Syntheses of (Diacetoxyiodo)arenes or Iodylarenes from Iodoarenes, with Sodium Periodate as the Oxidant†

- Easy, safe, and effective novel methods for preparing either (diacetoxyiodo)-arenes, ArI(OAc)2, or iodylarenes, ArIO2, from the corresponding iodoarenes, ArI, using sodium periodate as the oxidant are presented in this paper. In order to obtain 2- and 4-iodylbenzoic acids, the respective sodium salts of 2- and 4-iodobenzoic acids should be used as the starting substrates, because mixtures containing the corresponding iodosyl derivatives as the main products along with the intended iodyl compounds are produced from the free parent acids.


Introduction
(Diacetoxyiodo)arenes, ArI(OAc) 2 , and in particular the parent compound (diacetoxyiodo)benzene, PhI(OAc) 2 , have been known for a long time [1][2][3][4][5]. They are potent, often chemoselective, oxidizing agents. They are also used for the facile syntheses of, for example, [bis(trifluoroacetoxy)iodo]arenes or [hydroxy(tosyloxy)iodo]arenes (as selective oxidants), and aromatic iodonium salts (as arylating reagents) [2][3][4][5][6]. There are several preparative methods for these compounds. So far, the substrates used have generally been [2][3][4][5]: (i) iodosylarenes dissolved in glacial acetic acid; (ii) (dichloroiodo)arenes in which chloride is exchanged by acetoxy groups coming either from silver, lead(II) or sodium acetate, or from acetic acid in the presence of mercury(II) oxide in chlorinated solvents; (iii) iodoarenes which are oxidized in warm glacial acetic acid by either peracetic acid [5], sodium perborate [7], or chromium(VI) oxide (this last work accomplished in our laboratory) [8], or electrolytically [3,4]. The standard, and most general, method for the synthesis of ArI(OAc) 2 (the oxidative diacetoxylation of iodoarenes by warm peracetic acid solution) is, in fact, a very prolonged reaction (12-16 h), and the utmost care should be taken to exactly maintain the temperature required (40 o C) [5]. The ArI(OAc) 2 are generally crystalline compounds, fairly stable in the air, which may be stored for long periods by avoiding light; the same is true in respect to the iodylarenes, ArIO 2 , which are discussed below, but it should be noted that the melting points of many ArIO 2 are, in fact, their decomposition points, often accompanied by explosion (see Table 2).
The first organic compound of iodine (V), iodylbenzene, PhIO 2 , was prepared by Willgerodt 101 years ago [1]. Since then various iodylarenes, ArIO 2 , have been reported [1,[3][4][5]. For a long time rather very few ArIO 2 have been used as useful oxidants in organic synthesis. Their former limited applications were mostly due to their polymeric nature, which makes them insoluble in most ordinary solvents, except water. Moreover, they should always be handled with care, since violent decomposition of several ArIO 2 can been induced e.g. by scraping them with a spatula, on heating above 200 o C, and also upon impact. However, since ca. 1980 a quickly growing importance in modern organic synthesis of PhIO 2 , 3-iodylbenzoic acid, 4-(t-butyl)iodylbenzene, "2-iodylbenzoic acid" [which, in fact, has a cyclic structure, i.e. 1-hydroxy-1,2-benziodoxol-3(1H)-one 1-oxide], and particularly its more safe triacetoxy derivative, named the Dess-Martin reagent [1,1,1-triacetoxy-1,2benziodoxol-3(1H)-one] has been reported in numerous papers [3][4][5][6]9]. They are commonly used as moderate and very selective oxidants. ArIO 2 are also useful in the preparation of diaryliodonium salts. In the last short review [9] the opinion has been put forward that "2-iodylbenzoic acid" and the corresponding Dess-Martin reagent may, most satisfactorily, be prepared as illustrated in Scheme 1.
Sharefkin and Saltzman [10] and more recently Indian chemists [11] have briefly reviewed all the previously reported oxidative methods for preparing ArIO 2 from either the corresponding ArI, ArICl 2 , or ArIO. In their review [11] the latter authors provided the following critical assessment: "The reported methods of Lucas and Kennedy [12] (i.e. by the thermal disproportionation of ArIO to ArIO 2 and ArI) for the preparation of PhIO 2 from PhIO proved to be lenghty and irksome in our hands while that of Sharefkin and Saltzman [10] was neither economical nor free of danger for the use of commercial 40% peracetic acid, though the yield of the desired PhIO 2 (obtained there from PhI) in this single step reaction was excellent" (72-82%). The Indian chemists [11] oxidized several ArI using KBrO 3 under strongly acidic conditions, which afforded the respective ArIO 2 in 46-96% yields. In 1993, a two-phase method was reported [13], involving the direct oxidation of ArI using aqueous hypochlorite and a phase transfer catalyst, Bu 4 N + HSO 4 -. Italian chemists [14] instead used Oxone® to oxidize 2-iodobenzoic acid to "2-iodylbenzoic acid", and they criticized the "classic" KBrO 3 /H 2 SO 4 procedure [11] as follows: "The KBrO 3based oxidation methods are unappealing for the user in that the reaction is performed in hot aqueous H 2 SO 4 (0.73 M), KBrO 3 is classified as a carcinogen (R-45) in the international classification of substance toxicity, and obnoxious bromine vapors are copiously evolved (62 g/mol of "2-iodylbenzoic acid") from the reaction mixture, with personal and environmental contamination risk. Still, incomplete oxidation of 2-iodobenzoic acid is not totally avoided and recovered "2-iodylbenzoic acid" is contaminated with variable amounts of less-oxidized precursors". They prepared several batches of "2-iodylbenzoic acid" using the "classic" Indian method [11]. The purity of this reagent varied between 90 and 95%, the major contaminants being 2-iodobenzoic acid (1-3%) and "2-iodosylbenzoic acid" (5-10%). It was also suggested that the explosive properties of some samples of "2-iodylbenzoic acid" were due to the presence of bromate or other impurities.

