Iodonium Metathesis Reactions

Abstract

A metathesis reaction occurs when a diaryliodonium triflate is heated with an aryl iodide, resulting in formation of a new diaryliodonium triflate.

Graphical Abstract

Aryl iodides undergo I-arylation upon treatment with diaryliodonium triflates

This paper describes a thermally induced, metal-free metathesis reaction between diaryliodonium triflates and certain aryl iodides according to Eq. 1 (Scheme 1). While DiMagno observed fluoride-catalyzed aryl ligand exchange between diaryliodonium species (Eq. 2),^[1] and Koser reported oxygen ligand exchange between PhI(OH)OTf and aryl iodides (Eq. 3),^[2] the chemistry of Eq. 1 appears to be undocumented.

Scheme 1.

Iodonium metathesis reactions.

By way of background, diaryliodonium species are carriers of electrophilic aryl synthons, hence they arylate a diversity of nucleophiles,^[3] even very weak ones.^[4] For that reason, they are of significant interest in organic chemistry.^[5] On the other hand, Yamamoto has shown that organic iodides possess appreciable I-nucleophilicity.^[6] This raises the question of whether diaryliodonium agents might be able to I-arylate aryl iodides; i.e., whether an iodonium metathesis reaction might be possible. The ability to prepare diaryliodonium complexes by this method would have a favorable impact in organic, materials, and medicinal chemistry;^[3–5] however, the prognosis for the feasibility the transformation seemed poor. For instance, Olofsson, detected no aryl group exchange between bis(4-tolyl)iodonium triflate and 2,6-dimethyl-iodobenzene upon heating in DMF.^[7]

In contrast with the foregoing, we found that 4-iodotoluene, 4-iodoanisole, and 1-iodonaphthalene displace PhI from Ph₂IOTf, leading to mixtures of products of mono- and bis-substitution. With low-melting^[8] aryl iodides, the reaction may be induced by melting a mixture of the reactants.^[9] However, incomplete consumption of Ph₂IOTf was consistently observed under these conditions. Generally better results were obtained by operating in 1,2-dichloroethane (DCE) solution, in which case most of the starting Ph₂IOTf was consumed (Table 1). The solution protocol is thus the method of choice. Interestingly, such metathesis reactions failed in donor solvents such as (CF₃)₂CHOH, MeCN, DMF, or DMSO. This stands in contrast with DiMagno’s report that the chemistry of Eq.2 takes place in MeCN solution,^[1] but it is consistent with Olofsson’s observation that no aryl group exchange between diaryliodonium triflates and aryl iodides occurs in DMF.^[7] A rationale for all this segues from a possible mechanism for the metathesis reaction proposed herein.

Table 1.

Solution metathesis reactions of Ph₂IOTf.

entry	Ar’	time (h)	ratio 3:4:1[b]	yield[c]
a	4-Me-C6H4	24	9.8:28.5:1.0	48
b	4-MeO-C6H4	25	27.3:23.1:1.0	78
c	1-naphthyl	26	2.0:1.0:0.0	14

^[a]Conditions: 0.2 M solution of Ph₂IOTf in (CH₂Cl)₂, 5 equiv aryl iodide, thick-walled glass tube sealed with a Teflon screwcap and immersed in an oil bath kept at 120–125 °C.

^[b]Molar ratios calculated by integration of ¹H NMR spectra.

^[c]Percent yield after silica gel column chromatography (gradient 10% → 40% acetone-CH₂Cl₂) to remove nonpolar byproducts. The stated value is the sum of the yields of individual compounds present in the product mixture. A comparison of ¹H NMR spectra of crude and purified reaction mixtures indicated that insignificant changes in the ratio of iodonium triflate products had occurred upon chromatography.

