Convenient preparation of ytterbium(III) chalcogenolate complexes by insertion of ytterbium into chalcogen-chalcogen bonds. Application in the ring-opening of epoxides

Article abstract of DOI:10.1016/S0040-4039(00)00687-0

Convenient conditions are reported for the preparation of ytterbium(III) chalcogenolate complexes by insertion of ytterbium metal into the chalcogen- chalcogen bond of disulfides, diselenides, and ditellurides. The resulting complexes have been found to t

Full text of DOI:10.1016/S0040-4039(00)00687-0

Tetrahedron Letters 41 (2000) 4923±4927
Convenient preparation of ytterbium(III) chalcogenolate
complexes by insertion of ytterbium into chalcogen±chalcogen
bonds. Application in the ring-opening of epoxides
Jennifer Dowsland, Fiona McKerlie and David J. Procter*
Department of Chemistry, The Joseph Black Building, University of Glasgow, Glasgow G12 8QQ, UK
Received 3 April 2000; accepted 27 April 2000
Abstract
Convenient conditions are reported for the preparation of ytterbium(III) chalcogenolate complexes by
insertion of ytterbium metal into the chalcogen±chalcogen bond of disul®des, diselenides, and ditellurides.
The resulting complexes have been found to transfer arylsulfanyl, -selenanyl, and -telluranyl groups to
epoxides in a facile ring-opening reaction. The ytterbium(III) chalcogenolate complexes appear to perform
a dual role in both activating the epoxide, due to their Lewis acidic nature, and supplying a nucleophile to
the coordinated substrate. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: lanthanides; epoxides; sul®des; selenium and compounds; tellurium and compounds.
Organosulfur and selenium chemistry occupies an important place in organic synthesis.1 The
development of new methods for the introduction of sulfur-, selenium-, and tellurium-containing
groups into organic molecules, particularly in a stereocontrolled manner, remains a signi®cant
challenge. Our interest in organolanthanide-mediated transformations2 has led us to consider
lanthanide(III) chalcogenolate complexes as reagents for the transfer of organosulfanyl, -selena-
nyl, and -telluranyl groups. We believed such complexes would be capable of performing a dual
role in reactions with substrates: initial coordination of the substrate to the Lewis acidic lantha-
nide(III) centre would activate the substrate, and the complex would then provide a nucleophile
for reaction with the activated substrate. We report here a convenient route to ytterbium(III)
chalcogenolate complexes and their application in the ring-opening of epoxides.
Relatively few lanthanide(III) chalcogenolate complexes have been reported,3 5 and their
application in organic synthesis has been limited.6,7 Lanthanide(III) chalcogenolate complexes
have previously been prepared from dichalcogenides;4,6f h however, few reports of their preparation
by the insertion of lanthanide metals, with no metal co-reductant,6f h into the chalcogen±chalocogen
bond of dichalcogenides have appeared.5f,7 Unfortunately, in all previous examples, unsatisfactory
* Corresponding author. E-mail: davidp@chem.gla.ac.uk
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00687-0

4924
methods for metal-activation were employed. We wished to develop a more general and convenient
method for the preparation of complexes, not only from disul®des, but also from diselenides and
ditellurides, and examine their reactions with organic electrophiles.
The condition of the metal surface in these reactions is crucial. In a related reaction, activation
was achieved by the reversible complexation of the lanthanide metal to benzophenone in the
presence of HMPA.7a We found that excellent results can be achieved by a much simpler
approach which does not require the use of HMPA. Our approach involves mechanical activation8
followed by the addition of iodomethane and is outlined in Scheme 1.
Scheme 1.
The exact structure of the resultant lanthanide(III) complexes has not yet been determined;
however, homoleptic complexes as shown in Scheme 1 appear likely3 and have previously been
suggested.7a As we expected these lanthanide(III) chalcogenolate complexes to be Lewis acidic
and also to act as a source of nucleophiles, we chose to examine their reaction with epoxides.
Typically, in the reaction with dichalcogenides, the metal is completely consumed after 1±2 h
and a red/brown suspension is formed. Addition of neat epoxide to the complex results in an
instantaneous colour change and the reaction is complete in less than 15 min. In most cases,
approximately 0.5 equivalents of dichalcogenide was recovered from the reaction and up to 2
equivalents of the ring-opened product were obtained.9 It is therefore clear that each complex is
capable of transferring more than one ligand. From the apparent stoichiometry of the reaction,
we believe that each complex transfers at least two ligands. Untransferred ligands may remain on
the lanthanide centre until acidic work-up leads to the regeneration of the dichalcogenide
(Scheme 2).
Scheme 2.
In order to attain steric and electronic saturation, many lanthanide(III) chalcogenolate complexes
contain bridging chalcogenolate groups linking either two lanthanide centres, or a lanthanide and
an alkali metal centre (ate complexes).5 If a dimeric species, such as 1, were formed here, then it
might be expected that the bridging ligands would not be as readily transferred to the epoxide
substrate. Thus, the stoichiometry of the reaction may give an indication as to the nature of the
complex. Further studies aimed at con®rming the structure of these complexes are currently
underway.

