The Exploration of Aroyltrimethylgermane as Potent Synthetic Origins and Their Preparation.

The synthetic utilities of acylgermanes are surprisingly rarely explored compared with their analogues. In this communication, the survey of aroyltrimethylgermane as potent synthetic origins has been studied. A variety of novel chemical transformations have been realized, including using the acylgermane group as a directing group in Rh-catalyzed aromatic C-H alkenylation reaction and Ir-catalyzed aromatic C-H amidation reactions. Additionally, a general approach for acylgermanes preparation has been established as well. The catalytic system proceeds effectively in the presence of Pd(OAc)2/BINOL-based monophosphite (L11) and allows for the straightforward access to a wide range of functionalized acylgermanes in high yields.

Introduction

The acylsilanes, germanes, and stannanes have been well known as an electronically unique class of group-14 element compounds with remarkable n→π∗ redshifted transition band and their lower transition energy (Ramsey et al., 1974, Page et al., 1990). The electron inherent in these compounds leads to a distinct reactivity from other carbonyl compounds (Brook et al., 1960, Harnish and West, 1963, Yoshida et al., 1989, Yoshida et al., 1992). During the past decades, they have been explored as versatile synthetic intermediates in various novel chemical transformations (Brook, 1974, Moser, 2001, González et al., 2015). However, compared with the well understanding of the reactivity of acylsilanes and acylstannanes, the synthetic reactivity of acylgermanes is surprisingly much less explored (Cirillo and Panek, 1992, Galliford and Scheidt, 2008, Ito et al., 2011, Lettan et al., 2006, Matsuda et al., 2014, Mattson et al., 2004, Mattson et al., 2006, Obora et al., 2002, Schmink and Krska, 2011, Yu et al., 2016, Matsumoto and Shindo, 2012, Shindo et al., 2007, Zhang et al., 2013).

Acylgermanes have been of great interest recently, because they showed the unique advantages in the field of photo-initiated free-radical polymerization reactions (Ganster et al., 2008, Jöckle et al., 2017, Jöckle et al., 2018, Lalevee et al., 2009, Moszner et al., 2009, Neshchadin et al., 2013, Radebner et al., 2017a, Radebner et al., 2017b, Haas et al., 2018, Lappert et al., 1987, Jutzi and Hampel, 1986, Zhu et al., 2019). Due to the great achievements on chemical transformations of acylsilanes and acylstannanes, it is intriguing to discover the potential synthetic utilities of acylgermanes.

Results

With this idea in mind, testings were performed and a set of new transformations of benzoyltrimethylgermane were succeeded (Scheme 1), for example, the intermolecular Schmidt reaction; palladium-catalyzed acylation of allylic trifluoroacetate; synthesis of β-keto ester by using diazo ester and benzoyltrimethylgermane as the reaction partner; rhodium-catalyzed arylation of benzoyltrimethylgermane with sodium tetraphenylborate; thiazolium-catalyzed additions of acylgermane to access 1,4-dicarbonyl products; and one-pot assembly of pyrrole ring.

Scheme 1

New Synthetic Transformations of Benzoyltrimethylgermane

[a] (2-Azidoethyl)benzene (1.5 equiv.), TfOH (2.0 equiv.), DCM, rt, 5 min. [b] Allyl trifluoroacetate (1.2 equiv.), Pd(TFA)2 (5 mol%), THF, 70°C, 8 h. [c] (1) Ethyl diazoacetate (1.0 equiv.), LDA (1.0 equiv.), THF, −78°C; (2) MeOH, 0°C. [d] NaBPh4 (2.0 equiv.), [Rh(cod)Cl]2 (3.0 mol%), m-xylene, 130°C, 24 h. [e] (1) Thiazolium (30 mol%), butyl acrylate (1.0 equiv.), DBU (0.3 equiv.), i-PrOH (4.0 equiv.), THF, 70°C; (2) H2O. [f] (1) Thiazolium (30 mol%), chalcone (1.0 equiv.), DBU (0.3 equiv.), i-PrOH (4.0 equiv.), THF, 70°C; (2) H2O. [g] (1) Thiazolium (30 mol%), chalcone (1.0 equiv.), DBU (0.3 equiv.), i-PrOH (4.0 equiv.), THF, 70°C; (2) aniline (3.0 equiv.), p-toluenesulfonic acid (2.0 equiv.), EtOH, 4 Å MS, 70°C. See also Figures S9–S24.

With these promising results in hand, we start to look at the preparation of acylgermane compounds. Since the first acylgermane, benzoyl(triphenyl)germane, was synthesized in 1960 by Brook and co-workers (Brook et al., 1960), many synthetic methods toward acylgermanes were developed, including hydrolysis of germyldithianes (Brook et al., 1967, Corey and Seebach, 1965), reacting of acyl chlorides, esters, and amides with germyllithiums or other germylmetallic reagents (Bravo-Zhivotovskii et al., 1983, Castel et al., 1990, Castel et al., 1992, Iserloh and Curran, 1998, Kiyooka and Miyauchi, 1985, Yamamoto et al., 1987, Nanjo et al., 2001, Piers and Lemieux, 1995), and palladium-catalyzed transformation of alkynes with germanium hydride (Kinoshita et al., 2002). Given the importance of acylgermanes and its newly developed promising synthetic utilities in chemical transformations, we believe a practical approach to acylgermanes is under the current demand and could be achieved by carbonylative coupling of aryl halides with hexamethyldigermanium.