Results and Discussion
Previously, we have devised in our laboratory [8] a simple, two-step conversion of various ArI to ArI(OAc) 2 in the anhydrous CrO 3 /AcOH/Ac 2 O/conc.H 2 SO 4 liquid system, followed by mixing with excess 20% aqueous ammonium acetate solution which afforded, after purification, ArI(OAc) 2 in 58-82% yields. PhI(OAc) 2 , for example, was prepared from PhI in 79% yield. This procedure is 8-16 times faster and ca. 5 times less expensive than the method of McKillop and Kemp [7]. However, this method is hardly applicable for iodotoluenes [4-MeC 6 H 4 I(OAc) 2 was obtained from 4-MeC 6 H 4 I in only 20% yield], while iodoanisoles were quickly oxidized in the reaction mixtures, with the evolution of iodine vapors, and the formation of tarry products. In our present work we have developed a quick and efficient oxidative method, which allows us to obtain also the respective pure ArI(OAc) 2 from ArI substituted with the methyl or methoxy group (Table 1) -hence this complements nicely our former procedure [8]. The essence of our novel method is described in Scheme 2. The above oxidative reactions were carried out in boiling, anhydrous binary AcOH/Ac 2 O mixtures. The presence of sodium acetate (in stoichiometric quantities) in the reaction mixtures was indispensable -without its addition the oxidation reactions did not proceed; when sodium acetate was replaced by pyridine, then the final yields of ArI(OAc) 2 were lowered by ca. 10-20%. After completing the oxidative reactions, the reaction mixtures were poured into water, the collected solids were washed and air-dried in the dark, and were recrystallized from ethyl acetate admixtured with acetic anhydride to reacetylate the partly hydrolyzed crude products (cf. Ref. [8]). After cooling, excess hexane was added to improve the crystallization yields ( Table 1). The parent pure PhI(OAc) 2 was obtained from PhI in 73% yield, the other yields are given in Table 1. This method is not applicable for ArI substituted with strong electron-withdrawing groups. The purities of the purified products, ArI(OAc) 2 , were checked by mixed melting points with authentic specimens as well as with 1 H-NMR spectra and elemental analyses. For more details see Experimental and Table 1.
The heterophasic reactions of various iodoarenes ( When the above reactions were completed, the reaction mixtures were diluted with ice water. The collected precipitates were washed on the filter with cold water, followed by CHCl 3 or acetone (to remove the unreacted ArI), and then air-dried in the dark. Iodometric titrations [15] showed that our crude ArIO 2 had 98-99% purity, hence they may be used directly in various subsequent reactions. These crude products did not oxidize KI in a saturated sodium borate solution (a reaction which differentiates iodosyl compounds from iodyl compounds [15]). Small samples of the crude products were recrystallized from boiling water to give the analytical specimens, whose melting/detonation points were close to those reported in the literature (Table 2). Their IR solid spectra displayed the characteristic frequencies for ArIO 2 , in the spectral region 710-800 cm -1 [3,11].
