Stereo- and Regiospecific S_N2′ Reaction of MBH Adducts with Isocyanoacetates: en Route to Transition-Metal-Free α-Allylation of Isocyanoacetates

Abstract

Herein, we report that under mild and transition-metal-free conditions an unprecedented and practical S_N2′ reaction of Morita–Baylis–Hillman adducts with isocyanoacetates takes place in a stereo- and regiospecific manner. This reaction which tolerates a wide variety of functionalities delivers transformable α-allylated isocyanoacetates in high efficiencies. Preliminary studies on the asymmetric version of this reaction indicate that ZnEt₂/chiral amino alcohol combinations are an asymmetric catalytic system for this transformation, giving an enantioenriched α-allylated isocyanoacetate with a chiral quaternary carbon in a high yield.

Introduction

Isocyanoacetates have proven to be synthetically very useful building blocks for the assembly of a large array of functional organic molecules.¹ They have been widely used in the synthesis of many biologically active natural products,² pharmaceuticals,³ organometallics,⁴ and so on. Thereafter, there has been long-standing interests in the innovation of new reactivity and accordingly developing novel transformations involving them.

Being a type of activated methylene compounds, isocyanoacetates have been widely employed as α-nucleophiles to attack electrophiles leading to a variety of nucleophilic addition and cycloaddition. Typical electrophilic acceptors include alkyl (pseudo)halides,^2a electron-deficient alkenes^2c,5 and heterocycles,⁶ alkynes,⁷ cumulated double bonds,⁸ imines,⁹ carbonyls,^2d,10 aziridines,¹¹ nitrones,¹² 1,3-dipoles,¹³ aryl diazonium salts¹⁴ and so forth. Morita–Baylis–Hillman (MBH) adducts, which have proven to be a type of versatile electrophiles,¹⁵ however, have never been reported to react with isocyanoacetates to the best of our knowledge. Thereafter, the exploration on the interaction between isocyanoacetates and MBH adducts is of much significance and deserves investigation.

As a type of structures containing multiple transformable functionalities, α-allylic isocyanoacetates are assumed to be synthetically useful manifolds which can be used to construct diverse products. Nevertheless, existing methods for the synthesis of α-allylic isocyanoacetates have proven to be very scarce. Known methods that utilize allylic alcohol esters as allylation components under palladium catalysis conditions¹⁶ suffer from expensive and contaminative palladium metal, uncontrollable regioselectivity, and/or limited reaction scope (Scheme 1a). Hence, the development of novel, regiospecific, and especially transition-metal free allylation of isocyanoacetates is of much significance and worth investigation. In this work, we report that under transition-metal-free conditions isocyanoacetates can be regio- and stereospecifically allylated by employing MBH adducts as allylating components, giving α-allylic isocyanoacetates in high efficiency (Scheme 1b).

Scheme 1. α-Allylation of Isocyanoacetates.

