Dechlorination of Polychlorinated Biphenyls: A Kinetic Study of Removal of PCBs from Mineral Oils P. De Filippis,* M. Scarsella, and F. Pochetti Chemical Engineering Department, University of Rome “La Sapienza”, Via Eudossiana 18, 00184 Rome, Italy A kinetic study was done of the dechlorination of polychlorinated biphenyls (PCBs) eliminated from contaminated dielectric oils by using the potassium poly(ethylene glycolate) (KPEG) process. Experimental runs at laboratory scale showed that the kinetics of the removal reaction was first-order for each PCB present and first-order with respect to the KPEG concentration. The PCB elimination grade was also affected by the KOH/PEG ratio. An exponential correlation was found between the kinetic constant for each congener and its respective gas chromatographic relative retention time. Introduction It is widely recognized that polychlorinated biphenyls (PCBs) are one of mankind’s most dangerous pollut- ants. 1,2 The extremely high toxicity of this class of compounds results from the compounds’ high stability to chemical agents (resistance to degradation) and their propensity to bioaccumulation in living organisms. 3,4 In the past 20 years, many investigations focused on the removal and destruction of PCBs from contaminated products (mainly dielectric oils). Numerous methods were developed for processing these compounds. 5,6 High-efficiency incineration is the recommended tech- nology for destruction of PCB-contaminated products. 7 This technology converts PCBs to HCl, CO 2 , and H 2 O and minimizes the formation of toxic compounds such as dioxins. Unfortunately, high-temperature incinera- tion is expensive and is limited to pure PCBs mixtures and hydrocarbons containing very high concentrations of PCBs. Other problems arise with the handling and transporting of contaminated material and the wasteful destruction of reusable materials. Interest in recovery of reusable materials, and the necessity to treat contaminated products containing low concentrations of PCBs, have spurred renewed interest in the selective destruction or removal of PCBs to yield reusable products. 8-12 Several selective destruction technologies are now available, based on chemical dechlorinations, photochemical degradations, biological treatments, and electrochemical processes. Among nondestructive, industrial-scale decontamina- tion processes, the reaction of PCBs with potassium poly(ethylene glycolate) (KPEG) appears promising. This technology was initially applied by using molten sodium metal dispersed in poly(ethylene glycol)s. 13,14 The reaction is rapid and completely decomposes PCBs. However, this reaction is somewhat dangerous. Metallic sodium requires special handling precautions since trace amounts of water could activate dangerous side reac- tions. This problem was resolved by substituting metal- lic sodium with either sodium hydroxide or potassium hydroxide. This procedure is safe, effective, and tolerant of water and other contaminants. 15 Currently, this technology has been successfully ap- plied to PCBs decontamination of dielectric oils. The application of KPEG processes for the decontamination of lube-used oils is more difficult than that for dielectric oils, because of the various additives of lube oils (detergents, antioxidant and emulsifying agents) and the presence of water (up to 10 wt %). These additives can influence the kinetics of the PCB decontamination reactions by reducing the yields of dechlorination. Previous work showed the effectiveness of KPEG technology in the treatment of both dielectric and lube contaminated oils and the importance of the PEG/oil and KOH/PEG ratios, and of the PCBs chlorination degree, on the efficiency of the KPEG treatment. 16 This paper describes a kinetic study of the removal of PCBs from mineral oils by the KPEG dechlorination process at different temperatures. Experimental Procedure Experiments were performed on a dielectric oil con- taining a mixture of the PCBs Aroclor 1420, 1450, and 1460.The dechlorinating agent was poly(ethylene glycol) alkoxide. These reactions were conducted in a magnetically stirred, thermostated 300-mL glass vessel. The experi- mental procedure was as follows: contaminated dielec- tric oil containing 1015 ppm of PCBs was preheated in the vessel to a selected temperature while being stirred. After the addition of KOH-PEG, previously prepared by dissolving KOH in PEG at 70 °C, the run was started. Experimental conditions used in this work are listed in Table 1. All runs were batch modes ranging between 2.5 and 5 h. Approximately 5-mL aliquots of liquid samples were withdrawn from the reactor at fixed time intervals to determine the residual PCBs content in the oil. Prior to analysis, all the samples remained static for several hours to allow separation of the glycol and oil phases, i.e., the polar heavy phase and the nonpolar light phase, respectively. PCBs contents in the oil and glycol phases were determined using gas-liquid chromatography (GLC). * To whom correspondence should be addressed. E-mail: giava@pcgiava.ing.uniroma1.it. 380 Ind. Eng. Chem. Res. 1999, 38, 380-384 10.1021/ie9803422 CCC: $18.00 © 1999 American Chemical Society Published on Web 01/14/1999