The Heck reaction of allylic alcohols catalysed by an N-heterocyclic carbene-Pd(ii) complex and toxicity of the ligand precursor for the marine benthic copepod Amphiascoides atopus

The palladium-catalysed reaction of aryl halides and allylic alcohols is an attractive method for obtaining α,β-unsaturated aldehydes and ketones, which represent key intermediates in organic synthesis. In this context, a 1,2,3-triazol-5-ylidene (aNHC)-based palladium(ii) complex formed in situ has been found to be a selective catalyst for the syntheses of building blocks from the corresponding aryl halides and allylic alcohols, with yields ranging from 50% to 90%. The lack of toxic effects of the ligand precursor (1,2,3-triazolium salt) of the palladium(ii) complex for the harpacticoid copepod Amphiascoides atopus allowed us to contrast the efficiency of the catalytic system with the potential impact of the principal waste chemical in global aquatic ecosystems, which has not been previously addressed.


Introduction
Easy access to 1,2,3-triazoles via a simple copper-catalysed azide-alkyne cycloaddition reaction (CuAAC) has led to the development of a "post-click chemistry" strategy to obtain 1,2,3triazolium salts by selective alkylation at the N3 position of the triazole ring. 1 1,2,3-Triazolium salts represent an attractive group of chemical compounds given that they can be used as ionic liquids (ILs), 2 as hosts in anion recognition, as components of molecular machines and supramolecular assemblies 3 or as precursors of abnormal (mesoionic) 4 N-heterocyclic carbenes (aNHCs) as a consequence of their high stability. 5 As shown in Scheme 1, deprotonation and metalation of 1,2,3-triazolium salts give a 1,2,3-triazol-5-ylidene complex (aNHC), which may be a precursor for an efficient catalyst for organic reactions thanks to its unique electronic features and donor properties. 6 Recently, 1,2,3-triazolium iodide salts have been used as efficient precursors for 1,2,3-triazol-5-ylidene ligands 7 (aNHCs) for palladium-catalysed Suzuki-Miyaaura 8 and Heck-Mizoroki cross-coupling reactions. 9 In those cases, the aNHC and palladium(II) complex are formed in situ from triazolium salts under mild reaction conditions. With these relevant protocols, we focused on the use of a similar system in the synthesis of a-b unsaturated aldehydes and ketones via the Heck coupling of aryl halides and allylic alcohols, with the idea of developing an efficient and selective protocol using 1,2,3triazol-5-ylidene (aNHC)-based palladium(II) complexes.
The high thermal stability and low vapour pressure make 1,2,3triazolium salts attractive for industrial eco-friendly processes. It is assumed that their structural characteristics do not cause air pollution or damage to occupational health. However, their properties, such as resistance to photodegradation, water solubility and stability, suggest that they may be a threat to aquatic ecosystems due to bioaccumulation. 10 For that reason, we considered it necessary to compare the efficiency of the catalytic system (aNHC-based palladium(II) complexes) with the impact of Scheme 1 Synthesis and uses of 1,2,3-triazolium salt. the 1,2,3-triazolium salt precursor in aquatic ecosystems. 11 Thus, we evaluated its acute toxicity in the harpacticoid copepod Amphiascoides atopus, considering that it will be among the principal waste chemicals eliminated aer the catalytic process.
Sediments in aquatic ecosystems are rich in small crustaceans, including benthic invertebrates, which are effective indicators of impacts at higher levels of biological organization given their importance to overall ecosystem structure and function. 12 In particular, marine copepods belonging to the order Harpacticoida, one of the most abundant benthic invertebrates and an important food source for macroinvertebrates and sh, are suitable for use in tests that rapidly assess the acute, sublethal, or chronic effects of contaminants. 13 To our knowledge, there are no previous studies testing the toxicity of 1,2,3-triazolium salts in copepods or in other crustaceans.