4925
A range of dichalcogenides and epoxides have been used in the reaction and our results are
summarised in Table 1.10 Cyclic epoxides are opened eciently to give anti-b-hydroxy sul®des
and selenides in excellent yield. Entries 7±15 show the expected regiochemical tendency for
epoxide-opening at the least hindered position of the epoxide. One exception can be seen in entry
16, where the reaction of styrene oxide gives products arising mainly from attack at the benzylic
position. Reaction with epichlorohydrin (entry 8) occurs chemoselectively to give only the product
of epoxide-opening, even when an excess of reagent is used. This supports the idea that activation
of the substrate by the reagent is crucial to the overall reaction.
Table 1
The reaction of ytterbium(III) chalcogenolate complexes with enantiomerically pure epoxide
substrates has also been studied (entries 13±16). The enantiomeric purity of the products 5a, 5b,
5c, 6aa and 6ab has been determined.11 The reaction of complexes with (R)-(^)-glycidyl methyl
ether gives the expected products, 5a, 5b and 5c, with no loss of enantiomeric purity. The reaction

4926
of (S)-styrene oxide, however, gives 6ab of considerably lower enantiomeric purity than the
starting epoxide. From the same reaction, the minor product 6aa is obtained with no loss of stereo-
chemical integrity. This clearly illustrates that the reaction of styrene oxide proceeds, to a large
extent, via the benzylic cation.
As shown in Scheme 2, the product prior to aqueous work-up is a ytterbium(III) alkoxide. Due
to the oxophilicity of the lanthanides, the lanthanide±oxygen bond is relatively strong and its
formation at the expense of the lanthanide±sulfur bond appears to represent the driving force for the
epoxide insertion reaction. Previous reports have shown that successful cleavage of the lanthanide±
oxygen bond can be achieved by treatment with TMSCl.12 Trapping of the intermediate lantha-
nide(III) alkoxide, formed in reactions of (R)-(^)-glycidyl methyl ether, with TMSCl, was found
to give the protected b-hydroxy selenides in good yield, via a ring-opening/protection sequence.
Presumably the by-product from the reaction is a lanthanide(III) chloride species (Scheme 3).
Scheme 3.
Attempts to employ dialkyl disul®des in the epoxide-opening reaction gave less satisfactory
results. With dibenzyl and dibutyl disul®de, reaction did occur but the ytterbium metal was not
completely consumed. This suggests the reactions of dialkyldisul®des with ytterbium may follow
a di€erent stoichiometry. Separation of the resultant solution from the remaining metal, and
reaction with 1,2-epoxybutane gave the expected sul®des in low yield.13
In summary, we have described a convenient method for the preparation of ytterbium(III)
chalcogenolates by the insertion of ytterbium metal into the chalcogen±chalcogen bond of
dichalcogenides. These complexes have been found to transfer arylsulfanyl, -selenanyl, and -tell-
uranyl groups to epoxides in facile ring-opening reactions. We believe these Lewis acidic com-
plexes perform a dual role in both activating the epoxide, and delivering a nucleophile to the
coordinated substrate.
Studies aimed at con®rming the structure of the ytterbium(III) chalcogenolate complexes and
further exploring their applications in organic synthesis are currently underway in our laboratories.
Typical procedure: Ytterbium metal14 (50 mg, 0.29 mmol, 2 equiv.) was stirred vigorously in a
test tube under a stream of argon for 1 h with intermittent heating. Methyl iodide (4 ml) was
then added followed by diphenyldiselenide (135 mg, 0.43 mmol, 3 equiv.) and the reaction
mixture stirred vigorously at room temperature with intermittent heating. After 1±2 h, the metal
was completely consumed and an orange suspension was formed. Additional THF (0.5 ml) was
then added, followed by neat (R)-glycidyl methyl ether (78 ml, 0.87 mmol, 6 equiv.). After 15 min,
aqueous saturated NH4Cl (3 ml) and H2O (3 ml) were added, followed by citric acid (182 mg,
0.87 mmol, 6 equiv.). After 10 min, 30% ethyl acetate/hexane (5 ml) was added and the aqueous
layer separated and extracted with further portions of 30% ethyl acetate/hexane (2Â5 ml).
The combined organic layers were then dried (Na2SO4) and concentrated in vacuo. Puri®cation

