Gold-Catalyzed Double Migration-Benzannulation Cascade toward Naphthalenes

Abstract

A novel gold(I)-catalyzed cycloisomerization of propargylic esters leading to unsymmetrically substituted naphthalenes has been developed. This cascade reaction involves an unprecedented tandem sequence of 1,3- and 1,2-migration of two different migrating groups. It is believed that this transformation likely proceeds via the formation of 1,3-diene intermediate or its precursor, which upon cyclization and aromatization steps transforms into the naphthalene core.

In recent years, transition-metal-catalyzed transformations of propargylic esters¹ have received much attention. Particularly intriguing is reactivity of these easily accessible compounds in the context of gold catalysis,² which has been reflected in the development of a variety of diverse and elegant transformations leading to an immense array of complex organic molecules. Remarkable propensity of propargylic esters 1 to undergo 1,3-acyl migration,^1,2 through the formation of an activated allene equivalent, intermediate i,¹ allowed for efficient and expeditious assembly of various acyclic unsaturated synthons³ and complex carbo-⁴ and heterocycles⁵ (eq 1). Herein, we wish to report a gold(I)-catalyzed double 1,3-/1,2-migration-benzannulation cascade of propargylic esters 1 into naphthalenes 2 (eq 2).

(1)

(2)

Considering the enhanced electrophilicity of the sp³ center^1b in intermediate i, we envisioned that incorporation of a suitable 1,2-migrating group (MG) in ii would provoke a subsequent 1,2-shift⁶ into iii, which upon proton loss and protiodeauration would afford 1,3-diene⁷ 3. It is reasonable to propose that the latter may undergo 6π-electrocyclization into naphthalene^8,9 2, analogously to the known cyclization of acyloxy-1,3,5-trienes¹⁰ (eq 3).

(3)

To this end, a possible isomerization of propargyl phosphate 1a (MG = H) in the presence of different catalysts has been tested (Table 1). It was found that employment of Ag triflate gave corresponding allene 4a in good yield (entry 1). Remarkably, switching to cationic Au(I) triflate led to the formation of target 1,3-diene 3a in 86% yield (entry 2). Monitoring of the reaction course revealed that this transformation proceeded through allenic intermediate 3a.¹² Employment of Au(III) complexes (entries 3 and 4), non-cationic Au(I) halides (entries 5 and 6), Cu(I) and Cu(II) triflates, as well as Brønsted or Lewis acids, resulted in no reaction.¹²

Table 1.

Optimization of Reaction Conditions

entry	catalyst	yield 4a, %a	yield 3a, %a
1	10% AgOTf	73	0
2	5% Ph3PAuCl, 5% AgOTf	0	0
3	5% AuCl3	0	0
4	5% AuCl3, 15% AgOTf	0	86
5	5% AuI	0	86
6	5% R3PAuCl (R = Et, Ph)	0	86

^aIsolated yield of product for reaction performed on 0.1–0.2 mmol scale.

It deserves mentioning that the isomerization of acetates 1 into 1,3-dienes 5 in the presence of Ag-catalysts was reported (eq 4).¹¹ However, the nature of the second step (1,2-H-shift or proton elimination) remained unclear. To address this issue, we performed mechanistic studies¹² for MG = H (eq 3) employing Au(I) catalyst. Experiments revealed that the reaction proceeds exclusively via a 1,3-shift^13–15 –elimination¹⁶ sequence.

(4)

Thus, we hypothesized that successful incorporation of a 1,2-migration into this cascade can only be achieved when a migrating group resides at a “proton-free” quaternary C-4 center. Therefore, isomerization of acetate 1d, possessing a strained cyclobutane ring, was examined. Indeed, a tandem 1,3-migration and ring expansion via a 1,2-shift occurred leading to 1,3-diene 3d in a moderate yield (eq 5). Moreover, isomerization of cyclopentyl homologs 1e and 1f afforded target naphthalenes 3e and 3f respectively (eq 5), thus providing a proof of concept for this cascade transformation.¹⁷

(5)

Next, cycloisomerization of substrates possessing better 1,2-migrating groups was examined (Table 2). Thus, a tandem acyloxy- or phosphatyloxy- and Ph-group migration/benzannulation of propargylic esters 1g–n proceeded smoothly to provide naphthalenes 2g–n in good to excellent yields. Terminal, alkyl-, and aryl-substituted acetylenes were nearly equally efficient in this transformation. Unexpectedly, cy-clization of dimethylphenyl-substituted acetate 1k proceeded via exclusive 1,2-Me-group migration to give 3k in good yield (entry 5). Notably, a variety of substituents, such as methoxy (entry 6), trifluoromethyl (entry 7), and 2-furyl (entry 8), were perfectly tolerated under these reaction conditions.

Table 2.

Gold(I)-Catalyzed Synthesis of Naphthalenes

entry	substrate	product	yield, %a,b
1		2g	86c
2		2h	94
3		2h	50
4		2j	75
5		2k	73
6		2l	90
7		2m	95
8		2n	94

^aIsolated yield.

^bReactions were performed on a 0.5 mmol scale.

^c5% Au-catalyst was used.

We propose the following plausible mechanisms for this novel cascade transformation (Scheme 1). Au-catalyzed 1,3-migration transforms 1 via cyclic intermediate^1b,3c,5b iv into allene 4.¹³ According to path A, 1,2-alkyl migration¹⁸ in iv produces benzylic cation v. The latter gives diene 3 upon protiodeauration or after proton transfer undergoes Friedel–Crafts alkylation to furnish naphthalene 2. Alternatively, a direct nucleophilic attack of vinyl-Au at the delocalized benzyl cation in v¹⁹ gives carbenoid intermediate vi, which upon 1,2-H-shift²⁰ and aromatization produces naphthalene 2 (Path B).²¹ According to path C, Au-catalyzed 6π-electrocyclization^14c,22 of 3 followed by elimination furnishes 2. In another scenario, direct intramolecular hydroarylation of 4 followed by 1,2-shift and proton loss in vii produces 2 (Path D). Unexpected exclusive Me- over Ph-migration in 1k is reasonably rationalized by stereoelectronic effect, according to which Ph group cannot accommodate requisite antiperiplanar orientation with the leaving group in iv.^12,23,24 Successful cycloisomerization of 3e, obtained via 1,3-migration/elimination cascade, into naphthalene 2e provided an additional support for possible intermediacy of 1,3-dienes in this transformation (eq 6).

Scheme 1.

Mechanistic Rationale for Au(I)-Catalyzed Cascade

(6)

In summary, we have developed a novel gold(I)-catalyzed approach toward polysubstituted naphthalenes, which features an unprecedented tandem sequence of 1,3- and 1,2-migration of different migrating groups in propargylic esters. A more detailed investigation on the mechanism, as well as the scope of this cascade, is ongoing and will be reported in due course.