As seen in Table 1, the reaction tends to afford mixtures of products. In an effort to improve selectivity, we examined the metathesis chemistry of mixed iodonium triflates, in which one of the aryl ligands is electron-poor, and the other, electron-rich.^[10] Such species react with common nucleophiles by what appears to be an S_NAr mechanism, resulting in transfer of the more electron-deficient aryl group^[11] — at least in the absence of transition metal catalysts.^[12] Greater aryl group transfer selectivity might obtain if such a preference were maintained in the context of the metathesis reaction.

The foregoing complexes did react selectively, but contrary to expectations, they preferentially transferred the more electron-rich group to 1-iodonaphthalene, 4-I-C₆H₄Me, 4-I-C₆H₄OMe, and Ph-I, (Tables 2–3), indicating that the metathesis reaction is unlikely to proceed by an S_NAr mechanism.^[13] On the other hand, thienyliodonium triflate 8 transferred the 4-MeOOCC₆H₄ group selectively (Table 4), leading to mixtures of products 9 and 4, but only traces of 10. In light of precedent,^[11] Coenen, who also observed preferential expulsion of 2-iodothiophene in S_NAr reactions of mixed 2-thienyliodonium salts,^[14] ascribed this selectivity to the electron-rich character of the thienyl group. In the present case, such an explanation would be untenable, given the results of Tables 2–3. A more satisfactory rationale for aryl group transfer selectivity in iodonium metathesis emerged from a computational estimate of the fractional positive charge on the I-atom of various aryl iodides (Table 5). Halides at the top of the list exhibit a smaller (+)-charge on the I-atom, and may thus be expected to be better I-nucleophiles than those at the bottom of the table, for which, increased I-(+)-charge translates into greater nucleofugality. A diaryliodonium triflate based on an aryl iodide with a greater extent of (+)-charge on the I-atom is likely to be destabilized relative to one centered on a less I-positive iodide, because the I-atom must acquire additional (+)-character in the hypervalent state.^[17] Such an accumulation of (+)-charge on an already positive I-atom is surely energetically unfavorable. Therefore, iodonium metathesis occurs so that the extent of (+)-charge on the hypervalent I-atom decreases; i.e., so that a more I-nucleophilic iodide displaces a less I-nucleophilic one. Table 5 seems to rationalize the formation of mixtures in reactions of Ph₂IOTf with 4-I-C₆H₄Me, 4-I-C₆H₄OMe, and 1-iodonaphthalene, given the similar charges on the I-atom of nucleophiles and nucleofuge (PhI). The selective transfer of the 4-MeOOCC₆H₄ group from 5 is consistent with the greater extent of (+)-charge on the I-atom of 2-iodothiophene (better nucleofuge) relative to methyl 4-iodobenzoate. Finally, the fact that no metathesis occurred between Ph₂IOTf and 5-iodouridine triacetate^[18–19] or 4-Br-C₆H₄-I may be ascribed to the unfavorable change in (+)-charge on the hypervalent I-atom that would be incurred during the reaction.^[20] On the basis of the foregoing, a possible mechanism for the iodonium metathesis reaction may be ventured as outlined in Scheme 2. Reversible dissociation of starting complex 11 and addition of a nucleophilic aryl iodide to iodonium ion 12 yields 13. The latter would form only in polar, nonnucleophilic media such as DCE: donor solvents (DMF, MeCN, etc.), surely more nucleophilic than aryl iodides, would outcompete Ar³I for 12, thus retarding / suppressing metathesis. Kinetically faster expulsion of the more nucleofugal aryl iodide from 13, perhaps via transition state 14,^[7] leads to iodonium ion 15, which then combines with TfO⁻ to give the final 16.

Table 2.

Metathesis reactions of iodonium triflate 5.

entry	Ar’	time	ratio 6:4:5:(other)[b]	yield[c]
a	Ph	15 h	13.9[d]:1.0:(−)[e]	51
b	4-MeC6H4	15 h	6.0:35.7:1.0:(−)[e]	59
c	4-MeOC6H4	15 h	2.3:2.2:1.0:(−)[f]	80
d	1-naphthyl	24 h	6.6:1.8:1.0:(−)[e]	35

^{[a], [b], [c]}Same as for Table 1.

