Selective sulfonylation of 1,2-diols and derivatives catalyzed by a recoverable fluorous tin oxide

Article abstract of DOI:10.1016/S0040-4039(00)01720-2

Fluorous tin oxide (C₆F₁₃CH₂CH₂)₂SnO is readily synthesized, exhibits spectra that are generally similar to dibutyltin oxide and appears to exist as an oligomer or polymer. The fluorous tin oxide can be used catalytically to effect the selective monotosylation of 1,2-diols with TsCl/Et₃N, and it can be readily recovered and reused. (C) 2000 Published by Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01720-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 9617–9621
Selective sulfonylation of 1,2-diols and derivatives catalyzed
by a recoverable fluorous tin oxide†
Brian Bucher and Dennis P. Curran*
Department of Chemistry, University of Pittsburgh, Pittsburgh, PA 15260, USA
Received 14 September 2000; accepted 3 October 2000
Abstract
Fluorous tin oxide (C6F13CH2CH2)2SnO is readily synthesized, exhibits spectra that are generally
similar to dibutyltin oxide and appears to exist as an oligomer or polymer. The fluorous tin oxide can be
used catalytically to effect the selective monotosylation of 1,2-diols with TsCl/Et3N, and it can be readily
recovered and reused. © 2000 Published by Elsevier Science Ltd.
The selective reaction of 1,2-diols with electrophiles mediated by dialkyltin oxides has proven
value both within and beyond carbohydrate chemistry.1 In general, a 1,2-diol like 1 is reacted first
with an equivalent amount of dibutyltin oxide 2a (R=Bu)2,3 under dehydrating conditions to
provide an intermediate stannoxane 3a (Eq. (1)). This is then reacted in a separate step with an
electrophile to provide a monofunctionalized product 4, often with high regioselectivity.
Tosylation is shown in Eq. (1), but acylation and a number of other transformations can be
accomplished by this procedure. Recently, Martinelli and co-workers described an important
advance in selective sulfonylation of 1,2-diols by introducing a one-step procedure that requires
only 2% of the tin oxidfle along with triethylamine (Eq. (1)).4 This procedure is attractive because
only a small amount of tin must be removed from the final product; however, chromatography is
still required, and the tin is not recovered for reuse. Even when catalytic amounts of tin are used,
environmental concerns dictate that it be recovered rather than discarded whenever possible.
Traditional stoichiometric tosylation
New catalytic tosylation
R2SnO
OTs
OH
O
Sn
O
OH
OH
OTs
OH
R
2% R2SnO
TsCl
2a,b
(1)
R
Ph
Ph
Ph
Ph
Et3N/TsCl
CH2Cl2
CH2Cl2
benzene
4
3a R = Bu
3b R = CH2CH2C6F13
1
4
* Corresponding author.
†
Dedicated to Professor Harry Wasserman on the occasion of his 80th birthday.
0040-4039/00/$ - see front matter © 2000 Published by Elsevier Science Ltd.
PII: S0040-4039(00)01720-2

