Asymmetric halocyclization is an important class of transformation to construct valuable halogenated compounds which are widely present in natural products and biological molecules [
1-
5]. The halogen handles in these products were readily manipulated using well-developed methods to give useful and attractive building blocks [
6-
8]. Moreover, the insertion of chlorine into the particular site of bioactive molecule could provide favorable pharmacokinetic and pharmacological properties [
9,
10]. Since more than 200 drugs including chlorine were approved by FDA, the use of chlorine to the optimization of lead compounds has received great interest from pharmaceutical chemists [
3]. Although great achievements have been made in the field of enantioselective chlorination with various organocatalytic strategies, low temperature and high catalyst loading are often required for excellent enantio– and diastereoselectivity [
11-
22].
Phase-transfer catalysis (PTC) is considered as one powerful and versatile methodology for organic synthesis because of the character of green chemistry and unique advantage over homogeneous alternatives [
23]. The development of new ligands and catalyst variants for efficient PTC has received continuous attention from synthetic chemists [
24,
25]. Among the existing methods, quaternary ammonium salts with chiral cation as PTC catalysts are prevalent in chiral ion-pair catalysis. Based on this strategy, considerable effort has been devoted to develop several synthetic methods for asymmetric halogenation (
Scheme 1A) [
26-
29]. However, this chiral cationic PTC system suffers from low enantioselectivity, harsh condition and limited reaction types. Recently, the asymmetric fluorination process represents a significant milestone in PTC by employing chiral phosphate anion to form an ion pair with the ammonium-based electrophilic fluorinating cation [
30]. In this PTC system, nonpolar solvents and Selectfluor were required to exclude the undesired background reaction. Later, Toste
et al. extended this strategy to other electrophiles for asymmetric transformation [
31,
32]. Unstable and insoluble bromo–, iodo analogues of Selectfluor were developed as electrophilic reagents for enantioselective halogenations under the chiral anion phase-transfer conditions (
Scheme 1B) [
33-
40]. Very recently, Sun
et al. demonstrated electricity-driven catalytic asymmetric bromocyclization with NaBr as bromine source with high catalyst loading by chiral phosphate anion PTC (
Scheme 1B) [
41]. As for the fledgling nature of this area, relatively large catalyst loading is required to achieve meaningful levels of enantioselectivity. Furthermore, efficient asymmetric conversions with chlorine and other electrophile have remained underdeveloped. The highly reactive nature of chloronium ions makes the development of enantioselective chlorination a formidable challenge by the strategy of anionic PTC [
42-
44].
Inspired by the previous work of highly
ortho-chlorination of aniline and phenol, asymmetric bromination and iodination with secondary ammonium salt as catalyst [
45-
47], we rationalized that more bulky tertiary ammonium salt could exchange with the chlorine and selenylation source to generate stable ammonium ion-pair, employed as key stereodetermining intermediate for asymmetric induction [
48]. This strategy would allow us to develop electrophilic halogen-initiated asymmetric cyclization to furnish chiral heterocycles. Herein, we are pleased to describe a mild and highly practical protocol for intramolecular asymmetric chlorocyclization of 2-alkenyl anilides
via anionic PTC to provide expedient access to chiral chlorinated 4
H-3,1-benzoxazine framework (
Scheme 1C). However, in sharp contrast to the general success achieved for the bromocyclization, enantioselective chlorocyclization of 2-alkenyl anilides remains challenging [
49]. A breakthrough was made by the Deng group, who developed asymmetric chlorocyclization utilizing cinchonidine-derived chiral esters as catalyst. However, high catalyst loading and harsh conditions were required, and substrate scope was restricted to 1,1-disubstituted alkenes in this protocol, which somehow limit its practical application [
50]. The commonly motifs of 4
H-3,1-benzoxazine are the key cores of bioactive natural products, agrochemicals, biological active pharmaceuticals, functional materials, chiral ligands and synthetic building blocks [
51-
54]. Hence, numerous endeavors have been made over years toward asymmetric synthesis of optical 4
H-3,1-benzoxazine [
55-
58].
The chiral ammonium salts
1 are air and moisture-stable, nontoxic, easy to handle, which easily be prepared by the reaction between amines and chiral phosphoric acid (CPA) under mild conditions. At the initial stage of investigation, the ammonium salt
1 catalyzed chlorocyclization of aniline
2a was examined at room temperature using
n-hexane and 1,3-dichloro-5,5-dimethylhydantoin (DCDMH) as the solvent and chlorinating reagent, respectively (
Table 1). The reaction was sluggish in the absence of catalyst owing to the insolubility of DCDMH in nonpolar solvent (entry 1). The chlorination worked well using tertiary ammonium salt
1a as catalyst to provide desired product
3a in 96% yield and 95%
ee (entry 2). Then different ammonium salts were tested for this chlorocyclization. A survey of different chiral counter anions revealed that H
8-TRIP phosphate was the catalyst of choice. The more bulky ammonium salt
1d gave the best result (entries 2–5). However, poor enantioselectivity was obtained in the presence of secondary ammonium salt
1e (entry 6). Subsequent screening of more solvents failed to give better results (entries 7–9). Cinchonine-derived quaternary ammonium salt, CPA and its derivatives which previously used as typical organocatalysts for asymmetric halogenation were also evaluated for the chlorination of
2a [
59-
63]. It is interesting to realize that the catalytic power of tertiary ammonium salt
1 was much higher than that of some
N-benzyl-cinchoninium bromide
C1-
C2, and quaternary ammonium salt
C3 (entries 10–12). H
8-TRIP, H
8-TRIP-Na, co-catalyst mixture (1:1) of H
8-TRIP with Lewis base
LB1 and
LB2 were examined in this reaction, poor conversions or enantioselectivities were observed (entries 13–16). Moderate reaction yield of
3a with good enantioselectivity was obtained when a catalyst blend (1:1) of H
8-TRIP and
LB3 was used (entry 17). The efficiency with tertiary ammonium phosphate
1d was found to be considerably higher than quaternary ammonium salt, phosphoric acid and its derivatives. These results highlighted the excellent catalytic performance and unique usefulness of the newly developed ammonium phosphate salt catalyst
1.
