In the wastewater treatment system, the sewage passes through a series of treatment steps that use physical, biological, and chemical processes to remove pollutants from water. Many different treatment processes of industrial wastewater effluents are studied [1-13].

In any case, the final products are the purified water and sludge. Common disposal processes for sewage sludge from tannery wastewater are land-filling and thermal conversion (combustion, incineration, pyrolysis). The most used thermal processes are carried out without the presence of catalysts. For municipal sludge, several catalysts were tested, such as TiO₂, Al₂O₃, Fe₂O₃, MnO₂/CuO/CaO. Only in the case of MnO₂/KOH, it has been found a clear demonstration that, with the addition of catalysts, the ignition of MSW is enhanced [14].

The influence of catalysts was also investigated in the thermal incineration of industrial sludge. In this case K₂CO₃, NaCl and Al₂O₃ were used. The catalytic performance of K₂CO₃ was better than NaCl and Al₂O₃ [15].

Very recently, we worked on the thermal treatment of tannery sewage sludge using CeO₂ as catalyst and studying the influence of catalyst loading by thermogravimetric analysis coupled with mass spectrometry (TG-MS). The catalyst mixed with the dried sludge improved either the selectivity of the sludge combustion process, reducing the emission of cyclic and aromatic substances, either the combustion peak temperature, yielding a higher oxidation rate of the sludge organic fraction [16].

Moreover by adding the catalysts in the step of sludge formation, i.e. immediately after the addition of flocculants into the coagulation-flocculation step of tannery wastewater, the combustion peak of the sludge organic fraction occurred at lower temperature, about 300°C, while in the absence of catalyst the onset was at about 525°C). This points out to a significant improvement of the oxidation process due to the catalytic material [17]. The reaction occurs in a solid-solid-gas system and to this purpose, it is necessary to improve the contact among the solid phases. A way to obtain an improved physical contact among the sludge particles and catalysts during the combustion is to place in rotation the reactor. So, this work was focused on the catalytic combustion of tannery sludge by using ferrites and CeO₂ in a stainless steel fixed bed reactor in comparison with the same reactor set in rotation.

Materials and characterization methods

Tannery sewage sludge samples were supplied by an Italian tannery wastewater treatment plant. Sludge conditioning was accomplished by preliminary sample milling. In particular sludge was dried in a stove at 80°C for 8 hours and then pounded in an electric mortar to a size between 180-250 m.

Dried sludge was characterized by different techniques. Moisture content was evaluated by gravimetric analysis of samples before and after an isothermal treatment at 100°C ± 2°C up to constant weight. Total ashes percentage was obtained by weighing the inorganic residue after isothermal treatment at 550°C for 8 hours. Finally, the organic matter was calculated by difference with the moisture and inorganic matter content in the samples. The mineral content was analysed by ICP-AES (Inductively Coupled Plasma Atomic Emission Spectrometry) Liberty Series II Varian with axial torch after sludge acidic digestion. The organic fraction of sludge was analysed by GC-MS after solvent extraction (hexane). By cool-on-column injector the sample was introduced into GC column (Restek Rtx-5ms, 5% diphenyl and 95% dimethyl polysiloxane, 0.25 mm ID, 30 meter), then the separated compounds were analysed by a Quadrupole Mass Selective Detector (HP 5973). For the analysis of volatile compounds a Purge and Trap (P&T) device was used to introduce samples into the GC-MS, equipped with a GC column (Restek Rtx-624, 6% cyanopropylphenyl and 94% polysiloxane, 0.25 mm ID, 60 meter) via split/splitless inlet system.

CeO₂, LaFeO₃ and LnFeO₃ (Ln=Ce, La, Nd, Pr, Sm) as catalysts were mixed with sludge by gently grinding in an agate mortar to obtain a contact between the two solid phases. The obtained samples are named Cex, Lay and Lnz, where where x, y and z represent the weight percentage of CeO₂, LaFeO₃ and LnFeO₃ in the mixture catalyst-sludge, respectively. The catalyst loading was in the range 3-6 wt%. Physico-chemical characterization of sludge and catalysts was performed by different techniques. Specific surface areas were obtained by N₂ adsorption measurement at -196°C with a Costech Sorptometer 1042 after pre-treatment at 120°C for 1 h in He flow (99.9990%). X-ray diffraction (XRD) was carried out using the X-ray microdiffractometer Rigaku D-max-RAPID.

