Introduction

The elimination of sulfur-containing organic molecules present in fuel is carried out through a hydrodesulfurization (HDS) reaction (Ojeda et al. 2005). This process can desulfurize compounds like thiophene, benzothiophene, and dibenzothiophene (DBT), but it is unsuccessful in treating dibenzothiophene derivatives that have alkyl groups near the sulfur atom, especially 4,6-dimethyldibenzothiophene (4,6-DMDBT) (Baeza et al. 2015). In this context, adsorptive desulfurization is a promising alternative approach to remove sulfur-containing organic molecules using inorganic materials as adsorbents (Kilbane 2006). Despite numerous studies on the adsorptive desulfurization process, it still possesses unresolved difficulties, the foremost of which is the regeneration of the adsorbent material. This is because adsorbent regeneration releases SO2 into the environment (Li et al. 2009). In this context the biodesulfurization (BDS) (Li et al. 2006), which is a biocatalytic process performed by microorganisms that remove selectively sulfur from hydrocarbon fractions without the generation of gaseous pollutants, is an alternative that could be applied for the elimination of the adsorbed sulfur molecules. In this regard, this work has evaluated the environmentally-friendly removal of sulfur-containing organic molecules adsorbed on inorganic materials through the use of BDS. Also, the influence of physical parameters of the inorganic materials on the processes of adsorption and removal of sulfur-containing organic molecules was studied.

Materials and methods

Bacterial strain

Rhodococcus rhodochrous (ATCC 53968) was grown in sulfur-free Medium A (Maghsoudi et al. 2001) containing sodium succinate (30 mM) and citrate (0.1 % w/v) as energy and carbon sources, respectively. Dibenzothiophene (DBT) (62 mM) (Merck) was used as the only sulfur source.

Sulfur-containing organic molecules and support materials

DBT (100 mg l−1) and 4,6-dimethyldibenzothiophene (4,6-DMDBT) (100 mg l−1) were used as sulfur-containing organic molecules. Silica (Si) D11-10 BASF (specific area of 80 m2 g−1), alumina (Al) T-126 Girdler (specific area of 200 m2 g−1), and sepiolite (Sep) 120 NF Tolsa (specific area of 300 m2 g−1) were used as inorganic supports (Dinamarca et al. 2014).

Selection of support particle size and determination of adsorption of sulfur-containing organic molecules

To study the effect of the particle size of the support on biodesulfurization (BDS), DBT and 4,6-DMDBT were mixed with Si, Al or Sep with two different particle size distributions of 0.063–0.425 and 1.18–2 mm in 25 ml flasks at 25 °C and shaken at 200 rpm for 24 h. To study the adsorption of sulfur-containing organic molecules, each support (1 g) was placed in a 25 ml flask containing 10 ml DBT (100 mg l−1) or 4,6-DMDBT (100 mg l−1) and mixed for 80 h. Samples of the solution were taken for analysis during the adsorption process. The amounts of adsorbed sulfur-containing organic molecules were analyzed by GC (Baeza et al. 2015).

Adsorption kinetics

The adsorption kinetics of DBT and 4,6-DMDBT over the three different support materials were studied considering non-dissociating molecular adsorption of these molecules on the support materials. The quantity adsorbed at any given time (qt) and equilibrium adsorption capacity (qe), could be predicted by either a pseudo-first-order or pseudo-second-order kinetic model (Srivastav and Srivastava 2009). The experimental data were fitted with the two models to determine the model kinetic parameters by non-linear regression analysis using Origin 8.0 statistical software.

Removal of adsorbed sulfur-containing organic molecules

To remove the adsorbed sulfur-containing organic molecules, free cells, collected by centrifugation at 4000×g for 30 min at 4 °C (from 109 to 25 × 109 cells), were placed in a 25 ml flask containing sulfur-free Medium A (10 ml) with 1 g support containing adsorbed DBT or 4,6-DMDBT. The reaction was carried out at 30 °C at 200 rpm for 48 h. The cultures were centrifuged and the residual sulfur-containing organic molecules adsorbed on the supports were extracted with ethyl acetate. The contents of DBT and 4,6-DMDBT were analyzed by GC (Baeza et al. 2015). The systems are designated as sulfur-containing organic molecule/solid systems, where the sulfur-containing organic molecules were DBT or 4,6-DMDBT, and the solid was the inorganic support. The activity of each experiment is expressed as the percentage removal of sulfur-containing organic molecules.

