the majority of the second silylation occurred between −30 °C and 0 °C. The stability observed for 11a at −30 °C under the current reaction conditions is noteworthy, given the generally accepted thermal lability of oxiranyl anions.^2b However, Pulido and co-workers have shown that ethylene oxide bearing a tert-butyldiphenylsilyl group could be lithiated, with BuLi–TMEDA complex in THF followed by efficient trapping (93%) with MeI at −50 °C.¹⁸ At 0 °C, a lower yield of methylated epoxide (31%) together with alkene (48%) from reductive alkylation was obtained.

Application of the optimised disilylation conditions (warm to −50 °C over 5 min, maintain for 1 h then re-cool protocol) to a range of terminal epoxides gave the corresponding epoxydisilanes in good yields (Table 3). It was demonstrated that ^sBuLi–5 could be used to bis-silylate either enantiomer of epoxide 1a (entries 6 and 7).

Table 3 Direct synthesis of epoxydisilanes 10 from epoxides 1

Entry	1		10		Yield^a (%)
a Yield following chromatography (mean value of two runs for entries 1–5).
1		1a		10a	61
2		1b		10b	58
3		1c		10c	53
4		1d		10d	71
5		1e		10e	50
6		(S)-1a		(S)-10a	55
7		(R)-1a		(R)-10a	47

---

Interestingly, subjection of two 2,2-disubstituted-1-epoxides (1f and 1g, Table 2) to the reaction conditions led only to the corresponding monosilylated epoxysilanes 9 and it proved impossible to push these reactions past the monosilylation stage.

The introduction of two different electrophiles to a terminal epoxide in a one-flask operation was also possible, albeit in modest yield (37%) by initial silylation followed by addition of benzaldehyde at −30 °C (Scheme 5).

Scheme 5 One-flask introduction of two different electrophiles.

Having established methods for mono-silylation (Me₃SiCl present during generation of the oxiranyl anion) and deuteration (CD₃OD as external electrophile), attention returned to the reaction of oxiranyl anions with other (external) electrophiles, including C–C bond formation. In the original screen of diamine ligands (Table 1) the use of 3,7-dibutyl-3,7-diazabicyclo[3.3.1]nonane 6 (dibutylbispidine, DBB)–^sBuLi^9a,19 (2.5 eq. each) in the lithiation–deuteration of 1a in hexane at −90 °C for 15 min had afforded 82% recovered epoxide 1a with only 45% D incorporation (entry 5).

Compared to ^sBuLi–(−)-sparteine, ^sBuLi–DBB 6 was found to deprotonate 1 at a considerably slower rate, and as other bispidine-type ligands were more efficient, 6 was not studied in more detail at the time. However, further extensive investigations with 6 have revealed that 6 is the preferred ligand for lithiation–electrophile trapping of terminal epoxides.

Initially it was found that employing ^sBuLi–6 over a prolonged reaction time afforded surprising stabilisation of the otherwise labile oxiranyllithium 2a.² Indeed, using ^sBuLi–6 (2.5 eq. each) in hexane at −90 °C for 16 h afforded epoxide 1a in good isolated yield (66%) with a high level of D incorporation (90%, as judged by ¹H NMR), alongside alkene 3a and alcohol 4a in relatively small amounts (16% and 10%, respectively). This result was reproducible; under the same conditions a repeat experiment gave an almost identical result (65% yield of 1a with 90% D, alongside 12% of alkene 3a and 10% alcohol 4a). Subsequent studies showed that a good level of D incorporation could be achieved using ^sBuLi–6 (2.5 eq. each) in hexane at −90 °C after 4 h, accompanied by fairly small levels of alkene 3a and alcohol 4a by-products. Although the use of Et₂O as solvent allowed an enhanced rate of epoxide deprotonation, this was at the expense of slightly increased levels of by-products 3a and 4a, while other solvents (cumene, tert-butylmethyl ether and THF) and different amounts (eq.) of ^sBuLi–6 gave inferior results.

Pleasingly, it was found that following lithiation of epoxide 1a using ^sBuLi–6 (2.5 eq. each) in hexane at −90 °C for 4 h, addition of Bu₃SnCl (2.5 eq., added neat) afforded trans-α,β-epoxystannane 13a in good yield (73% after column chromatography on Florisil^®). Since α,β-epoxystannanes are useful for indirect oxiranyl anion generation via transmetalation,^4a and may also allow subsequent introduction of other electrophiles using cross-coupling techniques,²⁰ an assessment of the stannylation scope was instigated using a range of epoxides.²¹ These studies revealed that conducting the reactions at a reduced reactant concentration (0.025 M for 1 rather than 0.050 M) together with using a shorter lithiation period (1–1.5 h) led to reduced by-product formation and returned the epoxystannane products 13 in good yield (Scheme 6, Table 4).

Table 4 Direct synthesis of stannylepoxides 13 from epoxides 1

Entry^a	1		13		Yield^b (%)
a Solution of epoxide in hexane was added to ^sBuLi–6 (2.5 eq. each) in hexane at −90 °C, and then stirred at −90 °C for 1 h (1.5 h, entry 1; 4 h, entry 4) before addition of Bu3SnCl. b Yield following chromatography on Florisil^® (neutral alumina, entry 4). c Epoxide added neat to ^sBuLi–6 (2.5 eq. each) at −100 °C, then reaction continued at −90 °C.
1		1a		13a	78
2		1b		13b	67
3		1c		13c	72
4		1f		13f	64
5		1h		13h	83
6		1i		13i	78
7		1j		13j	70
8		1k		13k	72
9		1l		13l	90
10		1m		13m	87
11		1n		13n	73
12		(S)-1a		(1S,2S)-13a	65
13^c		(S)-1o		(1S,2S)-13o	54

---

Scheme 6 Direct stannylation of terminal epoxides 1.

Isopropyl and tert-butyl oxiranes gave the best yields of epoxystannanes (entries 9 and 10), but epoxides bearing side-chain functionality were also tolerated (entries 2, 3, 5, 6 and 8). Lithiation of a 2,2-disubstituted epoxide 1f (entry 4) was significantly slower than the monosubstituted epoxides, requiring a 4 h lithiation period. Commercially available (S)-propylene oxide was successfully stannylated (54%, entry 13), as was (S)-1,2-epoxydodecane (65%, entry 12).

An investigation of lithiation–electrophile trapping to achieve C–C bond formation was then undertaken. Lithiation of 1,2-epoxydodecane 1a for 1.5 h, followed by addition of benzaldehyde gave a separable mixture of diastereoisomeric epoxy alcohols²² (75% combined yield, dr 1 : 1, Table 5, entry 1). The lack of diastereoselectivity is consistent with that observed for other oxiranyl anion reactions.¹² The methodology was extended to other aromatic aldehydes as the electrophile (Table 5, entries 2–-4). Non-enolisable aldehydes (cinnamaldehyde and pivaldehyde) were also trapped in good yields (73 and 70% respectively, entries 5 and 6). Potentially enolisable carbonyl electrophiles (isobutyraldehyde and 3-pentanone) were also incorporated, albeit in slightly lower yields (56 and 63% respectively, entries 7 and 8). Acylation using N,N-dimethylbenzamide also proceeded (50%, entry 9).²³

Table 5 Direct C–C bond formation from 1,2-epoxydodecane 1a

Entry^a	Electrophile	Product ^b		Yield^c (%)
a Solution of epoxydodecane 1a in hexane was added to ^sBuLi–6 (2.5 eq. each) in hexane at −90 °C, and then stirred at −90 °C for 1.5 h before addition of the electrophile. b Secondary alcohols obtained with dr 1 : 1. c Yield following chromatography (combined yields given for diastereomeric epoxy alcohols which, with the exception of entry 4, were largely chromatographically separable).
1	PhCHO		14a	75
2	3-ClC6H4CHO		14b	72
3	4-MeOC6H4CHO		14c	57
4	Furfural		14d	67
5	Cinnamaldehyde		14e	73
6	^tBuCHO		14f	70
7	^iPrCHO		14g	56
8	Et2CO		14h	63
9	PhCONMe2		14i	50

---

In conclusion, we have developed a methodology for the deprotonation–electrophile trapping of unfunctionalised terminal epoxides 1. This methodology provides a direct and stereocontrolled route to deuterated epoxides, epoxysilanes 9 and epoxystannanes 13, in addition to epoxydisilanes 10. The methodology has also been applied to C–C bond formation using carbonyl-based electrophiles. The wide availability of terminal epoxide starting materials, and the robustness of methods for their resolution to highly enantiomerically enriched form,^11e renders the methodology described herein as particularly useful. Further research concerning direct epoxide functionalisation and the synthetic utility of the substituted epoxide products is currently underway in our laboratories.

Experimental

General

All reactions requiring anhydrous conditions were conducted in flame- or oven-dried apparatus under an atmosphere of argon. Ethers were distilled from sodium benzophenone ketyl under argon; hexane from CaH₂ under argon. External reaction temperatures are reported unless stated otherwise. Reactions were monitored by TLC using commercially available (Merck) aluminium-backed plates, pre-coated with a 0.25 mm layer of silica containing a fluorescent indicator. Following aqueous work-up, organic layers were dried over MgSO₄ unless stated otherwise. Light petroleum refers to the fraction with bp 40–60 °C. Specific rotation [α]_D values were recorded at 25 °C and are given in 10⁻¹ deg cm² g⁻¹. IR spectra were recorded as thin films unless stated otherwise. NMR spectra were recorded in CDCl₃ unless stated otherwise with Bruker DPX 250, JEOL EX400 or Bruker AMX500 spectrometers. ¹H NMR chemical shifts are reported relative to CHCl₃ (δ_H 7.27), ¹³C NMR chemical shifts are reported relative to CDCl₃ (δ_C [central line of three] 77.0) and ¹¹⁹Sn chemical shifts are reported relative to Bu₄Sn as external reference. Coupling constants (J) are given in Hz.

