Selective synthesis of methyl dithienyl-glycolates

An efficient selective synthesis of methyl dithienyl-glycolates has been developed. The interest of this two steps protocol resides in the possibility of synthesized either methyl 2,2-dithienyl glycolate – the target intermediate for the preparation of anticholinergic agents – or its regio-isomer methyl 2,3-dithienyl glycolate – the most critical precursor of anticholinergic drug impurity.


Results and Discussion
The difficult separation of 1a and 1b leads to pollute drugs, due to the formation of pharmacological regioisomer impurities. 31Thus, to find an alternative method to selectively obtain either 1a or 1b, we decided to modulate the reactivity of the thienyl anion species by changing the nature of the counter cation.To that aim 2-thienyl lithium and 3-thienyl lithium (derived, respectively, from lithiation of 2-bromothiophene 2a and 3bromothiophene 2b), were reacted with oxalate 3 to obtain methyl 2-oxo-2-(thiophen-2-yl)acetate 4a and methyl 2-oxo-2-(thiophen-3-yl)acetate 4b.The latter compounds are intermediates for methyl dithienylglycolate 1a and 1b production.Preliminary runs (entries 1-2, Table 1), evidenced that a mixture of regioisomers 4a and 4b could be obtained when oxalate 3 was added to 3-thienyl lithium: in according to the anion equilibration, the longer the metalation time of 2b, the higher the yield of 4a.A: Base was dropped in 5 min to a solution of 2b and stirred for 15 min before addition of a solution of 3; B: Base was dropped in 5 min to a solution of 2b and stirred for 30 min before addition of a solution of 3; C: 2b was added to a solution of BuLi and 3 in THF; D: Base was added to a solution of 2b and 3.
In fact, a complete regio-selectivity in favor of 4b was reached by adding 2b to a cold mixture of base (n-BuLi) and 3 (entry 4).Furthermore, the best yield of 4b (66%, entry 4) was achieved by adding the strong, nonhindered base n-BuLi at -78 °C, to a mixture of 2b and a slightly excess of 3 in THF as the solvent.Other solvents (entries 5-6), gave worst results while, the use of non-nucleophilic base LDA, resulted in the formation of a series of by-products (entry 7).3][34][35][36] This latter, in turn, evolves in the 4c product in presence of 3 (entry 1, Table 2).Furthermore, 1a reacted under the best reaction conditions found for 1b, giving 4a in good yield (entry 2); we were able to increase the yield of 4a by generating the stable lithium anion at the thiophene C-2 position and then adding oxalate 3 to the reaction mixture (entry 3).Similar behavior (Scheme 3) was found when the 2-thienyl anion specie, generated by metalation of 2a at low temperature, was then trapped using as electrophiles 4a or 4b: under these reaction conditions, compounds 1a (derived from intermediate 4a, path a.) and 1b (derived from 4b, path b.) were isolated as pure isomers and fully characterized.On the contrary, our attempts to isolate 1b or its regio-isomer methyl 3,3-dithienylglycolate by 3bromothiophene 2b lithium halogen exchange in the presence of either oxo-acetate 4b or 4a, gave a mixture of degradation compounds (path c.).Comparison between 1 H NMR spectra of compounds 1a and 1b (Table 3), exhibited a slightly highfield chemical shifts for almost all 1a signals (only 1a and 1b H3 protons resonate together at ~7.19 ppm); furthermore, 1b 1 H NMR showed two additional signals at 7.39 ppm (H7) and 7.10 ppm (H9).Even for 13 C NMR spectra, all the peaks of 1a and b are quite good shifted; the major difference arise in the presence on substrate 1b of both the isolated signal at 142.5 ppm (C6) and 123.2 ppm (C7).

Conclusions
We have described a complete regio-selective protocol for methyl 2,2-dithienylglycolate 1a and methyl 2,3dithienylglycolate 1b synthesis.Depending on both the nature of the bromothiophene derivative used and the condensation conditions, it was possible to obtain either 1a -the key starting material in the preparation of important anticholinergic agents -or 1b, precursor of pharmacological impurities.By this way 1b was fully characterized, giving the characteristic signals that permit its differentiation for the target compound.

Experimental Section
General.All available chemicals and solvents were purchased from commercial sources and were used without any further purification.Thin layer chromatography (TLC) was performed using 0.25 mm silica gel precoated plates Si 60-F254 (Merck) visualized by UV-254 light and CAM staining.Purification by flash column chromatography (FCC) was conducted by using silica gel Si 60, 230-400 mesh, 0.040-0.063mm (Merck).
Melting points were determined on a Büchi B450 apparatus and are corrected. 1H and 13 C NMR spectra were recorded on a Bruker Fourier 300 (recorded at: 300.13 MHz for 1 H; 75.00 MHz for 13 C) or Bruker Avance Spectrometer (recorded at: 400.13 MHz for 1 H; 100.62 MHz for 13 C); chemical shifts are indicated in ppm downfield from TMS, using the residual proton (CHCl3 7.28 ppm; acetone 2.05 ppm) and carbon (CDCl3 77.0 ppm; acetone 207.1 and 30.9 ppm) solvent resonances as internal reference.Coupling constants values J are given in Hz.

Table 1 .
Reactivity of bromothiophene 2b a b Isolated yields.

Table 2 .
Reactivity of bromothiophene 2a. a Base was dropped in 5 min to a solution of 2a and stirred for 30 min before addition of a solution of 3.
b Isolated yields.A: Base was added to the mixture of 2a and 3; B: