Tellurium(II)/Tellurium(III)‐Catalyzed Cross‐Dehydrogenative C−N Bond Formation

Abstract The TeII/TeIII‐catalyzed dehydrogenative C−H phenothiazination of challenging phenols featuring electron‐withdrawing substituents under mild aerobic conditions and with high yields is described. These unexpected TeII/TeIII radical catalytic properties were characterized by cyclic voltammetry, EPR spectroscopy, kinetic experiments, and DFT calculations.

Therefore, it possesses a low oxidation potential and various stable oxidation states, while forming relatively stable CÀTe bonds. [9] In this context, recent developments in metal catalysis highlight the use of ligands which actively take part in the reaction via electron transfer events. These ligands are socalled "redox non-innocent" ligands. [10] Non-innocent ligands act in several ways. For example, they can act as electron reservoir. This allows strategic metals to bypass unfavorable oxidation states while maintaining catalytic activity. They can also actively take part in bond breaking/forming events via hydrogen abstraction. [11,12] Thus, we propose herein an unprecedented Te II /Te III catalysis approach containing a bidentate, nitrogen-bridged redox non-innocent ligand (Scheme 1), in the aim of unlocking new catalytic properties.
In order to proceed with this objective, selenium and tellurium azine derivatives were targeted as prospective catalysts. PSeZH (phenoselenazine, X = Se) was easily accessed with a simple two-step procedure from the literature (Scheme 2 a). [13] PTeZH (phenotellurazine, X = Te), however, proved slightly more challenging. After testing and optimizing various retrosynthetic approaches, we eventually estab-lished a route through 2,2'-diiododiphenylamine (Scheme 2 b). Indeed, the latter substance could smoothly react with elemental tellurium under simple basic conditions to afford PTeZH in 67 % yield of isolated product (Scheme 2 b).
This new method was found reliable and scalable, affording an easy and serviceable route to the target tellurium metalloid-based heterocyclic catalyst. Once with the PSeZH and PTeZH catalyst candidates in hand, we set out to optimize the tellurium-catalyzed cross-dehydrogenative phenothiazination of unprecedented and typically challenging phenols featuring electron-withdrawing substituents (7 < pK a < 10). We finally selected and optimized a mild O 2 -mediated basic oxidation method, for a limited reaction time of 3 hours. [14] K 2 HPO 4 was found to be an optimal base in comparison to K 3 PO 4 , NaHCO 3 , or AcOK, although good results were also obtained with K 2 CO 3 . [14] Importantly, the Te catalyst (PTeZH) was found significantly superior to the Se catalyst candidate (PSeZH). Moreover, while 5 mol % of PTeZH catalyst loading provided encouraging results, 10 mol % was found optimal. The optimized tellurium-catalyzed conditions are shown in Scheme 3 (product 3 aa, 97 % isolated).
This Te-catalyzed reaction was found to tolerate a number of unprecedentedly acidic phenols (pK a down to 7.5, 3 ea, 73 %), with high yields. Challenging functional groups such as ketones, a pyridine, and an aldehyde were moreover well tolerated (3 ha, 3 hb, 3 he, 3 pa, 3 oa, 3 qa). Importantly, control experiments omitting the Te catalyst systematically led to very poor conversions, thus highlighting by contrast the strong catalytic role of the Te II organometalloid PTeZH complex (Scheme 3).
Moreover, the non-catalyzed reaction (absence of PTeZH catalyst) does not perform much better at longer reaction times.
For example, the uncatalyzed method afforded 3 aa in only 66 % yield after 24 h reaction time, versus 97 % in 3 h for the optimized Te-catalyzed conditions. In some cases, the uncatalyzed reaction yielded only traces of the expected coupling product (3 ab, 3 bb, 3 bf, 3 hb < 5 %). Thus, the use of these highly sustainable O 2 -based reaction conditions requires the presence of the PTeZH catalyst. Finally, this Te II -catalyzed method allowed the straightforward scale-up of the reaction without any loss of yield (3 ia, pK a = 7.8, 86 %), therefore demonstrating its robustness.