With the optimized reaction conditions being established, we turned to evaluate the substrate scope of the tertiary ammonium salt catalyzed chlorocyclization, and the results are shown in
Scheme 2. A wide range of
o-vinylanilides were compatible in this reaction, giving the chlorocyclized products in high yields and excellent enantioselectivities. The compatibility of various substituents at the alkene moiety was then examined. The aliphatic substrates with cyclic and uncyclic substituents deliver the chlorinated products
3a-
3c in excellent yields and enantioselectivity. Then the substrates with various substituents on the phenyl ring at the alkene moiety were explored. Both electron-donating groups and electron-withdrawing groups were tolerant in the reaction conditions, affording the corresponding cyclized compounds
3d-
3i in up to 99%
ee. It is noteworthy that the steric hindrance plays little influence on the chlorocyclization, excellent enantioselectivity was observed when the
o-methyl group was substituted (
3e). The 1-naphthyl-substituted substrate
2j gave rise to the chlorocyclized compound
3j in 93% yield and 99%
ee, and the absolute configuration was confirmed by X-ray analysis. The substrates of different substituted aniline derivatives were further explored. Electron-donating groups such as methyl and phenyl and electron-withdrawing groups such as fluoride, chloride and bromide were well tolerated, providing the desired products
3k–
3t in excellent results. The substrate
2u with other aromatic core was successfully evaluated for the chlorination to furnish
3u in excellent outcomes. Next, various amides were also explored for the scope compatibility of this chlorocyclization. The position and electron feature of the substituent to the aromatic ring had negligible impact on the enantioselectivity, the corresponding cyclized products
3v-
3af were also obtained smoothly in up to 99%
ee. Notably, the aliphatic amide can be applied in this reaction, providing the corresponding product
3ag in excellent yield and high enantioselectivity. Switching the benzene ring to other aromatic cores such as thiophene, pyridine and naphthalene gave the desired products
3ah-
3aj in excellent yields and enantioselectivities. Furthermore, the internal and cyclic olefins were well amenable in the optimized condition, leading to the desired products
3ak-
3am in high enantioselectivity and diastereoselectivity.
The late-stage functionalization of complex molecules illustrated the utility of this catalytic system (
Scheme 3). The amides containing the motif of drug molecules derivatives successfully engage in this catalytic asymmetric chlorination to afford adapalene
3an and probenecid
3ao in generally good yields with excellent diastereomeric excess. Natural products derivatives containing menthol, borneol, geraniol, epiandrosterone and cholesterol worked well in the reactions, yielding the desired products
3ap-
3au with excellent results. The modification of these important compounds, many of which contain sensitive alkene, ketal and carbonyl groups, suggest the good functional group tolerance of this method.
The recycling stability of this type of PTC catalyst was also investigated. This developed PTC protocol could be easily scaled up and the catalyst loading could be reduced to 1 mol% although a longer reaction time was applied. For instance, 4 mmol of aniline
2h was subjected to the reaction and 1.39 g (99% yield and 99%
ee) of the desired chlorinated product
3h was obtained (
Scheme 4A). After the completion of this chlorination, the tertiary ammonium salt catalyst
1d could be recovered by column chromatography on silica gel and the comparable results in terms of reaction conversion and enantioselectivity were obtained during the halogenation/catalyst recycling process for 3 times. These results indicating that the catalyst worked with negligible loss in both the activity and selectivity. Since the chlorine carrier hydantoin is insoluble in hexane, we attempted to isolate the product under column-free condition. Upon completion, the chlorocyclization was filtered and the insoluble hydantoin was washed with hexane to furnish the chlorinated compound
3h in 99% yield with purities of up to 99% (determined by
1H NMR) and 98%
ee (
Scheme 4B). The practical and environmental benign features of this catalytic system are attractive for the modern manufacture sectors.
