Longrui
Chen
a,
Devonna
Leslie
b,
Michael G.
Coleman
*b and
James
Mack
*a
aDepartment of Chemistry, University of Cincinnati, Cincinnati, Ohio 45221-0037, USA. E-mail: james.mack@uc.edu
bSchool of Chemistry and Materials Science, Rochester Institute of Technology, Rochester, New York 14623-5604, USA. E-mail: mgcsch@rit.edu
First published on 24th April 2018
Silver and copper foil were found to be effective, versatile and selective heterogeneous catalysts for the cyclopropenation of terminal and internal alkynes under mechanochemical reaction conditions. This methodology enables the functionalization of a wide range of terminal or internal alkynes under ambient, aerobic, and solvent-free conditions. Finally, we have demonstrated a unique and versatile one-pot domino Sonogashira-cyclopropenation mechanochemical reaction for the formation of complex cyclopropenes.
Mechanochemistry is a ‘chemical reaction that is induced by the direct absorption of mechanical energy.’72 While it is unlikely that the solvent-free mechanochemical laboratory procedure comes to the mind of contemporary synthetic chemists as a useful and simple experimental procedure, it is one of the earliest recorded chemical methods known to convert physical energy, namely heat and pressure, into chemical energy.73,74 This ancient synthetic method has been extensively used over the past century for the formation of new macromolecular products, metal–organic frameworks, compositions and cocrystals by grinding, extrusive and milling forces from poorly soluble solid–solid, solid–liquid and liquid–liquid inorganic reactants.75–91 Mechanochemical methods have also been utilized for the dehalogenation of pollutants that are resistant to degradation through chemical, biological, and photolytic processes.92–101 In recent years, solvent-free mechanochemistry has emerged as a ‘green’ alternative to the unsustainable use of petroleum-based solvents in classic organic chemistry transformations.102–106 With respect to fine chemical manufacturing, it is also an effective experimental procedure, known as medicinal mechanochemistry, for the synthesis of many classes of organic compounds, active pharmaceutical ingredients (APIs), and commercially-available drugs.75,89,105–113
Unlike solution-phase organic chemistry, solvent-free ball milling conditions are generally conducted without the need for inert gas, external heat, or solvent. The stainless-steel reactors vials are machined from stainless-steel rods and the organic/inorganic reactants are added in the open atmosphere with one or more stainless-steel balls. In our previous work, diazoacetate and an excess of the unsaturated hydrocarbon (trap) were added to a stainless steel vial lined with a silver foil catalyst and a single stainless steel ball. Similar to the analogous solution-phase cycloaddition reactions, the excess trap serves to suppress any undesired diazoacetate dimerization pathways and does not serve as a solvent in this mechanochemical process. This work demonstrated that solid–solid ball milling of a diazoacetate in the presence of excess alkene was as effective as liquid–liquid ball milling which suggested that efficient mixing of the reaction media was achieved. The silver foil lined reaction vial was mechanochemically ball milled using an SPEX 8000M vibratory mixer/mill® at 18 Hz to furnish cyclopropane compounds in high yields (Fig. 1).71 Moreover, the stainless-steel vials, metal foil, and excess unsaturated hydrocarbons were easily recovered and recycled for later use.
Having established the general boundaries for the catalytic activity and selectivity of metal foil-catalyzed mechanochemical cyclopropenation reactions, various terminal alkynes were investigated using copper foil (Table 2). Electron-rich and electron-poor terminal acetylenes were all obtained in good to excellent yields (entries 1–10). 1-Cyclohexenyl acetylene afforded the cyclopropene product 13 exclusively in good yield (entry 11). 3-Thiophenyl acetylene afforded the corresponding cyclopropene product 14 in 90% yield (entry 12). Remarkably, 4-ethynylaniline cleanly underwent the mechanochemical cyclopropenation in the presence of the unprotected aniline in excellent isolated yield (entry 13). To the best of our knowledge, this is the first instance where an unprotected N–H bond was chemically inert to carbene transfer and afforded cyclopropene 15 without additional protection and deprotection steps.22,115,116 Next, we surveyed a range of electronically diverse diazoacetate compounds and determined that dimethyl diazomalonate and ethyl diazoacetate compounds were ineffective for the copper foil-catalyzed cyclopropenation of phenyl acetylene under solvent-free mechanochemical milling conditions (entries 14–16).
