Synlett 2021; 32(05): 521-524
DOI: 10.1055/s-0040-1707161
cluster
The Power of Transition Metals: An Unending Well-Spring of New Reactivity
© Georg Thieme Verlag Stuttgart · New York

Zinc-Catalyzed Transacetalization of N,O-Acetals into N,N-Acetals with Benzotriazoles, Indazoles, and Azides

Sang Ik Shin
,
Nguyen H. Nguyen
,
Jangbin Im
,
Department of Chemistry, Center for New Directions in Organic Synthesis (CNOS) and Research Institute for Natural Sciences, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 04763, Korea   Email: sshin@hanyang.ac.kr
› Author Affiliations

This work was supported by the National Research Foundation of Korea (Grant Numbers NRF-2012M3A7B4049653, NRF-2014–011165, and NRF-2017R1A2B4010888).
Further Information

Publication History

Received: 02 May 2020

Accepted after revision: 31 May 2020

Publication Date:
23 June 2020 (online)

 


This paper is dedicated to Professor Barry M. Trost to celebrate his career on the occasion of 20 years of Science of Synthesis.

Abstract

N,O-Acetals obtained from β-oxidation of ynamides underwent transacetalization with benzotriazoles, leading to N,N-acetals. The Zn(OTf)2 efficiently catalyzed the process, and the reaction is further accelerated in hexafluoroisopropanol, providing a single N1-regiosiomer. The transacetalization conditions developed could be extended to other N-donors, such as 1H-indazole and TMSN3 to afford the corresponding N,N-acetals.


#

Benzotriazoles are found as a core scaffold of a broad range of medicinal agents, such as anticancer, antifungal, antibacterial, antitubercular, and antiviral compounds.[1] They can be derivatized in a variety of methods, including addition to π-bonds[2] or metal carbene,[3] alkylation through nucleophilic substitution[4] or by radical mechanism,[5] and arylation.[6] However, a mixture of N1- and N2-regioisomers are typically produced, and few of the above methods are completely regioselective.[2a] [b] [3]

Oxidation of ynamides mediated by pyridine-N-oxides have gained intense interest because the resulting α-oxidation intermediates displayed an umpoled enolonium reactivity,[7] mediating addition of unmodified nucleophilic arenes, furnishing α-substituted carbonyl compounds I.[8] In contrast, we recently reported β-oxidation of ynamides promoted by mCPBA, furnishing β-keto-N,O-acetals II (Scheme [1], A).[9] We demonstrated that the subsequent transacetalization with secondary alcohols occurred in a highly enantioselective manner in the presence of a chiral phosphoric acid (Scheme [1], B), furnishing transacetalization product III.[9] In continuation of exploring the reactivity of N,O-acetals II, we report herein transacetalization with benzotriazoles to form benzotriazole derivatives IV, occurring in a highly regioselective manner (Scheme [1], C).

Zoom Image
Scheme 1 Regiodivergent oxidation of ynamides and subsequent transformations

We set out to examine the substitution, employing N,O-acetal 1a and benzotriazole 2 for the optimization study (Table [1]). The reaction did not proceed in the absence of catalyst in CH2Cl2 at 60 ℃ (entry 1). However, addition of Lewis acids significantly improved the conversion (entries 2–6). Among them, Zn(OTf)2 turned out to be the most potent Lewis acids, furnishing a mixture of N1- and N2-adducts (3a and 4a, respectively) in a combined 80% yield with N1/N2 (3a/4a) ratio of 5.7:1, upon heating at 60 ℃ for 60 h. The isomeric ratio (N1/N2) was not affected much by different Lewis acids, which later turned out to be a thermodynamic ratio.

