Thioether S-Ligation in a Side-on μ-η²:η²-Peroxo Dicopper(II) Complex

Abstract

[(ANS)Cu^I(CH₃CN)]⁺ (1A), with N₂S_thioether ANS ligand, reacts with O₂ giving a side-on peroxodicopper(II) species [{(ANS)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (1P), ν_O-O = 731 cm⁻¹. 1P is shown to possesses S-thioether ligation, following chemical-spectroscopic comparisons with analogues having all N-ligands or a –S(Ph) group. The finding is a rare occurrence and new for side-on O₂²⁻ binding.

Copper(I)–dioxygen adducts or derived species pertinent to copper protein O₂-binding and activation are currently of considerable interest in bioinorganic chemistry.^1-4 The focus of this paper relates to the discovery and influence of thioether sulfur ligation in copper(I)-dioxygen adducts, a topic which has attracted more recent attention. This is due to the existence of important copper monooxygenases, namely dopamine β-monoxygenase (DβM) and peptidyl glycine α-hydroxylating monooxygenase (PHM), which serve important roles in neurotransmitter or regulatory horomone biosynthesis.⁵ The active-site in these proteins, where substrate binding and copper-dioxygen activation occur, is the so-called Cu_M site, which is ligated by two His and one Met ligands.^5-7 Biochemical and chemical evidence suggests that O₂ binds here and that a Cu^II-superoxo (O₂^·−) dioxygen adduct or a derived Cu^II-hydroperoxo or cupryl (Cu^II-O^·) entity initiates substrate oxidation chemistry.^{5, 8-12} In fact, the observance of an X-ray structure on a protein Cu^II-O₂^·− complex.⁷ biochemical evidence.^{5, 8-10} and theoretical calculations^9,10 that the Cu^II-superoxo species is the one effecting a substrate hydrogen-atom abstraction reaction.

The vast majority of the literature deals with copper(I)/dioxygen chemistry and adducts and reactivity with systems containing all nitrogen ligands, typically bi- tri- or tetradentate entities.^{1-4, 13, 14} However, it is clearly important to develop an understanding of how thioether ligation influences the structure(s), spectroscopy, and reactivity of Cu^I_n-(O₂)-derived species.

Some work in this general area has occurred. A dicopper(I) complex with Met-based ligand was synthesized by Casella and co-workers;^{15, 16} reaction with O₂ led to hydroxylation of a xylyl spacer group. Tolman and co-workers reported Cu^I-O₂ chemistry with a N₂S β-diketiminate ligand, but the S-donor does not coordinate to copper in derived O₂-adduct.¹⁷ Quite recently, Nicholas and co-workers synthesized a bis-imidazole-thioether tridentate N₂S ligand while Cu^I-complex oxidation occurs, they did not observe Cu^I-O₂ adducts.^18-20

Recently, we reported the first copper-dioxygen adduct with thioether ligation, an “end-on” μ-1,2-peroxodicopper(II) complex, each copper ion possessing a tetradentate N₃S ligand.^{21, 22}

As DβM and PHM employ N₂S rather than N₃S protein ligand donors, we were interested to look further, at new tridentate ligands. In this report, we describe copper(I)/O₂ reactivity with ANS (Scheme 1), a linear N₂S tridentate ligand. We also had previously studied copper(I)/O₂ chemistry with the N₃ ligand MeAN (Scheme 1), which readily forms a side-on binuclear peroxo complex [{(MeAN)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (2P), λ_max = 360 nm (ε=22,000 M⁻¹cm⁻¹) and 540 nm (ε = 2,500 M⁻¹cm⁻¹).²³ Our hypothesis was that an analogue, ANS, in which one alkylamine arm of MeAN is replaced by a thioether donor, might exhibit related chemistry. This turns out to be true, as we described here.

Scheme 1.

Ligands and their Cu₂O₂ adducts.

