New enantioselective method for hydration of alkenes using cyclodextrins as phase transfer catalyst

Article abstract of DOI:10.1016/j.tet.2005.09.064

A new enantioselective/inverse phase transfer catalysis (IPTC) reaction for the Markovnikov hydration of double bounds by an oxymercuration-demercuration reaction with cyclodextrins as catalysts was disclosed. Moderate ee (up to 32%) and yields (14-60%) w

Full text of DOI:10.1016/j.tet.2005.09.064

Tetrahedron 61 (2005) 11986–11990
New enantioselective method for hydration of alkenes using
cyclodextrins as phase transfer catalyst
a
Artur R. Abreu, Iva Costa, Carla Rosa, Luısa M. Ferreira, Ana Lourenc¸o
a
a
a
a
´
and Pedro P. Santosb,
*
a
´
REQUIMTE, Departamento de Quımica, FCT-UNL, 2829-516 Caparica, Portugal
b
a
´
´
CQE, Instituto Superior Tecnico, Departamento de Eng Quımica, UTL, Avenida Rovisco Pais, 1049-001 Lisboa, Portugal
Received 4 July 2005; revised 9 September 2005; accepted 12 September 2005
Available online 14 October 2005
Abstract—A new enantioselective/inverse phase transfer catalysis (IPTC) reaction for the Markovnikov hydration of double bounds by an
oxymercuration–demercuration reaction with cyclodextrins as catalysts was disclosed. Moderate ee (up to 32%) and yields (14–60%) were
obtained for allylic amines and protected allylic alcohols as starting materials.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
demercuration process. The reaction of mercuric acetate
with unsaturated substrates leads to the addition of a
hydroxyl group to one side of the double bound and mercury
to the other side, via a mercurinium ion intermediate.
Reduction of the C–Hg bond with sodium borohydride in
aqueous sodium hydroxide yields the alcohol corresponding
to a Markovnikov addition of water to the double bound.2
The enantioselective version of this reaction was achieved
by the use of optically active mercuric salts but low yields
and moderate ee were observed.3 Another drawback of this
method is that the mercuric salts are not readily available
and have to be synthesized. Oxymercuration of (1:1)
complexes of olefin–b-cyclodextrin is also not a practical
proposition.4
The enantioselective funcionalization of alkenes is
extremely important in synthetic organic chemistry and
much effort has been expended for the development of new
synthetic methods.1 The epoxidation of allylic alcohols with
titanium tartarate complexes (Sharpless) and of styrene-like
olefins with porphyrin complexes (Katsuki–Jacobsen) are
two prominent ways of asymmetric funtionalization of
olefins.1 For asymmetric hydration of a terminal double
bond, Sharpless or Jacobsen asymmetric epoxidation
followed by hydride opening of the resulting epoxide ring
might be a satisfactory solution (Scheme 1). However, the
scope of this strategy is limited by the epoxidation step as
the asymmetric epoxidation of some olefins cannot be
accomplished.1
The present new method of hydration of alkenes using
cyclodextrins was developed to overcome difficulties
observed with some alkenes where the Sharpless/Jacobsen
epoxidation failed to produce enantioselective epoxides.
2. Results and discussion
To improve synthetic efficiency of the hemisynthesis of
homopumiliotoxins5 from jussiaeiines6 the enantioselective
hydration of a double bond was needed (Scheme 2).
Scheme 1. Main synthetic strategies for asymmetric hydration of terminal
double bonds.
Alternatively, Markovnikov-like hydration of double bonds
may be performed by the classical oxymercuration–
Piperidine model compounds were used as starting
materials to develop the synthetic procedure. A piperidine
ring containing an exocyclic double bond (1) was produced
from a protected 3-hydroxymethylene piperidine, similar to
the starting natural products. The enantioselective
Keywords: Reduction; IPTC; Oxymercuration; Cyclodextrin; Hydration.
*
Corresponding author. Tel.: C351 218417878; fax: C351 218417122
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.09.064

A. R. Abreu et al. / Tetrahedron 61 (2005) 11986–11990
11987
Scheme 2. Proposed pathway for the hemisynthesis of homopumiliotoxins from jussiaeiines.
epoxidation/hydride opening by Sharpless and Jacobsen/
Katsuky catalysts were assayed but both failed.
transfer reagents.9 Stereochemical (size of the hydrophobic
cavity) and stereoelectronic effects may have strong
influence on the behaviour of the cyclodextrin as IPTC
catalyst.