Silylation of terminal epoxides employing ligand 8

trans-1-(Trimethylsilyl)-1,2-epoxydodecane (9a). To a solution of ^sBuLi (1.3 M in cyclohexane, 0.45 mL, 0.59 mmol) in hexane (4 mL) at −90 °C was added a solution of freshly distilled 8 (116 mg, 0.60 mmol) in hexane (1 mL) and the reaction mixture was allowed to warm to 0 °C. After a few seconds at 0 °C, the mixture was cooled to −90 °C and a solution of 1,2-epoxydodecane (44.1 mg, 0.24 mmol) and chlorotrimethylsilane (36 µL, 0.29 mmol) in hexane (1 mL) was added dropwise over a period of 10 min and the mixture was stirred for 2 h at −90 °C. The reaction mixture was then allowed to warm to −50 °C over 30 min and was quenched with MeOH (1 mL) at −50 °C and then with aqueous 1 N HCl (2 mL) at 0 °C. The two phases were separated and the aqueous one was extracted with Et₂O (2 × 5 mL). The combined organic extracts were washed with brine (1 × 5 mL), dried and concentrated in vacuo. The crude material was purified by flash chromatography (SiO₂, 0.5% Et₂O in pentane) to afford trans-1-(trimethylsilyl)-1,2-epoxydodecane as a colourless oil (45.1 mg, 73%). υ_max/cm⁻¹ 2957, 2926, 2855, 1467, 1418, 1249, 870, 841; δ_H (400 MHz, CDCl₃) 2.75 (1H, ddd, J 6.0, 5.0, 3.5), 1.96 (1H, d, J 3.5), 1.64–1.25 (18H, m), 0.88 (3H, t, J 7.0), 0.05 (9H, s); δ_C (100 MHz, CDCl₃) 56.3, 51.8, 34.2, 32.0, 29.7, 29.7, 29.6, 29.4, 26.5, 22.8, 14.2, −3.6; MS (CI⁺, NH₃) 257 ([M + H]⁺, 82%), 129 (100%), 90 (30%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 257.2300, C₁₅H₃₃OSi requires 257.2301.

trans-1-(Trimethylsilyl)-1,2-epoxydec-9-ene (9b). The reaction was performed on 1,2-epoxydec-9-ene (61.7 mg, 0.40 mmol) according to the procedure described for 9a above but after 2.5 h at −90 °C, the mixture was allowed to warm to −80 °C. Purification by flash chromatography (SiO₂, 1% Et₂O in pentane) afforded trans-1-(trimethylsilyl)-1,2-epoxydec-9-ene as a colourless oil (0.0648 g, 71%). υ_max/cm⁻¹ 3074, 2957, 2923, 2857, 1636, 1466, 1440, 1415, 1291, 1247, 990, 907, 874, 836, 749, 699; δ_H (400 MHz, CDCl₃) 5.90–5.80 (1H, m), 5.05–4.90 (2H, m), 2.80–2.70 (1H, m), 2.05 (2H, m), 1.97 (1H, d, J 3.5), 1.70–1.25 (10H, m), 0.06 (9H, s); δ_C (100 MHz, CDCl₃) 139.1, 114.2, 56.1, 51.7, 34.0, 33.7, 29.3, 29.0, 28.8, 26.3, −3.7; MS (CI⁺, NH₃) 244 ([M + NH₄]⁺, 25%), 90 (100%); HRMS (CI⁺, NH₃ for [M + NH₄]⁺) found 244.2101, C₁₃H₃₀NOSi requires 244.2097.

trans-4-Phenyl-1-(trimethylsilyl)-1,2-epoxybutane (9c). The reaction was performed on 4-phenyl-1,2-epoxybutane (48.8 mg, 0.33 mmol) according to the procedure described for 9a above but after 4 h at −90 °C, the mixture was allowed to warm to −80 °C. Purification by flash chromatography (SiO₂, 2% Et₂O in pentane) afforded trans-4-phenyl-1-(trimethylsilyl)-1,2-epoxybutane as a colourless oil (44.3 mg, 61%). υ_max/cm⁻¹ 3017, 2975, 2954, 2943, 2858, 1604, 1496, 1454, 1416, 1292, 1249, 1031, 864, 840, 747, 699; δ_H (400 MHz, CDCl₃) 7.35–7.10 (5H, m), 2.95–2.75 (3H, m), 2.10–1.80 (3H, m), 0.04 (9H, s); δ_C (50 MHz, CDCl₃) 141.9, 128.9, 128.8, 128.7, 126.4, 56.1, 52.5, 36.3, 33.1, −3.2; MS (CI⁺, NH₃) 238 ([M + NH₄]⁺, 50%), 90 (100%); HRMS (CI⁺, NH₃ for [M + NH₄]⁺) found 238.1625, C₁₃H₂₄NOSi requires 238.1627.

trans-5-(tert-Butyldimethylsiloxy)-1-(trimethylsilyl)-1,2-epoxypentane (9d). The reaction was performed on 5-(tert-butyldimethylsiloxy)-1,2-epoxypentane (87.1 g, 0.40 mmol) according to the procedure described for 9a above but after 2 h at −90 °C, the mixture was allowed to warm to −80 °C. Purification by flash chromatography (SiO₂, 2% Et₂O in pentane) afforded trans-5-(tert-butyldimethylsiloxy)-1-(trimethylsilyl)-1,2-epoxypentane as a colourless oil (86.4 mg, 74%). υ_max/cm⁻¹ 2970, 2960, 2930, 2856, 1470, 1247, 1097, 838, 775; δ_H (400 MHz, CDCl₃) 3.75–3.60 (2H, m), 2.85–2.75 (1H, m), 1.98 (1H, d, J 3.5), 1.80–1.60 (4H, m), 0.89 (9H, s), 0.07 (15H, apparent s); δ_C (100 MHz, CDCl₃) 68.3, 55.6, 51.2, 30.0, 29.9, 25.6, 18.3, −3.4, −5.5; MS (CI⁺, NH₃) 289 ([M + H]⁺, 50%), 273 (100%), 217 (10%), 164 (10%), 132 (20%), 90 (90%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 289.2023, C₁₄H₃₃O₂Si₂ requires 289.2019.

trans-6-Chloro-1-(trimethylsilyl)-1,2-epoxyhexane (9e). The reaction was performed on 6-chloro-1,2-epoxyhexane (54.6 mg, 0.41 mmol) according to the procedure described for 9a above but after 3 h at −90 °C, the mixture was allowed to warm to −80 °C. Purification by flash chromatography (SiO₂, 3% Et₂O in pentane) afforded trans-6-chloro-1-(trimethylsilyl)-1,2-epoxyhexane as a colourless oil (54.5 mg, 65%). υ_max/cm⁻¹ 2969, 2858, 1456, 1418, 1289, 876, 839; δ_H (400 MHz, CDCl₃) 3.56 (2H, t, J 6.3), 2.80–2.75 (1H, m), 1.98 (1H, d, J 3.6), 1.90–1.85 (2H, m), 1.80–1.50 (4H, m), 0.07 (9H, s); δ_C (100 MHz, CDCl₃) 55.8, 51.5, 44.8, 33.2, 32.2, 23.7, −3.9; MS (CI⁺, NH₃) 224 ([M + NH₄]⁺, 50%), 90 (100%); HRMS (CI⁺, NH₃ for [M + NH₄]⁺) found 224.1232, C₉H₂₃NOSiCl requires 224.1237.

2-Cyclododecyl-1-(trimethylsilyl)ethylene oxide (9f). The reaction was performed on cyclododecyl ethylene oxide (100 mg, 0.51 mmol) according to the procedure described for 9a above but after 4.5 h at −90 °C, the mixture was allowed to warm to −80 °C. Purification by flash chromatography (SiO₂, pentane) afforded 2-cyclododecyl-1-(trimethylsilyl)ethylene oxide as a colourless oil (91.6 mg, 67%). Data consistent with literature.^13a

trans/cis-2-Phenyl-1-(trimethylsilyl)-1,2-epoxybutane (9g). The reaction was performed on 2-phenyl-1,2-epoxybutane (54.9 mg, 0.37 mmol) according to the procedure described for 9a above but after 3 h at −90 °C, the mixture was allowed to warm to −50 °C. Purification by flash chromatography (Al₂O₃, pentane) afforded trans- and cis-2-phenyl-1-(trimethylsilyl)-1,2-epoxybutane as a colourless oil (97 : 3 inseparable mixture of trans and cis isomers respectively, 57.4 mg, 71%). Data for the major trans product consistent with literature,^13b minor component assigned tentatively as the cis isomer.

Silylation of terminal epoxides employing ligand 5

(1S,2S)-1-(Trimethylsilyl)-1,2-epoxydodecane [(1S,2S)-9a]. To a solution of ^sBuLi (1.3 M in cyclohexane, 1.0 mL, 1.3 mmol) in hexane (10 mL) at −90 °C was added (−)-sparteine 5 (0.30 mL, 1.3 mmol). The reaction mixture was allowed to warm to 0 °C over 5 min and then re-cooled to −90 °C. A solution of 1,2-epoxydodecane (184 mg, 1.0 mmol) and chlorotrimethylsilane (0.16 mL, 1.3 mmol) in hexane (2 mL) was added dropwise over a period of 15 min and the mixture was stirred for 15 min at −90 °C. The reaction mixture was then allowed to warm to −50 °C over 30 min and was quenched with MeOH (1 mL) at −50 °C and then with aqueous 1 N HCl (5 mL) at 0 °C. Work-up and purification as detailed for the synthesis of 9a above afforded (1S,2S)-1-(trimethylsilyl)-1,2-epoxydodecane [(1S,2S)-9a] as a colourless oil (169 mg, 66%). [α]²⁵_D −3.8 (c = 1.1, CHCl₃); other data as for racemate 9a. Specific rotation value indicated 30% ee in favour of the (1S,2S)-isomer based on comparison with results using enantiopure (>99% ee) (S)-1,2-epoxydodecane (see below).

