Guo-ping
Lu
ab,
Chun
Cai
b and
Bruce H.
Lipshutz
*a
aDepartment of Chemistry & Biochemistry, University of California, Santa Barbara, California 93106, USA. E-mail: lipshutz@chem.ucsb.edu
bChemical Engineering College, Nanjing University of Science & Technology, Nanjing, Jiangsu 210094, P. R. China
First published on 1st October 2012
A nonionic amphiphile, TPGS-750-M, enables efficient Stille couplings between a wide range of substrates to be conducted in water as the only medium, in most cases at room temperature.
Ideally, Stille couplings would be carried out at room temperature, in water as the only medium. Several nanometal catalysts have been designed to enable Stille couplings to be run under such conditions.7 However, opportunities for variation in substrate type are limited. More commonly, high temperatures are required in the presence of various Pd catalysts.8 Recently, we have reported that TPGS-750-M (polyethanol-α-tocopherylsuccinate; shown below) is an excellent commercially available “designer” surfactant that self-assembles in water to form nanomicelles within which several cross-couplings efficiently take place.9 Herein, we describe a new protocol for conducting Stille couplings in an aqueous solution containing TPGS-750-M, in most cases at room temperature.
Scheme 1 Stille couplings with aryl bromides in water at room temperature. |
Based on these initial results, both 3-chlorotoluene and 2-furyltributyltin were selected to investigate potential couplings between aryl chlorides, and to further optimize reaction conditions to be applied to couplings involving more challenging organotin partners (Table 1). Initially, several commonly used catalysts and additives were screened (entries 1–14). The combination of Pd(P(t-Bu)3)2 and DABCO4c,11 emerged as the best choice. Remarkably, the limited conversion initially observed could be increased dramatically by adding one equivalent of NaCl (entries 15, 16), but not NaF (entry 17), presumably due to the enlarged, reorganized micelles which offer increased surface area and thus, increased binding constants for substrates and catalysts.9 Alternatively, the positive effects of chloride on the palladium center may be operative.12 A good yield of product 4 (85%) could be achieved by slight heating to 50 °C. The corresponding coupling run “on water” (i.e., in the absence of TPGS-750-M), led to lower conversion (entry 16 vs. entry 18).13 Screening various surfactants indicated that both CTAB, an ionic surfactant (entry 22), and the neutral amphiphile TPGS-750-M were equally effective in this model reaction.
Entry | Surfactant | Additive | Conversionb (%) |
---|---|---|---|
a Conditions: aryl halide (0.250 mmol), organotin reagent (0.275 mmol), Pd(P(t-Bu)3)2 (0.005 mmol for entries 1–22, 0.010 mmol for entries 23–28), solution of aqueous surfactant (2 wt.%, 1 mL), additive (0.750 mmol) rt, 24 h. b Conversion determined by GC. c The catalyst is Pd(P(t-Bu)2OH)2Cl2 (2 mol%). d The catalyst is Pd(OAc)2/2Xphos (2 mol%). e The catalyst isPd2(dba)3/4P(o-Tol)3 (1 mol%). f 20 mol%. g 1.0 equiv. h At 50 °C. i Isolated yield. j At 40 °C. | |||
1c | TPGS-750-M | / | NR |
2d | TPGS-750-M | / | 20 |
3e | TPGS-750-M | / | NR |
4 | TPGS-750-M | / | 27 |
5 | TPGS-750-M | CuIf | NR |
6 | TPGS-750-M | LiCl | 14 |
7 | TPGS-750-M | CsF | 4 |
8 | TPGS-750-M | Bu4NF | 17 |
9 | TPGS-750-M | Net3 | 30 |
10 | TPGS-750-M | NaOH | 7 |
11 | TPGS-750-M | K2CO3 | 34 |
12 | TPGS-750-M | K2CO3 + CuClf | 17 |
13 | TPGS-750-M | K2CO3 + ZnBr2f | 12 |
14 | TPGS-750-M | DABCO | 37 |
15 | TPGS-750-M | NaClg | 32 |
16 | TPGS-750-M | DABCO + NaCl g | 50, 85 h , i |
17 | TPGS-750-M | DABCO + NaFg | 35 |