A new synthetic method for direct enantioselective
hydration of the double bond was searched and an inverse
phase transfer catalysis (IPTC) process with a chiral catalyst
looked promising and therefore, investigated.
Surprisingly or not, enantioselectivity was observed when
the oxymercuration–demercuration process of 1 (Scheme 3)
was conducted in a two phase water–n-hexane (1/1) system
with a-and b-cyclodextrin as IPTC catalyst (Table 1, entries
1,2).
PTC reactions have been extensively applied to promote a
variety of interfacial reactions with small molecules and
several enantioselective methods have been developed (for
example, Michael additions and reductions).7 IPTC
reactions—where a lipophilic reactant is transported to the
aqueous phase by the catalyst—are still rare and only
limited examples have been reported. We now report the
first example of an enantioselective IPTC hydration
reaction, promoting the hydration of alkenes via an
oxymercuration–demercuration process.
Scheme 3. Enantioselective oxymercuration–demercuration of the piper-
idine derivative 1.
Following this result, several other alkenes were tested. The
alkenes were reacted with different molar quantities of
mercuric salts in a water–n-hexane mixture (1/1) containing
a phase transfer catalyst with formation of oxymercurials.
These intermediates were not isolated but submitted to
demercuration with alkaline borohydride.
The intermediate of the oxymercuration reaction results
from the attack of the neutral alkene to the water soluble
mercuric salt in the usual solvent system aqueous THF.8 A
two phase PTC reaction would allow to control the contact
between the alkene and the mercuric salt, as the former will
stay in the organic phase and the latter in the aqueous phase.
A neutral PTC catalyst would be the right choice to carry the
alkene to the aqueous phase and to create an asymmetric
environment where the reaction would take place (Fig. 1).
Depending on the demercuration conditions the dihydroxyl-
ation product was also formed10. The demercuration
reaction was conducted under inert atmosphere to increase
the yield of monoalcohol product. The resulting alcohols
were purified by preparative thin-layer chromatography and
the ees were determinated by HPLC on chiral stationary
phase or by 1H NMR experiments (using chiral shift
reagents or Mosher’s esters) (Tables 1 and 2).
Unfunctionalized alkenes like styrene failed to react under
the IPTC conditions. However, hydration of allylic amines
and allylic protected alcohols was successful and moderate
enantioselectiviy was observed. As usual, this reaction gave
excellent chemoselectivity with formation of the more
substituted alcohol.
A study of the reaction conditions was carried out with the
allylic alcohol 2 as starting material (Table 2). The use of
ultrasound, previously applied to prepare the mercuric salt
in situ,11 increased significantly the reaction rate (entries
1–4). Even though Hg(OCOCF3)2 gives better yields then
Hg(OCOCH3)2, enantioselectivity was only achieved with
the latter (entries 5 and 7). The yield of the process was
increased using an excess of the mercuric compound
(entries 8–11). For this substrate, b-cyclodextrin was a
better catalyst (entries 5–9) and higher enantioselectivity
Figure 1. Simplified mechanism of an IPTC process mediated by
cylodextrins.
Neutral cyclodextrins were tested as phase transfer
catalysts. They are cyclic polymeric sugars with a basket-
like structure and are known to behave as chiral phase

11988
A. R. Abreu et al. / Tetrahedron 61 (2005) 11986–11990
Table 1. Reaction conditions, yields and ee for IPTC oxymercuration–demercuration of allylic amines
Substrate
Entry
IPTC catalyst
(cyclodextrin)
Time (h)
Yield (%)
ee (%)
1
2
3
4
b
a
4
4
4
4
0.7
0.7
0.7
0.7
58
36
36
38
60
27
53
46
25a
32a
18a
11a
15a
17a
11a
14a
2,6-Di-O-methyl-b
Random methyl-b
b
5b
6b
7b
8b
1
a
2,6-Di-O-methyl-b
Random methyl-b
9
b
a
2
2
2
2
31
40
60
43
25a
11a
11a
10a
10
11
12
4
5
6
2,6-Di-O-methyl-b
Random methyl-b
13
14
15
16
b
a
3
3
3
3
40
42
25
55
30c
11c
28c
31c
2,6-Di-O-methyl-b
Random methyl-b
17
18
19
20
b
a
2
2
2
2
30
32
27
14
0c
0c
6c
0c
2,6-Di-O-methyl-b
Random methyl-b
1
a ee calculated by H NMR experiment by addition of chiral shift reagent (Eu(hfpc)3).
b Hg(OCOCH3)2 as oxymercuration agent.
