Further improvements of the dibutyl tin oxide-catalyzed regioselective diol tosylation

Article abstract of DOI:10.1016/j.tetlet.2009.11.026

In this Letter, we report that selective monotosylation of a 1,2-diol is possible using only 0.1 mol % of Bu₂SnO. More interestingly, we found that the corresponding tin acetal 3b gave faster conversions and more reproducible reaction times. Moreover, the loading of this catalyst could be as low as 0.05-0.005 mol %.

Full text of DOI:10.1016/j.tetlet.2009.11.026

Tetrahedron Letters 51 (2010) 579–582
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Further improvements of the dibutyl tin oxide-catalyzed
regioselective diol tosylation
*
Michel Guillaume , Yolande Lang
Chemical Process Research, Johnson & Johnson Pharmaceutical Research and Development, Turnhoutseweg, 30, 2340 Beerse, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
In this Letter, we report that selective monotosylation of a 1,2-diol is possible using only 0.1 mol % of
Bu2SnO. More interestingly, we found that the corresponding tin acetal 3b gave faster conversions and
more reproducible reaction times. Moreover, the loading of this catalyst could be as low as 0.05–
0.005 mol %.
Received 2 October 2009
Revised 29 October 2009
Accepted 9 November 2009
Available online 18 November 2009
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
the uncatalyzed reaction gave only 76% of desired monotosyl
derivative 6a, accompanied by 8% of the starting material 2a and
The use of dibutyl tin oxide 1 (Bu2SnO) is well known for the
regioselective derivatization of vicinal diols. Since Shanzer’s origi-
nal Letter,1 the stoichiometric process has been widely applied in
carbohydrate chemistry.2,3 The reaction can be generalized to func-
tionalize other diols as well.4
16% ditosylated product 7a.
Given these disappointing results, we decided to use 1 in order
to improve the selectivity of the tosylation. Some of our results are
summarized in Table 1.
In order to limit the Sn content in the API,11 we first decreased
the quantity of the tin catalyst. To our delight, we noticed that
0.1 mol % Bu2SnO 1 was sufficient for the reaction to proceed
smoothly (entry 14). However, when the amount of 1 was de-
creased to 0.05%, the reaction became too slow (50% conversion
after 16 h, entry 15), as in the case of the uncatalyzed reaction
for which only 60% conversion was noticed after 4 days at 25 °C.
Unlike Martinelli, we observed that the use of diisopropyl ethyl
amine (iPr2NEt, Hünig’s base) gave product of better purity than
with Et3N.
The catalytic process, described in 1999 by Martinelli et al.,5 en-
ables the selective tosylation of a diol with only 2 mol % of Bu2SnO
in the presence of a classical base such as triethylamine (Et3N). In a
more recent paper,6 the same author detailed the mechanistic as-
pects of the reaction (Scheme 1): the intermediate formation of
the corresponding tin acetal 3 was postulated in accordance with
the reactivity of tin acetals isolated in the stoichiometric process.
Quite recently, Servi described the use of 2 mol % catalytic Sn
acetal derived from the reacting diol itself.8
In this Letter, we wish to report the results from our own re-
search group.9 Most notably, we have observed: (i) that the quan-
tity of Bu2SnO 1 could be decreased from 2 mol % to 0.1 mol %, (ii)
an important reactivity difference between 1 and the correspond-
ing Sn acetal, (iii) that the generic Sn acetal catalyst 3b could be
used for selective diol tosylation, (iv) that the use of 3b led to
diminished catalyst loading from 0.1 mol % to 0.05 and even
0.005 mol % in some cases, and (v) the generality of the process
on different commercial diol substrates.
More surprisingly, in the course of the optimization work, a
rather sharp temperature peak was observed after 5 h of reaction,
as shown in Scheme 3.
We assume that this induction period is due to the slow forma-
tion of the reactive tin acetal intermediate of type 3. The subse-
quent steps of the catalytic cycle (selective TsCl addition, metal
exchange) probably proceed much faster, as shown in Scheme 4.
To verify our assumption, we synthesized the tin acetal 3b de-
rived from Bu2SnO and ethylene glycol3 (Scheme 5).
In turn, the catalyst 3b was used in the tosylation reaction of 2a
into 6a (Table 3). As expected from our mechanistic insight, the
reaction began immediately and went to completion with a cata-
lyst loading as low as 0.05%. (Table 2).
In order to test the generality of our new methodology, several
commercially available diols were subjected to the reaction condi-
tions. Some pertinent results are outlined in Table 3.
It appears that the Sn-derivative-catalyzed reaction allows
higher yields and better selectivities than the uncatalyzed reaction.
Moreover, the use of tin acetal 3b allows further decrease of
2. Results and discussion
During the synthesis of the antipsychotic R209130, we wanted
to selectively tosylate diol 2a as shown in Scheme 2.10
Initially, we wanted to avoid the use of tin derivatives because
of the potential toxicity associated with organotins: unfortunately,
* Corresponding author.
E-mail address: mguillau@its.jnj.com (M. Guillaume).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.11.026