The corresponding chlorocyclized compounds are very useful building blocks for the synthesis of functional molecules (
Scheme 4C). Dechlorination of
3b was achieved
via nBu
3SnH reduction, affording product
4 in high yield. The compound
3d underwent hydrogenation to give chiral compound
5 in 99% yield, followed by remove of the benzoyl group to provide diarylamine
6 in high enantioselectivity, which is useful in drug discovery and chiral catalyst synthesis [
64,
65]. Simple reduction of
3d by LiAlH
4 readily furnished secondary alcohol
7 in 72% yield. Subsequent deprotection of the benzyl group of
7 with hydrogen to form chiral amino alcohol
8, which is used as an important intermediate for the synthesis of chiral ligand, base and benzoxazine nucleus [
66]. Treatment of benzoxazine
3h with NaBH
3CN in acetic acid yield amino alcohol intermediate, which cyclized with triphosgene in the base condition, providing 1,4-dihydro-2
H-3,1-benzoxazine-2-one derivatives
9 in 69% yield without any loss of stereochemistry.
3h was subjected to the site-selective bromination and Suzuki coupling, affording the functionalized benzoxazine
10 in 85% yield and 98%
ee.
Further evaluation of chiral tertiary ammonium salt showed that catalyst
1a could catalyze the selenocyclization of
o-vinyl anilides
2 with
N-(phenylseleno)phthalimide (NPSP) as the electrophile. As shown in
Scheme 5, the substrates
2 with diverse functionalities were proved compatible under this catalytic protocol and could react smoothly with the selenium reagent NPSP to furnish the corresponding selenylated compounds
11 in moderate yields and good enantioselectivities.
To get a better understanding of the catalytic system, we attempted to probe the active species (
Scheme 6). Upon mixing
1a and DCDMH in CDCl
3, a new proton signal at 3.40 ppm appeared which might be due to species
1a-Cl, further evidenced using ESI high-resolution mass spectrometry. This active species
1a-Cl was vital for the asymmetric induction which the phosphate anion interacts with tertiary ammonium cation tightly
via electrostatic and multiple electropositive hydrogens interact with the phosphate oxygens.
To gain insights in the reaction mechanism and the origins of enantioselectivity, we have performed density functional theory studies. The ammonium salt
1d, aniline
2a, and DCDMH were used as the catalyst, substrate, and chlorinating reagent, respectively, in our calculations. As shown in
Fig. 1, the complexation of
1d and DCDMH would lead to
IN1 and
IN2, in which the chlorine has bound to nitrogen in the latter. This step is slightly endergonic by 4.1 kcal/mol and there are multiple C-H
…O hydrogen bonding interactions between the phosphate anion and the ammonium cation. Following that, the binding of substrate
2a would form isomeric species
IN3 and
IN3’, which are exergonic by 1.4 and 5.5 kcal/mol, respectively. Starting from
IN3, the asymmetric chlorocyclization occurs
via transition state
TS4 to generate
IN5. This process has an activation free energy barrier of 15.0 kcal/mol and is exergonic by 31.7 kcal/mol (with respect to
IN3’). On the other hand, from
IN3’, the minor pathway takes place through
TS4’ to provide
IN5’. This pathway has an activation free energy barrier of 16.1 kcal/mol and is exergonic by 34.9 kcal/mol. By comparison, the pathway leading to the major enantiomer (
via TS4) is 1.1 kcal/mol more favorable than that leading to the minor enantiomer (
via TS4’), which is consistent with experiment (
i.e., 99%
ee). By detailed analyses on the selectivity-determining transition state, we found that there was favorable hydrogen bonding interaction between phosphate oxygen and the ammonium C-H bond (O
…H = 2.24 Å) in
TS4. In addition, there were much pronounced steric repulsions between phosphate anion and the aniline substrate part in
TS4’ (
i.e., the shortest H
…H distances are 2.23 and 2.26 Å). Therefore, the hydrogen bonding interaction and the steric effect account for the observed high enantioselectivity.
In our previous ion pair catalysis work [
47], the strong bidentate noncovalent interaction of the hydrogen bonding (N-H
…O) and ion pair interaction between phosphate anion and secondary ammonium cation was responsible for the asymmetric induction of bromination and iodation with high directionality, and the hydrogen bonding of C-H
…O (2.66 Å) was not observed. However, as shown in
TS4, the phosphate anion interacts with tertiary ammonium cation tightly
via strong attractive force between ammonium cation and phosphate ion, and multiple electropositive hydrogens interact with the phosphate oxygens (C-H
…O = 2.24, 2.51, 2.54 Å) in this work. Consequently, the resulting highly directional noncovalent interaction would result in effective asymmetric induction for the chlorocyclization.
In conclusion, we have developed the chiral anion PTC for asymmetric chlorination and selenylation of o-vinylanilides in high enantioselectivities under mild reaction conditions. This catalytic protocol is easily scale up and catalyst loading could be further reduced to 1 mol%. The robust tertiary ammonium salt could be easily recovered and reused for 3 times without notable loss of catalytic performance. The features of excellent reactivity, selectivity, stability and recyclability, and column-free purification of this chiral anion PTC system are significant for practical application in organic synthesis. In addition, computational mechanistic studies were used to elucidate the origin of the stereoselectivity. Further investigation on the application of this PTC system to other asymmetric reactions is underway.
PDF
Full
56
32
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