The copper(I) complex [(ANS)Cu^I(CH₃CN)]⁺ (1A) (as B(C₆F₅)₄⁻ salt) was generated from [Cu^I(CH₃CN)₄]B(C₆F₅)₄ addition to ANS, and characterized as a greenish white air-sensitive solid. [(ANS)Cu^II(Cl)₂] (1B) was also synthesized. Both of these Cu^I and Cu^II complexes were characterized by X-ray diffraction (Fig. 1). Noteably, the S(thioether) coordination to copper ion occurs in both Cu^I and Cu^II complexes.

Fig. 1.

ORTEP diagram view (50% probability ellipsoids) of 1A and 1B. Left: X-ray structure of 1A: Cu1-N1, 2.056 Å; Cu1-N2, 2.122 Å; Cu1-N3, 1.935 Å; Cu1-S1, 2.292 Å. Right: X-ray structure of 1B: Cu1-N1, 2.071 Å; Cu1-N2, 2.136 Å; Cu1-S1, 2.387 Å, τ = 0.61.²⁴

In order to describe the electronic environment of the ligand copper coordination complexes, while also comparing the environment of sulfur containing ANS _(N2S) with MeAN (N₃), CO adducts of [(ANS)Cu^I(CH₃CN)]⁺ (1A) and [(MeAN)Cu^I]⁺ (2A) were generated and the electrochemistry via cyclic voltammetry of 1A and 2A were evaluated (Table 1). The IR spectrum of [(ANS)Cu^I-CO]⁺ (1-CO) shows that ν_CO = 2092 cm⁻¹, significantly higher than that observed for [(MeAN)Cu^I-CO]⁺ (2-CO, 2079 cm⁻¹). It is noteable that ν_CO for the Cu_M sites in PHM and DβM (Table 1)^{25, 26} are essentially the same as that of [(ANS)Cu^I-CO]⁺ (1-CO), indicating that 1A possesses a chemical environment reasonably mimicking that of protein active sites.

Table 1.

IR data for Cu^I-CO adducts and E_1/2 values of Cu^I complexes

	MeAN	ANS	PHM[a]	DβM[a]
Cu-CO νCO (cm−1), THF solution	2079	2092	2093	2089
E1/2 (mV) (vs. Fe(Cp)2+/0) DMF solution under Ar	−195	−180

^[a]data for the protein Cu_M site with His₂Met coordination.

As for the ligand comparison, the results clearly indicate that MeAN is a better donor to copper(I) which then back-donates to its CO ligand to a greater extent, lowering ν_CO. The electrochemical behavior for 1A vs 2A is consistent with findings for CO ligation and ν_CO values, as the strong donor ligand MeAN better stabilizes the higher oxidation state, giving a more negative E_1/2 value (Table 1).

Oxygenation of 1A at −94 °C in acetone gives a dark orange species, λ_max = 375 nm (ε =10,000 cm⁻¹M⁻¹) and 495 nm (ε =1,000 cm⁻¹M⁻¹) (Fig. 2), features consistent with those of an μ-η²:η²-peroxo-dicopper(II) species.¹ Thus, the complex is formulated as [{(ANS)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (1P) Confirmation derives from resonance Raman spectroscopic data with λ_ex=379.5 nm employing either ¹⁶O₂ or ¹⁸O₂ for complex generation: ν_Cu-Cu = 273 cm⁻¹ (Fig. 3A) and ν_O-O = 731 cm⁻¹ (Δν(¹⁸O₂) = −40 cm⁻¹) (Fig. 3B). These vibrations are characteristic of a μ-η²:η²-peroxo-dicopper(II) species.^1,27

Fig. 2.

UV-Vis changes upon reaction of O₂ with [(ANS)Cu(CH₃CN)]⁺ (1A) (yellow) in Acetone at −94 °C giving [{(ANS)Cu^II}₂(μ-η²:η²-O₂)]²⁺ (1P) (red); warming to RT leads to a decomposition product (green).