(1S,2S)-1-(Trimethylsilyl)-1,2-epoxydodecane [(1S,2S)-9a]. The reaction was performed on (S)-1,2-epoxydodecane (130 mg, 0.71 mmol) according to the procedure described for (1S,2S)-9a above to afford (1S,2S)-1-(trimethylsilyl)-1,2-epoxydodecane as a colourless oil (130 mg, 72%). [α]²⁵_D −12.5 (c = 1.1, CHCl₃); other data as for racemate 9a.

(1R,2R)-1-(Trimethylsilyl)-1,2-epoxydodecane [(1R,2R)-9a]. The reaction was performed on (R)-1,2-epoxydodecane (184 mg, 1.0 mmol) according to the procedure described for (1S,2S)-9a above to afford (1R,2R)-1-(trimethylsilyl)-1,2-epoxydodecane as a colourless oil (156 mg, 60%). [α]²⁵_D + 11.2 (c = 1.0, CHCl₃); other data as for racemate 9a.

bis-Silylation of terminal epoxides

1,1-bis-(Trimethylsilyl)-1,2-epoxydodecane (10a). To a solution of ^sBuLi (1.2 M in cyclohexane, 1.1 mL, 1.3 mmol) in hexane (10 mL) at −90 °C was added (−)-sparteine 5 (0.30 mL, 1.3 mmol). The reaction mixture was allowed to warm to 0 °C over 5 min and then re-cooled to −90 °C. A solution of 1,2-epoxydodecane (0.184 g, 1.0 mmol) and chlorotrimethylsilane (0.16 mL, 1.3 mmol) in hexane (2 mL) was added dropwise maintaining the internal temperature at −90 °C. After 15 min the mixture was warmed to −50 °C over a period of 5 min. The mixture was stirred at −50 °C for 1 h and then re-cooled to −90 °C and chlorotrimethylsilane (0.25 mL, 2.0 mmol) was added. ^sBuLi–(−)-sparteine 5 complex (2 mmol) [prepared by adding (−)-sparteine 5 (0.46 mL, 2.0 mmol) dropwise to a solution of ^sBuLi (1.2 M in cyclohexane, 1.7 mL, 2.0 mmol) in hexane (8 mL) at −90 °C and allowing the mixture to warm to 0 °C] was added dropwise to the reaction maintaining the internal temperature at −90 °C. After 15 min the mixture was warmed to room temperature over 18 h. Aqueous 1 N HCl (5 mL) was added dropwise at 0 °C and the layers were separated. The aqueous phase was extracted with Et₂O (3 × 5 mL) and the combined organic phases were washed with brine (5 mL), dried and concentrated in vacuo. Purification by flash chromatography (SiO₂, 0.5% Et₂O in light petroleum) afforded 1,1-bis-(trimethylsilyl)-1,2-epoxydodecane (10a) as a pale yellow oil (203 mg, 62%). A second run returned 10a in 60% yield. [α]²⁵_D −1.2 (c = 1.1, CHCl₃); υ_max/cm⁻¹ 2957, 2926, 2855, 1466, 1408, 1261, 1250, 1051, 891, 859, 842; δ_H (400 MHz, CDCl₃) 2.94 (1H, dd, J 7.0, 4.5), 1.69–1.23 (18H, m), 0.88 (3H, t, J 7.0), 0.12 (9H, s), 0.04 (9H, s); δ_C (100 MHz, CDCl₃) 62.0, 51.5, 32.0, 31.6, 29.7, 29.7, 29.7, 29.6, 29.4, 27.6, 22.8, 14.2, 0.3, −2.1; MS (CI⁺, NH₃) 329 ([M + H]⁺, 68%), 313 (24%), 257 (88%), 90 (100%), 73 (38%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 329.2704, C₁₈H₄₁OSi₂ requires 329.2696. Specific rotation value indicated 12% ee in favour of the (S)-isomer based on comparison with results using enantiopure (>99% ee) (S)-1,2-epoxydodecane (see below).

1,1-bis-(Trimethylsilyl)-1,2-epoxydec-9-ene (10b). The reaction was performed on 1,2-epoxydec-9-ene (154 mg, 1.0 mmol) according to the procedure described for 10a above to afford 1,1-bis-(trimethylsilyl)-1,2-epoxydec-9-ene as a pale yellow oil (182 mg, 61%). A second run returned 10b in 55% yield. [α]²⁵_D −1.1 (c = 1.1, CHCl₃); υ_max/cm⁻¹ 2959, 2927, 2856, 1414, 1249, 1052, 907, 841; δ_H (400 MHz, CDCl₃) 5.92–5.68 (1H, m), 5.07–4.87 (2H, m), 3.01–2.87 (1H, m), 2.16–1.97 (2H, m), 1.29–1.50 (10H, m), 0.12 (9H, s), 0.04 (9H, s); δ_C (100 MHz, CDCl₃) 139.0, 114.2, 61.8, 51.3, 33.7, 31.5, 29.0, 20.0, 28.9 27.0, 0.7, −1.9; MS (CI⁺, NH₃) 299 ([M + H]⁺, 60%), 255 (20%), 164 (25%), 147 (20%), 90 (100%), 73 (25%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 299.2227, C₁₆H₃₅OSi₂ requires 299.2226.

4-Phenyl-1,1-bis-(trimethylsilyl)-1,2-epoxybutane (10c). The reaction was performed on 4-phenyl-1,2-epoxybutane (148 mg, 1.0 mmol) according to the procedure described for 10a above to afford 1,1-bis-(trimethylsilyl)-1,2-epoxy-4-phenylbutane as a pale yellow oil (161 mg, 55%). A second run returned 10c in 51% yield. [α]²⁵_D −3.4 (c = 1.1, CHCl₃); υ_max/cm⁻¹ 3027, 2955, 1495, 1454, 1250, 1052, 895, 841; δ_H (400 MHz, CDCl₃) 7.38–7.15 (5H, m), 3.10–3.04 (1H, m), 2.95–2.75 (2H, m), 2.07–1.85 (2H, m), 0.14 (9H, s), 0.05 (9H, s); δ_C (100 MHz, CDCl₃) 141.2, 61.2, 128.5, 128.4, 125.9, 52.1, 33.6, 33.5, 0.6, −1.9; MS (CI⁺, NH₃) 293 ([M + H]⁺, 70%), 277 (55%), 201 (70%), 147 (60%), 90 (100%), 73 (40%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 293.1745, C₁₆H₂₉OSi₂ requires 293.1757.

5-(tert-Butyldimethylsiloxy)-1,1-bis-(trimethylsilyl)-1,2-epoxypentane (10d). The reaction was performed on 5-(tert-butyldimethylsiloxy)-1,2-epoxypentane (216 mg, 1.0 mmol) according to the procedure described for 10a above to afford 5-(tert-butyldimethylsiloxy)-1,1-bis-(trimethylsilyl)-1,2-epoxypentane as a pale yellow oil (263 mg, 73%). A second run returned 10d in 69% yield. [α]²⁵_D −3.4 (c = 1.1, CHCl₃); υ_max/cm⁻¹ 2955, 2858, 1472, 1251, 1101, 838; δ_H (400 MHz, CDCl₃) 3.73–3.57 (2H, m), 2.99–2.93 (1H, m), 1.49–1.80 (4H, m), 0.91 (9H, s), 0.14 (9H, s), 0.06 (6H, s), 0.04 (9 H, s); δ_C (100 MHz, CDCl₃) 62.7, 61.6, 32.5, 30.4, 30.3, 25.8, 0.2, −1.4, −5.3; MS (CI⁺, NH₃) 361 ([M + H]⁺, 95%), 271 (60%), 229 (40%), 164 (35%), 147 (15%), 90 (100%), 73 (20%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 361.2400, C₁₇H₄₁O₂Si₃ requires 361.2414.

6-Chloro-1,1-bis-(trimethylsilyl)-1,2-epoxyhexane (10e). The reaction was performed on 6-chloro-1,2-epoxyhexane (135 mg, 1.0 mmol) according to the procedure described for 10a above to afford 6-chloro-1,1-bis-(trimethylsilyl)-1,2-epoxyhexane as a pale yellow oil (142 mg, 51%). A second run returned 10e in 49% yield. [α]²⁵_D −2.3 (c = 1.1, CHCl₃); υ_max/cm⁻¹ 2956, 2855, 1451, 1411, 1250, 1052, 840; δ_H (400 MHz, CDCl₃) 3.57 (2H, t, J 7.0), 3.00–2.93 (1H, m), 1.63–1.86 (6H, m), 0.14 (9H, s), 0.06 (9H, s); δ_C (100 MHz, CDCl₃) 61.6, 52.4, 44.7, 32.3, 30.7, 24.9, 1.2, −1.9; MS (CI⁺, NH₃) 279 ([M + H]⁺, 50%), 263 (30%), 164 (35%), 147 (40%), 90 (100%), 73 (35%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 279.1364, C₁₂H₂₈ClOSi₂ requires 279.1367.

(S)-1,1-bis-(Trimethylsilyl)-1,2-epoxydodecane [(S)-10a]. The reaction was performed on (S)-1,2-epoxydodecane (184 mg, 1.0 mmol) according to the procedure described for 10a above to afford (S)-1,1-bis-(trimethylsilyl)-1,2-epoxydodecane as a pale yellow oil (180 mg, 55%). [α]²⁵_D −10.3 (c = 1.0, CHCl₃); other data as for racemate 10a.

(R)-1,1-bis-(Trimethylsilyl)-1,2-epoxydodecane [(R)-10a]. The reaction was performed on (R)-1,2-epoxydodecane (184 mg, 1.0 mmol) according to the procedure described for 10a above to afford (R)-1,1-bis-(trimethylsilyl)-1,2-epoxydodecane as a pale yellow oil (155 mg, 47%). [α]²⁵_D + 10.6 (c = 1.0, CHCl₃); other data as for racemate 10a.

