Amides as a model system of low molar mass algal organic matter. Influence on the adsorption of p-nitrophenol on activated carbon

2.1 Adsorbates and sorbent

The p-nitrophenol used in the research was purchased from Fluka (≥99,5%). Stock and work solutions of p-nitrophenol were prepared in phosphate buffer (pH 7.4) organic carbon free water (<0.2 mg TOC/L). Acylation of the corresponding amines with alkyl chloride, known as Schotten-Baumann reaction, was used to synthesize N-methylformamide (NMF), N-ethylformamide (NEF), N-t-butylacetamide (NTBA), N,N-dimethylacetamide (NNDMA), N-n-butylbenzamide (NBB), N-2-phenylethylpropanamide (N2PP), N-hexylpropanamide (NHP), N-n-butylbenzamide (NMB). The purity of these N-monosubstituted amides was checked by gas chromatography (GC) and mass spectrometry (MS). Based on the GC/MS results, all the amides used in this work had a purity of 99.2% or better. The physical-chemical properties of the investigated secondary amides are listed in Table 1. Structural formulas of selected amides are shown in Fig. 1.

Close

Figure 1 Structural formulas of selected amides.

Export to PPT

The powdered activated carbon (PAC) Norit SA 2 (Cabot Norit Nederland B.V) was used as the adsorbent. An AutosorbiQ Surface Area Analyzer (Quantochrome Instruments, USA) was used for measurements of the average pore radius, the total pore volume and the micropore test. Samples were outgassed at 293 K for 5 h before running isotherms. The physical-chemical characteristics of the PAC used are listed in Table 2.

2.2 Kinetic and equilibrium adsorption experiments in single and binary systems

The powdered activated carbon was prepared as an adsorbent by drying at 378 K for 24 h. The adsorption equilibrium of p-nitrophenol on activated carbon was tested at a constant temperature of 298 K. The equilibrium and kinetics of the adsorption of a single-solute p-nitrophenol system were tested initially, which was followed by an investigation of the p-nitrophenol adsorption in the presence of the investigated amides (binary systems). In the single-solute system the concentration of p-nitrophenol was 5 mg/dm³. In the binary systems, the concentration of p-nitrophenol was the same, while the concentration of amides was 5, 50 and 100 mg/dm³.

Batch equilibrium experiments were performed on a rotary shaker at 180 rpm using 250 cm³-shaking flasks at room temperature for 180 min. After shaking, solutions were filtered through 0.45 μm glass fiber filters and the absorbances were measured on a Shimadzu UV 1800 UV–VIS spectrometer at the characteristic wavelength (400 nm) in 1 and 2 cm cuvettes at 298 K. Synthetic pH adjusted (7.4) organic carbon free water (<0.2 mg/dm³) was used as a blank. All of the experiments were performed in duplicate. For all investigated amides the initial pH value and the pH value after the addition of the activated carbon suspension were in the range of 7.4–8.2 and did not change during the experiment.

Prior to the investigation of the effect of the amide on the adsorption kinetics of p-nitrophenol, a set of tests was conducted to determine the impact of speed mixing on the kinetics of adsorption of p-nitrophenol. The tests were performed at 5 different stirring speeds (0, 50, 100, 240, 380 rpm). Based on the obtained values for film diffusion coefficients it was found that from the speed of 100 rpm stirring velocity has no more influence on the film diffusion.

The kinetic experiments were performed in a glass reaction vessel with a volume of 1 dm³. Previously prepared aqueous solution containing p-nitrophenol and amide (500 cm³) in the specific concentration ratio (1:1, 1:10, 1:20) was added into the reaction vessel and stirrer was turned on. Continuous mixing of the solution was maintained by means of a magnetic stirrer with a controlled number of revolutions. Kinetic experiment was started by addition of a certain amount of activated carbon suspension. Samples were taken at defined time intervals over a period of 180 min. After reaching certain time, the adsorbent was removed for determination of p-nitrophenol concentration. Filtration of samples was performed using 0.45 μm glass fiber filters and the absorbances were measured by a UV–VIS spectrometer at the characteristic wavelength (400 nm) in 1 and 2 cm cuvettes at 298 K. All experiments were performed in duplicate.

Solid-phase solute concentrations were calculated from a mass balance of solute between the solid and aqueous phases.

2.3 Data analysis

The simple film diffusion model and the HSDM model were applied in order to assess and compare external and internal diffusion coefficients of p-nitrophenol in the presence of different amides.

4.1 Amide impact on the diffusion kinetics of p-nitrophenol

Experimental kinetic curves for the systems investigated are shown in Fig. 2. The equilibrium adsorption capacity, external mass (k_f) and surface diffusion (D_S) coefficients for p-nitrophenol in the presence of investigated amides are presented in Table 3. The determination coefficients obtained for fitting the experimental kinetic data were in the range of 0.9217 to 0.9897. The coefficients were tested by Student's (n−2) t-test, and it was found that the coefficients are statistically significant at the significance level of p ≤ 0.05, i.e. with a certainty of 95%.

Close

Figure 2 Kinetics of the adsorption of p-nitrophenol in the presence of investigated amides (●–p-nitrophenol, ○–5 mg/dm³ of amide, Δ–50 mg/dm³ of amide, □–100 mg/dm³ of amide).

Export to PPT

On the basis of the obtained values for the adsorption capacity it can be estimated that the secondary amides with the lower molar mass and less voluminous substituents (NMF, NEF, NNDMA, NTBA) show no significant influence on the adsorption capacity of p-nitrophenol, whereas for amides with the voluminous substituents (NHP, NMB, NBB, N2PP) a decrease in adsorption capacity with increasing concentration of the amide in the system can be noticed.