^[d]Compounds 7 and 4 are identical in this case.

^[e]Other products — if at all detectable — were present in insignificant amounts.

^[f]A trace amount of Ph₂IOTf, too small to be measured accurately and perhaps arising as per ref. 1, was apparent by ¹H-NMR and MS.

Table 3.

Metathesis reactions of iodonium triflates 7.

entry	Ar	Ar’	ratio 6:4:7:(other)[b]	yield[c]
a[d]	Ph	Ph	12.9[e]:1.0:(−)[f]	65
b[d]	“	4-MeC6H4	1.0:1.3:0.0:(−)[f]	73
c[d]	“	4-MeOC6H4	60.1:35.1:1.0:(−)[f]	76
d[d]	“	1-naphthyl	33.3:5.9:1.0:(−)[f]	59
e[d]	4-MeOC6H4	Ph	11.5:8.2:1.0:(1.0)[g]	60
f[d]	“	4-MeC6H4	1.3:1.0:0.0:(−)[f]	79
g[d]	“	4-MeOC6H4	6.9[e]:1.0:(−)[f]	89
h[d]	“	1-naphthyl	8.4:1.2:1.2:(1.0+0.6)[h]	64
i[d]	mesityl	4-MeC6H4	1.0:7.0:0.0:(−)[f]	26
j[d]	“	4-MeOC6H4	1.0:13.0:0.0:(−)[f]	37
k[d]	“	1-naphthyl	1.0:8.4:8.4:(−)[i]	27

^{[a], [b], [c]}: same as in Table 1.

^[d]Reaction time: 12 h.

^[e]Compounds 7 and 4 are identical in this case.

^[f]Other products — if at all detectable — were present in insignificant amounts.

^[g]This other product was bis-(4-anisyl)iodonium triflate (¹H-NMR, MS), perhaps arising as per ref. 1.

^[h]The major other product was 1-naphthyl-4-nitrophenyl-iodonium triflate; the minor other product was bis-(4-anisyl)iodonium triflate (¹H-NMR, MS).

^[i]A trace amount of 1-naphthyl-4-nitrophenyl-iodonium triflate, too small to be measured accurately, was apparent by ¹H-NMR and MS.

Table 4.

Metathesis reactions of thienyliodonium triflate 8.

entry	Ar’	ratio 9:10:4:8:(other)[b]	yield[c]
a[d]	Ph	1.0:trace:15.1:0.0:(−)[e]	40
b[d]	4-MeC6H4	1.0:trace:7.0:0.0:(−)[e]	66
c[d]	4-MeOC6H4	1.3:trace:1.0:trace:(−)[e]	91
d[d]	1-naphthyl	1.0:trace:4.1:trace:(−)[e]	38

^{[a], [b], [c]}: same as in Table 1.

^[d]reaction time: 12h.

^[e]Other products — if at all detectable — were present in insignificant amounts.

Table 5.

Calculated charge on the I-atom of various aryl iodides.

substance	charge on the I-atom[a]
4-iodotoluene	+ 0.068
iodobenzene	+ 0.068
1-iodonaphthalene	+ 0.070
4-iodoanisole	+ 0.071
4-bromo-1-iodobenzene	+ 0.088
methyl 4-iodobenzoate	+ 0.093
2-chloro-5-iodopyridine	+ 0.120
4-nitro-1-iodobenzene	+ 0.120
5-iodo-1-methyluracil	+ 0.162
2-iodothiophene	+ 0.166

^[a]Geometric mean (ref. 15) of values (units of electron charge, e) calculated by MNDO, MNDO-d, AM1 and PM3 after geometry optimization (ref. 16).

Scheme 2.

Mechanistic hypothesis for the iodonium metathesis reaction.

In summary, iodonium metathesis reactions are feasible. This engenders numerous opportunities that are actively being researched. Pertinent results will be disclosed in due time.