1
c ee calculated by H NMR experiment of the Mosher’s esters.
was obtained by lowering the temperature of the reaction
mixture (entry 12–14). The best conditions were found to be
3 mol equiv of mercuric acetate in H2O/n-hexane with
b-cyclodextrin as IPTC catalyst, at 0 8C with ultrasound
conditions. Moderate ees were observed. A non-catalyzed
hydration process always operates and reduces the observed
ee (entry 15, 19% yield at 0 8C obtained without PTC
catalyst).
process for 2 was applied to four different allylic amines
with four different cyclodextrins as catalysts. It looks like
that an aromatic ring must be present in the substrate to
achieve enantioselectivity (Table 1, entries 17–20). The best
results are always obtained with a- or b-cyclodextrin.
Modified cyclodextrins (2,6-di-O-methyl and random
methyl cyclodextrins) neither improve yield nor enantio-
selectivity. N-allyl-N-benzylmethylamine is an interesting
example as the benzyl group may be removed by catalytic
hydrogenation.
Allylic amines showed similar behaviour. The optimized
Table 2. Reaction conditions, yields and ee for IPTC oxymercuration–demercuration of allylic alcohol (2)
Entry Mercuric salt
Solvent
Mixing
IPTC catalyst Temperature
(cyclodextrin) (8C)
Time
(h)
Yield
(%)
ee
(%)
1
2
3
4
5
6
7
8
Hg(OCOCF3)2 (1 equiv)
H2O/THF
Magnetic stirring
Magnetic stirring
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
Ultra sound
—
—
—
—
b
a
b
a
b
b
b
b
b
b
—
20
20
20
20
20
20
20
20
20
20
20
25
0
6
24
1.5
1.5
0.5
0.75
0.5
0.75
0.75
0.75
0.75
0.70
0.75
5
41
35
52
42
52
38
20
18
35
52
52
52
51
0
0a,b
—
Hg(OCOCF3)2 (1 equiv)
Hg(OCOCF3)2 (1 equiv)
Hg(OCOCF3)2 (1 equiv)
Hg(OCOCF3)2 (1 equiv)
Hg(OCOCF3)2 (1 equiv)
Hg(OCOCH3)2 (1 equiv)
Hg(OCOCH3)2 (1 equiv)
(2 equiv)
(3 equiv)
(4 equiv)
(3 equiv)
(3 equiv)
H2O/n-hexane
H2O/THF
H2O/THF
—
—
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
H2O/n-hexane
0a
0a
16a
0a
9
15a
15a
14a
15a
25a,b
—
10
11
12
13
14
15
(3 equiv)
(3 equiv)
K75
0
0.75
19
—
1
a ee calculated by H NMR experiment by addition of chiral shift reagent (Eu(hfpc)3).
b ee calculated by HPLC on chiral stationary phase.

A. R. Abreu et al. / Tetrahedron 61 (2005) 11986–11990
11989
3. Experimental
was quenched with water and extracted with ethyl ether, the
organic phase dried and concentrated to dryness under
vacuum. The residue was chromatographed by column
chromatography with ethyl acetate–n-hexane (1/2). Com-
Melting points were recorded on a Reichert-Thermovar hot
stage apparatus and are reported uncorrected. Infrared (IR)
spectra were recorded on Perkin Elmer Spectrum 1000 as
KBr pellets or as film over NaCl plates.
pound (1)14 was obtained as a colourless oil, IV: nmax (cmK1
)
1
1698, 1656; H NMR (CDCl3) d 7.36–7.28 (5H, m, ArH),
5.13 (2H, s, CH2benzylic), 4.84–4.77 (2H, m, CCH2), 3.96
(2H, s, N(Cbz)CH2C), 3.52 (2H, t, JZ5.6 Hz, N(Cbz)CH2-
CH2), 2.27 (2H, t, JZ5.6 Hz, CCH2CH2), 1.64 (2H, m,
CCH2CH2); 13C NMR (CDCl3) d 155.1, 142.4, 136.9,
128.4–127.8, 110.2, 66.9, 50.5, 44.2, 32.6, 26.5.