To verify our hypothesis, we decide to choose iodobenzene (1a) and hexamethyldigermanium (2) as the model substrates to establish the catalyst system. Initially, different palladium catalyst and phosphine ligands were tested; however, these reactions could give only traces of the desired product (see in Table S1). Gratifyingly, the desired product 3a was furnished in a promising yield of 54% when P(OMe)3 (L1) was used as the ligand (Table 1, entry 1). This led us to examine the effect of different R groups (L2-L4, Table 1, entries 2–4). Notably, when ligand L4 with R of 2,4-ditBuPh group was used, 3a was obtained in 80% yield (Table 1, entry 4). The testing of the other palladium catalyst resulted in a decreased yield of 3a (Table 1, entries 5–7). We then turned our attention to the other ligands based on different backbones (L5-L12, Table 1, entries 8–13). When the modified ligands L11 and L12 based on the binaphthol (BINOL) backbone bearing R = 1-Ad and R = 2-Ad were used, resulting similar high yields in 87% and 89%, respectively (Table 1, entries 13–14). Other commonly used phosphite ligands such as monophos L12 only afforded the product in poor yield. The catalyst loading can be decreased to 2.5 mol% Pd(OAc)2 and 5.0 mol% L11 as well and furnished 3a in 90% GC yield with 83% isolated yield (Table 1, entry 16). Remarkably, 3a also can be achieved in 80% isolated yield even under 1 bar CO pressure (Table 1, entry 17). Here it is also important to mention that Me3GeI can be obtained as the byproduct.

Table 1

Investigation of Reaction Conditions

Entry	Ligand	Palladium	Yield (%)a
1	L1	Pd(OAc)2	54
2	L2	Pd(OAc)2	58
3	L3	Pd(OAc)2	60
4	L4	Pd(OAc)2	80
5	L4	Pd2(dba)3	75
6	L4	[PdCl(C3H5)]2	70
7	L4	Pd(CH3CN)2Cl2	58
8	L5	Pd(OAc)2	30
9	L6	Pd(OAc)2	69
10	L7	Pd(OAc)2	59
11	L8	Pd(OAc)2	45
12	L9	Pd(OAc)2	70
13	L10	Pd(OAc)2	87
14	L11	Pd(OAc)2	89
15	L12	Pd(OAc)2	41
16b	L11	Pd(OAc)2	90 (83)c
17d	L11	Pd(OAc)2	80d

See also Table S1, Figures S1–S8.

Reaction conditions: 1a (0.2 mmol), 2 (0.24 mmol), Pd(OAc)2 (5.0 mol %), L (10.0 mol %), CO (20 bar), 100°C and toluene (1.0 mL).

Yields were determined by GC with hexadecane as an internal standard.

Pd(OAc)2 (2.5 mol %), L (5.0 mol %).

Isolated yield.

CO (1 bar).

Discussion

Having established the optimal conditions, we subjected the hexamethyldigermanium to the reaction with different (hetero)aryl iodides (Scheme 2). To our delight, the scope of this transformation is significantly broad. A variety of aryl iodides bearing electron-donating substituents at the para positions were successfully converted to the desired products 3a-3g in good yields. Various electron-withdrawing functional groups such as halogen, N-heterocycles, formyl, esters as well as ketones at the para positions of aryl iodides were all well tolerated and afforded the corresponding substituted products in 56%–86% isolated yields. Notably, the reaction proceeds smoothly with aryl iodides bearing cyano (3i), nitro (3j), azidomethyl (3r), and vinyl (3u). Ortho- or meta-substituted aryl iodides were able to give the corresponding products in high yield as well (3v-3ac). Moreover, di-, tri-, and tetra-substituted aryl iodides also reacted smoothly to furnish the desired products 3ad-3ag in good to excellent yields. Importantly, 1-iodonaphthalene, 2-iodothiophene, and indole-containing substrate were successfully compatible under the reaction conditions (3ah-3aj, 71%–75% yields). Interestingly, when applying 1,4-diiodobenzene as the substrate, mono- (3ak) and di-substituted (3ai) acylgermanes could be obtained respectively by controlling the equivalents of hexamethyldigermanium added. Nevertheless, 3-iodopyridines and 4-hydroxyiodobenzene did not work well under the reaction conditions. To demonstrate the potential applications, late-stage modification of various biologically active molecules, natural products, and pharmaceuticals derivatives were also conducted. Menthol derivative 3am and 3an can be isolated in 72% and 82% yield, respectively. Clofibrate-, glucose-, nerol-, and cholesterol-derived 3ao-3ar were all obtained in good yields (74%–98% yield, Scheme 2).

