David Lorenzo

Phenol abatement using persulfate activated by nZVI, H2O2 and NaOH and development of a kinetic model for alkaline activation

ABSTRACT Three persulfate (PS) activation methods (nanoparticles of zero-valent iron (nZVI), hydrogen peroxide and alkali) were compared using phenol as target pollutant. Firstly, four experiments were conducted at 25°C in a batch way using the same initial phenol and oxidant concentrations (10 mM and 420 mM, respectively), being the molar ratio activator/PS fixed to 0.005 with nZVI (mass ratio 0.0011 nZVI/PS), to 2 using hydrogen peroxide and to 2 and 4 with NaOH. Phenol and PS conversions and aromatic byproducts profiles during 168 h reaction time were measured and compared, as well as mineralization and ecotoxicity of the samples. It was found that both phenol and aromatic byproducts (catechol and hydroquinone) totally disappeared using PS activated by alkali before 24 h, while a significant amount of aromatic intermediates was obtained with nZVI and H2O2. Additional runs were carried out using shorter times (0–2 h) to discriminate the oxidation route and the kinetic model of phenol abatement by using PS activated by alkali. Different initial concentrations of phenol (5–15 mM), PS (210 and 420 mM) and molar ratio NaOH/PS (2 and 4) were employed. The kinetic model obtained predicts accurately the evolution of phenol, persulfate, hydroquinone and catechol.

Abatement of chlorinated compounds in groundwater contaminated by HCH wastes using ISCO with alkali activated persulfate

In this work, in situ chemical oxidation (ISCO) with alkali activated persulfate has been tested for the elimination of HCH isomers and other chlorinated compounds in groundwater from Sabiñanigo (Sardas landfill), which was contaminated by solid and liquid wastes illegally dumped in the area by a company producing lindane. Due to the site lithology and the type of pollutants found in groundwater (HCHs and chlorobenzenes) alkali (NaOH) activated persulfate (PS) was selected as oxidant. The influence of variables such as PS concentration (42-200mM) and NaOH:PS molar ratio (2:1 to 4:1) on chlorinated compound abatement has been studied and a kinetic model to predict the composition of all chlorinated organic compounds (COCs) in the aqueous phase with time was obtained. It was found that a fast initial hydrodechlorination reaction took place in which HCH isomers reacted to trichlorobenzenes (mainly 1,2,4 TCB) at pH≥12. Mono-, di-, tri and tetrachlorobenzenes remaining were oxidized without producing aromatic intermediates. At the condition tested a first order kinetic model for COCs and PS concentration was obtained. Zero order alkali concentration was obtained while pH was being kept at 12 for the whole reaction time.

Reply to Behrman

We would like to thank Prof. Behrman for his comments on our paper ‘Activated persulfate by nZVI, H2O2 and NaOH in phenol abatement and development of a kinetic model for alkaline activation’ and we appreciate the opportunity to respond. In his letter, Prof. Behrman states that ‘the authors discuss only the radical chemistry of peroxydisulfate. Peroxydisulfate undergoes many reactions involving free radicals, but there is also a large body of non-radical polar reactions’. We want to point out that the aim of our work was to compare the efficiency of persulfate activation using three different methods (nanoparticles of zero-valent iron, H2O2 and under alkaline conditions). Subsequently, in the alkaline activation, the effect of different parameters (initial concentration of phenol, persulfate and activator:oxidant ratio) on phenol abatement was evaluated. In this way, and in accordance with the byproducts generated (hydroquinone and catechol), we focused on the evolution of these compounds using the three methods, but we did not go into detail about the mechanisms by which they were produced. Moreover, Prof. Behrman states that ‘Under conditions of low temperatures [Lominchar et al. used 25°C] and in the absence of catalysts, polar reactions predominate because of the relatively high activation energies required for radical formation. In particular, the reaction between phenols and peroxydisulfate under alkaline conditions has been wellstudied. Elbs discovered this reaction in 1893 and it has been extensively reviewed. There is no radical involvement’. The radical mechanism summarized in the paper was based on three recent publications. The first was the work of Furman et al. [1] which demonstrated the activation of persulfate under alkaline conditions and subsequent production of hydroxyl radicals without the interaction with organic compounds. The second was the study by Ahmad et al. [2] which demonstrated that phenoxide activation of persulfate was produced by the reduction of persulfate by the phenoxide rather than through nucleophilic attack by the phenoxide (the Elbs persulfate oxidation reaction). The alkaline activation of persulfate and the activation of persulfate through the reduction of persulfate by the phenoxide can both produce hydroxyl radicals [2] with hydroquinone being the main byproduct of phenol oxidation [3] as observed in our study. Taken together, we believe that these reports indicate that the formation of hydroxyl radicals is entirely feasible under the conditions used in our work. We would like to thank Prof. Behrman again for his feedback to our paper and we hope that this reply improves the clarity of our article.