Then, we were interested in evaluating the catalytic properties of aNHC-based palladium(II) complexes formed in situ from 4a and 4b in the Heck reaction between bromo pyridine 5a and allylic alcohol 6a to give ketone 7a. Aer various solvent systems were examined, we concluded that DMF was the best solvent ( Table 1, entries 1-6). A higher reaction yield was obtained in the presence of Pd(OAc) 2 and NaOAc ( Table 1, entry 1). Furthermore, the yield was lowered when the reaction was carried out in the presence of PdCl 2 ( Table 1, entries 2 and 4) or Scheme 2 Synthesis of 1,3,4-trisubstituted-1,2,3-triazolium iodide salts 4a and 4b.  when t-BuOK or K 2 CO 3 was used as the base (Table 1, entries 7 and 8). When the reaction was performed in THF under reux, only the starting material was observed in the 1 H NMR spectrum of the reaction mixture (Table 1, entry 5). Product 7a was obtained in 55% and 50% yields at 50 C (Table 1, entry 9), and it was not observed at room temperature (Table 1, entry 10) in the presence of 4a or 4b. When the reaction was carried out with 5 mol% 4a or 4b and 5 mol% Pd(OAc) 2 , the yield did not increase considerably (Table 1, entry 11). Notably, a complex reaction mixture was obtained in the absence of triazolium salts 4a or 4b.
Using the best reaction conditions, we next examined the application of a 1,2,3-triazolium iodide and palladium salt system from the 4a aNHC precursor for the cross-couplings of a variety of aryl bromides with primary or secondary allylic alcohols ( Table 2). The results indicate that the aNHC-based palladium(II) complex formed in situ constituted an efficient and selective catalyst system for the synthesis of saturated aldehydes 7b-7e and ketones 7f-7r in good yields. In the case of tertiary allylic alcohols, the only nal products were a-b unsaturated alcohols (7s-7v, Table 2). Unfortunately, the reaction does not evolve successfully when it was carried with aryl chlorides. One of the principal drawbacks of the reaction studied here is the formation of mixtures with both saturated and unsaturated carbonyl compounds as principal products, 14,15 which was not observed under our reaction conditions. Saturated carbonyl compounds were obtained as the only product, and unsaturated products (a-b unsaturated carbonyl compounds) were not observed when crude products were analysed by NMR.
Regarding the reaction mechanism, we propose the direct metalation of 1,2,3-triazolium salt 4a with Pd(AcO) 2 in the presence of NaOAc to obtain in situ 1,2,3-triazol-5-ylidene (aNHC)-based palladium(II) complexes 8 (Scheme 3a), which should be the principal precursor to the catalytic species in the reaction. The catalytic process starts with the oxidative addition of bromo pyridine 5a, followed by coordination and regioselective alkene insertion (allylic alcohol 6a). Finally, Pd-H belimination involving carbinol hydrogen furnished desired carbonyl compound 7a. 16 Toxicity of 1,2,3-triazolium salt 4a Then, we used the harpacticoid copepod Amphiascoides atopus as a model to evaluate the toxicity (LC50 at 24, 48, 72 and 96 h) of 1,2,3-triazolium salt 4a. The results obtained were based on the mortality of adult copepods, which varied signicantly among triazolium salt concentrations at every exposure time (Fig. 1 There was practically no mortality in unexposed (control) organisms and those exposed to 30 mg L À1 4a. Some mortality was observed at higher concentrations ( Fig. 1); however, compared to the control, signicant differences occurred above 250 mg L À1 at 24 h, 200 mg L À1 at 48 h, 150 mg L À1 at 72 h, and 100 mg L À1 at 96 h (all P < 0.05). The estimated LC50 values were 250.4 mg L À1 (CI 215.4 to 290.5 mg L À1 ) at 24 h, 173.9 mg L À1 (CI 150.4 to 196.4 mg L À1 ) at 48 h, 155 mg L À1 (CI 135.6 to 173.9 mg L À1 ) at 72 h, and 111.5 mg L À1 (CI 93.4 to 129.8 mg L À1 ) at 96 h. From these results and following the Scheme 3 (a) Proposed structure of 1,2,3-triazol-5-ylidene (aNHC)based palladium(II) complex 8. (b) Proposed mechanism for synthesis of a-b unsaturated aldehydes and ketones via the Heck coupling of aryl halide 5a and allylic alcohol 6a.  The lack of toxic effect of triazolium salt 4a tested herein could be attributed to its hydrophilic properties, which limit its biocompatibility and adsorption onto or intercalation into the cellular membrane. 19 Our results are based only on the mortality of adult copepods. This may be a shortcoming considering that early developmental stages of copepods as well as sublethal endpoints are more sensitive indicators of the toxicity of contaminants. 20 Additionally, further studies are required to test the effect of the triazolium salt used in the present work on other aquatic organisms, including marine and freshwater species. However, it is noteworthy that there are no previous studies testing the toxicity of triazolium salts in marine eukaryotes or copepods because we consider that the LC50 values from toxicity bioassays described in the present work are fundamental for future more ecologically relevant studies. 21