580
M. Guillaume, Y. Lang / Tetrahedron Letters 51 (2010) 579–582
Bu
Bu
Bu
O
Bu
Cl
OH
Sn
TsCl, Et3N
Bu2SnO 1
Sn
O
OH
O
-H2O
R
OTs
R
R
2
3
4
Et3N.HCl
OH
OH
R
Bu
O
Bu
Cl
Et3N
Sn
OH
OTs
OH
R
R
6
5
Scheme 1. Bu2SnO-catalyzed monotosylation of a diol.7
OH
OTs
OH
NHMe
OH
TsCl
conditions
O
OAc
OAc
F
F
F
R209130
2a
6a
+ ditosylated derivative
7a
Scheme 2. Reaction of diol 2a with TsCl.
Table 1
Optimization of reaction conditions for Bu2SnO-catalyzed tosylation of diol 2a
Entry
Conditions
T (°C)
Conversiona
Selectivity [6a]:[7a]a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
No Bu2SnO, TsCl (1.5 equiv), pyridine (10 equiv)
Bu2SnO (10%), TsCl (1.5 equiv), Et3N (1.5 equiv)
Bu2SnO (10%), TsCl (1.5 equiv), Et3N (1.5 equiv)
Bu2SnO (2%), TsCl (1.1 equiv), Et3N (1.1 equiv)
Bu2SnO (1%), TsCl (1.05 equiv), Et3N (1.05 equiv)
Bu2SnO (2%), TsCl (1.05 equiv), Et3N (1.05 equiv)
Bu2SnO (5%), TsCl (1.05 equiv), Et3N (1.05 equiv)
Bu2SnO (1%), TsCl (1 equiv), Et3N (1.05 equiv)
Bu2SnO (0.5%), TsCl (1 equiv), iPr2NEt (1.05 equiv)
Bu2SnO (1%), TsCl (1 equiv + 0.1 equiv), iPr2NEt (1.05 equiv + 0.1 equiv)
Bu2SnO (0.5%), TsCl (1.1 equiv), iPr2NEt (1.2 equiv)
Bu2SnO (0.5%), TsCl (1.05 equiv), iPr2NEt (1.2 equiv)
No Bu2SnO, TsCl (1.05 equiv), iPr2NEt (1.2 equiv)
Bu2SnO (0.1%), TsCl (1.05 equiv), iPr2NEt (1.2 equiv)
Bu2SnO (0.05%), TsCl (1.05 equiv), iPr2NEt (1.2 equiv)
25
40
25
25
25
25
25
25
25
25
25
25
25
25
25
92%
86:14
67:33b
90:10
94:6
94:6
96:4
96:4
92:8
98:2
91:9
>99%
>99%
>99%
>99%
>99%
>99%
87%
97%
93%
98%
>99%
96:4
>99:1
Only 60% conversion after 4 days; presence of 7a too!
97% 96:4
Only 50% conversion after 16 h at rt
a
HPLC, area%. Unless stated otherwise, samples were analyzed after 16 h which generally corresponds to a steady state.
Formation of impurity arising mainly from ring-closure of monotosylated derivative.
b
catalyst loading, from 0.1 mol % to 0.005 mol %. Below this limit, a
35
30
25
20
15
10
5
negative influence on the yield and/or selectively is observed.
In summary, we have shown that the loading of Bu2SnO 1 in the
tin-catalyzed selective tosylation of diols 2a could be diminished
from 2 mol % to 0.1 mol %. Moreover, the sharp exothermicity ob-
served after 5 h led us to the hypothesis that the reactive species,
the Sn acetal 3, was formed slowly, whereas subsequent steps of
the catalytic cycle were fast. In accord with this hypothesis, addi-
tion of the generic acetal 3b instead of Bu2SnO led to a rapid and
highly selective reaction with catalyst loading that have been re-
duced from 0.1 mol % to 0.05 mol % and even further down to
0.005 mol % with some diol substrates.
0
0
2
4
6
8
10
12
reaction time (hours)
Scheme 3. Temperature evolution of the reaction.