1-Hydroxy-1-phenyl-2-trimethylsilyl-2,3-epoxytridecene (12-1 and 12-2; two diastereoisomers, separable). To a solution of ^sBuLi (1.2 M in cyclohexane, 1.1 mL, 1.3 mmol) in hexane (10 mL) at −90 °C was added (−)-sparteine 5 (0.30 mL, 1.3 mmol). The reaction mixture was allowed to warm to 0 °C over 5 min and then re-cooled to −90 °C. A solution of 1,2-epoxydodecane (0.184 g, 1.0 mmol) and chlorotrimethylsilane (0.16 mL, 1.3 mmol) in hexane (2 mL) was added dropwise maintaining the internal temperature at −90 °C. After 15 min the mixture was warmed to −50 °C over a period of 5 min. The mixture was stirred at −50 °C for 1 h and then re-cooled to −90 °C and ^sBuLi–(−)-sparteine 5 complex (2 mmol) [prepared by adding (−)-sparteine 5 (0.46 mL, 2.0 mmol) dropwise to a solution of ^sBuLi (1.2 M in cyclohexane, 1.7 mL, 2.0 mmol) in hexane (8 mL) at −90 °C and allowing the mixture to warm to 0 °C] was added dropwise to the reaction maintaining the internal temperature at −90 °C. After 15 min the mixture was warmed up slowly to −30 °C (ca. 4 h). Benzaldehyde (0.2 mL, 2 mmol) was added and the mixture was allowed to warm to room temperature over 18 h. Aqueous 1 N HCl (5 mL) was added dropwise at 0 °C and the layers were separated. The aqueous phase was extracted with Et₂O (3 × 5 mL) and the combined organic phases were washed with brine (5 mL), dried and concentrated in vacuo. Purification by flash chromatography (SiO₂, 5% Et₂O in light petroleum) afforded alcohols 12-1 (76 mg, 21%) and 12–2 (56 mg, 16%) as pale yellow oils (combined yield 132 mg, 37%). Data for 12–1: υ_max/cm⁻¹ 3469, 2925, 2854, 1248, 1046, 841, 700; δ_H (400 MHz) 7.49–7.28 (5H, m), 4.84 (1H, s), 2.94 (1H, m), 2.17 (1H, s), 1.73–1.28 (18H, m), 0.90 (3H, t, J 7.0), −0.15 (9H, s); δ_C (100 MHz) 141.6, 128.1, 127.5, 126.4, 68.1, 61.1, 59.0, 34.1, 31.9, 30.3, 29.6, 29.5, 28.9, 28.4, 27.4, 22.9, 14.1, −2.6; HRMS (FI⁺ for M⁺) found 362.2642, C₂₂H₃₈O₂Si requires 362.2641. Data for 12-2: υ_max/cm⁻¹ 3470, 2925, 2855, 1248, 1047, 841, 701; δ_H (400 MHz) 7.43–7.27 (5H, m), 4.77 (1H, s), 3.06 (1H, m), 2.58 (1H, s), 1.69–1.28 (18H, m), 0.89 (3H, t, J 7.0), −0.17 (9H, s); δ_C (100 MHz) 139.9, 128.2, 127.3, 125.8, 74.8, 62.4, 61.9, 31.9, 29.7, 29.7, 29.5, 29.4, 29.3, 28.9, 27.6, 22.7, 14.1, −2.0; HRMS (FI⁺ for M⁺) found 362.2629, C₂₂H₃₈O₂Si requires 362.2641.

Stannylation of terminal epoxides: general procedure

To a solution of DBB 6 (162 mg, 0.68 mmol) in hexane (8.5 mL) at −90 °C was added dropwise ^sBuLi (1.40 M in cyclohexane, 0.48 mL, 0.67 mmol) and the mixture was allowed to warm to 0 °C over 5 min. On re-cooling to −90 °C, a solution of the appropriate epoxide 1 (0.27 mmol) in hexane (1 mL + 1 mL wash) was added dropwise over a period of 15 min and the mixture was stirred at −90 °C for 1 h. Tributyltin chloride (0.68 mmol) was then added neat in one portion and the mixture was manually agitated at −90 °C for 10 min before warming to 0 °C over 14 h. After quenching with aqueous 0.5 M H₃PO₄ (2.5 mL), the aqueous layer was extracted with Et₂O (3 × 10 mL) and the combined organic phases were washed with H₂O (5 mL), brine (5 mL), dried and concentrated in vacuo. Purification of the residue by flash chromatography (Florisil^® unless otherwise stated, eluent, petrol–Et₂O) gave the α,β-epoxystannane 13.

trans-1-(Tributylstannyl)-1,2-epoxydodecane (13a). The reaction was performed on 1,2-epoxydodecane (50 mg, 0.27 mmol) according to the general procedure described above but with a 1.5 h lithiation period. Flash chromatography (0.5% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxydodecane as a colourless oil (100 mg, 78%). υ_max/cm⁻¹ 2961, 2925, 2865, 2854, 1464, 1416, 1377, 1340, 1263, 1073, 866; δ_H (400 MHz, CDCl₃) 2.83–2.79 (1H, m), 2.46 (1H, d, J 3.6, J_Sn–H 99), 1.70–1.20 (30H, m), 0.94–0.86 (18H, m); δ_C (100 MHz, CDCl₃) 56.2, 52.8, 34.8, 31.9, 29.6, 29.6, 29.5, 29.3, 29.0, 27.0, 26.3, 22.6, 14.1, 13.6, 8.6 (J_Sn–C 330); δ¹¹⁹_Sn (186 MHz, CDCl₃) −33.2; MS (EI⁺) 474(M⁺, <1%), 417 (5%), 291 (10%), 235 (42%), 179 (68%), 121 (20%), 69 (20%), 57 (61%), 41 (100%); HRMS (EI⁺ for M⁺) found 474.2884, C₂₄H₅₀OSn requires 474.2884.

trans-1-(Tributylstannyl)-1,2-epoxy-9-decene (13b). The reaction was performed on 1,2-epoxy-9-decene (50 µL, 0.27 mmol) according to the general procedure described above. Flash chromatography (0.5% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxy-9-decene as a colourless oil (80 mg, 67%). υ_max/cm⁻¹ 3077, 2956, 2927, 2871, 2854, 1641, 1464, 1416, 1376, 1340, 1292, 1266, 1182, 1145, 1073, 909, 873, 864, 745; δ_H (400 MHz, CDCl₃) 5.87–5.75 (1H, m), 5.03–4.90 (2H, m), 2.85–2.77 (1H, m), 2.46 (1H, d, J 3.8, J_Sn–H 98), 2.09–2.00 (2H, m), 1.69–1.26 (22H, m), 1.00–0.85 (15H, m); δ_C (100 MHz, CDCl₃) 139.1, 114.1, 56.1, 52.8, 34.8, 33.7, 29.4, 29.0, 28.8, 27.3, 26.3, 13.6, 8.7 (J_Sn–C 328); δ¹¹⁹_Sn (186 MHz, CDCl₃) −33.1; MS (CI⁺, NH₃) 462 ([M + NH₄]⁺, 2%), 386 (2%), 308 (100%), 291 (7%), 268 (4%), 252 (3%), 172 (10%), 154 (7%); HRMS (CI⁺, NH₃ for [M + NH₄]⁺) found 462.2751, C₂₂H₄₈NOSn requires 462.2752.

trans-1-(Tributylstannyl)-1,2-epoxy-4-phenyl-butane (13c). The reaction was performed on 1,2-epoxy-4-phenyl-butane (40 mg, 0.27 mmol) according to the general procedure described above. Flash chromatography (1% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxy-4-phenyl-butane as a colourless oil (86 mg, 72%). υ_max/cm⁻¹ 3063, 3027, 2957, 2926, 2865, 2845, 1604, 1586, 1496, 1455, 1416, 1376, 1340, 1268, 1181, 1073, 1030, 870, 748, 698; δ_H (400 MHz) 7.32–7.27 (2H, m), 7.24–7.18 (3H, m), 2.91–2.71 (3H, m), 2.48 (1H, d, J 3.7, J_Sn–H 96), 2.02–1.81 (2H, m), 1.62–1.40 (6H, m), 1.32 (6H, m), 1.01–0.81 (15H, m); δ_C (100 MHz) 141.6, 128.4, 128.3, 125.9, 55.6, 53.1, 36.6, 32.6, 29.0, 27.3, 13.7, 8.7 (J_Sn–C 332); δ¹¹⁹_Sn (186 MHz, CDCl₃) −33.4; MS (ES⁺, NH₃) 289 ([M + NH₄]⁺, 12%), 308 (42%), 166 (100%); HRMS (ES⁺ for [M + NH₄]⁺) found 456.2285, C₂₂H₄₂NOSn requires 456.2283.