TG-MS analyses were performed in air flow by a thermogravimetric analyzer (SDTQ600, TA Instruments) studying the evolved gas by a mass detector in the range m/z=29-209 (Pfeiffer Vacuum Benchtop Thermostar). Measurements were carried out on about 30 mg of sample in 100 (stp) cm³/min air flow (chromatographic grade) at 10°C/min heating rate, in the temperature range 20-1000°C. The parameters used for the evaluation of the catalytic activity from thermal analysis (TG-MS) were the first derivative maximum value of weight loss with respect to time (DTG_max), to describe the rate of conversion for the combustible matter, and the value of peak temperature (T_max) associated to DTG_max. The emission of aromatic substances was monitored by analyzing the mass fragment m/z=78, characteristic of benzene.

Stainless steel reactor for constant temperature oxidation

A microreactor was employed in order to evaluate the reactivity of the mixture sludge-catalyst. The microreactor comprises a 500 mm length, 17 mm ID stainless steel tubular flow reactor, heated by an electrical furnace. Oxidation tests at constant set temperatures were carried out in air flow. ABB continuous analyzers determined exhaust gas concentrations: URAS 14 (for CO, CO₂ and SO₂ concentrations) and MAGNOS 106 (for oxygen). Signals from the analyzers were sent to a personal computer for data processing. Samples to be tested were previously diluted with quartz particles to avoid local raise of temperature during the tests. The diluted mixture was loaded to the reactor filling about 3 cm in length of its central zone. The test started raising the temperature to the desired value (T=350°C) in static air. Then the air stream was fed to the reactor starting the oxidation. The feeding air flow rate was 30 L/h (STP). Operating pressure was 1 atm. The initial mass of the sample was about 70 mg. In the case of tests with rotating reactor, the rotating rate was equal to 3 rpm.

Sludge and catalysts characterization

The analytical results obtained for furnished sludge are shown in Table 1.

Table 1: Results of analytical tests on sludge.

The sample is mainly constituted by water (the portion loss at 102°C) and organic matter and the inorganic matter (residue at 550°C).

The abundance of chromium in the sludge depends of the initial composition of wastewater, as they are collected and equalized, i.e. from the prevalent tanning process in the relative industrial district. The presence of Al and Fe is mainly due to the use of Al₂(SO₄)₃ as coagulant agent for wastewater, although it may derive, in small quantities, also from the tanning industry.

The values of specific surface area for sludge and catalysts are reported in Table 2.

Table 2: Values of specific surface area for catalysts.

The surface area of CeO₂ is strongly higher. Consequently, the latter would seem able to make good contact with the sludge. XRD analysis carried out on CeO₂ and LaFeO₃ (Figure 1 and Figure 2) revealed the presence of only the signal due to the crystalline structure of the two substances. Since the catalyst LnFeO3 is a solid that includes several rare earth elements (La, Ce, La, Nd, Pr, Sm), the XRD peaks are not easy to attribute. From Figure 3 the diffraction patterns of the sample LnFeO₃ are clearly related, to those of La, Ce, Nd, Pr, Sm ferrites, Fe₂O₃ and Fe₃O₄ [18]. Finally, the signals obtained for values of 2 equal to 32.1 and 55.5, are likely due to CeO₂ present in the structure of the sample analyzed.

Figure 1: XRD spectrum of CeO2 catalyst.

Figure 2: XRD spectrum of LaFeO₃ catalyst.

Figure 3: XRD spectrum of LnFeO₃ catalyst.

I^*: integral of mass fragment m/z=78 profile as a function of run time; n.d: not detectable.

As general remarks on TG-MS results, the addition of catalytic material to dried sludge determined effects either with regard to the selectivity of the process, by reducing the emission of aromatic substances, either with respect to the operating conditions, resulting in lower values of T_max. Table 3 summarizes the results obtained from the analysis carried out on all samples tested in terms of T_max and the amount of aromatic compounds emitted, evaluated by the integral (I) of the signal associated to the mass fragment m/z=78 as a function of run time. T_max decreased from 457°C for sludge alone to about 285°C for a CeO₂ loading of 5 wt%. The amount of aromatic compounds emitted during the combustion decreased to zero for a CeO₂ loading of 6 wt% and for all the mixtures containing LaFeO₃ and LnFeO₃ catalysts.

Table 3: Values of T_max and aromatic compounds emission for sludge and catalysts.

Constant temperature oxidation tests

For all the samples, during the combustion tests in the reactor, there was the emission of CO₂, CO and SO₂. The behaviour of CO₂ produced from the oxidation of sludge in the absence of catalyst is depicted in Figure 4.

Figure 4: Behavior of CO₂ produced during the oxidation of sludge in the absence of catalyst.

Two main stages in the behaviour of CO₂ can be distinguished; the first, with high intensity, occurring in the first five minutes of reaction, is due to the combustion of volatiles, while the second, lower in intensity, is associated to the char combustion. From these data it is possible to observe that gas-phase combustion of volatiles is the dominant reaction in sludge combustion [19] while char combustion is the slowest reaction step [20]. Because the surface flux of water and volatiles, a probable prevention to the transfer of oxygen towards the exhaust char pellet, occurs [21]. A similar behaviour was obtained for all the samples.