Proton and carbon nuclear magnetic resonance spectra (1H
and 13C NMR) were recorded on Bruker ARX (400 MHz)
spectrometer. Chemical shifts are expressed in ppm,
downfield from TMS (dZ0) as an internal standard; J
values are given in Hz. The exact attribution of NMR
signals was preformed using two dimensional NMR
experiments.
3.1.2. N-allyl-N-benzylmethylamine (4).15 To N-benzyl-
methylamine and NaH (1.1 equiv) in dry DMF (0.8 M) at
0 8C allylbromide (1.1 equiv) was added. The reaction was
completed in 1 h at room temperature. Water was added and
the mixture extracted 3! with ethyl acetate, the organic
phases dried and concentrated to dryness under vacuum.
Mass spectra were taken with a Micromass GC-TOF (GCT)
mass spectrometer. Microanalyses were performed on a
Thermo Finnigan-CE Instruments Flash EA 1112 CHNS
series microanalyser. The analyses were performed by the
analytical services laboratory of REQUIMTE.
1
Compound (4) was obtained as a colourless oil (76%), H
NMR (CDCl3) 7.36–7.26 (5H, m, ArH), 5.92 (1H, m, CH),
5.17 (2H, m, CH2CH), 3.50 (2H, s, CH2benzylic), 3.03 (2H,
d, JZ6.3 Hz, NCH2CH), 2.19 (3H, s, NCH3).
All reagents and solvents were reagent grade and were
purified and dried by standard methods.
3.1.3. N-allyl-N-cyclohexylmethylamine (6).16 This com-
pound was prepared from cyclohexylmethylamine by the
same procedure as compound (4). Compound (6) was
purified by extraction to an acidic phase (HCl 10%),
neutralization and recovered with dichloromethane. By
evaporation of the dried organic phase the pure compound
Organic extracts were dried over anhydrous sodium sulfate.
Analytical thin-layer chromatography was performed on
Merck Kieselgel 60, F254 silica gel 0.2 mm thick plates. For
preparative TLC (PTLC) Merck Kieselgel 60, F254 silica
gel 0.5, and 1 mm thick plates (20!20 cm) were used.
Column chromatography was eluted over Merck Kieselgel
60 (240–400 mm) silica gel.
1
was obtained as a colourless oil (79%), H NMR (CDCl3)
5.84 (1H, m, CHallyl), 5.13 (2H, m, CH2CHallyl), 3.10 (2H,
d, JZ6.0 Hz, NCH2CH), 2.37 (1H, m, NCH), 2.22 (3H, s,
NCH3), 1.78 (4H, m, H-cyclohexyl), 1.61 (1H, m,
H-cyclohexyl), 1.21 (4H, m, cyclohexyl), 1.07 (1H, m,
cyclohexyl).
HPLC was performed on a system equipped with a Dionex
P680 pump, UVD340S detector and Carolcel OD column.