1-(Tributylstannyl)-2-cyclododecyl oxirane (13f). The reaction was performed on cyclododecyl ethylene oxide (53 mg, 0.27 mmol) according to the general procedure described above but with a 4 h lithiation period. Flash chromatography (neutral alumina, 0.5% Et₂O in light petroleum) afforded 1-(tributylstannyl)-2-cyclododecyl oxirane as a colourless oil (84 mg, 64%). υ_max/cm⁻¹ 2951, 2929, 2865, 2851, 1470, 1446, 1376, 1346, 1287, 1073, 1023, 960, 847; δ_H (400 MHz) 2.57 (1H, s, J_Sn–H 94), 1.84–1.25 (34H, m), 1.05–0.84 (15 H, m); δ_C (100 MHz) 62.6, 61.2, 32.6, 32.1, 29.0, 27.3, 26.2, 26.1, 25.7, 22.7, 22.4, 22.3, 21.1, 21.0, 13.7, 9.6; δ¹¹⁹_Sn (186 MHz) −40.7; MS (EI⁺) 486 (M⁺, 2%), 429 (4%), 291 (12%), 235 (37%), 179 (65%), 149 (9%), 135 (25%), 121 (50%), 57 (100%); HRMS (EI⁺ for M⁺) found 486.2868, C₂₅H₅₀OSn requires 486.2878.

trans-1-(Tributylstannyl)-1,2-epoxy-10-(tert-butyldimethylsilyloxy)decane (13h). The reaction was performed on 10-(tert-butyldimethylsiloxy)-1,2-epoxydecane (80 mg, 0.27 mmol) according to the general procedure described above. Flash chromatography (1% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxy-10-(tert-butyldimethylsilyloxy)decane as a colourless oil (131 mg, 83%). υ_max/cm⁻¹ 2956, 2928, 2856, 1464, 1412, 1377, 1360, 1255, 1102, 1006, 836, 775; δ_H (400 MHz) 3.60 (2H, t, J 6.6), 2.83–2.79 (1H, m), 2.46 (1H, d, J 3.8), 1.70–1.24 (26H, m), 1.01–0.82 (24H, m), 0.05 (6H, s); δ_C (100 MHz) 63.3, 56.2, 52.8, 34.8, 32.9, 29.6, 29.5, 29.3, 29.0, 27.3, 26.3, 26.0, 25.8, 18.4, 13.7, 8.7 (J_Sn–C 328), −5.3; δ¹¹⁹_Sn (186 MHz) −33.2; MS (ES⁺) 594 ([M + NH₄]⁺, 2%), 308 (40%), 287 (100%), 271 (20%), 252 (6%), 229 (15%), 137 (5%), 132 (20%), 91 (30%); HRMS (ES⁺ for [M + NH₄]⁺) found 594.3727, C₂₈H₆₄NO₂SiSn requires 594.3723.

trans-1-(Tributylstannyl)-1,2-epoxy-10-chlorodecane (13i). The reaction was performed on 1,2-epoxy-10-chlorodecane (52 mg, 0.27 mmol) according to the general procedure described above. Flash chromatography (1% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxy-10-chlorodecane as a colourless oil (102 mg, 78%). υ_max/cm⁻¹ 2955, 2928, 2855, 2845, 1464, 1416, 1376, 1340, 1291, 1266, 1073, 1021, 960, 866, 726; δ_H (400 MHz) 3.54 (2H, t, J 6.8), 2.83–2.79 (1H, m), 2.46 (1H, d, J 3.7, J_Sn–H 98), 1.82–1.71 (2H, m), 1.68–1.22 (24H, m) 1.02–0.82 (15H, m); δ_C (100 MHz) 56.2, 52.8, 45.1, 34.8, 32.6, 29.5, 29.4, 29.0, 28.8, 27.3, 26.8, 26.3, 13.7, 8.7 (J_Sn–C 330); δ¹¹⁹_Sn (186 MHz) −33.2; MS (ES⁺) 498 ([M + NH₄]⁺, 2%), 308 (100%), 291 (27%), 271 (20%), 252 (10%), 208 (58%), 190 (12%), 155 (5%), 138 (7%), 95 (8%); HRMS (ES⁺ for [M + NH₄]⁺) found 498.2519, C₂₂H₄₉ClNOSn requires 498.2519.

trans-1-(Tributylstannyl)-1,2-epoxyhexane (13j). The reaction was performed on 1,2-epoxyhexane (33 µL, 0.27 mmol) according to the general procedure described above. Flash chromatography (0.5% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxyhexane as a colourless oil (74 mg, 70%). υ_max/cm⁻¹ 2957, 2928, 2872, 2855, 1465, 1416, 1377, 1340, 1292, 1266, 1152, 1072, 1045, 1021, 863, 750; δ_H (400 MHz) 2.86–2.77 (1H, m), 2.46 (1H, d, J 3.8, J_Sn–H 97), 1.71–1.24 (18H, m), 1.01–0.83 (18H, m); δ_C (100 MHz) 56.1, 52.8, 34.5, 29.0, 28.5, 27.3, 22.6, 14.0, 13.6, 8.7 (J_Sn–C 329); δ¹¹⁹_Sn (186 MHz) −33.2; MS (ES⁺) 408 ([M + NH₄]⁺, 2%), 332 (2%), 308 (100%), 291 (11%), 268 (5%), 252 (5%), 196 (3%), 155 (3%), 118 (3%); HRMS (ES⁺ for [M + NH₄]⁺) found 408.2280, C₁₈H₄₂NOSn requires 408.2283.

trans-1-(Tributylstannyl)-1,2-epoxy-5-(5,5-dimethyl[1,3]dioxanyl)hexane (13k). The reaction was performed on 1,2-epoxy-5-(5,5-dimethyl[1,3]dioxanyl)hexane (54 mg, 0.27 mmol) according to the general procedure described above. Flash chromatography (5% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxy-5-(5,5-dimethyl[1,3]dioxanyl)hexane as a colourless oil (96 mg, 72%). υ_max/cm⁻¹ 2955, 2922, 2870, 1464, 1416, 1395, 1374, 1273, 1251, 1212, 1153, 1103, 1043, 1022, 872; δ_H (400 MHz) 3.59–3.41 (4H, m), 2.91–2.82 (1H, m), 2.49 (1H, d, J 3.5, J_Sn–H 98), 1.97–1.64 (4H, m), 1.62–1.40 (6H, m), 1.37 (3H, s), 1.30 (6H, m), 1.01 (3H, s), 0.99–0.81 (18H, m); δ_C (100 MHz) 98.6, 70.3, 56.1, 52.8, 34.3, 29.9, 29.0, 28.9, 27.3, 22.7, 22.5, 20.4, 13.7, 8.7 (J_Sn–C 330); δ¹¹⁹_Sn (186 MHz) −33.2; MS (ES⁺) 491 ([M + H]⁺, 4%), 337 (8%), 308 (56%), 291 (16%), 252 (6%), 201 (100%), 183 (90%), 129 (37%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 491.2543, C₂₃H₄₇O₃Sn requires 491.2542.

trans-1-(Tributylstannyl)-1,2-epoxy-3-methylbutane (13l). The reaction was performed on 1,2-epoxy-3-methyl-butane (28 µL, 0.27 mmol) according to the general procedure described above. Flash chromatography (0.5% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxy-3-methylbutane²⁴ as a colourless oil (92 mg, 90%). υ_max/cm⁻¹ 2957, 2928, 2872, 2854, 1464, 1416, 1377, 1288, 1219, 1072, 937, 879; δ_H (400 MHz) 2.62–2.55 (1H, m), 2.53 (1H, d, J 3.8, J_Sn–H 98), 1.66–1.22 (13H, m), 1.07 (3H, d, J 6.6), 1.01–0.83 (18H, m); δ_C (100 MHz) 61.8, 51.8, 33.2, 29.0, 27.3, 19.6, 18.3, 13.7, 8.7 (J_Sn–C 328); δ¹¹⁹_Sn (186 MHz) −32.6; MS (ES⁺) 394 ([M + NH₄]⁺, 5%), 308 (100%), 291 (10%), 268 (8%), 252 (5%); HRMS (ES⁺ for [M + NH₄]⁺) found 394.2127, C₁₇H₄₀NOSn requires 394.2126.

trans-1-(Tributylstannyl)-1,2-epoxy-3,3-dimethylbutane (13m). The reaction was performed on 1,2-epoxy-3,3-dimethylbutane (33 µL, 0.27 mmol) according to the general procedure described above. Flash chromatography (0.5% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxy-3,3-dimethylbutane as a colourless oil (92 mg, 87%). υ_max/cm⁻¹ 2956, 2929, 2872, 2855, 1464, 1418, 1377, 1362, 1293, 1072, 937, 881; δ_H (400 MHz) 2.64 (2H, d, J 0.8, J_Sn–H 99), 1.65–1.42 (6H, m), 1.38–1.26 (6H, m), 1.02–0.83 (24H, m); δ_C (100 MHz) 64.2, 48.7, 31.5, 29.0, 27.3, 25.6, 13.7, 8.6 (J_Sn–C 328); δ¹¹⁹_Sn (186 MHz) −31.4; MS (ES⁺) 408 ([M + NH₄]⁺, 5%), 332 (2%), 308 (100%), 291 (8%), 252 (2%); HRMS (ES⁺ for [M + NH₄]⁺) found 408.2284, C₁₈H₄₂NOSn requires 408.2283.

trans-1-(Tributylstannyl)-2-cyclohexyl oxirane (13n). The reaction was performed on cyclohexyl oxirane (34 mg, 0.27 mmol) according to the general procedure described above. Flash chromatography (0.5% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-2-cyclohexyl oxirane as a colourless oil (82 mg, 73%). υ_max/cm⁻¹ 2951, 2925, 2852, 1465, 1450, 1418, 1377, 1340, 1259, 1073, 1024, 960, 869; δ_H (400 MHz) 2.60 (1H, dd, J 6.6 and 3.8), 2.53 (1H, d, J 3.8, J_Sn–H 99), 1.99–1.90 (1H, m), 1.80–1.63 (4H, m), 1.62–1.45 (6H, m), 1.38–1.06 (12H, m), 0.94–0.87 (15H, m); δ_C (100 MHz) 60.7, 51.6, 42.8, 30.3, 29.0, 27.3, 26.4, 25.8, 25.6, 13.7, 8.7 (J_Sn–C 328); δ¹¹⁹_Sn (186 MHz) −32.7; MS (EI⁺) 416 (M⁺, 3%), 359 (19%), 291 (35%), 235 (64%), 179 (100%), 121 (14%); HRMS (EI⁺ for M⁺) found 416.2097, C₂₀H₄₀OSn requires 416.2096.

(1S,2S)-1-(Tributylstannyl)-1,2-epoxydodecane (1S,2S)-13a. The reaction was performed on (S)-1,2-epoxydodecane (50 mg, 0.27 mmol) according to the general procedure described above (1 h lithiation period employed in this reaction, unlike the 1.5 h lithiation used in the synthesis of racemic 13a). Flash chromatography (0.5% Et₂O in light petroleum) afforded trans-1-(tributylstannyl)-1,2-epoxydodecane as a colourless oil (83 mg, 65%). [α]²⁵_D −27.8 (c = 1.0, CHCl₃); other data as for racemate 13a.