18 | None | DABCO + NaClg | 39 |
19 | SDS | DABCO + NaClg | 50 |
20 | Trition X-100 | DABCO + NaClg | 40 |
21 | Brij 30 | DABCO + NaClg | 39 |
22 | CTAB | DABCO + NaClg | 55 |
23 | TPGS-750-M | / | 12 |
24 | TPGS-750-M | K2CO3 | 14 |
25 | TPGS-750-M | DABCO | 44 |
26 | TPGS-750-M | DABCO + NaCl g | 76 i , 85 i , j |
27 | CTAB | DABCO + NaClg | 68i |
28 | None | DABCO + NaClg | 12 |
On the other hand, in the reaction between 2-bromo-1,3-dimethylbenzene and (E)-3-(tributylstannyl)allyl acetate, TPGS-750-M proved to be more effective than CTAB, run under otherwise identical conditions (entry 26 vs. 27). Very low conversion to the desired product 5 was observed when the reaction was conducted “on water” (entry 28), indicative of the vagaries associated with this type of approach to cross-couplings, as observed previously on many occasions.13
With optimized conditions in hand, several combinations of aryl halides and organotin reagents were investigated to ascertain the scope of the protocol (Table 2). In general, aryl bromides coupled smoothly at room temperature. Tributylphenyltin appeared to be a less active coupling partner than other organotin reagents. Although in the case of p-bromoanisole (entry 5), P(P(t-Bu)3)2 led to full conversion at 50 °C, a poor yield of desired product was obtained due to homocoupling of the bromide. Switching to Pd2(dba)3/P(o-Tol)3 with increased catalyst loading (4 mol%) was found to enhance the yield at the expense of homocoupling.
Entry | X | Time (h) | T (°C) | Product | Yieldb (%) |
---|---|---|---|---|---|
a Conditions: aryl halide (0.250 mmol), organotin reagent (0.275 mmol), Pd(P(t-Bu)3)2 (0.005 mmol), aqueous TPGS-750-M solution (2 wt.%, 1 mL), DABCO (0.750 mmol), NaCl (0.250 mmol). b Isolated yield. c 2 mol% Pd2(dba)3 and 8 mol% P(o-Tol)3 are used as catalyst. d 4 mol% Pd(P(t-Bu)3)2 is used. e A mixture of 15 and 4-butyl-2-nitrotoluene (3:1) determined by 1H NMR. f Trimethylphenyltin (1.1 equiv) used. | |||||
1 | Br | 24 | rt | 93 | |
2 | Br | 4 | rt | 88 | |
3 | Br | 4 | rt | 94 | |
4 | Cl | 24 | rt | 94 | |
5 | Br | 24 | 50 | 39, 80c | |
6 | Br | 4 | rt | 91 | |
7 | Br | 5 | rt | 92 | |
8 | Br | 1 | rt | 95 | |
9 | Cl | 24 | 50 | 75d | |
10 | Br | 4 | rt | 97 | |
11 | Cl | 24 | 50 | NRd | |
12 | Br | 5 | rt | 97 | |
13 | Cl | 24 | 60 | 78d,e | |
14 | Cl | 24 | 60 | 92d,f | |
15 | Cl | 24 | 50 | 45d | |
16 | Cl | 24 | 60 | 73d |
The analogous reactions of aryl chlorides, not surprisingly, were more sluggish. However, in some cases, e.g., 2-furyltributyltin and p-chlorobenzonitrile, the coupling was successful even at room temperature (entry 4). Generally, the couplings took place under the influence of mild heat, along with an increase in catalyst loading to 4 mol% (entries 9, 13–16). No product was formed, however, in the reaction of an electron-rich aryl chloride (e.g., o-chloroanisole) with tributyl(phenylethynyl)-stannane (entry 11). It should be noted that the byproduct 4-butyl-2-nitrotoluene was formed when 4-chloro-2-nitrotoluene was coupled with either of two tributylstannanes (entries 13, 15). The by-product in the case of Bu3SnPh could be avoided by switching to the trimethylstannyl analog (entry 14).
To compare and contrast these couplings, guided by the principles of green chemistry,5 with traditional Stille coupling conditions, a direct comparison was performed between 3-bromobenzothiophene and 2-furyltributyltin (Scheme 2). Rather than an organic solvent (dioxane), additives (excess cesium fluoride), and heat (80 °C),14 the “green” protocol afforded a considerably higher yield, done in water at room temperature, and in a shorter time frame.