Entry | R1 | Product | Isolated yield (%) |
---|---|---|---|
a Reation conditions: diazoacetate (1.00 eq.), alkyne (5.00 eq.), and copper foil added to a SPEX 8000M mixer/mill (18 Hz) for 16 h. b 1 eq. of methyl p-bromophenyldiazoacetate was used. c 1 eq. of ethyl diazoacetate was used. d 1 eq. of dimethyl diazomalonate was used. e Recovered starting material. | |||
1 | 4-(OCH3)Ph | 3 | 88 |
2 | 4-(CH3)Ph | 4 | 95 |
3 | 3-(CH3)Ph | 5 | 88 |
4 | 4-(C(CH3)3)Ph | 6 | 81 |
5 | 4-(F)Ph | 7 | 85 |
6 | 4-(Br)Ph | 8 | 85 |
7 | 2-(Cl)Ph | 9 | 82 |
8 | 3-(Cl)Ph | 10 | 84 |
9 | 2,4,6-(CH3)Ph | 11 | 78 |
10 | 2-(CH3)-4-(OCH3)Ph | 12 | 90 |
11 | 1-Cyclohexenyl | 13 | 86 |
12 | 3-Thiophenyl | 14 | 90 |
13 | 4-(NH2)Ph | 15 | 94 |
14b | Ph | 16 | 81 |
15c | Ph | 17 | Tracee |
16d | Ph | 18 | Tracee |
Next, we investigated the scope of the silver foil-catalyzed mechanochemical cyclopropenation reaction with various internally substituted acetylenic compounds (Table 3). 1-Phenyl substituted aliphatic alkynes (entries 1 and 2) were isolated in excellent yields (>92%). Interestingly, 1-TMS-2-phenylacetylene underwent cyclopropenation in relatively low yield by GC analysis (entry 3). It is important to note that the analogous solution phase, homogeneous, Ag(I)-catalyzed reaction is completely unreactive.114 Increasing the steric bulk of the silyl group had a detrimental effect on the reactivity of the mechanochemical cyclopropenation methodology (entries 4 and 5), while diaryl-substituted acetylenes were efficiently converted into the cyclopropene products 24–26 in isolated yields up to 85% (entries 6–8). Various electron rich and electron poor aryl diazoacetates were also tested and all afforded excellent yields (>80%) of the corresponding cyclopropene products (entries 10–14). As stated earlier, the electronic character of the diazoacetate plays an important role in its reactivity with the transition metal catalyst, and ultimately, the efficiency of the carbene transfer.117,118 For example, homogeneous Ag(I) salts effectively catalyze an interesting C–Cl insertion with ethyl diazoacetate when using methylene halide or haloform as solvents.114,119 We discovered that eliminating dichloromethane as the commonly used solvent afforded the corresponding cyclopropene products 34 and 35, in the presence of ethyl diazoacetate, under solvent-free silver-foil catalyzed mechanochemical reaction conditions (entries 16 and 17). Finally, diethyl diazomalonate was completely unreactive and only starting materials were recovered (entry 18). This finding is in agreement with previous work that stable metal–carbene complexes are observed in the solution phase reactions of diazomalonate compounds and silver(I) salts.120,121
Entry | R1 | R2 | R3 | Product | Yield (%)b |
---|---|---|---|---|---|
a Reation conditions: diazoatate (1.00 eq.), alkyne (5.00 eq,), and silver foil added to a SPEX 8000M mixer/mill (18 Hz) for 16 h. b Isolated yields unless otherwise stated. c GC yield. d 4-Octyne was used. e Recovered starting material. | |||||
1 | CH2CH3 | Ph | CH3 | 19 | 92 |
2 | (CH2)3CH3 | Ph | CH3 | 20 | 95 |
3 | Si(CH3)3 | Ph | CH3 | 21 | 20c |
4 | Si(CH(CH2))3 | Ph | CH3 | 22 | Tracee |
5 | Si(Ph)3 | Ph | CH3 | 23 | Tracee |
6 | Ph | Ph | CH3 | 24 | 84 |