In order to understand this remarkable catalytic effect, the cyclic voltammetry (CV) plots of all four chalcogen congeners are presented in Figure 1. The first three congeners (X = O, S, Se) were found to have a similar oxidation potential (E8 (1/2ox) =+ 0.24, + 0.22, + 0.24 V, respectively). In contrast, the oxidation potential of the largest congener (PTeZH, X = Te) deviates significantly from the other three chalcogens, at only + 0.08 V.  The simulated spectra were obtained with EasySpin, [15] via the cwEPR GUI plugin, using the simulation parameters listed in Table S3 (see  SI). [16] This oxidation potential difference for tellurium has important consequences for its reactivity, as will be discussed below. Next, the radical character of each oxidized chalcogenazine congener was investigated by electron paramagnetic resonance (EPR) spectroscopy. The radical species were generated by bubbling air through a solution of POZH, PSZH, PSeZH, and PTeZH in [D 6 ]benzene at room temperature. The corresponding EPR profiles are shown in Figure 2. Interestingly, while we had expected similar N-centered neutral radicals [17] for all four investigated chalcogens, only the first three (X = O, S, Se) showed an EPR signal that is compatible with an N-centered neutral radical species PXZC (Figure 2). In contrast, phenotellurazine (PTeZH) delivered a very different EPR signal, which, according to simulations and supporting DFT property calculations, corresponds to a (protonated) radical cation species: PTeZHC + . This difference between tellurium and the other chalcogens presumably arises from a lower oxidation potential (+ 0.08 V) and subsequent weaker acidity of PTeZHC + compared to the other three chalcomers (X = O, S, Se, E8 (1/2ox) =+ 0.22-0.24 V). Indeed, one-electron oxidation of POZH, PSZH, and PSeZH apparently leads to strongly acidified PXZHC + radical cations, which spontaneously deprotonate at nitrogen to the corresponding persistent neutral radicals. In contrast, the (NPA) charge and spin density of PTeZHC + are significantly shifted from N to Te (see SI, Table S4).
Next, we measured the relative initial rates of conversion (5 min reaction time) of the various chalcogenazines as Nsubstrates, with common phenothiazine PSZH (X = S) as the reference (k rel = 1). In those four parallel experiments, POZH was found to be the fastest azine (k rel = k X /k S = 4.4), and PTeZH the slowest (k rel = 0.7, Scheme 4 a). In a competition set-up however (Scheme 2 b), POZH becomes 20 times faster, while PTeZH becomes circa 100 times-two orders of magnitude-slower than competing PSZH (k rel = 0.01). Moreover, in the latter case, the PSZH initial conversion rate has been multiplied by 4 in comparison to the non-catalyzed reaction (absence of PTeZH).
PTeZH is a good catalyst in this reaction because it combines a significantly lower oxidation potential compared to PSZH (+ 0.08 V versus + 0.22 V, respectively), such that it must oxidize first, with a very reactive neutral N-centered radical (PTeZC). Indeed, the H-atom transfer (HAT) process was calculated to be very favorable from PSZH to PTeZC. The latter species therefore serves as radical catalyst [18] which is generated from the in situ deprotonation of PTeZHC + , facilitated by the basic reaction conditions and/or peroxide anions resulting from O 2 reduction (Scheme 5, see also SI). This process would thus increase the rate of formation as well as the concentration of the key persistent PSZC neutral radical species, which is a known intermediate in the dehydrogenative phenothiazination reaction. [17] This favorable HAT process would therefore lead to a reaction acceleration. Reoxidation of the PTeZH Te II catalyst would then occur again towards the Te III PTeZHC + intermediate, thus closing the catalytic cycle.
In conclusion, we have demonstrated that a Te II organometallic complex could catalyze the dehydrogenative CÀH phenothiazination of challenging phenols bearing electron-withdrawing substituents, with acidities as low as pK a = 7.5 (3 ea, 73 %). In all cases, the absence of Te II catalyst leads to dramatically lower conversions. This unexpected catalytic effect essentially arises from a combination of two important properties: a lower oxidation potential of the PTeZH catalyst towards the PTeZHC + radical cation and a significantly higher spin density at the tellurium center compared to the sulfurbased substrates. It is thus probable that PTeZH will find further applications as radical catalyst [18,19] for the development of innovative (radical-catalyzed) cross-dehydrogenative couplings.