What do iron salts precipitate

The presence of ferric ions may render the water inadequate for consumption in some industrial, commercial and residential applications. In concentrations above 0. Several techniques for the removal of metals are known. The most usual consists of iron precipitation in the hydroxide form, followed by sedimentation. However, the process is slow, becoming inadequate in situations of large liquid volumes or of high rates of pollutant dilution, due to the need of a large tank, besides requiring a subsequent filtration stage that delays the process.

The possibility of reducing the iron grade by flotation was studied. This separation technique was suggested initially by Sebba and has been indicated for the recovery of metals Zouboulis, ; Ghazy et al. Several advantages have been attributed to ion flotation: 1 efficiency in removal; 2 speed of the operation; 3 small space occupation; 4 production of small sludge volume; 5 possibility of selective removal; 6 application flexibility and 8 moderate costs McIntyre et al.

Ion removal by flotation can take place: 1 through an electrical interaction between the polar part of the molecule of the collector and the contaminating species, forming a sublet that is transported to the surface by air bubbles introduced into the column; 2 through the formation of a precipitate, in the form of hydroxide or sulphate Capponi et al.

It is considered that the positive charge of the metallic ion exercises a strong influence on the electrons of the molecules associated to the water, resulting in polarization and a tendency for proton dissociation Gregory, Fe OH 2. The last product of the reaction is a precipitate of a colloidal hydroxide of iron, Fe OH 3 S.

In the level of concentration used 50 ppm or about 8. In agreement with Rubin and Johnson , the precipitate flotation of iron hydroxide can be efficient in systems with a concentration of metallic ions times greater than that of the collector, contrarily to what happens in the flotation of soluble iron that requires at least a collector concentration similar to that of the metallic ions.

In all tests, ultra purified water through the Millipore "Academic" System was used. The system consists of pre-treatment through a filter and purification by reverse osmosis River-5 , followed by an ultra purification apparatus Milli-Q. The flotation tests were carried out in a glass column, 65 cm high and 5 cm in diameter, which had a porous plate and was fed by an airflow supplier, with rate controlled by a rotameter Figure 1-A.

The determination of the iron content was made by colorimetry, using a UV spectrophotometer, Spectronics model Genesis 2PC. The methodology used consisted of placing the solution of hexahydrate iron chloride in a mL beaker, for conditioning with the SDS, for a pre-determined period, in a mechanical agitator with a constant speed of rpm.

In sequence, the sample was placed in the flotation column and a constant air flow rate of 0. At the end of the test, the froth containing the iron ions was removed. The purification of the water was evaluated through the determination of the iron content in the water and by the determination of the surface tension in the residual product of each test.

Transmission electron microscopy was used to observe the precipitate morphology and size. According to the results presented in Tables 1 and 2 , the efficiency of the process is influenced by the concentration of the collector SDS and by the conditioning time. No removal of iron was observed with collector concentrations below M.

With 10 -5 M, the removal was partial Removal was practically total with 5 x 10 -5 M. For a concentration of SDS M, removal of That removal level allowed purified water with less than 0. The results, presented in Table 3 , show that the process is improved by the increase in conditioning time. The residual iron content is reduced from That fact suggests an increase in the precipitate size during the conditioning period.

The precipitated growth was favored by the fact that the conditioning was carried out at pH 8, close to the isoelectric point of the colloid Weber Jr. Larger size precipitate floats more readily. Comparing Figures 2 and 3, it is observed that the colloidal particle resultants offing from the precipitation of iron form larger aggregates with 20 minutes of conditioning Figure 3.

The increased size of precipitate implicates in a smaller value in the ratio collector ions precipitate necessary for efficient removal. As the concentration of SDS was maintained constant, the reduction in the surface area of precipitate results in an increase of the residual concentration of the collector corresponding to the excess.

That implicates in a reduction of the surface tension Figure 4. The results presented in Figure 5 show that the surface tension of the treated water decreases from In agreement with data presented in Figure 5 , surface tensions of The process was shown to be very efficient. Removal of The collector conditioning time, a parameter usually not considered strongly relevant, exerted an important influence on the flotation process.