The reaction mixtures discussed above were nearly neutral. When they were made alkaline by the addition of an aqueous NaOH, the unreactive Na 3 H 2 IO 6 precipitated out, and the reactions did not proceed. Negative results were also obtained, when NaIO 4 was replaced for NaIO 3 as the oxidant under the same reaction conditions. 3-Iodobenzoic acid gave 3-iodylbenzoic acid without problem in boiling aqueous NaIO 4 solution ( Table 2). But 2-and 4-iodobenzoic acids gave mainly either cyclic "2-iodosylbenzoic acid" [i.e. 1-hydroxy-1,2-benziodoxol-3(1H)-one] in 76% crude yield, or 4iodosylbenzoic acid in ca. 84% crude yield. Iodometric titrations [15] showed that the crude cyclic "2iodosylbenzoic acid" was only slightly admixtured with the iodyl derivative (ca. 5%), while the crude 4-iodosylbenzoic acid contained ca. 40-50% of the corresponding iodyl derivative. However, when the sodium salts of the two acids (prepared in situ, see Experimental) were reacted as above with boiling aqueous NaIO 4 solutions, then this resulted in the formation of either the cyclic "2-iodylbenzoic acid" (vide supra) in 71% crude yield, or 4-iodylbenzoic acid in 88% crude yield. Iodometric titrations [15] showed that they were 98-99% pure. After recrystallizations from boiling water, their chemical structures were supported analytically, spectrally (IR), and by their characteristic melting/detonation points. In our opinion, these two crude products may also be used directly in many subsequent reactions, because they have a satisfactory purity.
The proposed multistage mechanism of converting iodoarenes to iodylarenes, with using NaIO 4 as the oxidant in boiling water, would be as follows:
.ArI + 2 NaIO 4 ArIO 2 + 2 NaIO 3 In the course of the above reactions (Scheme 3), at first some yellowish colored solids were observed to copiously precipitate out (their color being characteristic of ArIO), which were then gradually transformed into the final colorless products, ArIO 2 . It is known [1,[3][4][5] that thermally unstable ArIO cannot be recrystallized from hot or boiling water, in contrast to ArIO 2 , since they readily disproportionate as follows [1,12]: 2ArIO → ArIO 2 + ArI. However, when 2-iodobenzoic acid was oxidized with NaIO 4 in boiling water, then the formed 2-OIC 6 H 4 COOH would momentarily transform into its cyclic (and fairly thermostable) desmotropic form, 1-hydroxy-1,2-benziodoxol-3(1H)-one, and its subsequent thermal disproportionation was hardly possible; this was not the case when sodium 2-iodobenzoate (and sodium 4-iodobenzoate) was oxidized with NaIO 4 in boiling water (see Experimental). From the above it is seen that NaIO 4 used as the oxidant either in boiling AcOH/Ac 2 O/AcONa solutions (Scheme 2) or in boiling water (Schemes 3 and 4) does, in fact, oxidize the iodine atoms in iodoarenes, ArI, to some iodine(III) intermediates, which in the subsequent reactions gave the desired final products, viz. ArI(OAc) 2 (Scheme 2) or ArIO 2 (Scheme 3) in good yields. As previously seen [16], in our present work the successful use of sodium periodate as the oxidant also resulted in easy, safe, and effective preparations of twelve ArI(OAc) 2 (Table 1) and fourteen ArIO 2 ( Table 2) from the respective ArI. The oxidative reactions of ArI proceeded smoothly under the conditions given in Experimental, and this positively extends the organic chemistry of polyvalent iodine compounds. Our novel method for preparing ArI(OAc) 2 from iodotoluenes and iodoanisoles nicely complements our former method [8]. Also, our novel method for preparing ArIO 2 from ArI seems to be interesting, because taking both into account the safety aspects as well as good yields and satisfactory purities (98-99%) of the crude iodyl compounds obtained by us, this may compete in full with the "classic" results attained formerly by the Indian chemists [11] and their followers (see references in Ref. 14). Our present paper also extends, very interestingly, the known [16] oxidative applicability of NaIO 4 in organic synthesis.