Scheme 2

Synthesis of Aroyltrimethylgermanes

Reaction Conditions: 1a (0.2 mmol), 2 (0.24 mmol), Pd(OAc)2 (2.5 mol%), L11 (5.0 mol %), CO (20 bar), toluene (1.0 mL), 100°C, 12 h, isolated yield. [a] 2 (0.2 mmol). [b] 2 (0.5 mmol). See also Figures S25–S111 and S140–144.

The practicability of a synthetic methodology is the possibility for easy scale up. Hence, we performed the reaction in a 2 mmol scale in the presence of 1.5 mol% palladium catalyst and 3.0 mol% ligand (L11), and the desired product 3a was obtained in 355 mg, 80% yield (Scheme 3).

Scheme 3

Scale-up Experiment

Furthermore, encouraged by the recent work on acylsilane-directed aromatic C-H functionalization by rhodium (Becker et al., 2014), iridium (Becker et al., 2015), and ruthenium catalysis (Lu et al., 2019), we investigated the possibility of preparing the desired ortho-olefinated acylgermanes by utilizing acylgermane as a reactant. By using [(RhCp∗Cl2)2] (2.5 mol%), AgOTf (10 mol%), and Cu(OAc)2 (1.5 equiv.) in DCE at 70°C for 24 h, the desired ortho-olefinated benzoyltrimethylgermane 12a can be obtained in 83% isolated yield. We then roughly examined the scope of this acylgermane-directed aromatic C-H alkenylation reaction. Various acrylates such as methyl acrylate, ethyl acrylate, butyl acrylate, tert-butyl acrylate, and even phenyl vinyl sulfone were all efficiently reacted with benzoyltrimethylgermane (3a) to produce the corresponding products in moderate to good yields (12a-12e, Scheme 4). Two different substituted acylgermanes 3 were also investigated, and both were able to afford the desired products in good yields (12f-g, Scheme 4).

Scheme 4

Rhodium-Catalyzed Alkenylation of Acylgermanes

[a] Reaction conditions: 3 (0.2 mmol), 11 (0.4 mmol), [{Cp*RhCl2}2] (2.5 mol %), AgOTf (10.0 mol %), and Cu(OAc)2 (1.5 equiv) in DCE (0.5 mL) under 70°C for 24 h. Isolated yield for all products.

See also Figures S112–S125.

Meanwhile, inspired by the iridium-catalyzed ortho-amidation reaction of aroylsilanes with sulfonyl azides by Bolm and co-workers, benzoyltrimethylgermane (3a) and benzenesulfonyl azide were selected as representative substrates. A combination of [(IrCp∗Cl2)2] (2.5 mol %), AgBF4 (10 mol%), and AgOAc (5.0 mol%) was applied as the catalytic system in DCE at 60°C. To our delight, the catalytic system was very efficient, and the desired product 14a was obtained in 87% yield after 1.5 h. Subsequently, sulfonyl azide (13) substrates were tested, and the corresponding products (14b-c, Scheme 5) were all obtained in high yields. Menthol-, glucose-, and cholesterol-derived acylgermanes can be well tolerated as well and resulting the desired products 14d-f in excellent yields (89%–93% yield, Scheme 5). Additionally, under the irradiation of light (415 nm), acylgermane can be activated to generate acyl radical and then captured by ((phenylethynyl)-sulfonyl)benzene to produce alkynone product (Scheme 6).

Scheme 5

Iridium-Catalyzed Amidation of Acylgermanes

[a] Reaction conditions: 3 (0.2 mmol), 13 (0.24 mmol), [IrCp*Cl2]2 (2.5 mol %), AgBF4 (10.0 mol %), AgOAc (5.0 mol %) in DCE (0.5 mL) under 60°C for 1.5 h. Isolated yield for all products. See also Figures S126–S137.

Scheme 6

Alkynone Synthesis from Acylgermane

[a] Reaction conditions: 3a (0.1 mmol), B (0.15 mmol) in MeCN (1.5 mL) under room temperature under light (415 nm) for 12 h, isolated yield.

See also Figures S138–S139.

Conclusions

In summary, we have demonstrated the versatility of acylgermanes in various new synthetic transformations and also developed a general approach to a variety of synthetically useful acylgermanes by palladium-catalyzed carbonylative reaction. The using of new BINOL-based monophosphite (L11) ligand enables the effective transformation of a broad range of aryl iodides, including drug-like molecules. It is noteworthy that, for the first time, the acylgermane group has been successfully explored as directing group in Rh-catalyzed aromatic C-H alkenylation reaction and Ir-catalyzed ortho-amidation reaction. These new synthetic applications highlight the usefulness of the obtained aroylgermanes products. Further studies are ongoing in our laboratory to investigate the properties and reactivity of the acylgermanes compounds.

Limitations of the Study

Aryl bromides and aryl chlorides are still not applicable as substrates in this system. The obtained acylgermanes is not stable under light and cannot be stored for long term.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.