7 | 4-(CH3)Ph | Ph | CH3 | 25 | 85 |
8 | 4-(Br)Ph | Ph | CH3 | 26 | 65 |
9 | 3-Thiophenyl | Ph | CH3 | 27 | 86 |
10 | CH3 | 4-(OCH3)Ph | CH3 | 28 | 86 |
11 | CH3 | 4-(C(CH3)3)Ph | CH3 | 29 | 84 |
12 | CH3 | 4-BrPh | CH3 | 30 | 89 |
13 | CH3 | 4-CF3Ph | CH3 | 31 | 88 |
14 | (CH2)3CH3 | 4-BrPh | CH3 | 32 | 80 |
15 | (CH2)3CH3 | 4-(CH3)Ph | CH3 | 33 | 79 |
16 | CH3 | H | CH2CH3 | 34 | 80c |
17 | (CH2)3CH3 | H | CH2CH3 | 35 | 85c |
18 | (CH2)3CH3 | CO2CH3 | CH3 | 36 | Tracee |
19d | Ph | CH3 | 37 | 85 |
Having established the efficiency of this mechanochemical reaction, we set out to explore several intermolecular experiments to measure the chemoselectivity of the mechanochemical reaction using an equimolar mixture of olefins and acetylenes in the presence of methyl phenyldiazoacetate (Table 4). After 16 h of mechanochemical milling, we discovered that silver foil highly favors the cyclopropanation of olefins in the presence of terminal acetylenes (entry 1). These findings suggest that the silver foil does not lose its catalytic activity despite known silver acetylide catalyst deactivation pathways for the solution-phase, homogeneous silver-catalyzed counterpart.114 On the other hand, copper foil was mildly chemoselective and favored the cyclopropenation pathway for a mixture monosubstituted alkenes and terminal alkynes (entry 1). The silver foil-catalyzed [2 + 1] cycloaddition was observed to be less selective for 1,2-disubstituted alkenes over their internally substituted acetylenic counterparts (entries 2–5). While the copper foil-catalyzed [2 + 1] cycloaddition of 1,2-disubstituted alkenes and internal alkynes favored the cyclopropanation reaction pathway (entries 2–5). Finally, the mechanochemical milling of diphenyl-substituted olefins and diphenylacetylene resulted in a highly chemoselective ratio of cycloaddition products (entries 6 and 7), where silver foil favored cyclopropenation and copper foil favored cyclopropanation.
Entry | R1 | R2 | Ag foil | Cu foil | ||
---|---|---|---|---|---|---|
Yieldb (%) | Product (a![]() ![]() |
Yieldb (%) | Product (a![]() ![]() |
|||
a Reaction Conditions: diazoacetate (1.00 eq.), alkyne (1.50 eq.), alkyne (1.5 eq.) and metal foil added to a SPEX 8000M mixer/mill (18 Hz) for 16 h. b GC yield. c Determined by GC analysis. d Diastereomeric ratios were not measured. | ||||||
1 | Ph | H | 91 |
1![]() ![]() ![]() ![]() |
92 |
1![]() ![]() ![]() ![]() |
2 | n-Butyl | n-Butyl (trans) | 80 |
39![]() ![]() ![]() ![]() |
40 |
39![]() ![]() ![]() ![]() |
3 | n-Butyl | n-Butyl (cis) | 83 |
39![]() ![]() ![]() ![]() |
48 |
39![]() ![]() ![]() ![]() |
4 | Ph | CH3 (trans) | 90 |
2![]() ![]() ![]() ![]() |
86 |
2![]() ![]() ![]() ![]() |
5 | Ph | CH3 (cis) | 90 |
2![]() ![]() ![]() ![]() |
88 |
2![]() ![]() ![]() ![]() |
6 | Ph | Ph (trans) | 80 |
24![]() ![]() ![]() ![]() |
30 |
24![]() ![]() ![]() ![]() |
7 | Ph | Ph (cis) | 82 |
24![]() ![]() ![]() ![]() |
42 |
24![]() ![]() ![]() ![]() |
In order to further test the chemoselectivity of this mechanochemical [2 + 1] cycloaddition reaction, we extended this study to include compounds containing enyne moieties. The mechanochemical metal foil-catalyzed cycloaddition of enyne moieties afforded astonishingly chemoselective transformations (Table 5).