9618
We have recently introduced a family of fluorous tin reagents exemplified by tin hydrides that
bear either one [Rf(CH2)nSn(Me)2H] or three [(Rf(CH2)n)3SnH] fluorous chains.5 If suitable
alkylene spacers are used, these reagents mimic their organic parents in reactivity but are readily
separated from crude products by fluorous-organic liquid–liquid or solid–liquid extractions.6
Recovery and reuse of the tin reagents is often possible. The long range goal of this work is to
understand the reaction and separation features of a complete family of fluorous reagents. To
further advance this goal, we decided to make a fluorous analog 2b (R=CH2CH2C6F13) of
dibutyltin oxide 2a and to test its use in selective diol functionalization with a focus on the
Martinelli catalytic sulfonylation procedure. We summarize herein the results of this study.
Bis-perfluorohexylethyltin oxide 2b was synthesized as shown in Eq. (2).7 The Grignard
reagent derived from 5 was coupled with diphenyltin dichloride 6 to give about a 7:1 mixture of
fluorous tin reagent 7 and the Wurtz coupled dimer (C6F13(CH2)4C6F13). These were not easily
separable, so the mixture was treated with chloroacetic acid following the procedure of Kong.8
Recrystallization of the crude product from hexane gave pure bis-chloroacetate 8. This was then
stirred in ether with 2.5 M sodium hydroxide. The ether phase was washed, dried and
evaporated to provide tin oxide 2b as a free-flowing white powder. This product was used
without further purification.
1) Mg/THF
ClCH2CO2H
NaOH
ICH2CH2Rf
Ph2Sn(CH2CH2Rf)2
(ClCH2CO2)2Sn(CH2CH2Rf)2
[OSn(CH2CH2Rf)2]n
ether
2) Ph2SnCl2
(2)
5 Rf = C6F13
6
7
8
2b
Dibutyltin oxide 2a is not monomeric and is thought to exist as a polymer.1–3 Fluorous tin
oxide 2b generally resembles dibutyltin oxide in its spectroscopic behavior. For example, 2b
1
exhibits broad resonances in the H NMR spectrum, but resonances in its 13C and 19F spectra
are sharp and interpretations are straightforward.9 Based on these similarities, we suspect that
2b is oligomeric or polymeric. The following partition coefficients (FC-72/organic solvent) were
measured for 2b: CH2Cl2, 1.7; CH3CN, 0.28; toluene, 1.9.
We first probed the usefulness of tin oxide 2b in a standard two-step procedure (Eq. (1)).7 A
benzene solution of 1-phenyl-1,2-ethanediol 1 and 2b was refluxed for 24 h in the presence of
,
4 A molecular sieves. Evaporation of the benzene then gave stannylidene 3b in 83% yield.
Reaction of this crude product with tosyl chloride in dichloromethane gave tosylate 4 in 80%
yield after aqueous workup and eight washings with FC-72 to remove the tin.10 No resonances
could be detected in the 19F NMR spectrum of 4. The results of this preliminary experiment
were very encouraging. The yields and rates of reactions of the fluorous tin reagent 2b were
roughly comparable to dibutyltin oxide. The reaction succeeded in a standard organic solvent;
fluorinated cosolvents were not required.
Fluorous tin oxide 2b served equally well in Martinelli’s catalytic procedure.4 In two identical
reactions, diol 1 (1 mmol, 1 equiv.), tosyl chloride (1 equiv.), triethylamine (1 equiv.) and
fluorous tin oxide 2b were stirred in dichloromethane for 1 h.11 After acidic workup, one crude
product was taken up in dichloromethane and washed with FC-72 (eight times) to provide the
tosylate 4 in 80% yield. The other product was charged to a column of fluorous reverse phase
silica gel, which was eluted with 9:1 MeOH/H2O to give the tosylate 4 in 80% yield. Like
Martinelli, we observed only the monotosylate in these catalyzed reactions, and this tosylate was
free of fluorous tin products as assessed by 19F NMR spectroscopy.

9619
The generality of the method was tested by reacting the substrates (1 mmol) shown in Table
1 under the catalytic procedure with fluorous tin oxide 2b. These reaction products were purified
by liquid–liquid extraction and the tin oxide was recovered. Only monotosylate products of the
1-alcohol were observed in each case. The reaction in Entry 1 was conducted three times with
the recovered tin oxide. Yields of all reactions were comparable and an average of 92% of the
initial fluorous tin oxide was recovered each time. This reaction was also run on 10 mmol (ꢀ1
g) scale with comparable results. For the triol (Entry 5), only the 1°-alcohol of the 1,2-diol was
tosylated. Reaction times were roughly of the order observed by Martinelli when using
dibutyltin oxide.4a 1,2-Diols reacted very quickly (about 1 h), while alkoxyalcohols (Entries 6
and 7) reacted considerably more slowly (about 1 day).
Since only a small amount of tin is added to the reaction (2 mol%), it is difficult to assess the
effectiveness of the workup in removing the tin. To clearly show that the workup did remove the
Table 1
Monotosylation of alcohols with 2% fluorous tin oxide
Entry
1c
Substrate
Reaction timea (h)
Yield of
Structure of
monotosylate
monotosylateb (%)
1–2
75–80
77
2
1
3
4
1
1
72
70
5
6
1
74
84
23
7
23
94
a Reactions were followed by TLC.
b Isolated yields of purified product.
c Reaction time and yield ranges for multiple runs with recovered tin oxide.