M. Guillaume, Y. Lang / Tetrahedron Letters 51 (2010) 579–582
581
Bu
Bu
Bu
O
Bu
Cl
OH
Sn
Bu2SnO
TsCl,Et 3N
Sn
O
OH
O
-H2O
R
OTs
R
R
SLOW
2
3
4
FAST
Et3N.HCl
OH
OH
R
Bu
O
Bu
Cl
Et3N
Sn
OH
OTs
OH
R
R
6
Scheme 4. Catalytic cycle for the TsCl addition on diol 2 catalyzed by Bu2SnO 1.
HO OH
toluene
Sn
3. Experimental section
Bu2SnO
O
O
General procedures: All materials were purchased from commer-
cial suppliers and used without further purification. All reactions
were conducted under an atmosphere of nitrogen. Whereas in
the lab only glass vessels were used, both steel and glass-lined ves-
sels were used in the pilot plant. For each reaction, a sample of the
reaction mixture was collected and analyzed by means of HPLC.
Scheme 5. Synthesis of tin acetal derivative 3b.
Table 2
Optimization of reaction conditions for tin acetal-catalyzed tosylation of diol 2a
Entry Catalyst loading [3a]
Conversion after 24 h
(%)
Selectivity
(%)
[6a]:[7a]*
1
2
3
0.1
0.05
0.005
97
97
84
96:4
>99:1
>99:1
3.1. 2,2-Dibutyl-[1,3,2]dioxastannolane (3b)
The following experiment was only performed at lab scale
(0.1 mol).
*
HPLC area %.
Table 3
Monotosylation reaction of commercial diols
Sn
Uncatalyzed
Bu2SnO (mol %)
0.1
O
O
Entry
Substrates
0
0.1
0.05
0.01
0.005
0.001
0.0005
%Conversion selectivity
OH
OH
O
11%
64:36
96%
95:5
97%
97:3
97%
97:3
92%
98:2
94%
95:5
85%
95:5
81%
89:11
1
OMe
OH
OH
OH
40%
>99:1
95%
99:1
97%
>99:1
98%
>99:1
92%
99:1
87%
99:1
70%
96:4
2
3
/
/
OH
O
41%
54:46
95%
99:1
95%
99:1
95%
98:2
96%
95:5
92%
93:7
63%
70:30
OH
4
5
0%
2%
93%
85%
88%
86%
88%
86%
84%
68%
83%
/
0
/
/
/
OH
*
OH
OH *
Substrates 1–3: % conversion and selectivity are based on HPLC area%.
Substrates 4 and 5: % conversion and selectivity are based on GC area%.
*
The GC chromatogram did not show any ditosylation or starting material, analysis with HPLC showed some ditosylation.