The silver foil-catalyzed process resulted in the cyclopropenation exclusively for internal alkynes (entries 1 and 2). After simply switching the silver to copper foil, we observed the exclusive formation of the corresponding cyclopropanation products (entries 1 and 2). Interestingly, 1-cyclohexenyl acetylene resulted in complete deactivation of the silver foil catalyst and no cycloaddition product was isolated (entry 3). This finding may suggest that the substrate was deactivated on the silver foil which is contrary to the intermolecular process observed earlier (Table 4, entry 1). Copper foil effectively and selectively converted the trisubstituted olefin into the cyclopropane product 13 in good yield (entry 3). However, the cyclopropanation of 1,1-disubstituted alkene moiety yielded that same cyclopropane product 50 for both silver and copper foil (entry 4). Overall, these highly chemoselective mechanochemical transformations gave rise to interesting building blocks that are useful for other organic transformations.
In addition to developing ‘solvent-free’ mechanochemical organic chemistry methods, we are also interested in realizing strategies for the construction of complex organic compounds from two or more synthetic steps in a single highly atom-economical ‘one-pot’ procedure. In recent years, several multicomponent coupling reactions have been developed by combining well-established individual reactions without the need for extraneous use of solvent in workups or purification steps.122–128 Based on the mechanochemical Sonogashira coupling reaction previously developed by our group,129 we envisioned a three-component coupling of a terminal alkyne, aryl halide and a diazoacetate compound as the first example of a transition-metal catalyzed domino mechanochemical reaction. At the outset, we were cognizant that the terminal alkynes were susceptible to cyclopropenation with copper foil, while silver(I) salts are inert to terminal alkynes but remain catalytically active. Previous work suggests that silver foil may be capable of serving as an effective catalyst130 and co-catalyst131–138 for the Sonogashira coupling reaction, and later, catalyzed the cyclopropenation of the newly formed internal alkynes in the presence of a diazoacetate compound. To test this hypothesis, we first examined the feasibility of the Pd(II)/silver foil-catalyzed Sonogashira mechanochemical reaction. Gratifyingly, we observed good to excellent yields of the internal alkyne product by GC analysis (Table 6).
Entry | X | R1 | R2 | Product | Yieldb (%) |
---|---|---|---|---|---|
a Reaction conditions: halide (1.05 mmol), alkyne (1.00 mmol), base (1.5 mmol), Pd catalyst and silver foil added to a SPEX 8000M mixer/mill (18 Hz) for h. b GC yields. | |||||
1 | I | 4-(Cl)Ph | Ph | 51 | 90 |
2 | I | 4-(Br)Ph | Ph | 52 | 88 |
3 | I | 4-(F)Ph | Ph | 53 | 87 |
4 | Br | 4-(F)Ph | Ph | 53 | 80 |
5 | I | 4-(CH2CH3)Ph | Ph | 54 | 93 |
6 | Br | 4-(CH2CH3)Ph | Ph | 54 | 95 |
7 | Br | 4-(C(CH3)3)Ph | Ph | 55 | 90 |
8 | Br | 4-(CH3)Ph | Ph | 56 | 90 |
9 | I | 4-(CH3)Ph | Ph | 56 | 95 |
10 | I | 4-(Cl)Ph | CH2CH2CH2CH3 | 57 | 92 |
11 | I | 4-(Br)Ph | CH2CH2CH2CH3 | 58 | 88 |
12 | I | 4-(F)Ph | CH2CH2CH2CH3 | 59 | 90 |
13 | I | 4-(CH3)Ph | CH2CH2CH2CH3 | 60 | 93 |
14 | Br | 4-(CH3)Ph | CH2CH2CH2CH3 | 60 | 93 |
15 | Br | 4-(C(CH3)3Ph | CH2CH2CH2CH3 | 61 | 93 |
16 | I | Ph | 4-(Cl)Ph | 51 | 85 |
17 | Br | Ph | 4-(Cl)Ph | 51 | 77 |
18 | I | Ph | 4-(Br)Ph | 52 | 88 |
19 | Br | Ph | 4-(Br)Ph | 52 | 80 |
20 | I | Ph | 4-(F)Ph | 53 | 89 |