(1S,2S)-1-(Tributylstannyl)-1,2-epoxypropane (13o). The reaction was performed on (S)-propylene oxide (19 µL, 0.27 mmol) according to the general procedure described above but with addition of the neat epoxide to frozen ^sBuLi–6 complex at −100 °C, followed by allowing the mixture to warm to −90 °C, before continuing with the general procedure. Flash chromatography (2% Et₂O in light petroleum) afforded (1S,2S)-1-(tributylstannyl)-1,2-epoxypropane as a colourless oil (51 mg, 54%). [α]²⁵_D −43.0 (c = 1.0, CHCl₃); υ_max/cm⁻¹ 2957, 2926, 2872, 1464, 1403, 1376, 1267, 1231, 1072, 1039, 842; δ_H (400 MHz) 2.91 (1H, m), 2.43 (1H, d, J 3.8, J_Sn–H 98), 1.62–1.45 (6H, m), 1.39 (3H, d, J 5.1), 1.37–1.21 (6H, m), 0.98–0.85 (15H, m); δ_C (100 MHz) 54.0, 52.0, 29.0, 27.3, 20.3, 13.7, 8.7 (J_Sn–C 329); δ¹¹⁹_Sn (186 MHz) −33.6; MS (ES⁺) 366 ([M + NH₄]⁺, 10%), 349 ([M + H]⁺, 4%), 308 (100%), 291 (16%), 268 (27%), 250 (8%), 196 (4%); HRMS (ES⁺ for [M + NH₄]⁺) found 366.1818, C₁₅H₃₆NOSn requires 366.1813.

Lithiation–electrophile trapping with carbonyl-based electrophiles:general procedure

To a solution of DBB 6 (162 mg, 0.68 mmol) in hexane (8.5 mL) at −90 °C was added dropwise ^sBuLi (1.40 M in cyclohexane, 0.48 mL, 0.67 mmol) and the mixture was allowed to warm to 0 °C over 5 min. On re-cooling to −90 °C, a solution of 1,2-epoxydodecane 1a (50 mg, 0.27 mmol) in hexane (1 mL + 1 mL wash) was added dropwise over a period of 15 min and the mixture was stirred at −90 °C for 1.5 h. The electrophile (0.68 mmol) was then added neat in one portion [or dropwise in Et₂O (1 mL) for solids] and the mixture was manually agitated at −90 °C for 10 min before warming slowly to 0 °C (over 14 h). After quenching with aqueous 0.5 M H₃PO₄ (2.5 mL), the aqueous layer was extracted with Et₂O (3 × 10 mL) and the combined organic phases were washed with H₂O (5 mL), brine (5 mL), dried and concentrated in vacuo. Purification of the residue by column chromatography (SiO₂ unless otherwise stated) gave the substituted epoxide 14.

(1RS,2RS,3RS)-1-Phenyl-2,3-epoxytridecan-1-ol (14a-1) and (1SR,2RS,3RS)-1-phenyl-2,3-epoxytridecan-1-ol (14a-2). The general procedure was followed using benzaldehyde as the electrophile. Flash chromatography (10% EtOAc in light petroleum) gave the epoxy alcohols 14a-1 and 14a-2 (59 mg, 75% combined yield, dr 1 : 1). Data for 14a-1: υ_max/cm⁻¹ 3436 (br, OH), 2955, 2926, 2855, 1494, 1466, 1455, 1198, 1065, 1027, 904, 756 700; δ_H (400 MHz) 7.41–7.31 (5H, m), 4.89 (1H, d, J 3.0) 3.19 (1H, td, J 6.0, 2.5), 2.99 (1H, t, J 3.0), 2.36 (1H, broad s), 1.65–1.15 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 140.1, 129.0, 128.6, 126.8, 71.4, 61.8, 55.6, 32.3, 31.8, 30.0, 29.9, 29.7, 26.3, 23.1, 14.5; MS (ES⁺) 308 ([M + NH₄]⁺, 100%), 290 (M⁺, 7%), 273 (13%), 257 (7%), 202 (2%), 170 (2%), 138 (13%), 120 (5%), 105 (5%); HRMS (ES⁺ for [M + NH₄⁺]⁺) found 308.2587, C₁₉H₃₄NO₂ requires 308.2590. Data for 14a-2: υ_max/cm⁻¹ 3432 (br, OH), 2948, 2926, 2854, 1720, 1466, 1453, 1273, 1113, 1026, 908, 841, 735, 700; δ_H (400 MHz) 7.44–7.30 (5H, m), 4.51 (1H, t, J 5.0) 3.05 (1H, td, J 5.5, 2.5), 2.98 (1H, dd, J 5.0, 2.5), 2.49 (1H, d, J 5.0), 1.65–1.16 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 140.8, 129.0, 128.5, 126.6, 74.5, 62.7, 57.8, 32.3, 31.8, 30.0, 29.9, 29.7, 26.3, 23.1, 14.5; MS (ES⁺) 308 ([M + NH₄]⁺, 96%), 290 (M⁺, 13%), 273 (14%), 257 (100%), 202 (8%), 170 (7%), 138 (33%), 124 (8%), 105 (12%); HRMS (ES⁺ for [M + NH₄]⁺) found 308.2593, C₁₉H₃₄NO₂ requires 308.2590.

(1RS,2RS,3RS)-1-(3-Chlorophenyl)-2,3-epoxytridecan-1-ol (14b-1) and (1SR,2RS,3RS)-1-(3-chlorophenyl)-2,3-epoxytridecan-1-ol (14b-2). The general procedure was followed using 3-chlorobenzaldehyde as the electrophile. Flash chromatography (10% EtOAc in light petroleum) gave the epoxy alcohols 14b-1 and 14b-2 as colourless oils (32 mg of each, 72% combined yield, dr 1 : 1). Data for 14b-1: υ_max/cm⁻¹ 3429 (br, OH), 2925, 2855, 1599, 1576, 1467, 1432, 1197, 1100,1077, 884, 786; δ_H (400 MHz) 7.40 (1H, s), 7.32–7.24 (3H, m), 4.85 (1H, d, J 2.5), 3.14 (1H, td, J 6.0, 2.5), 2.96 (1H, t, J 3.0), 2.45 (1H, br s), 1.65–1.14 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 141.7, 134.5, 129.8, 128.3, 126.4, 124.4, 70.3, 61.1, 55.2, 31.9, 31.3, 29.5, 29.3, 29.2, 25.9, 22.7, 14.1; MS (EI⁺) 326 ([M(³⁷Cl)]⁺, 6%), 324 ([M(³⁵Cl)]⁺, 14%), 306 (26%), 289 (7%), 208 (18%), 185 (33%), 183 (100%), 169 (35%), 167 (74%); HRMS (EI⁺ for [M(³⁵Cl)]⁺) found 324.1850, C₁₉H₂₉ClO₂ requires 324.1851. Data for 14b-2: υ_max/cm⁻¹ 3415 (br, OH), 2925, 2855, 1599, 1575, 1467, 1431, 1195, 1100,1078, 901, 785; δ_H (400 MHz) 7.43 (1H, s), 7.32–7.27 (3H, m), 4.49 (1H, t, J 5.0), 3.05 (1H, td, J 5.5, 2.5), 2.95 (1H, dd, J 5.5, 2.5), 2.60 (1H, br s), 1.67–1.17 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 142.3, 134.6, 129.9, 128.2, 126.4, 124.2, 73.3, 62.0, 57.4, 31.9, 31.4, 29.5, 29.3, 25.8, 22.7, 14.1; MS (ES⁺) 344 ([M(³⁷Cl) + NH₄]⁺, 35%), 342 ([M(³⁵Cl) + NH₄]⁺, 100%), 326 (9%), 308 (10%), 291 (28%), 200 (5%), 139 (5%); HRMS (ES⁺ for [M(³⁵Cl) + NH₄]⁺) found 342.2196, C₁₉H₃₃ClNO₂ requires 342.2194.

(1RS,2RS,3RS)-1-(4-Methoxyphenyl)-2,3-epoxytridecan-1-ol (14c-1) and (1SR,2RS,3RS)-1-(4-methoxyphenyl)-2,3-epoxytridecan-1-ol (14c-2). The general procedure was followed using 4-methoxybenzaldehyde as the electrophile. Flash chromatography (20% EtOAc in light petroleum) gave the epoxy alcohols 14c-1 and 14c-2 as colourless oils (25 mg of each, 57% combined yield, dr 1 : 1). Data for 14c-1: υ_max/cm⁻¹ 3435 (br, OH), 2925, 2854, 1613, 1586, 1514, 1465, 1249, 1174, 1037, 907, 832; δ_H (400 MHz) 7.31 (2H, d, J 9.0), 6.91 (2H, d, J 9.0), 4.84 (1H, br s), 3.82 (3H, s), 3.18 (1H, td, J 5.5, 2.5), 2.95 (1H, t, J 2.5), 2.26 (1H, d, J 1.5), 1.64–1.16 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 159.5, 131.7, 127.8, 114.0, 70.6, 61.3, 55.3, 55.2, 31.9, 31.5, 29.6, 29.5, 29.3, 25.9, 22.7, 14.1; MS (ES⁺) 338 ([M + NH₄]⁺, 14%), 320 (M⁺, 7%), 303 (35%), 287 (20%), 168 (100%), 154 (50%), 137 (23%), 121 (12%); HRMS (ES⁺ for [M + NH₄]⁺) found 338.2687, C₂₀H₃₆NO₃ requires 338.2690. Data for 14c-2: υ_max/cm⁻¹ 3381 (br, OH), 2955, 2918, 2849, 1613, 1587, 1515, 1469, 1378, 1251, 1177, 1034, 906, 892; δ_H (400 MHz) 7.34 (2H, d, J 8.5), 6.91 (2H, d, J 8.5), 4.45 (1H, t, J 5.0), 3.82 (3H, s), 3.02 (1H, td, J 5.5, 2.5), 2.96 (1H, dd, J 5.5, 2.5), 2.44–2.39 (1H, m), 1.65–1.16 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 159.4, 132.6, 127.5, 114.0, 73.7, 62.3, 57.3, 55.3, 31.9, 31.4, 29.6, 29.5, 29.3, 25.9, 22.7, 14.1; MS (ES⁺) 338 ([M + NH₄]⁺, 8%), 320 (M⁺, 7%), 303 (53%), 287 (25%), 168 (60%), 154 (100%), 137 (47%); HRMS (ES⁺ for [M + NH₄]⁺) found 338.2686, C₂₀H₃₆NO₃ requires 338.2690.