582
M. Guillaume, Y. Lang / Tetrahedron Letters 51 (2010) 579–582
To a suspension of 1 (24.9 g, 0.1 mol) in toluene (100 ml, 1 L/
To 2a in toluene (34 L) was added 1 (9 g, 0.25 g/mol) and the
mixture was stirred at 25 °C for 1 h. N,N-Diisopropyl ethyl amine
(5.0 kg, 38.4 mol) was added, followed by tosyl chloride (7.0 kg,
36.8 mol). After stirring the reaction mixture at 25 °C for 16 h,
HCl 1 N (1.35 equiv) in water was added. The pH of the aqueous
layer was in the range of 1–2. The organic layer was dried over
Na2SO4 (3.5 kg), filtered, and the crude solution was used as such
in the next step.
mol), ethylene glycol (28 ml, 5 equiv) was added at 25 °C. Water
was removed azeotropically at 110–114 °C and the reaction mix-
ture was stirred at that temperature for 5 h. After gradual cooling
(110 °C?20 °C over 12 h), the precipitate 3b was filtered, washed
with toluene (25 ml; 0.25 L/mol), and dried at 40 °C under vacuum.
Yield: 27.4 g (93%).Mp 230–231 °C.
Compound 3b was used as such in subsequent experiments. For
analytical purposes, a 10 g sample was recrystallized from toluene
(40 ml, 4 ml/g) with gradual cooling (110 °C?20 °C over 10 h).
Anal. Calcd for C10H22O2Sn: C, 41.00; H, 7.57. Found: C, 40.68; H,
7.70.
Estimated yield: 80%.
References and notes
1H NMR—CDCl3: d 1.0 (6H, t, J = 7.3 Hz), 1.3 (m, 4H), 1.4 (m, 4H),
1.6 (m, 4H), 3.6 (s, 4H).
1. Shanzer, A. Tetrahedron Lett. 1980, 21, 221.
2. David, S.; Hanessian, S. Tetrahedron 1985, 41, 643.
3. Walkup, R. E.; Vernon, N. M.; Wingard, R. E., Jr. EP 448413A1, 1991.
4. Boons, G.-J.; Castle, G. H.; Clase, A.; Grice, P.; Ley, S. V.; Pinel, C. Synlett 1993,
913.
5. Martinelli, M. J.; Nayyar, N. K.; Moher, E. D.; Dhokte, U. P.; Pawlak, J. M.;
Vaidyanathan, R. Org. Lett. 1999, 1, 447.
6. Martinelli, M. J.; Vaidyanathan, R.; Pawlak, J. M.; Nayyar, N. K.; Dhokte, U. P.;
Doecke, C. W.; Zollars, L. M. H.; Moher, E. D.; Khau, V. V.; Košmrlj, B. J. Am. Chem.
Soc. 2002, 124, 3578.
7. For sake of simplicity, we omitted the dimeric structures indicated in
Martinelli’s paper.
8. Fasoli, E.; Caligiuri, A.; Servi, S.; Tessaro, D. J. Mol. Catal. A 2006, 244, 41.
9. Guillaume, M.; Lang, Y. WO 058902, 2008.
10. Compernolle, F.; Toppet, S.; Brossette, T.; Mao, H.; Koukni, M.; Kozlecki, T.;
Medaer, B.; Guillaume, M.; Lang, Y.; Leurs, S.; Hoornaert, G. J. Eur. J. Org. Chem.
2006, 1586–1592.
11. The maximum allowed Sn content in the drug substance was 20 ppm. Because a
recrystallization is performed in a subsequent step, >99.9% Sn is removed and
after crystallization of the final compound, we only find 2 ppm of Sn, which
corresponds to an overall removal yield of >99.99%. Sn content is determined
with Inductively Coupled Plasma (ICP) methodology.
3.2. Acetic acid 8-fluoro-11-[2-hydroxy-3-(toluene-4-
sulfonyloxy)-propyl]-10,11-dihydro-5H-dibenzo[a,d]cyclo-
hepten-10-yl ester (6a)
Laboratory procedure (0.1 mol scale).
To a solution of 2a (34.4 g, 0.1 mol) in toluene (1.2 L/mol) was
added 1 (25 mg, 0.1 mol %) at 25 °C. The mixture was stirred for
1 h. Diisopropyl ethylamine (21 ml, 1.2 equiv) was added and the
reaction mixture was stirred for 5 min. Tosyl chloride (20 g,
1.05 equiv) was added and the reaction mixture was stirred at that
temperature for 16 h. Hydrochloric acid 1 N (150 ml) was added
and the mixture was stirred vigorously. pH of the aqueous
phase = 1–2; the organic phase was filtered over sodium sulfate
and used further in the next step.
12. Compound 6a was an oil and we did not purify or isolate it. The yield was
estimated based on the 65% yield over two steps.
Estimated yield: 80%.12
Pilot plant procedure (35 mol scale).