Table 1 Screening of Conditions for Transacetalization of 1a a

Enry

Catalyst

Solvent

Time (h)

Yield of 3a/4a (%)b

 1

none

CH2Cl2

 48

 0/0c

 2

Fe(OTf)2

CH2Cl2

 72

64/11

Yb(OTf)3

CH2Cl2

 72

48/8

 3

Cu(OTf)2

CH2Cl2

 72

38/5

 4

In(OTf)3

CH2Cl2

 60

49/7

 5

Sc(OTf)3

CH2Cl2

 60

55/8

 6

Zn(OTf)2

CH2Cl2

 60

68/12

 7

HNTf2

CH2Cl2

 72

54/9

 8

HBF4·OEt2

CH2Cl2

 84

50/6

 9

CF3COOH

CH2Cl2

120

58/7

10

H3PO4

CH2Cl2

144

50/13

11

Zn(OTf)2

C6H6

 72

65/13

12

Zn(OTf)2

Et2O

 60

69/12

13

Zn(OTf)2

MeOH

 48

88/0

14

Zn(OTf)2

i-PrOH

 30

82/0

15

Zn(OTf)2

HFIP

  9

91/0

a The reaction was conducted with 1a (0.1 mmol) and 2 (0.3 mmol).

b Yield was determined by crude 1H NMR spectra, with CH2Br2 as an internal reference.

c Starting 1a remained intact.

Brønsted acids also catalyzed the transacetalization, but the reaction took a much longer time, compared to Zn(OTf)2 (Table [1], entries 7–10). We then screened solvents for the transacetalization. While CH2Cl2, C6H6, and Et2O gave similar N1/N2 ratio, alcoholic solvents exclusively produced N1-adduct (entries 11–14). Finally, hexafluoroisopropanol (HFIP) gave a drastically improved conversion as well producing only 3a in 91% yield as measured by crude 1H NMR spectrum.

With the conditions optimized above, we inspected substitution of various N,O-acetal substrates 1 with benzotriazole (BT, 2, Scheme [2]). Substitution of 1a could be repeated in 4 mmol scale with an identical efficiency (9 h), affording the product 3a in 87% isolated yield. Polycyclic aromatic (3b,c) and heteroaromatic (3d,g) groups could be accommodated in the ketone moiety. Electron-rich (3hj) as well as electron-poor (3kt) aryl groups also underwent smooth transacetalization including sterically demanding ones (3i,k). Substrates with an aliphatic ketone (3tv) afforded the transacetalization products smoothly, with a slightly decreased yield. Change of R1 and R2 groups as in 1wy were also well tolerated, providing the corresponding N,N-acetals 3wy in good yields.

Zoom Image
Scheme 2 Substitution with benzotriazole. Reagents and conditions: 1 (0.1 mmol), 2 (0.3 mmol), Zn(OTf)2 (10 mol %), 60 °C, 9 h. Isolated yield is reported.

Importantly, the currently developed conditions for the transacetalization to N,N-acetals can be extended to other N-donor compounds (Scheme [3]). For example, substitution of 1a with 1H-indazole 5 proceeded as efficiently under otherwise identical conditions, affording N-substitution product 6a in 75% yield in 10 h. The identity of 6q, i.e., whether it was the C- or N-adduct, was determined from the DEPT spectrum, and whether it was the N1- or N2-isomer, the identity of 6q was based on the NOE spectra. The generality with 1H-indazole was briefly examined, and the products 6b,i,q were obtained smoothly in 10–16 h. Employing TMSN3, an azide could also be incorporated into the N,N-acetal 7a (68%) in 12 h. Synthesis of other N,N-acetals bearing an azide moiety also went smoothly, affording 7b,f,h,i,p in good yields in 6–18 h.

Zoom Image
Scheme 3 Substitution with 1H-indazole and TMSN3. Reagents and conditions: 1 (0.1 mmol), 1H-indazole or TMSN3 (0.3 mmol), Zn(OTf)2(10 mol %), 60 °C, 12 h. Isolated yield is reported.