(1RS,2RS,3RS)-1-(2-Furyl)-2,3-epoxytridecan-1-ol and (1SR,2RS,3RS)-1-(2-furyl)-2,3-epoxytridecan-1-ol 14d (diastereoisomers not separated). The general procedure was followed using 2-furaldehyde as the electrophile. Flash chromatography (15% EtOAc in light petroleum) gave a mixture of diastereoisomeric epoxy alcohols 14d as a pale brown oil (21 mg, 67% combined yield, dr 1 : 1). υ_max/cm⁻¹ 3414 (br, OH), 2925, 2855, 1598, 1504, 1467, 1378, 1226, 1149, 1010, 910, 735; δ_H (400 MHz) 7.44–7.41 (1H, m), 6.38–6.36 (2H, m), 4.90 (0.5H, d, J 3.5), 4.63 (0.5H, d, J 4.5), 3.19 (0.5H, td, J 5.5, 2.5), 3.13 (0.5H, dd, J 4.5, 2.5), 3.09–3.03 (1H, m), 2.39 (1H, broad s), 1.69–1.19 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 153.5, 152.4, 142.8, 142.5, 110.3, 108.0, 107.0, 67.1, 65.1, 59.8, 58.8, 56.3, 55.5, 31.9, 31.4, 29.6, 29.5, 29.3, 25.9, 22.7, 14.1; MS (ES⁺) 298 ([M + NH₄]⁺, 85%), 280 (M⁺, 30%), 263 (53%), 249 (91%), 247 (100%), 214 (18%), 202 (20%), 128 (34%), 114 (95%), 111 (42%); HRMS (ES⁺ for [M + NH₄]⁺) found 298.2376, C₁₇H₃₂NO₃ requires 298.2377.

(1RS,2RS,3RS)-1-(3-Decyl-oxiranyl)-3-phenyl-prop-2-en-1-ol (14e-1) and (1SR,2RS,3RS)-1-(3-decyl-oxiranyl)-3-phenyl-prop-2-en-1-ol (14e-2). The general procedure was followed using trans-cinnamaldehyde as the electrophile. Flash chromatography (10% EtOAc in light petroleum) gave the epoxy alcohols 14e-1 and 14e-2 as pale yellow oils (30 and 33 mg respectively, 73% combined yield, dr 1 : 1). Data for 14e-1: υ_max/cm⁻¹ 3412 (br, OH), 2956, 2925, 2854, 1601, 1579, 1495, 1467, 1452, 968, 749; δ_H (400 MHz) 7.45–7.23 (5H, m), 6.72 (1H, d, J 16.0), 6.19 (1H, dd, J 16.0, 6.5), 4.53–4.47 (1H, m), 3.07 (1H, td, J 6.0, 2.5), 2.94 (1H, t, J 2.5), 2.11 (1H, d, J 2.5), 1.67–1.17 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 136.3, 132.7, 128.6, 128.0, 126.7, 126.6, 70.0, 60.2, 55.2, 31.9, 31.5, 29.6, 29.5, 29.4, 29.3, 26.0, 22.7, 14.1; MS (ES⁺) 334 ([M + NH₄]⁺, 35%), 316 (M⁺, 15%), 299 (M − OH⁺, 100%), 283 (18%), 208 (13%), 164 (69%), 150 (73%), 133 (22%), 104 (12%); HRMS (ES⁺ for [M + NH₄]⁺) found 334.2737, C₂₁H₃₆NO₂ requires 334.2741. Data for 14e-2: υ_max/cm⁻¹ 3410 (br, OH), 2956, 2925, 2854, 1601, 1495, 1465, 1453, 968, 749; δ_H (400 MHz) 7.44–7.24 (5H, m), 6.72 (1H, d, J 16.0), 6.28 (1H, dd, J 16.0, 6.0), 4.23–4.17 (1H, m), 3.02 (1H, td, J 5.5, 2.5), 2.92 (1H, dd, J 5.5, 2.5), 2.15–2.09 (1H, m), 1.64–1.17 (18H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 136.3, 131.9, 128.6, 128.0, 127.4, 126.6, 72.2, 61.1, 56.8, 31.9, 31.5, 29.6, 29.5, 29.4, 29.3, 25.9, 22.7, 14.1; MS (ES⁺) 334 ([M + NH₄]⁺, 23%), 316 (M⁺, 13%), 299 (M–OH⁺, 100%), 283 (14%), 208 (11%), 164 (43%), 150 (100%), 133 (32%), 104 (13%); HRMS (ES⁺ for [M + NH₄]⁺) found 334.2742, C₂₁H₃₆NO₂ requires 334.2741.

(1RS,2RS,3RS)-1-tert-Butyl-2,3-epoxytridecan-1-ol (14f-1) and (1SR,2RS,3RS)-1-tert-butyl-2,3-epoxytridecan-1-ol (14f-2). The general procedure was followed using pivaldehyde as the electrophile. Flash chromatography (7.5% EtOAc in light petroleum) gave the epoxy alcohols 14f-2 and 14f-2 as colourless oils (27 mg and 25 mg respectively, 70% combined yield, dr 1 : 1). Data for 14f-2: υ_max/cm⁻¹ 3469 (br, OH), 2951, 2926, 2855, 1480, 1467, 1364, 1287, 1246, 1216, 1184, 1117, 1052, 1016, 904; δ_H (400 MHz) 3.50 (1H, t, J 2.5), 3.03 (1H, td, J 5.5, 2.5), 2.87 (1H, t, J 3.0), 1.86 (1H, d, J 2.0), 1.64–1.21 (18H, m), 0.99 (9H, s), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 75.5, 58.3, 55.0, 34.2, 31.9, 31.7, 29.6, 29.5, 29.3, 26.0, 22.7, 14.1; MS (ES⁺) 288 ([M + NH₄]⁺, 100%), 271 ([M + H]⁺, 5%), 253 (22%), 183 (11%), 129 (4%); HRMS (ES⁺ for [M + NH₄]⁺) found 288.2909, C₁₇H₃₈NO₂ requires 288.2897. Data for 14f-2: υ_max/cm⁻¹ 3397 (br, OH), 2959, 2918, 2852, 1470, 1362, 1318, 1256, 1192, 1100, 1021, 1007, 906, 895; δ_H (400 MHz) 3.08 (1H, dd, J 6.0, 5.0), 2.91–2.84 (2H, m), 2.02–1.98 (1H, m), 1.65–1.19 (18H, m), 0.99 (9H, s), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 78.1, 58.5, 57.2, 34.7, 31.9, 31.6, 29.6, 29.5, 29.3, 25.9, 25.8, 22.7, 14.1; MS (ES⁺) 288 ([M + NH₄]⁺, 100%), 270 (M⁺, 5%), 253 (15%), 183 (10%), 117 (3%); HRMS (ES⁺ for [M + NH₄]⁺) found 288.2892, C₁₇H₃₈NO₂ requires 288.2897.

(1SR,2RS,3RS)-1-iso-Propyl-2,3-epoxytridecan-1-ol (14g-1) and (1RS,2RS,3RS)-1-iso-propyl-2,3-epoxytridecan-1-ol (14g-2). The general procedure was followed using iso-butyraldehyde as the electrophile. Flash chromatography (10% EtOAc in light petroleum) gave the epoxy alcohols 14g-1 and 14g-2 as colourless oils (18 mg and 16 mg respectively, 56% combined yield, dr 1 : 1). Data for 14g-1: υ_max/cm⁻¹ 3446 (br, OH), 2958, 2926, 2855, 1467, 1379, 1367, 1256, 1177, 1153, 1078, 1025, 906; δ_H (400 MHz) 3.62–3.57 (1H, m), 3.02 (1H, td, J 5.5, 2.5), 2.82 (1H, t, J 3.0), 1.89–1.78 (2H, m), 1.61–1.20 (18H, m), 1.02 (3H, d, J 4.0), 1.00 (3H, d, J 4.0), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 72.9, 59.6, 54.9, 31.9, 31.7, 31.6, 29.6, 29.5, 29.4, 29.3, 26.0, 22.7, 18.5, 17.8, 14.1; MS (ES⁺) 274 ([M + NH₄]⁺, 100%), 256 (M⁺, 35%), 240 (70%), 223 (20%), 202 (30%), 158 (12%), 104 (18%); HRMS (ES⁺ for [M + NH₄]⁺) found 274.2743, C₁₆H₃₆NO₂ requires 274.2741. Data for 14g-2: υ_max/cm⁻¹ 3437 (br, OH), 2958, 2926, 2855, 1467, 1380, 1364, 1246, 1148, 1080, 1025, 902; δ_H (400 MHz) 3.17 (1H, q, J 6.0), 2.90 (1H, td, J 6.0, 2.5), 2.78 (1H, dd, J 5.0, 2.5), 1.97–1.93 (1H, m), 1.83 (1H, sextet, J 7.0), 1.61–1.21 (18H, m), 1.02 (3H, d, J 7.0), 0.98 (3H, d, J 7.0), 0.89 (3H, t, J 7.0);δ_C (100 MHz) 76.0, 60.3, 57.1, 32.5, 31.9, 31.6, 29.6, 29.5, 29.4, 29.3, 25.9, 22.7, 18.4, 18.1, 14.0; MS (ES⁺) 274 ([M + NH₄]⁺, 94%), 256 (M⁺, 27%), 240 (35%), 202 (22%), 158 (34%), 104 (37%); HRMS (ES⁺ for [M + NH₄]⁺) found 274.2739, C₁₆H₃₆NO₂ requires 274.2741.