Results and Discussion

We commenced our investigation by selecting readily available MBH adduct 1a(17) and ethyl isocyanoacetate 2a as starting materials. Under an inert atmosphere, this reaction in the presence of Cs₂CO₃ in DCM (dichloromethane) at room temperature for 24 h produced α-allylic isocyanoacetate 3aa in a 73% yield (Table 1, entry 1). In the case that Cs₂CO₃ was replaced by K₂CO₃ or Na₂CO₃, the reaction did not take place, implying the basicity and/or the solubility of the bases might play critical roles in the reaction (Table 1, entries 2–3). While K₃PO₃ gave the α-allylic isocyanoacetate 3aa in a 37% yield, organic bases such as Et₃N and DBU were fruitless with the starting materials being recovered in nearly quantitative yields (Table 1, entries 4–6). A control experiment revealed that the reaction did not take place in the absence of any base, showing the critical role of the bases in the reaction (Table 1, entry 7). The results that a catalytic amount of Cs₂CO₃ (20 mol %, Table 1, entry 8) just furnished a 5% yield revealed that a stoichiometric amount of the base was necessary for the reaction.¹⁸ Further optimization on the loading of the base revealed that 2 equiv amount was the best, with the highest yield of 3aa being obtained (Table 1, entries 9–10). The screening on the reaction solvents demonstrated that DCE (dichloroethane) was the best choice, with the highest yield (76%) being afforded (Table 1, entries 11–15). Because neither a lower (0 °C) nor a higher temperature (45 °C) provided a better result, room temperature was proven to be the most suitable temperature for the reaction (Table 1, entries 16–17). As both lengthening and shortening the reaction time led to lower yields, 24 h were proven to be the most proper reaction time (Table 1, entries 18–19). Finally, it was revealed that raising the loading amount of 2a was beneficial for the reaction (Table 1, entries 20–22). When 2a was used in three equivalence of 1a, the highest reaction yield (87%) was obtained (Table 1, entry 21). Moreover, we also tested some strong inorganic bases, such as NaOH and t-BuOK, as the base by employing MBH adduct 1a′ as an allylation component (Table 1, entries 23–25), finding that NaOH was completely incompetent with no desired product 3aa was obtained (Table 1, entry 23). On the other hand, t-BuOK was proven to be a competent base for the reaction with the desired product 3aa was produced in a 70% yield (Table 1, entry 24). However, it once again proved that the base must be used in a stoichiometric amount. When the used amount of t-BuOK was reduced to a catalytic amount (20%), only a 14% yield of product 3aa was furnished (Table 1, entry 25).

Table 1. Optimization Work^a.

entry	base	[S]	T (°C)	t h	yieldb (%)
1	Cs2CO3	DCM	RT	24	73
2	K2CO3	DCM	RT	24	0
3	Na2CO3	DCM	RT	24	0
4	K3PO4	DCM	RT	24	37
5	Et3N	DCM	RT	24	0
6	DBU	DCM	45	24	0
7	no	DCM	RT	24	0
8c	Cs2CO3	DCM	RT	24	5
9d	Cs2CO3	DCM	RT	24	58
10e	Cs2CO3	DCM	RT	24	66
11	Cs2CO3	MeCN	RT	24	15
12	Cs2CO3	toluene	RT	24	71
13	Cs2CO3	THF	RT	24	69
14	Cs2CO3	DMF	RT	24	75
15	Cs2CO3	DCE	RT	24	76
16	Cs2CO3	DCE	0	24	25
17	Cs2CO3	DCE	45	24	73
18	Cs2CO3	DCE	RT	12	68
19	Cs2CO3	DCE	RT	48	59
20f	Cs2CO3	DCE	RT	24	80
21g	Cs2CO3	DCE	RT	24	87
22h	Cs2CO3	DCE	RT	24	83
23g,i	NaOH	DCE	RT	24	0
24g,i	t-BuOK	DCE	RT	24	70
25c,g,i	t-BuOK	DCE	RT	24	14

^aReaction conditions: 1a (0.2 mmol), 2a (0.24 mmol), base (0.4 mmol), [S] (1 mL).

^bIsolated yields.

^cBase (0.04 mmol).

^dBase (0.2 mmol).

^eBase (0.6 mmol).

^f2a (0.4 mmol).

^g2a (0.6 mmol).

^h2a (0.8 mmol).

ⁱ1a′ was used instead of 1a.