3.1. Synthesis of non commercial olefins
3.2. Oxymercuration–demercuration reactions
3.1.1. 3-Methylenepiperidine-1-carboxylic acid benzyl
ester (1). 3-Hydroxymethylenepiperidine was treated with
benzylchloroformate by standard method12 to obtain the
Cbz derivative: yellow oil, IV: nmax (cmK1) 3436, 1678; 1H
NMR (CDCl3) d 7.40–7.26 (5H, m, ArH), 5.12 (2H, s,
CH2benzylic), 3.98–3.78 (2H, m, N(Cbz)CaHeq), 3.48 (2H,
m, CH2OH), 3.09–2.93 (2H, m, N(Cbz)CaHax), 1.81–1.65
(3, m, CHCH2OH, N(Cbz)CbHeq, N(Cbz)CgHeq), 1.45 (1H,
m, N(Cbz)CbHax), 1.26–1.19 (1H, m, N(Cbz)CgHax); 13C
NMR (CDCl3) d 155.0, 136.8, 128.6–126.9 (5C), 67.0, 64.3,
46.5, 44.8, 33.0, 26.8, 24.0. This compound was tosylated
by standard method13 to obtain the tosylated derivative:
white solid, mp 60–62 8C, IV: nmax (cmK1) 1694, 1354; 1H
NMR (CDCl3) d 7.76 (2H, d, JZ7.48 Hz, HTs), 7.38–7.26
(8H, m, ArH), 5.10 (2H, s, CH2benzylic), 3.96–3.84 (4H, m,
N(Cbz)CaHeq, CH2OTs) 2.85 (1H, m, N(Cbz)CaHax), 2.68–
2.60 (1H, m, N(Cbz)CaHax), 2.44 (3H, s, CH3), 1.89–1.84
(1H, m, CHCH2OTs), 1.75 (1H, m, N(Cbz)CgHeq) 1.66–
1.61 (1H, m, N(Cbz)CbHeq), 1.45–1.42 (1H, m, N(Cbz)Cb-
Hax), 1.26–1.21 (1H, m, N(Cbz)CgHax); 13C NMR (CDCl3)
d 155.2, 144.9, 136.7, 132.7, 129.9–127.9, 71.6, 67.1, 46.3,
44.3, 35.3, 26.6, 23.9, 21.6. Anal. Calcd for C21H25NO5S: C,
62.51; H, 6.25; N, 3.47; S, 7.95 found C, 62.15; H, 6.17; N,
3.41; S, 7.75. This compound was treated with potassium
t-butoxide (1.1 equiv) in DMSO (1.6 M) under inert
atmosphere at room temperature for 2 h. Then the reaction
Standard homogeneous conditions. To 10 ml of H2O–THF
mixture (1/1) the mercuric reagent was added followed by
the addition of the alkene (0.4 mmol). After alkene
consumption, 4 ml of NaOH (3 M) were added under inert
atmosphere followed by the addition of NaBH4 (1 equiv) in
2 ml of NaOH (3 M). The mixture was stirred until complete
flocculation of Hg0. The THF was evaporated, the aqueous
phase extracted with ethyl acetate, the organic phase dried
and concentrated to dryness under vacuum. The resulting
alcohol was purified by plate chromatography.
Heterogeneneous conditions. The alkene (0.4 mmol) was
dissolved in 5 ml of n-hexane followed by the addition, at
the pretended temperature, of cyclodextrin (0.04 mmol),
5 ml of water and the mercuric reagent. After alkene
consumption, 4 ml of NaOH (3 M) were added under inert
atmosphere followed by the addition of NaBH4 (1 equiv) in
2 ml of NaOH (3 M). The mixture was stirred until complete
flocculation of Hg0 and then extracted with ethyl acetate, the
organic phase dried and concentrated to dryness under
vacuum. The resulting alcohol was purified by plate
chromatography.
By the above procedures the following compounds were
prepared. The spectral data were in accordance with

11990
A. R. Abreu et al. / Tetrahedron 61 (2005) 11986–11990
literature: 1-benzyloxypropan-2-ol (3),17 1-(benzylmethyl-
amino)-propan-2-ol,18 1-imidazol-1-yl-propan-2-ol,19
3-hydroxy-3-methylpiperidine-1-carboxylic acid benzyl
ester.20
Acknowledgements
We would like to thank Carlos M. Afonso and Nuno
Lourenc¸o from IST, UTL, Lisbon, for chiral HPLC
analyses. We thank to FEDER and POCTI programs for
financial support.
3.2.1. 1-(Cyclohexylmethylamino)-propan-2-ol. The
alcohol was obtained as a colourless oil, 1H NMR
(CDCl3) 3.71 (1H, m, CHOH), 2.82 (1H, br l, OH), 2.38–
2.34 (2H, m, CH2N), 2.25 (3H, s, NCH3), 2.22 (1H, m,
NCH), 1.80 (3H, m, H-cyclohexyl), 1.65 (2H, m,
H-cyclohexyl), 1.61 (1H, m, H-cyclohexyl), 1.33–1.01
(5H, m, cyclohexyl), 1.11 (3H, d, JZ6.1 Hz, CH3CH);
13C NMR (CDCl3) d 63.5, 62.4 (CHOH, CHN), 61.2
(CH2N), 37.2 (CH2N), 29.7, 29.3, 28.2, 26.3, 26.0, 19.9
(CH3C).
References and notes
1. Katsuki, T. In Catalytic asymmetric synthesis; Ojima, I, Ed.
2nd ed.; Wiley-VCH: New York, 2000; pp 287–325.
