DETERMINATION OF LOW ARSENIC CONCENTRATIONS IN LIGHT GASOLINE BY GFAAS TECHNIQUE

Osnovu postrojenja za izomerizaciju lakog benzina u rafinerijama nafte čine reaktori koji su napunjeni platinskim katalizatorom. Glavni katalitički otrovi za ovu vrstu katalizatora su već u malim količinama voda, sumpor, oksigenati i dušik. Međutim, negativni učinak na aktivnost katalizatora imaju i metali. Arsen, kao jedan od metalnih otrova, štetan je već u μg/l te ga je potrebno precizno detektirati kako bi se skupi platinski katalizator mogao pravodobno zaštititi. U radu je prikazan postupak određivanja niskih koncentracija arsena u lakom benzinu (red veličine μg/l). Korištena je metoda atomske apsorpcijske spektrometrije uz primjenu grafitne peći (GFAAS). GFAAS tehnika je pogodna za određivanje tragova metala u frakcijama nafte i naftnim produktima i dobra je alternativa razmjerno skupoj spektrometriji masa s induktivno spregnutom plazmom (ICP-MS). Uzorak je stabiliziran miješanjem propisanih volumena uzorka s propan-1-olom i vodenom otopinom nitratne kiseline, pri čemu nastaje stabilna mikroemulzija (jedna faza). Kao modifikator je korištena otopina Pd(NO3)2. Validacija metode uključuje stabilnost mikroemulzije, područje linearnosti, iskorištenje, preciznost, granicu detekcije i granicu kvantifikacije.The main parts of the light naphtha isomerization unit in oil refinery are reactors loaded with platinum base catalyst. Even small amount of water, sulfur, oxygenates and/or nitrogen is poisonous for the catalyst. In addition, the catalyst activity is negatively affected by metals. Arsenic, one of the metal poisons, decreases the catalyst activity in μg/l quantity. Therefore, it’s crucial to detect it precisely in order to protect the expensive platinum based catalyst in time. A procedure for determination of arsenic in light gasoline at μg/l levels by graphite furnace atomic absorption spectrometry (GFAAS) is proposed. GFAAS is good alternative for the determination of trace metals in petroleum fractions and distillates to relatively expensive ICP-MS. Sample stabilization was achieved by mixing appropriate volumes of the sample, propan-1-ol and nitric acid aqueous solution. This mixture resulted in one-phase medium (microemulsion), which was indefinitely stable. Pd(NO3)2 solution was used as modifier. Validation of the method includes microemulsion stability, linearity, yield, precision, limit of detection and limit of quantification