(2RS,3RS)-1,1-Diethyl-2,3-epoxytridecan-1-ol (14h). The general procedure was followed using 3-pentanone as the electrophile. Flash chromatography (7.5% EtOAc in light petroleum) gave recovered 1,2-epoxydodecane 1a (13 mg, 26%) and epoxy alcohol 14h as a colourless oil (46 mg, 63%). υ_max/cm⁻¹ 3481 (br, OH), 2963, 2925, 2855, 1464, 1378, 1304, 1262, 898; δ_H (400 MHz) 2.99 (1H, td, J 6.0, 2.5), 2.69 (1H, t, J 2.5), 1.73 (1H, s), 1.64–1.20 (22H, m), 0.98–0.84 (9H, m); δ_C (100 MHz) 71.0, 62.9, 54.8, 31.9, 31.7, 29.6, 29.5, 29.3, 28.8, 26.1, 22.7, 14.1, 7.6; MS (ES⁺) 288 ([M + NH₄]⁺, 100%), 270 (M⁺, 8%), 254 (31%), 237 (53%), 170 (11%), 104 (7%); HRMS (ES⁺ for [M + NH₄]⁺) found 288.2898, C₁₇H₃₈NO₂ requires 288.2897.

(2RS,3SR)-1-Phenyl-2,3-epoxytridecan-1-one (14i). The general procedure was followed with N,N-dimethylbenzamide as the electrophile, added as a solution in Et₂O. Flash chromatography (7.5% EtOAc in light petroleum) gave recovered 1,2-epoxydodecane 1a (15 mg, 30%) and epoxy ketone 14h as a white amorphous solid (39 mg, 50%). Mp 100–102 °C; υ_max/cm⁻¹ 2954, 2918, 2850, 1690 (C=O), 1598, 1580, 1469, 1450, 1232, 1077, 877; δ_H (400 MHz) 8.02 (2H, d, J 7.5), 7.63 (1H, d, J 7.5), 7.51 (2H, d, J 7.5), 4.03 (1H, d, J 2.0), 3.15 (1H, td, J 5.5, 2.0), 1.84–1.66 (2H, m), 1.59–1.20 (16H, m), 0.89 (3H, t, J 7.0); δ_C (100 MHz) 194.7, 135.6, 133.8, 128.8, 128.3, 60.1, 57.4, 32.0, 31.9, 29.5, 29.4, 29.3, 25.8, 22.7, 14.1; MS (CI⁺, NH₃) 289 ([M + H]⁺, 50%), 273 (100%), 159 (8%), 133 (5%), 120 (12%), 105 (58%); HRMS (CI⁺, NH₃ for [M + H]⁺) found 289.2168, C₁₉H₂₉O₂ requires 289.2168.

Acknowledgements

We thank the EPSRC for Research Grants (GR/N26302 and GR/S46789/01), the European Union for a Marie Curie Fellowship (HPMF-CT-1999-00264), the EPSRC and GlaxoSmithKline for a CASE award (to N.J.R.), and the EPSRC National Mass Spectrometry Service Centre for mass spectra. We also thank Dr P. O'Brien (York) for initial test samples of ligands 7 and 8.

References

1. I. Erden, in Comprehensive Heterocyclic Chemistry II, vol. 1A, ed. A. R. Katritzky, C. W. Rees and E. F. V. Scriven, Pergamon, Oxford, 1996, pp. 97.
2. (a) T. Satoh, Chem. Rev., 1996, 96, 3303; (b) D. M. Hodgson and E. Gras, Synthesis, 2002, 1625; (c) D. M. Hodgson, K. Tomooka and E. Gras, in Topics in Organometallic Chemistry: Organolithiums in Enantioselective Synthesis, vol. 5, ed. D. M. Hodgson, Springer-Verlag, Heidelberg, 2003, pp. 217–250; (d) F. Chemla and E. Vrancken, in The Chemistry of Organolithium Reagents, ed. Z. Rappoport and I. Marek, Wiley & Sons, New York, 2004, pp. 1165–1242.
3. J. J. Eisch and J. E. Galle, J. Am. Chem. Soc., 1976, 98, 4646.
4. (a) P. Lohse, H. Loner, P. Acklin, F. Sternfeld and A. Pfaltz, Tetrahedron Lett., 1991, 32, 615; (b) T. Satoh, S. Kobayashi, S. Nakanishi, K. Horiguchi and S. Irisa, Tetrahedron, 1999, 55, 2515.
5. D. M. Hodgson, G. P. Lee, R. E. Marriott, A. J. Thompson, R. Wisedale and J. Witherington, J. Chem. Soc., Perkin Trans. 1, 1998, 2151.
6. K. M. Morgan and S. Gronert, J. Org. Chem., 2000, 65, 1461.
7. Preliminary communications: (a) D. M. Hodgson and S. L. M. Norsikian, Org. Lett., 2001, 3, 461; (b) D. M. Hodgson and E. H. M. Kirton, Synlett, 2004, 1610; (c) D. M. Hodgson, N. J. Reynolds and S. J. Coote, Org. Lett., 2004, 6, 4187.
8. E. Doris, L. Dechoux and C. Mioskowski, Synlett, 1998, 337.
9. (a) K. M. Bertini Gross, Y. M. Jun and P. Beak, J. Org. Chem., 1997, 62, 7679; (b) J. R. Harrison, P. O'Brien, D. W. Porter and N. M. Smith, J. Chem. Soc., Perkin Trans. 1, 1999, 3623; (c) P. Scheiber and P. Nemes, Liebigs Ann. Chem., 1994, 1033.
10. (a) P. F. Hudrlik and A. M. Hudrlik, in Advances in Silicon Chemistry, vol 2, ed. G. L. Larson, JAI Press Inc., Greenwich, 1993, pp. 1; (b) G. H. Whitham, in Science of Synthesis: Houben-Weyl Methods of Molecular Transformations, vol. 4, ed. I. Fleming, Thieme, Stuttgart, 2001, pp. 633–646.
11. Epoxides 1 in Table 2 were commercially available or prepared according to (a) J. A. Elings, R. S. Downing and R. A. Sheldon, Eur. J. Org. Chem., 1999, 837 (entry 3); (b) L. H. Yang, A. E. Weber, W. J. Greenlee and A. A. Patchett, Tetrahedron Lett., 1993, 34, 7035 (entry 4); (c) I. Rothberg, L. Schneider, S. Kirsch and R. Ofee, J. Org. Chem., 1982, 47, 2675 (entry 5); (d) T. J. Michnick and D. S. Matteson, SYNLETT, 1991, 631 (entries 6–7); (e) S. E. Schaus, B. D. Brandes, J. F. Larrow, M. Tokunaga, K. B. Hansen, A. E. Gould, M. E. Furrow and E. N. Jacobsen, J. Am. Chem. Soc., 2002, 124, 1307 (entries 9–10).
12. G. A. Molander and K. Mautner, J. Org. Chem., 1989, 54, 4042.
13. (a) C. Burford, F. Cooke, G. Roy and P. Magnus, Tetrahedron, 1983, 39, 867 (Table 2, entry 6); (b) J. D. Warren and Y. Shi, J. Org. Chem., 1999, 64, 7675 (Table 2, entry 7).
14. (a) J. J. Eisch and J. E. Galle, J. Organomet. Chem., 1988, 341, 293; (b) C. Courillon, J. C. Marie and M. Malacria, Tetrahedron, 2003, 59, 9759.
15. D. M. Hodgson, P. J. Comina and M. G. B. Drew, J. Chem. Soc., Perkin Trans. 1, 1997, 2279.
16. (a) P. C. B. Page and M. J. McKenzie, in Science of Synthesis: Houben–Weyl Methods of Molecular Transformations, vol. 4, ed. I. Fleming, Thieme, Stuttgart, 2001, pp. 513; (b) A. E. Mattson, A. R. Bharadwaj and K. A. Scheidt, J. Am. Chem. Soc., 2004, 126, 2314; (c) X. Linghu, J. R. Potnick and J. S. Johnson, J. Am. Chem. Soc., 2004, 126, 3070.
17. (a) Y. Ukaji, A. Yoshida and T. Fujisawa, Chem. Lett., 1990, 157; (b) C. Agami, L. Dechoux and C. Mioskowski, Tetrahedron Lett., 1997, 38, 4071.
18. A. Barbero, P. Cuadrado, I. Fleming, A. M. González, F. J. Pulido and A. Sánchez, J. Chem. Soc., Perkin Trans. 1, 1995, 1525.
19. D. M. Hodgson, T. J. Buxton, I. D. Cameron, E. Gras and E. H. M. Kirton, Org. Biomol. Chem., 2003, 1, 4293.
20. J. R. Falck, R. K. Bhatt and K. M. Reddy, Synlett, 1997, 481.
21. Epoxides 1 in Table 4 were either commercially available, prepared as described previously (ref. 11, above) or prepared according to: (a) 1,2-epoxy-5-(5,5-dimethyl-[1,3]dioxanyl)hexane was prepared in two steps from 5-hexen-2-one (entry 8); (b) cyclohexyl oxirane was prepared according to ref. 11d, above.
22. Tentative stereochemical assignments for epoxy alcohol diastereoisomers 14 were made by comparison of NMR spectral data with those of structurally similar compounds in the following reports: (a) ref. 12 above; (b) T. Kawakami, I. Shibata, A. Baba and H. Matsuda, J. Org. Chem., 1993, 58, 7608; (c) A. E. Lurain, P. J. Carroll and P. J. Walsh, J. Org. Chem., 2005, 70, 448.
23. Trapping of other electrophiles (allyl bromide, MeI, EtOTf, PhSO₂F) has so far been unsuccessful.
24. J. A. Soderquist and C. Lopez, Tetrahedron Lett., 1991, 32, 6305.

---