2. Larock, R. C. Angew. Chem., Int. Ed. Engl. 1978, 17, 27–37.
3. Sugita, T.; Yamasaki, Y.; Itoh, O.; Ichikawa, K. Bull. Chem.
Soc. Jpn. 1974, 47, 1945–1947.
3.3. Methods for ee determination
4. Rao, K. R.; Sampathkumar, H. M. Synth. Commun. 1993, 23,
1877–1885.
By the addition of a chiral shift reagent. To a CDCl3
solution of the alcohol (4.7!10K2 mmol), Eu(hfpc)3
(0.2 equiv) was added and the resulting solution was
5. Garraffo, H. M.; Spande, T. F.; Daly, J. W.; Baldessari, A.;
Gros, E. G. J. Nat. Prod. 1993, 56, 357–373.
1
analysed by H NMR.
´
6. Maximo, P.; Lourenc¸o, A. J. Nat. Prod. 2000, 63, 201–204.
7. Takahashi, K. Chem. Rev. 1998, 98, 2013–2033.
8. Brown, H. C.; Geoghegan, P. J.; Kurek, J. T. J. Org. Chem.
1981, 46, 3810–3812.
By HPLC on a chiral column. A mixture of 2 and 3 was
eluted with n-hexane–isopropanol (99/1) resulting on the
retention times: 14 min for 2, and 38, 42 min for each
enantiomer of 3.
9. Cabou, J.; Bricout, H.; Hapiot, F.; Monflier, E. Catal.
Commun. 2004, 5, 265–270.
By formation of Mosher’s esters. The Mosher’s esters were
prepared by the addition of 1 equiv of Mosher’s acid
chloride to the alcohol in a CH2Cl2 solution in the presence
of base (K2CO3). The diastereomer pair was purified by
10. Kitching, W. In Trost, B., Fleming, I., Eds. 1st ed.;
Comprehensive organic synthesis; Pergamon: Oxford, 1991;
Vol. 7, pp 613–638.
11. Einhorn, J.; Einhorn, C.; Luche, J. L. J. Org. Chem. 1989, 54,
4479–4481.
1
thin-layer chromatography and analysed by H NMR.
12. Greene, T. W.; Wuts, P. G. M. Protective groups in organic
synthesis. 3rd ed.; Wiley: New York, 1999; pp 494–653.
13. Organic syntheses,; Wiley: New York, 1988; Collect. Vol. 6, p
586; Collect. Vol. 55, p 57.
4. Conclusions
A completely new and simple method of asymmetric
hydration of terminal double bonds of allylic amines and
protected allylic alcohols was discovered. The readily
available a- and b-cyclodextrins where able to induce
enantioselectivity to the hydration method via an IPTC
process. Moderate yields and ees were obtained. Results
were found to be dependent on the starting alkene and on
reaction conditions. Although there were previously
reported methods for enantioselective hydration of allylic
alcohols, no equivalent method was up-to-date available for
the direct enantioselective functionalization of allylic
amines. The present method was able to induce enantio-
selectiviy to the direct hydration of allylic amines. Further
developments to improve this new method (yields and
enantioselectivity) are underway and will be reported.
14. Deamici, M.; Frolund, B.; Hjeds, H.; Krogsgaardlarsen, P.
Eur. J. Med. Chem. 1991, 26, 625–631.
15. Aggarwal, V. K.; Fang, G. Y.; Charmant, J. P. H.; Meek, G.
Org. Lett. 2003, 5, 1757–1760.
16. Tweedie, V. L.; Allabash, J. C. J. Org. Chem. 1961, 26,
3676–3681.
17. Gaunt, M. J.; Yu, J. Q.; Spencer, J. B. J. Org. Chem. 1998, 63,
4172–4173.
18. Freifelder, M.; Moore, M. B.; Vernsten, M. R.; Stone, G. R.
J. Am. Chem. Soc. 1958, 80, 4320–4323.
19. Cooper, G.; Irwin, W. J. Org. Mass Spectrom. 1975, 10,
885–895.
20. Aitken, S. J.; Grogan, G.; Chow, C. S. Y.; Turner, N. J.;
Flitsch, S. L. J. Chem. Soc., Perkin Trans. 1 1998, 3365–3370.