21 | I | Ph | 4-(CH2CH3)Ph | 54 | 93 |
22 | Br | Ph | 4-(CH2CH3)Ph | 54 | 86 |
23 | I | Ph | 4-(CH3)Ph | 56 | 95 |
24 | Br | Ph | 4-(CH3)Ph | 56 | 85 |
25 | I | Ph | CH2CH2Ph | 62 | 83 |
26 | I | 4-(Cl)Ph | CH2CH2Ph | 63 | 80 |
27 | Br | 4-(CH3)Ph | CH2CH2Ph | 64 | 85 |
Our initial investigations of the cross coupling product of aryl iodides and bromides and phenyl acetylene afforded excellent yields by GC analysis (80–95%) (entries 1–9). Similar to our original report of the Sonogashira reaction under mechanochemical conditions, the reaction is much faster with iodine than bromine and is virtually unreactive with chlorine.129 Such that entries 2 and 11 only provided coupling across the carbon iodide bond. 1-Hexyne proved to be an equally effective cross coupling partner with aryl iodides or bromides (entries 10–15). A diverse range of p-substituted arylacetylene compounds (51–56) were observed by GC analysis from either iodobenzene or bromobenzene starting materials in good yields (>77%). No diyne byproducts resulting from homo coupling or multiple Sonogashira couplings were detected. Para-substituted aryl iodides and bromides were easily coupled to 4-phenyl-1-butyne to afford the corresponding Sonogashira products (62–64, entries 25–27) in good yield (>78%) by GC analysis.
Given the fact that silver foil is an effective co-catalyst for the cross coupling of aryl halides and terminal acetylenic compounds via mechanochemical milling, we then turned our attention back towards our initial hypothesis that a domino one-pot Sonogashira coupling followed by cyclopropenation would be a straightforward and general method for the synthesis of a diverse library of fully substituted cyclopropene compounds (Table 7).
Entry | Palladium | Vial catalystb | Approach | Reaction time (h) | Yieldc % |
---|---|---|---|---|---|
a Reactions conditions: diazoacetate (1 eq.), aryl halide (3 eq.), K2CO3 (3.5 eq.) and Pd (0.026) catalyst added in a custom made reaction vial. b Lined metal foil in vials as active catalyst during grinding. c All yields are the average of isolated yield of two experiments. d Recovered only starting material and small amount of dimer of diazo. e Only Sonogashira-coupling product indicated. f Mostly alkyne homo-coupling product and small amount of cyclopropenation product with terminal alkyne was observed. g Conducted in inert atmosphere. | |||||
1 | None | Stainless steel vial | 1 | 16 | 0d |
2 | PdCl2(PPh3)2 | Stainless steel vial | 1 | 16 | 0e |
3 | Pd(PPh3)4g | Stainless steel vial | 1 | 16 | 0e |
4 | PdCl2(PPh3)2 | Nickel vial | 1 | 16 | 0d |
5 | PdCl2(PPh3)2 | Ag foil (S.S.) | 1 | 16 | 40 |
6 | Pd(OAc)2 | Ag foil (S.S.) | 1 | 16 | Trace |
7 | Pd(PPh3)4g | Ag foil (S.S.) | 1 | 16 | 38 |
8 | Pd(dba)2 | Ag foil (S.S.) | 1 | 16 | 10 |
9 | PdCl2(PPh3)2 | Cu foil (S.S.) | 1 | 16 | 0f |
10 | Pd(OAc)2 | Ag foil (S.S.) | 2 | 6 + 16 | Trace |
11 | Pd(PPh3)4g | Ag foil (S.S.) | 2 | 6 + 16 | 75 |
12 | Pd(dba)2 | Ag foil (S.S.) | 2 | 6 + 16 | 60 |
13 | PdCl2(PPh3)2 | Ag foil (S.S.) | 2 | 6 + 16 | 76 |
14 | PdCl2(PPh3)2 | Ag foil (S.S.) | 2 | 6 + 12 | 75 |
15 | PdCl2(PPh3)2 | Ag foil (S.S.) | 2 | 6 + 10 | 72 |
16 | PdCl 2 (PPh 3 ) 2 | Ag foil (S.S.) | 2 | 6 + 8 | 72 |
17 | PdCl2(PPh3)2 | Cu foil (S.S.) | 2 | 16 + 16 | 0f |