Results and discussion

Selection of support particle size

The results (Table 1) reveal that higher adsorption of both sulfur-containing organic molecules occurred when a solid support with particle sizes from 0.43 to 0.063 mm was used as adsorbent; i.e., the adsorption of dibenzothiophene (DBT) and 4,6-dimethyl-dibenzothiophene (4,6-DMDBT) increased as the particle size of the adsorbents decreased. A larger total surface area per volume and a shorter diffusion path for the adsorbate when using smaller particles can explain this behavior (Tsai et al. 2003; Lu et al. 2016). On the other hand, the differences observed in the adsorption values of DBT and 4,6-DMDBT on the different supports at 1 and 24 h, showed in the Table 1, are caused by the different interactions between these sulfur-containing organic molecules with the adsorbents (Kim et al. 2006).

Table 1 Dependence of the adsorption parameters of dibenzothiophene (DBT) and 4,6-dimethyl-dibenzothiophene (4,6-DMDBT) with the support particle size
Full size table

Adsorption of sulfur-containing organic molecules

The adsorption capacity of DBT on the different supports is shown in Fig. 1a. At the beginning of the adsorption process, all supports displayed a linear increase of DBT adsorption until a maximum value reached after approximately 10 h. At longer times the adsorption values then remained constant until 80 h. The maximum adsorption of DBT on silica (Si) was 35.5 mg l−1, while those on alumina (Al) and sepiolite (Sep) were 30 and 23 mg l−1, respectively. In the case of adsorption of 4,6-DMDBT, as illustrated in Fig. 1b, only Si exhibited similar behavior to DBT adsorption, while maximum adsorption on Sep and Al was reached after longer periods. Comparison of the adsorption of both sulfur-containing organic molecules revealed higher adsorption of DBT on Si and Al, while no marked difference was observed in the case of Sep. The variation of the adsorption of DBT and 4,6-DMDBT on the supports can be explained by the different specific areas and density and strength of acid sites of these materials (Kim et al. 2006). However, the higher adsorption of both sulfur-containing organic molecules on the Si support showed that the strength of acid sites is the main factor that affects the adsorption of recalcitrant sulfur-containing molecules.

Fig. 1
figure 1

a Adsorption capacity of dibenzothiophene (DBT) on Alumina (Al), silica (Si) and sepiolite (Sep). (Filled cirle DBT/Sep, triangle DBT/Al, square DBT/Si). b Adsorption capacity of 4,6-dimethyldibenzothiophene (4,6-DMDBT) on Alumina (Al), silica (Si) and sepiolite (Sep). (Filled cirle 4,6-DMDBT/Sep, triangle 4,6-DMDBT/Al, square 4,6-DMDBT/Si)

Full size image

Table 2 summarizes the kinetic rate constants obtained using the first- and second-order models (kf and ks, respectively). Considering the three support materials studied, the highest values for the coefficient of determination (r2) and lowest values of the residual sum of squares (RSS) suggest that the adsorption of DBT and 4,6-DMDBT follow second-order kinetics. The qe values of the samples (Table 2) confirm that the strongest interaction of DBT is with Sep while for the case of 4,6-DMDBT is with Si.

Table 2 Kinetic parameters adsorption of dibenzothiophene (DBT) and 4,6-dimethyl-dibenzothiophene (4,6-DMDBT) on Alumina (Al), silica (Si) and sepiolite (Sep)
Full size table

Removal of adsorbed sulfur-containing organic molecules by Rhodococcus rhodochrous

Table 3 presents the percentage removal of the sulfur-containing organic materials adsorbed on supports by BDS; the removal of DBT and 4,6-DMDBT clearly depends on the type of support. Higher degradation of both sulfur-containing organic materials was observed when Sep and Si were used as the. The differences observed for the removal of the sulfur-containing compounds by the same support can be explained by the different interactions between the cells and supports (Dinamarca et al. 2010). Both physical and biological factors can influence cell–support interactions, such as the size of the bacterial cells, ionic strength (Yee et al. 2000) and the support surface structure (Jeyachandran et al. 2006). We previously studied the BDS activity of gas oil of biocatalytic immobilized systems using Pseudomonas stutzeri adsorbed on inorganic materials (Dinamarca et al. 2010). Our results showed higher activity in systems that had stronger interactions between the bacterial cells and support. The greater removal of DBT compared with that of 4,6-DMDBT could be explained by the difficulty microorganisms have desulfurizing long-chain alkylated DBTs (Bhatia and Sharma 2010).

Table 3 Percentage removal of dibenzothiophene (DBT) and 4,6-dimethyldibenzothiophene (4,6-DMDBT) on Alumina (Al), silica (Si) and sepiolite (Sep), by Rhodococcus Rhodochrous spp. in a DBS process
Full size table

Conclusions

Biodesulfurization (BDS) is effective and can be used to remove sulfur-containing organic molecules adsorbed on inorganic materials and avoid the generation of gaseous pollutants. The efficient removal of sulfur-containing organic molecules depends on the interaction between the support and adsorbent and the adsorption capacity of the bacteria used.