Chlorinated organic compounds in liquid wastes (DNAPL) from lindane production dumped in landfills in Sabiñanigo (Spain)

α, β and γ-hexachlorocyclohexane (HCH) are persistent and bioaccumulative pollutants and they were included in the Stockholm Convention on Persistent Organic Pollutants (POPs). Old lindane factories generated high amounts of wastes with HCH and other Chlorinated Organic Compounds (COCS). These were often dumped in the surroundings of the production sites, polluting soil and groundwaters with the associated risk of surface pollution. This is the case of the Sardas and Bailin landfills, located in Sabiñánigo (Huesca, Spain). Among the waste from lindane production, a liquid residue was detected in the landfill subsurfaces, forming a dense non-aqueous phase liquid (DNAPL) composed of HCH isomers, benzene and chlorobenzenes, with a high impact on groundwater pollution. In this study, six DNAPL samples obtained from the Bailin and Sardas landfills were analyzed by GC/MSD and GC/FID/ECD. Compounds were identified using mass spectra and the retention index from pure standards and literature information. Pure positional isomers of dichlorobenzene (DCB), trichlorobenzene (TCB), tetrachlorobenzene (TetraCB), HCH and pentachlorocyclohexene (PentaCX) were distinguished and quantified. In addition, heptachlorocyclohexane (HeptaCH) isomers, precursors of hexacholorocylohexene (HexaCX), were also identified and quantified in the DNAPL samples, although the corresponding isomers could not be discriminated. Information about PentaCX, HexaCx and HeptaCH identification is very limited in the literature. HCH contents in the DNAPL ranged from 22% to 30% in weight, the major isomers being lindane and δ-HCH, followed by α-HCH. The β isomer was the least abundant. HeptaCH contents were present in the same order of magnitude as HCHs in the DNAPL. PentaCXs and HexaCXs could have appeared as dehydrochlorination derivatives of HCHs and HeptaCHs, respectively. Two of the DNAPLs analyzed showed a higher content of TCBs and TetraCBs, associated with lower HCH and HeptaCH contents. Variations of these compounds in the DNAPL could be related to an alkaline dehydrochlorination in the landfill conditions.

Modelling of a Reactive Distillation in the production process of high purity Cyclohexanone to produce caprolactam

Abstract Cyclohexanone (ONE) is an important raw material to promote caprolactam (CPL). It is used as a monomer in nylon industries. In the ONE production process, this component needs to be purified from a mixture of cyclohexanol (OL) and some impurities such as 2-cyclohexen-1-one (CXENONE). The concentration of these impurities must be reduced o in order to improve the quality of nylon fibres produced from CPL. The main scope of this work is to model a reactive distillation column where the cyclohexanone (ONE) is separated from a mixture of cyclohexanol (OL) and 2-cyclohexen-1-one (CXENONE). The model is developed in gPROMS, where the kinetic model is implemented alongside with the NRTL equation whose binary interaction parameters are also obtained experimentally. The aforementioned model is validated in a packed distillation column operating at continuous conditions. This experimental setup is used to prove the capacity of the model to explain the real operation conditions. The mathematical model has shown favourable predictions on temperature profiles, concentration profile and conversion of all the compounds in each stage of the column. Finally, the MINLP problem formulation for optimising real operation is suggested along with planned sensitivity studies which will next be tackled.