Conclusions
In summary, a 1,2,3-triazol-5-ylidene (aNHC)-based palladium(II) complex, formed in situ from a 1,2,3-triazolium salt, is an efficient catalytic system for the formation of C-C bonds (Heck coupling) in the selective synthesis of a-b unsaturated aldehydes and ketones from the corresponding aryl halides and allylic alcohols. The efficiency of this catalytic system is reinforced by the null toxicity of the catalytic precursor (1,2,3-triazolium salt) in the marine benthic copepod Amphiascoides atopus. This result, in addition to the high thermal stability and low vapour pressure of the catalytic precursor, allows a glimpse of the low impact in air and in aquatic ecosystems resulting from one of the main waste chemicals produced in the catalytic process studied here.

General methods
All reagents were purchased from Aldrich Chemical Co and used without further purication unless stated otherwise. Yields refer to the chromatographically and spectroscopically ( 1 H and 13 C) homogeneous materials. The organic reactions were monitored by TLC carried out on 0.25 mm E. Merck silica gel plates. The developed TLC plates were visualised under a short-wave UV lamp or by heating aer they were dipped in Ce(SO 4 ) 2 . Flash column chromatography (FCC) was performed using silica gel (230-400) and employed a solvent polarity correlated with the TLC mobility. NMR experiments were conducted on a Varian 300 and Bruker 500 MHz instruments in CDCl 3 (99.9% D) and CD 3 OD (99.8% D) as solvents; the chemical shis (d) were referenced to CHCl 3 (7.26 ppm 1 H, 77.00 ppm 13 C), CD 3 OD (4.87 ppm 1 H, 49.00 ppm 13 C), or TMS (0.00 ppm).
The chemical shis are reported in parts per million (ppm).

Toxicity test
Copepods (A. atopus) were obtained from a stock maintained in laboratory culture in a 15 L ask with ltered natural seawater (35&) under natural conditions of photoperiod (typically, 12 h light/12 h dark) and temperature (28 AE 0.5 C), and fed with microalgae diet. 22 To choose the test concentrations a range-nding test was performed. The denitive tests were performed in six-well culture plates. Ten copepods (ovigerous females) were randomly transferred to each well containing 5 mL of ltered seawater (35&) at 28 AE 0.5 C. Copepods were not fed and exposed to increasing concentrations of 1,3,4-trisubstituted-1,2,3-triazolium iodide salts 4a: 0, 30, 50, 100, 150, 200, 250, 300, 350, 450 and 550 mg L À1 for 96 h. Experiments were set up with six replicates. Mortality of copepods was checked under a stereomicroscope every 24 h. Copepods were considered dead if they did not show movement of appendages and did not show any reaction when transferred to wells with seawater without 1,3,4-trisubstituted-1,2,3-triazolium iodide salts 4a in a period of up to 20 s of observation.

General procedure
Synthesis of 1,2,3-triazoles 3a and 3b. 23 To a solution of alkyne 1 (1 mmol), benzyl halide 2a or 2b (1.2 mmol), NaN 3 (1.2 mmol) in a mixture of THF-H 2 O (20 mL, 1 : 1 v/v), were added sodium ascorbate (0.5 mmol%) and CuSO 4 (0.5 mmol%). The mixture was stirred at 50 C for 48 h. Aer the reaction time, the mixture was cooled to room temperature and THF was eliminated under vacuum. The resulting precipitate of the Cu complex was decomposed by addition of small portions of aqueous ammonia. Then, the mixture was diluted with ethyl acetate (25 mL), washed with NaHCO 3 (7% aqueous, 2 Â 10 mL) and brine (2 Â 10 mL), dried over Na 2 SO 4 and concentrated under vacuum. The crude product was puried through a silica gel column chromatography with a gradient AcOEt-hexane.

General procedure for the catalytic reactions
In all Heck reactions, a DMF solution (5 mL) of 1 mmol of aryl halide 5, 1.0 mmol of allylic alcohol 6, 2.0 mmol of NaOAc, 1% mmol of Pd(OAc) 2 and 1% mmol of 1,2,3-triazolium salt 4 were heated for 6 h in a 90 C silicon oil bath equipped with a condenser system. Aer the reaction time, the mixture was cooled to room temperature and diluted with ethyl acetate (15 mL), washed with brine (3 Â 10 mL), dried over Na 2 SO 4 and concentrated under vacuum. The crude product was puried through a silica gel column chromatography with a gradient AcOEt-hexane.