Fig. 3.

rR spectra of [{(ANS)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (1P) in acetone with ¹⁶O₂ (red) and ¹⁸O₂ isotopic substitution (blue) in the region of ν_Cu-Cu (A) and ν_O-O (B) (λ_ex = 379.5 nm, 77K, 5mW power).

These values are also similar to those observed for [{(MeAN)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (2P) (ν_O-O = 721 cm⁻¹ and ν_Cu-Cu = 268 cm⁻¹),²³ however the higher ν_O-O vibration frequency of [{(ANS)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (1P) (731 cm⁻¹), compared to that for MeAN complex 2P (721 cm⁻¹), indicates that the O-O bond in 1P is stronger. Compared to MeAN, ANS is a weaker donor to copper resulting in increased peroxo π*_σ donation to Cu and less backbonding into the peroxo σ* resulting in a stronger peroxo O-O bond.²⁸

The supposition that S(thioether) ligation occurs in [{(ANS)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (1P) derives from observations discussed above and the following argument. If the thioether donor did not bind to copper in 1P and was dangling, (as seen in Tolman’s N₂S system),¹⁷ the remainining “bidentate amine” ligand (N₂ portion of ANS) would result in formation of a bis-μ-oxo dicopper(III) species. As is already known, the copper(I) complex with N,N,N’,N’-tetramethyl-(1,3)-propanediamine (TMPD), [(TMPD)Cu^I(CH₃CN)]⁺ (4A), which was described by Stack and co-workers,²⁹ does exactly this (Scheme 1); ν_Cu-O = 609 cm⁻¹: Δν(¹⁸O₂) = −28 cm⁻¹. We also note that for a copper(I) complex with TMPD, E_1/2 is very negative (−393 mV, vs. Fe(Cp)₂^+/0), consistent with it’s ability to form a bis μ-oxo complex, stabilizing a high oxidation state copper(III) environment.

To provide further support for thioether ligation in 1P, we separately synthesized the ligand ANSPh (Scheme 1) and its Cu(I) complex [(ANSPh)Cu^I(CH₃CN)]⁺ (3A). The large phenyl group of the ANSPh ligand, directly on the sulfur atom, inhibits sulfur coordination to copper ion in a Cu₂/O₂ adduct. This in fact is the case. The reaction of dioxygen with 3A at −94 °C in acetone gives a yellow bis-μ-oxo dicopper(III) species (3O), λ_max = 409 nm (ε = 9,000 cm⁻¹M⁻¹) similar to typical bis-μ-oxo dicopper(III) species. Resonance Raman data for 3O (Supporting Material) show only a characteristic core vibration, ν_Cu-O = 606 cm⁻¹ (Δν (¹⁸O₂) = −28 cm⁻¹), suggesting a structure just like Stack’s TMPD complex (Scheme 1)²⁹ and indicating the –SPh moiety does not coordinate. By inference, these results indicate that the S(thioether) donor in ANS must be able to coordinate to Cu(II) in [{(ANS)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (1P).

In summary, a side-on μ-η²:η²-peroxodicopper(II) species has been generated from the oxygenation of LCu(I) complex with N₂S thioether ligand ANS. By comparison to the nature of the Cu₂O₂ species generated with closely related ligands where either the S(thioether) is unable to coordinate or is absent (a bis-μ-oxo dicopper(III) complex), we can conclude that coordination of the S(thioether) donor within [{(ANS)Cu^II}₂(μ-η²:η²-O₂²⁻)]²⁺ (1P) occurs. This is the first case of thioether coordination within a μ-η²:η²-peroxo-dicopper(II) structural type. This advance in copper-dioxygen chemistry further indicates that it is possible to study systems with S(thioether) ligation. Future investigations will focus on the effect of S(thioether) on reactivity of derived copper-dioxygen complexes,³⁰ and on the design and study of systems where only one copper ion is present.