A similar trend is observed for film diffusion coefficient values k_f of the investigated amides. Namely, more voluminous amides (having higher molar mass) significantly affect the reduction of the k_f value. Furthermore, it can be observed that this effect increases with the concentration of the amide. These experimental facts can lead to the assumption that the increase in amide concentration causes an increase in the thickness of the boundary diffusion film layer. Based on literature data, it can be assumed that the amides are concentrated on the of water/carbon interface and thus affect the external diffusion phase of the process (Arnold et al., 2008; Bhinde et al., 2011). Namely, for some peptide surfactants it was found that an increase in hydrophobic tail chain length enhances the hydrophobic affinity, resulting in the increase of the adsorbed amount (Pan et al., 2010) Also, a correlation of work of adhesion (decrease in water solution surface tension) and carbon loading was established for some pollutants (de Ridder et al., 2013).

The values obtained for the coefficient of internal diffusion D_S are all of the same order of magnitude and not significantly different for the p-nitrophenol adsorption with and without the presence of amides.

Based on the differences observed for the k_f and D_S values, it can be assumed that some of the investigated amides affect the external rather than the internal diffusion step of the adsorption process. Differences in the values for external and internal diffusion coefficients may be a result of various factors. In literature it can be found that some high molar mass model substances (poly(styrene sulfonate)) have impact on reducing the D_S coefficient of atrazine for several orders of magnitude (Li et al., 2003). Adsorption from multicomponent aqueous solutions (system phenols-cresol) is a surface diffusion-controlled process (Mijangos et al., 2001). An increase in the effective diffusion coefficients can occur also due to the in situ formation of complex species (Bautista-Toledo et al., 2014). The effects of natural organic matter loading and an aging powdered activated carbon slurry are responsible for a decrease of the D_S value for atrazine (Lebeau et al.,1999).

4.2 Amide impact on the equilibrium adsorption of p-nitrophenol

Table 4 presents the Freundlich isotherm parameters (K_F and n), the single point distribution coefficient (K_d at concentration of 2.5 mg/dm3) and the standard free Gibbs energy change (ΔG) for p-nitrophenol (single systems) and p-nitrophenol in the presence of different concentrations of the investigated amides (binary systems).

Adsorption isotherms of p-nitrophenol in the absence and in the presence of different concentrations of the investigated amides are shown in Fig. 3.

Close

Figure 3 Adsorption isotherms of p-nitrophenol in the absence and in the presence of investigated amides at different concentrations (•–p-nitrophenol alone, ○–5 mg/dm³ of amide, Δ–50 mg/dm³ of amide, □–100 mg/dm³ of amide).

Export to PPT

As it can be expected on the basis of kinetic experiments, the adsorption equilibrium of p-nitrophenol is affected much more by the more voluminous amides. The group of amides with less voluminous substituents (NMF, NEF, NNDMA, NTBA) showed a negligible influence on the adsorption capacity of p-nitrophenol, whereas voluminous amides (NHP, NMB, NBB, N2PP) affect the adsorption capacity reduction with increasing concentrations. Accordingly, the values obtained for the free Gibbs energy changes indicate a less spontaneous adsorption process of the p-nitrophenol in the presence of the group of more voluminous amides.

Furthermore, the calculated values of diffusion and thermodynamic parameters of adsorption of p-nitrophenol have been correlated with structural parameters of the amide: the molecular volume V, the octanol-water partition coefficient logK_ow value. The coefficients were tested by Student's (n−2) t-test, and it was found that some of coefficients are statistically significant at the significance level of p ≤ 0.1, i.e. with a certainty of 90%. The obtained correlations are presented in Table 5.

The following correlation can be observed: the volume and octanol-water partition coefficient correlated positively. Moreover, significant correlations were also observed between all amide parameters and external diffusion constants k_f. Furthermore, a positive correlation of the volume and the octanol-water partition coefficient indicate that more voluminous, more hydrophobic amides reduce the possibility of binding p-nitrophenol on activated carbon.

As for the established correlation with the external diffusion coefficient, it can be assumed that more voluminous and less polar amides have impact on the increase of thickness of the outer film and reduce the capacity of the adsorbed p-nitrophenol. In other words, the less polar amide molecules concentrated at the boundary layer of polar and non-polar phase (water/carbon), thus creating an external film that remains stable despite the relatively high stirring velocity. Decrease of external diffusion can be also explained by NOM accumulation on the outer carbon surface and by increasing viscosity and thickness of the external layer (Carter and Weber, 1994).

In literature about multicomponent adsorption systems, a decrease of the absorption capacity of some of the components is usually explained mainly by two phenomena: the effect of direct competition for active sites on the adsorbent and the effect of blocking of the pores. According to the obtained results, it could be estimated that the effect of pore blockage is responsible for the reduced adsorption of p-nitrophenol in the presence of an amide.

The pore size distribution of the adsorbent and the molar mass fractions of NOM molecules show the most significant impact on the adsorption process of organic pollutants. Many studies indicated that a direct competition for available adsorbent sites is the dominant mechanism if the pores are large enough to admit micropollutants and NOM molecules. Also, if the pores are wide enough to pass only micropollutants and not NOM molecules, then the pore blocking mechanism is dominant (Pelekani and Snoeyink, 1999). In this study, powdered activated carbon with an average pore size of 20 Å was used. This average pore size can be classified as a lower mesopore. Based on the mean molecular weight of amide molecules and the medium pore size it can be estimated that the pore blocking mechanism is responsible.

As far as the inability of making more significant estimations is concerned, it can be concluded that the research needs to be expanded for the purpose of better distinguishing the impact of amides on the diffusion parameters of p-nitrophenol at this concentration level. It is necessary to use more structurally/electronic different amides as model systems and also to perform experiments on water matrices with a wider range of amide concentrations.