*2.2. Substrate Scope Studies*

With the optimal reaction conditions, we first tested the 1,3-dipolar addition reaction for benzoyl nitromethane and allylbenzene and had a yield of product of 90%. Then, to examine the generality and scopes of this methodology, we took a variety of benzoylnitromethane derivatives **1** (Scheme 2) and allylbenzene **2a** as substrates and representative results were shown in Scheme 2. These results showed that a variety of electronically varied aromatic α-nitroketones were well compatible with the cycloaddition in all the reactions, and reaction generally obtains in moderate to good yields for the synthesis of isoxazoline derivatives. Moreover, in this reaction, we found a good regioselectivity, which was consistent with the work of Ken-ichi Itoh [45].

**Scheme 2.** Scopes of the substituted phenylnitroketones.

At first, it was found that the R1 substituents would affect the cycloaddition efficiency in these reactions. The electron-rich α-nitroketones (**1e**, **1g**, Scheme 2) provided products (**3e**, **3g**, Scheme 2) in slightly better yields in comparison to the electron-deficient ones (**1d**, **1f**, **1i**, Scheme 2). Different electron-withdrawing substituents at the same position of phenyl-α-nitroketone resulted in similar yields (**3d**, **3i**, Scheme 2). Additionally, surprisingly, isoxazoline derivative (**3f**, Scheme 2) were obtained in moderate yields when electron-donating group(**-OMe**) were used with a yield of 66%, which was close to the yield of isoxazoline obtained by electron-deficient α-nitroketone. In the case of electron-deficient α-nitroketone, the corresponding were obtained in good yields (**3b**–**3d, Scheme 2**), respectively 73%, 70%, and 67%. The results show that the position of the substituent has no effect on the reaction results. Additionally, aromatic substrate, such as benzene, behaved similar to an electron-withdrawing substituents and gave a yield of 68% for product **3h**. However, the reaction rate was slower than that of aliphatic substituted substrates.

Next, we also investigated the scope of the alkenes (Table 2). The reaction of **1a** with alkenes derivatives **4** was carried out under the optimum reaction conditions whose results are shown in Table 2. All the reactions gave **5a**–**5f** as product, respectively, in good to excellent yield, except **5f**. The type of reaction substrate alkenes was modified, and it was found that the reaction proceeded well with both aliphatic alkenes and aromatic alkenes affording isoxazolines in good yields from the same *α*-nitroketone (entries 1–5, Table 2). In addition, cycloaddition of cyclohexene (**4f**) with benzoylnitromethane (**1a**) could also be achieved in fairly good yields, the corresponding isoxazoline **5f** was obtained in 69% yield (entry 6, **5f, Table 2**).

In addition, in order to expand the applicability of the reaction, we further examined the types of reaction materials. Isoxazolines were synthesized using the dipolarophiles **2a** (allylbenzene) and alkyl nitroketones **6** (Scheme 3). The results showed that in the presence of *p*-TsOH, alkyl nitroketones were also able to react with dipolar reagents to obtain isoxazoline derivatives; unfortunately, compared with **3** and **5** phenylisoxazoline, **7a** and **7b** yields were lower, 23% and 20%, respectively.


**Table 2.** Scopes of alkenes.

<sup>a</sup> General conditions: **1a** (0.125 mmol), **4** (0.625 mmol), *p*-TsOH (0.5 mmol), ACN (0.2 mL) at 80 ◦C for 22 h, unless stated otherwise. <sup>b</sup> Isolated yield.

**Scheme 3.** Variation of alkyl nitroketones.

Finally, the alkyen **8** was used for the reaction with α-nitroketone (**1a**) under the optimized reaction conditions, which obtained in excellent yields Isoxazoles. Under the same conditions, the reaction rate with alkyen was quicker than with alkenes. Nevertheless, **9a** and **9b** were obtained in 85% and 88% yield, respectively (Scheme 4).

**Scheme 4.** Variation of alkyne.