Note
Our preliminary results were presented at the 3rd Symposium on Iodine Utilization, Chiba (Japan), November 24-25, 2000 [18]. They are also related in our latest review: "Organic Iodine (I, III, and V) Chemistry: 10 Years of Development at the Medical University of Warsaw, Poland" [19].

General
The starting commercial iodoarenes, ArI, were freshly purified prior to use. 1 H-NMR solution spectra of pure (diacetoxyiodo)arenes (Table 1) were taken with a Tesla (100 MHz) spectrometer, while solid IR spectra of iodylarenes (Table 2) were taken with a Perkin-Elmer spectrophotometer. Microanalyses were carried out at the Institute of Organic Chemistry, the Polish Academy of Sciences, Warsaw.

(Diacetoxyiodo)arenes: General Procedure
NaIO 4 (2.20 g, 10.3 mmol; 3% excess) and AcONa (1.8 g, 22 mmol) were suspended in a stirred mixture of glacial AcOH (15 mL) with Ac 2 O (1.5 mL). An appropriate iodoarene (10 mmol) was added, and the mixture was stirred and boiled under a reflux condenser for 2 h, cooled to r. t., and next poured into water (50 mL). The precipitates (unless otherwise stated below) were collected by filtration, washed thoroughly with 10% aq. AcOH (2 x 10 mL) [17] to remove possibly all soluble impurities, and air-dried in the dark. The yellowish crude products were recrystallized from EtOAc/Ac 2 O (9:1, v/v). After short boiling, the yellowish solutions faded (cf. Ref. 8) and, after cooling, hexane was added in excess (Table 1) with stirring. After 30 min, the crystals were collected by filtration, washed with hexane and air-dried in the dark. See Table 1 for more details.  2 were extracted with CH 2 Cl 2 (3 x 10 mL). The collected extracts were dried (Na 2 SO 4 ), the solvent was distilled off under reduced presure, and the residues were triturated with hexane (15 mL) until they solidified. The solids were collected by filtration, washed with hexane, and air-dried in the dark. Next, they were recrystallized as above to give the pure products (Table 1).

Iodylarenes: General Procedure
NaIO 4 (4.70 g, 22 mmol; 10% excess) was suspended in water (20 mL). An appropriate iodoarene (10 mmol) was added with stirring; for ArI which are steam-volatile, 0.1-0.3 mL of toluene was also added to prevent their accumulation in the reflux condenser. The mixtures were vigorously stirred and refluxed for 8-16 h (Table 2). Next, ice water (50 mL) was added with stirring, and the temperature was lowered to r.t. The crude products were collected by filtration, washed with cold water (2 x 10 mL) and CHCl 3 or Me 2 CO (3 x 10 mL) to remove the unreacted ArI and PhMe, and air-dried in the dark; the crude yields are given in Table 2. Small crude samples were recrystallized from boiling water to give the analytically pure products. For 2-and 4-iodobenzoic acids, their sodium salts (10 mmol) should be used instead of the parent acids to obtain the respective iodylbenzoic acids (see below).