With optimal conditions in hand, we next investigated the scope of this reaction, and the results are summarized in Table 2. It was found that besides tert-butyl 2-(acetoxy(phenyl)methyl)acrylate 1a, tert-butyl 2-(((tert-butoxycarbonyl)oxy) (phenyl)methyl)acrylate 1a′ was also a reliable substrate in the reaction, giving product 3aa in a 72% yield in the presence of two equivalence of Cs₂CO₃ (Table 2, entry 2). Methyl 2-(acetoxy(phenyl)methyl)acrylate 1b could likewise smoothly carry out the reaction, leading to allylic isocyanoacetate 3ba in a 88% yield (Table 2, entry 3). In addition to ethyl isocyanoacetate 2a, tert-butyl isocyanoacetate 2b was also able to proceed with this reaction, affording desired product 3bb in a high yield (91%, Table 2, entry 4). α-Substituted isocyanoacetates such as ethyl 2-isocyanopropanoate 2c, ethyl 2-isocyano-3-phenylpropanoate 2d, and ethyl 2-isocyano-2-phenylacetate 2e were all proven to be competent reactants in the reaction, producing α-allylic isocyanoacetates with quaternary carbon centers in high yields.¹⁹ For these α-substituted isocyanoacetates 2c–2e, the desired α-quaternary isocyanoacetates 3ac, 3cd, and 3ae were still afforded when one equivalence of Cs₂CO₃ was employed, albeit in slightly lower yields (56–86%) (Table 2, entries 5–7). Additionally, it was found that 1-((isocyanomethyl)sulfonyl)-4-methylbenzene (TosMIC) 2f was also a good reaction partner with tert-butyl 2-(acetoxy(phenyl)methyl)acrylate 1a, leading to allylation product 3af in a 84% yield (Table 2, entry 8). With respect to MBH adducts 1 derived from aryl aldehydes, various functionalities on phenyl rings could be tolerated in the reaction. For example, MBH adducts bearing an electron-withdrawing group such as a chlorine atom at their para and meta positions both proved to be reliable substrates in the reaction, furnishing allylic isocyanoacetates 3db and 3ea in 63 and 76% yields (Table 2, entries 9–10). MBH adducts having an electron-donating substituent like a methoxy group at their para positions (1c) also reacted smoothly with ethyl isocyanoacetate 2a, providing allylic isocyanoacetate 3ea in a 80% yield (Table 2, entry 11). Furthermore, it was found that MBH adducts derived from an aliphatic aldehyde, such as tert-butyl 3-acetoxy-2-methylenehexanoate 1f, could be engaged in the reaction, producing allylic isocyanoacetate 3fa in a 54% yield (Table 2, entry 12). MBH adducts derived from heteroaryl aldehydes such as tert-butyl 2-(acetoxy(furan-3-yl)methyl)acrylate 1g was also a competent substrate in the reaction, giving allylic isocyanoacetate 3ga in a 75% yield (Table 2, entry 13). Moreover, MBH adducts derived from formaldehyde 1h could likewise react smoothly with ethyl isocyanoacetate 2a, furnishing allylic isocyanoacetate 3ha with a terminal alkene in a 51% yield (Table 2, entry 14). MBH adducts bearing a substituent at their alkene termini such as ethyl (E)-2-(acetoxy(phenyl)methyl)-3-phenylacrylate 1i were also good reaction partners with isocyanoacetate 2a, giving α-allylic isocyanoacetate 3ia in a 50% yield (Table 2, entry 15).

Table 2. Scope of the Reaction^a.

^aReaction conditions: 1 (0.2 mmol), 2 (0.6 mmol), Cs₂CO₃ (0.4 mmol), DCE (1 mL), RT, 24 h.

^bIsolated yields.

^c2 (0.24 mmol).

^dCs₂CO₃ (0.2 mmol).

^e1a′ was used instead of 1a and DBU was used as the base.

Based on the above results and some literature precedents,²⁰ a possible reaction pathway is depicted in Scheme 2. Under basic conditions, isocyanoacetates 2 may be converted into nucleophilic anions I in which α-carbon anions are stabilized by two election-withdrawing groups²¹ (Scheme 2, NC and EWG). Then, an S_N2′ process between I and electrophilic MBH adducts 1 may regiospecifically take place at the more electrophilic β-postion of the MBH adduct, giving α-allylic isocyanoacetates 3, and concurrently liberating the conterions PGO^–.

Scheme 2. Proposed Mechanism of the Reaction.