Table 4 summarizes the results obtained as a function of type and catalyst loading in terms of percentage of carbon burned (%C), evaluated with respect to the total weight of sludge loaded in the reactor. The mass of carbon burned was calculated by the integration of the signal associated to the CO and CO₂ produced as a function of run time.

Table 4: Percentage of carbon burned evaluated with respect to the total sludge weight loaded in the fixed bed microreactor.

The amount of carbon burned increased from 18 to 25.6 wt% by increasing CeO₂ loading. The highest value of %C was obtained in presence of LaFeO₃ and LnFeO₃ catalysts. In particular it increased up to 32 wt%, relevant more with respect to 18 wt% of the sludge alone.

These results clearly indicate that the presence of catalyst enhanced the combustion process allowing to burn a higher amount of organic substances contained in the sludge.

The results are also analysed considering the reactivity defined as the rate of change of the carbon conversion degree dX/dt (where X=(m₀-m)/m₀, m and m₀ being the current and the total mass of carbon that can be burnt, respectively) as a function of X [22].

The behaviour of dX/dt for sludge, Ce3, La3 and Ln3 is shown in Figure 5a and 5b. In all cases, the presence of catalysts determined an increase of the reactivity with respect to the sludge. Ce3 sample presented the highest dX/dt up to X=0.24 in which the combustion of volatiles occurs.

Figure 5: a)Behaviour of dX/dt as function of X for sludge, Ln3, La3 and Ce3, b)Behaviour of dX/dt as function of X in the range 0-0.2 for sludge, Ln3, La3 and Ce3.

Figure 5b shows the profiles of dX/dt as a function of X in the range 0.2-1 for the same samples.

For X higher than 0.24, the reactivity of Ce3 sample is negligible, underlining that CeO₂ catalyst enhanced mainly the reaction rate associated to the combustion of volatiles and that it is not active in the combustion of the char. On the contrary, both LaFeO₃ and LnFeO₃ accelerated also the latter reaction. In fact the two perovskites showed reactivity higher than sludge alone and allowed to burn all the organic fraction of the sludge.

In Figure 6a and 6b, the reactivity of sludge alone obtained with fixed bed in comparison with the rotating system, is reported.

Figure 6: a)Behaviour of dX/dt as function of X for sludge with fixed and rotating reactor, b)Behaviour of dX/dt as function of X in the range 0-0.2 for sludge with fixed and rotating reactor.

Passing from the fixed to the rotating reactor, both the reactivity associated to the combustion of volatiles (Figure 6a) and that one of char combustion (Figure 6b) was markedly enhanced. Moreover the mass of carbon burned, increased from 18 wt% to 31 wt% and the ratio CO/CO₂ decreased from 0.33 to 0.17 indicating that the rotation system enhanced the oxygen mass transfer rate from gas to sludge particle surface. An improvement of the combustion process is so determined.

In the presence of CeO₂ catalyst, neither the reactivity nor the ratio CO/CO₂ show significant differences passing from fixed to rotating reactor. An important influence was instead obtained in presence of LaFeO₃ catalyst with a content of 4 wt% (Figure 7a and Figure 7b). In this case, the reactivity of the sample was enhanced (Figure 7a) and in particular also the combustion rate of the char is higher when the rotating reactor was used, confirming that the mass transfer rate was increased with respect to the case in which the fixed bed was used.

Figure 7: a) Behaviour of dX/dt as function of X for La4 with fixed and rotating reactor, b) Behaviour of dX/dt as function of X in the range 0-0.2 for La4 with fixed and rotating reactor.

In the present study the combustion of tannery sewage sludge was evaluated in the absence and in the presence of cerium oxide or perovskites in a stainless steel fixed bed reactor. The oxidation profiles evidenced two main stages. The first occurring in the initial time of the reaction, due to the combustion of volatiles, while the second associated to the char combustion. The presence of catalysts determined an increase of the reactivity with respect to the sludge alone. CeO₂ catalyst enhanced mainly the reaction rate associated to the combustion of volatiles while it is not active in the combustion of the char. On the contrary, both LaFeO₃ and LnFeO₃ perovskites also accelerated the latter step. Passing from the fixed to the rotating reactor, the mass of carbon burned, increased, while the ratio CO/CO₂ decreased from 0.33 to 0.17 indicating that the rotation system enhanced the oxygen mass transfer rate from gas to sludge particle surface determining an improvement of the combustion process.

Catalytic thermal conversion presents many advantages, such as reduction of disposed solid mass and volume and energy recovery from the organic sludge fraction, determining lower disposal costs.