To support the mechanism of the current regioselective transacetalization, we carried out the following experiments. From the regioisomeric mixture of 3a/4a (N1/N2 = 6:1), the N1 isomer 3a was purified and was subjected to the reaction conditions. As in Scheme [4, 3a] isomerized into the 8:1 mixture of 3a/4a, upon heating at 80 °C in the presence of Zn(OTf)2 in CH2Cl2. In contrast, a mixture of 3a/4a (3:1) isomerized into 3a exclusively in HFIP (Scheme [5]). These indicated the reversible interconversion between 3a and 4a, and the equilibrium is shifted depending on solvents, presumably through a keteniminium ion. We attempted an enantioselective trans-acetalization of 1a with benzotriazole in the presence of (R)-TRIP or (R)-TCyP (10 mol%) in CH2Cl2 at 60 °C; we obtained 3a in 60–62 % yield after 72 h, unfortunately as a racemic form (0% ee).

Zoom Image
Scheme 4
Zoom Image
Scheme 5

Considering the novelty of the obtained scaffolds, we explored synthetic transformations as described in Scheme [6]. Reduction of the ketone group in 3a with NaBH4 furnished 8a as 3:1 diastereomeric mixture in good yields (Scheme [6], eq. 1), reflecting stability of the product in the absence of the flanking ketone. We attempted modification of benzotriazole under radical conditions.[10] However, surprisingly, desulfuration of 3a into 9a occurred, presumably via hydrogen atom abstraction at the acetal center (Scheme [6], eq. 2). An attempted Cu-catalyzed [3+2] cycloaddition of azide 7a [11] resulted in reduction into an imidate 10 (Scheme [6], eq. 3). This transformation may have occurred through oxidative addition of Cu(I) to the N–N bond in the azides, followed by reductive elimination. These transformation demonstrates the utility of the N,N-acetals obtained in this study.

Zoom Image
Scheme 6 Synthetic application. Reagents and conditions: (1) 3a (0.1 mmol), NaBH4 (0.3 mmol), r.t, 30 min; (2) 3a (0.1 mmol), NaBH4 (0.5 mmol), Bu3SnH (0.06 mmol), AIBN (5 mol %), 60 °C, 12 h; (3) 7a (0.1 mmol), CuI (10 mol%), DIPEA (0.03 mmol), AcOH (0.03 mmol), r.t, 2 h. Isolated yield is reported.

To summarize, we developed herein transacetalization conditions converting N,O-acetals 1 into the N,N-acetals 3 using benzotriazole, indazole, and azides.[12] The reaction is significantly accelerated by the combination of Zn(OTf)2 catalyst and HFIP solvent under which conditions substitution with benzotriazole occurred with a complete regioselectivity. Further work on the enantioselective N,N-transacetalization is currently underway in the laboratory.


#

Supporting Information



Zoom Image
Scheme 1 Regiodivergent oxidation of ynamides and subsequent transformations
Zoom Image
Scheme 2 Substitution with benzotriazole. Reagents and conditions: 1 (0.1 mmol), 2 (0.3 mmol), Zn(OTf)2 (10 mol %), 60 °C, 9 h. Isolated yield is reported.
Zoom Image
Scheme 3 Substitution with 1H-indazole and TMSN3. Reagents and conditions: 1 (0.1 mmol), 1H-indazole or TMSN3 (0.3 mmol), Zn(OTf)2(10 mol %), 60 °C, 12 h. Isolated yield is reported.
Zoom Image
Scheme 4
Zoom Image
Scheme 5
Zoom Image
Scheme 6 Synthetic application. Reagents and conditions: (1) 3a (0.1 mmol), NaBH4 (0.3 mmol), r.t, 30 min; (2) 3a (0.1 mmol), NaBH4 (0.5 mmol), Bu3SnH (0.06 mmol), AIBN (5 mol %), 60 °C, 12 h; (3) 7a (0.1 mmol), CuI (10 mol%), DIPEA (0.03 mmol), AcOH (0.03 mmol), r.t, 2 h. Isolated yield is reported.