9620
tin, we carefully assessed the product of Entry 2 by using 19F NMR spectroscopy with the CF3
group of the monotosylated product as an internal standard. The 19F NMR spectrum of the
crude product prior to any purification showed seven peaks: one for the CF3 group of the
tosylate and six for the CF2 groups (5) and CF3 (1) of 2b in the expected 50:1 ratio. The mixture
was then divided into two equal parts, one of which was purified by liquid–liquid extraction and
the other of which was purified by solid–liquid extraction. Each of these final products showed
only one peak in its 19F NMR spectrum, which corresponded to the CF3 group of the tosylate;
the resonances of 2b were no longer detectable.
Fluorous tin oxide 2b is easy to make and handle, and the preliminary results suggest that it
will be a valuable alternative to the standard dibutyltin oxide 2a in both catalytic and
stoichiometric procedures. Accordingly, we recommend that the fluorous tin oxide be used in the
catalytic sulfonylation procedure of Martinelli and be tried on other procedures where dibutyltin
oxide is usually the reagent of choice.
Acknowledgements
This work was supported by grants from the National Institutes of Health. Brian Bucher
thanks the ACS Division of Fluorine Chemistry for a Moisson Undergraduate Summer
Research Fellowship.
References
1. (a) Shanzer, A. Tetrahedron Lett. 1980, 21, 221. (b) David, S.; Hanessian, S. Tetrahedron 1985, 41, 643. (c)
O’Donnel, C. J.; Burke, S. D. J. Org. Chem. 1998, 63, 8614. (d) Leigh, D. A.; Martin, R. P.; Smart, J. P.;
Truscello, A. M. J. Chem. Soc., Chem. Commun. 1994, 1373. (e) Reginato, G.; Ricci, A.; Roelens, S.; Scapecchi,
S. J. Org. Chem. 1990, 55, 5132.
2. (a) Pereyre, M.; Quintard, J.-P.; Rahm, A. Tin in Organic Synthesis; Butterworth: London, UK, 1987; p. 342. (b)
Davies, A. G. Organotin Chemistry; VCH: Weinheim, Germany, 1997; p. 176. (c) Chemistry of Tin, 2nd ed.;
Smith, P. J., Ed.; Blackie: London, UK, 1997; p. 290.
3. Rajanbabu, T. V. In Encyclopedia of Reagents for Organic Synthesis; Paquette, L. A., Ed.; Wiley: New York,
1995; Vol. 3, p. 1629.
4. (a) Martinelli, M. J.; Nayyar, N. K.; Moher, E. D.; Dhokte, U. P.; Pawlak, J. M.; Vaidyanathan, R. Org. Lett.
1999, 1, 447. (b) Martinelli, M. J.; Vaidyanathan, R.; Van Khau, V. Tetrahedron Lett. 2000, 41, 3773.
5. (a) Curran, D. P.; Hadida, S. J. Am. Chem. Soc. 1996, 118, 2531. (b) Curran, D. P.; Hoshino, M. J. Org. Chem.
1996, 61, 6480. (c) Horner, J. H.; Martinez, F. N.; Newcomb, M.; Hadida, S.; Curran, D. P. Tetrahedron Lett.
1997, 38, 2783. (d) Hadida, S.; Super, M. S.; Beckman, E. J.; Curran, D. P. J. Am. Chem. Soc. 1997, 119, 7406.
(e) Ryu, I.; Niguma, T.; Minakata, S.; Komatsu, M.; Hadida, S.; Curran, D. P. Tetrahedron Lett. 1997, 38, 7883.
(f) Hoshino, M.; Degenkolb, P.; Curran, D. P. J. Org. Chem. 1997, 62, 8341. (g) Curran, D. P.; Luo, Z.;
Degenkolb, P. Bioorg. Med. Chem. Lett. 1998, 8, 2403. (h) Ryu, I.; Niguma, T.; Minakata, S.; Komatsu, M.; Luo,
Z. Y.; Curran, D. P. Tetrahedron Lett. 1999, 40, 2367. (i) Curran, D. P.; Hadida, S.; Kim, S.-Y. Tetrahedron
1999, 55, 8997. (j) Curran, D. P.; Hadida, S.; Kim, S.-Y.; Luo, Z. Y. J. Am. Chem. Soc. 1999, 121, 6607.
6. Curran, D. P.; Hadida, S.; He, M. J. Org. Chem. 1997, 62, 6714.
7. Full experimental details will be posted at http://preprint.chemweb.com/.
8. Kong, X.; Grindley, B.; Bakshi, P. K.; Cameron, T. S. Organometallics 1993, 12, 4881.
9. 1H NMR (300 MHz, acetone-d6): l 2.50–2.61 (broad band, 4H), 2.77–2.84 (t, 4H); 13C NMR (75 MHz,
acetone-d6) l 123.56–108.52 (m, 12C), 27.13 (m, 2C), 20.48 (t, 2C); 19F NMR (282 MHz, acetone-d6 with CFCl3)
l −125.69, −122.84, −122.35, −121.37, −115.17, −80.56; 119Sn NMR (111.8 MHz, CDCl3 with (CH3)4Sn) l
−59.1.

9621
10. Based on the partition coefficient, this should remove about 99% of the tin. Since only 2 mol% tin is used in the
catalytic procedure, fewer washings may suffice.
11. 1-Phenyl-1,2-ethane diol (1 mmol) was dissolved in CH2Cl2 (5 mL). Triethylamine (1 mmol) and tin oxide 2b
(0.02 mmol) were added. Tosyl chloride was added and the solution was stirred at room temperature for 50 min.
Water (1 mL) was added and the aqueous layer was washed with dichloromethane (2×10 mL). The combined
organic layers were washed with H2O (2×25 mL) and brine (2×25 mL). The organic layer was dried over MgSO4.
Removal of the solvent yielded a mixture of 4 and tin oxide. One part of the resulting mixture was washed with
FC-72 (8×25 mL). The dichloromethane was evaporated to yield 4 and the FC-72 was evaporated to yield 2b. The
other part was transferred to
a column containing fluorous reverse phase silica gel (bonded phase
ꢀOSi(Me)2CH2CH2C6F13) (100 mg). The column was then washed with a mixture of 9:1 methanol:water (3 mL),
followed by THF (3 mL). Evaporation of the methanol:water mixture yielded 4.
.