Early reports stated that Au was a catalyst of choice for the BOR because it would yield a near complete faradaic efficiency. However, it has recently been suggested that gold could yield to some extent the heterogeneous hydrolysis of BH, therefore lowering the electron count per BH, especially at low potential. Actually, the blur will exist regarding the BOR mechanism on Au as long as no physical proof regarding the reaction intermediates is not put forward. In that frame, in situ physical techniques like FTIR exhibit some interest to study the BOR. Consequently, in situ infrared reflectance spectroscopy measurements (SPAIRS technique) have been performed in 1 M NaOH/1 M NaBH(4) on a gold electrode with the aim to detect the intermediate species. We monitored several bands in B-H (nu ∼ 1180, 1080 and 972 cm(-1)) and B-O bond regions (nu = 1325 and ∼1425 cm(-1)), which appear sequentially as a function of the electrode polarization. These absorption bands are assigned to BH(3), BH(2) and BO species. At the light of the experimental results, possible initial elementary steps of the BOR on gold electrode have been proposed and discussed according to the relevant literature data.