Since the predictive models of mechanochemical catalysis are still in their infancy, we used surveyed several reaction conditions and experimental procedures to determine which were optimal. First, we attempted a one-pot domino reaction where methyl phenyl diazoacetate, phenylacetylene, iodobenzene, 5% palladium tetrakis triphenylphosphine and potassium carbonate were all added to a stainless steel reactor and ball milled for 16 h (approach 1). In the absence of silver foil, no product was observed irrespective of the reaction conditions (entries 1–4, 9, and 17). However, upon the addition of silver foil, the cross coupled/cyclopropenation product was isolated albeit in modest yield (entries 5, 7 and 8). Catalytic amounts of palladium(II) acetate were found to be completely unreactive towards Sonogashira coupling by mechanochemical ball milling (entries 6 and 10). Copper foil was observed to be effective for the Sonogashira coupling reaction (approach 1, entry 9) but failed to yield any appreciable amount of the cyclopropene product. In an effort to increase the efficiency of the desired process, we elected to add the diazoacetate compound after ball milling the Sonogashira reactants for 6 hours without purification. This was then followed by additional ball milling of the solvent-free reaction mixture (approach 2). Under traditional solution based conditions both the Sonogashira and cyclopropenation reactions are typically conducted under strictly inert atmospheric reaction conditions. To test this, we conducted the reaction using Pd(PPh3)4 under inert conditions using a dry box (entries 7 and 11) and compared the reaction using PdCl2(PPh3)2 under aerobic conditions (entries 5 and 13). In our previous report, we demonstrated both catalysts gave rise to similar yields of Sonogashira products.129 Since both reactions under approach 1 and approach 2 gave similar reaction yields, we concluded there is no preference for inert atmospheric mechanochemical ball milling reaction conditions. After optimization of the milling time and palladium salt, the two-step reaction resulted in good yields (60–75%) of the desired fully substituted cyclopropene product (entries 10–16). To the best of our knowledge, this represents the first example of a catalytic one-pot multi-component coupling/cycloaddition domino methodology for the organometallic synthesis of highly functionalized organic compounds. Lastly, copper foil was an effective co-catalyst for only the Sonogashira coupling reaction and no cyclopropenation of the internal alkyne was observed (entry 17).
To survey the general nature for the ‘one-pot’ Pd/Ag foil-catalyzed cross coupling/cyclopropenation mechanochemical milling reaction, we studied a variety of alkynes, aryl halides and aryl diazoacetates (Table 8).
Aliphatic and aryl alkynes coupled well with aryl iodides or bromides to afford internal alkynes, which upon treatment with aryl diazoacetate, effectively produced the cyclopropene products 20, 24, 65–69 in good overall yields (55–80%). In order to test the reactivity of electron rich and electron poor aryl diazoacetates for cyclopropenation, we subjected the coupling products of 1-hexyne and various aryl halides to yield cyclopropene products (70–73) in good overall yields. Compounds 70 and 73 did not produce any Sonogashira type reactivity at the aryl bromide site, presumably due the significant effect substituents have on the rate of Sonogashira coupling of aryl bromides compared to iodides under mechanochemical conditions.129 Overall, this multi-component strategy is highly general and atom economical (e.g., atom economy ranges 68–73%) where no solvent or inert atmospheric conditions are required.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8sc00443a |
This journal is © The Royal Society of Chemistry 2018 |