Scheme 2 The reaction of 3-bromobenzothiophene with 2-furyltributyltin. |
Likewise, Stille couplings with alkenyl halides could also be effected under otherwise identical conditions (Table 3). Both alkenyl iodides and bromides readily participate at room temperature to afford products 18–21 (entries 1–4). In the case of (Z)-β-bromostyrene, significant undesired Z-to-E isomerization,15 as well as homocoupling, took place (entry 5). However, switching catalysts from Pd(P(t-Bu)3)2 to Pd2(dba)3/P(o-Tol)3 led to retention of stereochemistry giving Z-22 in high yield, while avoiding homocoupling (entry 6).
Entry | X | Time (h) | Product | Yieldb (%) | |
---|---|---|---|---|---|
a Conditions: alkenyl halide (0.250 mmol), organotin reagent (0.275 mmol), Pd(P(t-Bu)3)2 (0.005 mmol), aqueous TPGS-750-M solution (2 wt.%, 1.0 mL), DABCO (0.750 mmol), NaCl (0.250 mmol), rt. b Isolated yield. c The Z/E ratio of β-bromostyrene is 96/4. d Z/E ratio of product is determined by GC on the crude products. e 1 mol% Pd2(dba)3 and 4 mol% P(o-Tol)3 are used as catalyst. | |||||
1 | I | 6 | 18 | 88 | |
2 | I | 4 | 19 | 91 | |
3 | Br | 24 | 20 | 89 | |
4 | I | 2 | 21 | 94 (Z/E = 99/1) | |
5c | Br | 24 | 22 | 50 (Z/E = 90/10)d | |
6c | Br | 24 | 90 (Z/E = 96/4)d,e |
Interestingly, while the literature illustrates a traditional coupling between Z-alkenyl triflate 23 and tributyl-4-methoxyphenylstannane in NMP leading to isomerized E-product 24,16 the desired Z-24 is formed under micellar conditions (Scheme 3). To further demonstrate the potential of this methodology, dienoate 25, a crucial intermediate en route to mammalian V-ATPase inhibitor 26,17 was smoothly generated by reaction between (E)-ethyl 3-iodoacrylate and (E)-3-(tributylstannyl)-prop-2-en-1-ol, yielding the unprotected alcohol 25 in water at room temperature (Scheme 4).
Scheme 3 Stille couplings with a Z-alkenyl triflate. |
Scheme 4 A potential application of Stille couplings to total synthesis. |
Finally, the prospects for recycling aqueous solutions of TPGS-750-M used in Stille couplings were studied (Table 4). After the successful coupling of 2-bromo-1,3-dimethylbenzene and tributyl(phenylethynyl)stannane under optimized conditions (entry 1), in-flask extractions with minimal amounts of hexane led to facile product isolation. The extraction also removed much of the Pd catalyst, accounting for the reduced yield (22%; entry 2). The addition of fresh catalyst to each recycle increased the level of conversion and the resulting yields returned to that seen in the initial experiment (entries 3–5).
Entry | Run | Time (h) | Yieldb (%) |
---|---|---|---|
a Conditions: 2,6-dimethylbromobenzene (0.250 mmol), tributyl phenylethynyltin (0.275 mmol), Pd(P(t-Bu)3)2 (0.005 mmol), aqueous TPGS-750-M solution (2 wt.%, 1.0 mL), DABCO (0.750 mmol), NaCl (0.250 mmol), rt. b Isolated yield. c Additional 2 mol% Pd(P(t-Bu)3)2 is used. d Yield after 4 h with additional 2 mol% Pd(P(t-Bu)3)2. e No additional Pd(P(t-Bu)3)2 added; conversion determined by GC. | |||
1 | 1 | 4 | 91 |
2c | 2 | 4 | 80d (22)e |
3c | 3 | 6 | 92 |
4c | 4 | 6 | 89 |
5c | 5 | 6 | 88 |
Footnote |
† Electronic supplementary information (ESI) available: More experimental details and characterization data of all products. See DOI: 10.1039/c2gc36042j |
This journal is © The Royal Society of Chemistry 2013 |