4927
by Kugelrohr distillation (oven temp 75ꢀC, 0.2 mmHg) gave 5b (136 mg, 0.56 mmol, 96%) as a
colourless oil.
Acknowledgements
D.J.P. would like to thank P®zer (FM) for the funding of a ®nal year project.
References
1. Procter, D. J. J. Chem. Soc., Perkin Trans. 1 2000, 835.
2. (a) Johnston, D.; McCusker, C. M.; Procter, D. J. Tetrahedron Lett. 1999, 40, 4913. (b) Johnston, D.; McCusker,
C. F.; Muir, K.; Procter, D. J. J. Chem. Soc., Perkin Trans. 1 2000, 681.
3. For homoleptic lanthanide(III) thiolate complexes with terminal bound `SAr' ligands, see: (a) Cetinkaya, B.;
Hitchcock, P. B.; Lappert, M. F.; Smith, R. G. J. Chem. Soc., Chem. Commun. 1992, 932. (b) Tatsumi, K.;
Amemiya, T.; Kawaguchi, H.; Tani, K. J. Chem. Soc., Chem. Commun. 1993, 773. (c) Mashima, K.; Nakayama,
Y.; Fukumoto, H.; Kanehisa, N.; Kai, Y.; Nakamura, A. J. Chem. Soc., Chem. Commun. 1994, 2523.
4. For other examples of lanthanide(III) complexes with terminal chalcogenolate ligands, see: (a) Berg, D. J.;
Andersen, R. A.; Zalkin, A. Organometallics 1988, 7, 1858. (b) Recknagel, A.; Noltemeyer, M.; Stalke, D.;
Pieper, U.; Schmidt, H.-G.; Edelmann, F. T. J. Organomet. Chem. 1991, 411, 347.
5. For lanthanide(III) complexes with bridging chalcogenolate ligands, see: (a) Aspinall, H. C.; Bradley, D. C.;
Hursthouse, M. B.; Sales, K. D.; Walker, N. P. C. J. Chem. Soc., Chem. Commun. 1985, 1585. (b) Schumann, H.;
Albrecht, I.; Hahn, E. Angew. Chem., Int. Ed. Engl. 1985, 24, 985. (c) Schumann, H.; Albrecht, I.; Gallagher, M.;
Hahn, E.; Janiak, C.; Kolax, C.; Loebel, J.; Nickel, S.; Palamidis, E. Polyhedron 1988, 7, 2307. (d) Schumann, H.;
Albrecht, I.; Gallagher, M.; Hahn, E.; Munchmore, C.; Pickardt, J. J. Organomet. Chem. 1988, 349, 103. (e) Stults,
S. D.; Andersen, R. A.; Zalkin, A. Organometallics 1990, 9, 1623. (f) Mashima, K.; Nakayama, Y.; Kanehisa, N.;
Kai, Y.; Nakamura, A. J. Chem. Soc., Chem. Commun. 1993, 1847. (g) Mashima, K.; Nakayama, Y.; Nakamura,
A.; Kanehisa, N.; Kai, Y.; Takaya, H. J. Organomet. Chem. 1994, 473, 85.
6. For reactions of complexes formed using SmI2, see: (a) Jia, X.; Zhang, Y.; Zhou, X. Tetrahedron Lett. 1994, 35,
8833. (b) Toste, F. D.; LaRonde, F.; Still, I. W. J. Tetrahedron Lett. 1995, 36, 2949. (c) Still, I. W. J.; Toste, F. D.
J. Org. Chem. 1996, 61, 7677. (d) Still, I. W. J.; Martyn, J. P. Synth. Commun. 1998, 28, 913. (e) Zhang, Y.; Yu, Y.;
Lin, R. Synth. Commun. 1993, 23, 189. For chalcogenolate complexes formed using samarium and a co-reductant,
see: (f) Lu, G.; Zhang, Y. Synth. Commun. 1998, 28, 4479. (g) Zhou, L.-H.; Zhang, Y.-M. Synth. Commun. 1999,
29, 533. (h) Wang, L.; Zhang, Y. Heteroat. Chem. 1999, 10, 203.
7. (a) Taniguchi, Y.; Maruo, M.; Takaki, K.; Fujiwara, Y. Tetrahedron Lett. 1994, 35, 7789. (b) Wang, L.; Zhang, Y.
Synth. Commun. 1998, 598.
8. Baker, K. V.; Brown, J. M.; Hughes, N.; Skarnulis, A. J.; Sexton, A. J. Org. Chem. 1991, 56, 698.
9. In some reactions, traces (<5%) of the corresponding b-hydroxy iodides were detected in the crude product
mixtures. We presume these by-products are formed by ring-opening with a `Yb(III)-I' species arising from the
initial formation of `MeYb(II)I' during the activation process.
10. All new compounds were characterised by 1H and 13C NMR, IR and mass spectra, and gave satisfactory elemental
analysis and/or accurate mass spectra.
11. The Mosher's acid esters of 5a, 5b, 6aa and 6ab were prepared and their 1H, 13C and 19F NMR spectra compared,
where possible, to those of derivatives prepared from the corresponding racemic epoxides.
12. (a) Batey, R. A.; Motherwell, W. B. Tetrahedron Lett. 1991, 32, 6211. (b) Nomura, R.; Matsuno, T.; Endo, T.
J. Am. Chem. Soc. 1996, 118, 11666.
13. The reactions of dibenzyldisul®de and dibutyldisul®de, with ytterbium and 1,2-epoxybutane, gave 1-benzylsulfanyl-
butan-2-ol and 1-butylsulfanyl-butan-2-ol in 28 and 30% yields, respectively (based on ytterbium).
14. Ytterbium metal (40 mesh) was purchased from Acros Organics and stored under Argon.