Hypervalent ethynyl(phenyl)-λ³-iodane 1 is an efficient electrophilic ethynylating agent for a variety of nucleophiles (C, N, O, P, As, S, Se, and halides) in the presence or absence of transition metal catalysts (Figure 1A; Ochiai et al., 1990; Stang et al., 1990; Varvoglis, 1992; Ochiai, 2003; Waser, 2016; Yoshimura and Zhdankin, 2016). However, its synthetic utility is restricted by its high cost and heat/moisture-sensitive character: it gradually decomposes at room temperature in air (Ochiai et al., 2003; Yudasaka et al., 2019). Therefore, an inexpensive, rapid, and facile preparation method of 1 has long been highly desired. The current approach to the synthesis of 1 relies on electrophilic Si/Sn–I(III) exchange reaction on acetylenic carbon atoms, and has remained essentially unchanged since the early days. In 1990, Stang and Ochiai independently reported pioneering approaches for the synthesis of 1. Stang et al. prepared ethynyl(phenyl)(triflato)-λ³-iodane (1b) from ethynyl(tributyl)stannane (3) and Tf₂O-activated iodosylbenzene (2) (Figure 1B; Stang et al., 1990). On the other hand, a two-step procedure for the synthesis of 1a via [β-(trimethylsilyl)ethynyl](phenyl)-λ³-iodane 5a was developed by Ochiai and co-workers (Figure 1C; Ochiai et al., 1990). Then, in 2011, Kitamura reported another practical stepwise approach using PhI(OAc)₂ (6) and bis(tert-butyldimethylsilyl)acetylene (4b) (Figure 1D; Kitamura et al., 2011). Although these methods provide short-step approaches to 1, they have several disadvantages from the viewpoint of cost/safety of reagents. In particular, 3 is still expensive and concentrated aqueous HF is notoriously toxic (Mckee et al., 2014). We report here a safe, low-cost, two-step method for the synthesis of 1a using calcium carbide CaC₂ (7) as an ethynyl group source.

Calcium carbide CaC₂ (7) is a widely utilized industrial material that is very inexpensive (0.05 $/g; 3.2 $/mol) (Sigma-Aldrich Co., LLC.). However, its synthetic use has been limited by its poor solubility: it is not soluble in non-reactive common organic solvents (Barber and Sloan, 1961). In recent decades, several approaches using strongly coordinating solvents (DMSO, DMF, etc.) and/or coordinating additives (F⁻, CO32-, HO⁻, water, etc.) have been reported, for which 7 served as a practical C22- source (Rodygin et al., 2016). It occurred to us that the treatment of 7 with inexpensive chloro(tributyl)stannane (8) (0.3 $/g; 98 $/mol) (Sigma-Aldrich Co., LLC.), followed by Sn–I(III) exchange reaction, might provide more convenient and straightforward access to 1.

Calcium carbide (~80%), ethynyl(tributyl)stannane (95%), bis(tributylstannyl)acetylene (95%) were purchased from Sigma Aldrich and used as received. Chloro(tributyl)stannane (>97%), and boron trifluoride etherate (>98%) were purchased from TCI Japan and used as received. (Diacetoxyiodo)benzene (98+%) was purchased from FUJIFILM Wako Pure Chemical or prepared according to the literature (Watanabe et al., 2018). Potassium carbonate (>99.5%) was purchased from FUJIFILM Wako Pure Chemical and used as received. Iodosylbenzene was prepared from (diacetoxyiodo)benzene according to the literature (Sharefkin and Saltzman, 1963). Anhydrous grade of dimethyl sulfoxide, tetrahydrofuran, and N, N-dimethylformamide was purchased from Kanto Chemical and degassed by purging with argon and/or dried with a solvent purification system containing a one-meter column of activated alumina.

We commenced our study by trapping CaC₂ 7 with 8. Exposure of well-ground 7 (4 equiv) to 8 in DMSO at room temperature did not give any alkynylstannanes. Addition of small amount of water (2 equiv), which has been reported to be effective for the electrophilic trapping of 7, was fruitless (Rodygin et al., 2016). On the other hand, heating at 80°C resulted in smooth consumption of 8 and after 1 h, a 6:4 mixture of 3 and bis(tributylstannyl)acetylene (9) was obtained in 79% yield (Table 1, entry 2). The ratio of 3 and 9 has changed in a range of ca. 3:7–6:4 through several runs, partly due to the reaction scale and the surface area of 7 (entries 1 and 2). Use of longer reaction time increased the ratio of 9 (entries 3 and 4). Under the conditions, the addition of water did not significantly change the yield of 9, but it accelerated the bis-stannylation (entries 4–6). Interestingly, this stannylation did not occur in other aprotic solvents such as THF and DMF, even at elevated temperatures (≤110°C) (Cochran et al., 1990). It should be noted that these alkynylstannanes 3 and 9 could be separated from other organostannane impurities on a short column packed with K₂CO₃-silica gel (1:9) mixture (Harrowven et al., 2010). Other crystallogen analog, trimethylsilyl chloride did not afford corresponding ethynyl(trimethyl)silanes under optimized conditions, partly because of the more moisture sensitive character of silyl chloride.

Next, we focused on the synthesis of ethynyl-λ³-iodane 1a using a mixture of alkynylstannanes 3 and 9. After screening various reaction conditions, we found an efficient method. Exposure of a 6:4 mixture of 3 and 9 (obtained from the reaction shown in entry 1 in Table 1) to a combination of PhI(OAc)₂ 6 and BF₃-Et₂O in dichloromethane at −78°C resulted in smooth Sn–I(III) exchange, and after 1 h, 1a was selectively obtained in 73% yield (Figure 2A). The standard PhIO 2–BF₃-Et₂O system also afforded 1a in high yield. It should be emphasized that these methods do not require time-consuming work-up. Simple washing of the reaction mixture with hexane and Et₂O by decantation gave pure 1a and a mixture of Bu₃SnX-type organostannanes thus formed by I(III)–Sn exchange was recovered quantitatively in the supernatant. As we expected, these optimized conditions could also be applied to authentic 3 and 9 individually to provide 1a in moderate to high yields (Figure 2B). In these cases, the combination of I(III)–organostannane pairs (6–3 and 2–9) gave better yields of 1a than opposite pairs (6–9 and 2–3), although the reason remains unclear. From a mechanistic point of view, our results using 9 is somewhat surprising since the Sn–I(III) exchange of 9 with cyano(trifluoromethylsulfonyloxy)iodobenzene (10) selectively affords bis[phenyl(triflato)-λ³-iodanyl]acetylene (11) (Figure 2C; Stang and Zhdankin, 1990, 1991). The moderately electrophilic nature of iodine center of the intermediates such as PhI(OAc)₂-BF₃ (Izquierdo et al., 2016) or PhIO-BF₃ (Ochiai, 2007) might be partly responsible for the selective formation of 1a.

In summary, we have developed a safe and inexpensive two-step method for the synthesis of 1a using readily available CaC₂ 7 as an ethynyl group source. This method not only provides time-/cost-/labor-saving methodology to prepare unstable ethynyl-λ³-iodane 1, but also serves as an effective approach for synthetically useful but costly bis(stannyl)acetylene 9 (Brend'amour et al., 2018).

All datasets generated for this study are included in the article/Supplementary Material.

KM and MU conceived and designed the experiments and wrote the manuscript. TH and TO conducted the experiments. All authors participated in data analyses and discussions.