To verify the synthetic value of this reaction, we performed a variety of post-transformations of product 3aa (Scheme 3). First, the isocyano group in 3aa could be selectively hydrolyzed by dilute HCl at room temperature, giving unnatural aminoester 4aa in a 64% yield. Second, regarding the fact that 3aa itself also belonged to a substituted isocyanoacetate structure, we deduced that it could be engaged into the allylation reaction again under identical conditions. To validate this, we conducted the reaction between 3aa and MBH adduct 1b and found that this reaction smoothly took place, furnishing twofold allylation product 6 in a 73% yield. It should be noted that in 6, the two allyl substituents were different, demonstrating that the α-position of isocyanoacetates could be orthogonally allylated by two different MBH adducts.

Scheme 3. Post-Transformations.

At last, we also tried to develop an asymmetric version of this reaction. After a series of screening and optimization (see the Supporting Information for details), we found that under the promotion of ZnEt₂ and a catalytical amount of chiral amino alcohol enantioenriched α-allylated isocyanoacetates 3ed could be obtained in an 84% yield and 56% ee (Scheme 4). Although the enantioselectivity of this reaction remained at a moderate level and needed to be further improved, this result was of much significance since it supplied a good foundation for the assembly of chiral α-quaternary amino acids, which had proven to be versatile building blocks in a large array of fields,²² from readily available substrates.

Scheme 4. Asymmetric Version of the Transformation.

Conclusions

In summary, we showed that under mild and transition-metal-free conditions S_N2′ reactions between readily available MBH adducts and isocyanoacetates could efficiently take place, giving transformable α-allylated isocyanoacetates in high selectivity. Studies on the reaction scope showed that a variety of functional groups could be tolerated in the reaction. Post-transformations demonstrated that as-synthesized α-allylic isocyanoacetate products could be further diversely derived, giving an α-allylic aminoester and α,α-diallylated isocyanoacetate in high efficiencies. Preliminary studies on the asymmetric version of this reaction showed that chiral amino alcohol/ZnEt₂ combinations was a potential asymmetric catalytic system for the transformation. Further studies on the improvement of the enantioselectivity of the asymmetric version of this reaction, and the application of this reaction are currently ongoing in our lab.

Experimental Section

Representative Procedure of the Racemic Reaction

To an oven-dried flask, Cs₂CO₃ (0.4 mmol), MBH adduct 1a (0.2 mmol), and isocyanoacetate 2a (0.6 mmol) were added. The mixture was repeatedly degassed and refilled with N₂ three times. Then, dry DCE (1 mL) was injected by a syringe and the reaction mixture was allowed to stir at room temperature for 24 h. After the completion of the reaction, the mixture was neutralized with 1 N HCl and extracted with DCM three times. After being dried with Na₂SO₄, combined organic layers were filtered through a pad of celite. The filtrate was then concentrated until the solvent was completely removed. The residue was then separated on a silica gel column, and product 3aa was obtained as pale yellow oil (57 mg, 87%).

Representative Procedure of the Asymmetric Version of This Reaction

Into an oven-dried flask, MBH adduct 1e′ (0.6 mmol), isocyanoacetate 2d (0.2 mmol), and chiral amino alcohol L₁ (0.03 mmol) were charged. The mixture was repeatedly degassed and refilled with N₂ three times and then cooled to 0 °C and dry DCE (1 mL) and Et₂Zn (0.46 mmol, 1 M in n-hexane) were subsequently injected by a syringe. After being stirred at room temperature for 12 h, the system was neutralized with 1 N HCl and extracted with DCM three times. After being dried with Na₂SO₄, the combined organic layers were filtered through a pad of celite. The filtrate was then concentrated until the solvents were completely removed. The residue was then separated on a silica gel column, and product 3ed was obtained as pale yellow oil (75 mg, 84%).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c07581.

• Experimental details; spectral data of ¹H, ¹³C nuclear magnetic resonance, and high-resolution mass spectrometry of all products; chiral high-performance liquid chromatography for product 3ed; and optimization details on an asymmetric version of the reaction (PDF)

The authors declare no competing financial interest.