This effect is favored at pH 8 near to the isoelectric point. Fe III in contrary reacted slowly with sulfides in aerobic wastewater Nielsen et al. However, Fe III added to anaerobic wastewater containing sulfide in a ratio of 0. The formation of FeS from this reaction was observed to continue for a time span of a few hours Nielsen et al. This contrasts with the results of this study, where sulfide precipitation with both Fe II and Fe III depending on pH and ratio init was run to completion within minutes.

But, as observed in this study, for Fe II a decrease in ratio init results in a lower reaction rate constant. Rounding the reaction orders of 2. Dependencies of pH on sulfide oxidation rate has previously been observed.

The rate constant in seawater showed a maximum at pH 6. In wastewater at pH 8. Under these conditions, the fraction of S -II precipitated increased with an increasing ratio, but it never attained a fraction of 1, even though Fe II was added above the stoichiometric ratio required for FeS formation Nielsen et al. This is in line with findings of this study and a study done by Nielsen et al. At a ratio of 2.

Values of k obs , 20 and S -II end obtained by fitting to experimental data were modelled by multiple linear regression to yield coefficients for the independent variables of the reactions surface plots Figure 1 a —1 d.

To obtain normal distributed residuals between observed and dependent variables, the dependent variables were transformed by natural logarithm before fitting the regression models.

In some cases, S -II end became negative. As this was an artefact of the fitting process, an arbitrary constant of 10 was added to all the fitted values prior to ln-transformation of S -II end. When using Fe III , the intercept of the regression slope for k obs , 20 varied according to the matrix.

This suggests that there might be an effect from the matrix, most likely due to the difference in COD concentrations. This was confirmed by a correlation analysis, which showed that k obs,20 had a statistically significant negative correlation with COD for the matrices of Fe III.

As discussed in Nielsen et al. These effects could possibly have consumed some of the added Fe III , which then would make it unavailable for sulfide precipitation, and hence ratio init would, in principle, be lower in these cases.

This corresponds to a k obs , 20 equal to the wastewater of Aalborg, but is still higher than the wastewaters of Frejlev and Svenstrup. From Figure 2 , it can be seen that the time to reach half of initial sulfide concentration at low pH and low ratio init is considerably faster using Fe III than when using Fe II.

From the correlation patterns of k obs , 20 and S -II end with pH and ratio init Figure 3 , it can be seen in Figure 3 a that k obs , 20 for Fe II exhibits a statistically significant positive correlation with both pH and ratio init. Fe III , on the other hand, does not show a significant correlation between k obs , 20 and ratio init , but only a significant positive correlation with pH Figure 3 b.

The correlation coefficients are color-coded from perfect positive correlation light grey to perfect negative correlation dark grey , and the correlation coefficients are stated for each combination. Even though only marginally affected, the correlation between these two parameters is statistically significant, whereas ratio init is statistically uncorrelated herewith. The decrease in effectiveness of sulfide precipitation observed at low pH is in line with Boon , who states that the effect of iron salts below pH 6.

The reason that Fe III yields good precipitation efficiency over the whole pH range, might be a combination of an additional 0. Findings of the present study are applicable for the water industry when designing sulfide abatement solutions or when using conceptual sewer process models for planning sulfide abatement strategies. In cases where end-of-pipe abatement is to be applied, knowledge about reaction kinetics is essential.

Based on this work, the required reaction times can be calculated to achieve proper abatement before discharge of sewage from a force main. This study furthermore shows that, when applying Fe II for sulfide abatement, both the amount of iron added and the pH is of significant importance, whereas for Fe III it is only pH that influences kinetics.

Sulfide precipitation rates were demonstrated to be faster using Fe III compared to using Fe II under anaerobic conditions when sulfide was present in the matrix before the addition of the iron salts.

It furthermore showed that, for Fe II , there was no statistically significant effect of the matrix on the rate constants. However, the specific compounds responsible for this observation were not determined.

For both iron salts, precipitation rates increased with increasing pH. The lack of correlation between k obs,20 and S -II end with ratio init for Fe III implies that sulfide control should be performed solely by adjusting pH while keeping ratio init close to stoichiometric requirements.

For Fe II , on the other hand, an increase in ratio init increased precipitation rates, and pH as well as ratio init should therefore be increased to obtain proper sulfide control. Model equations, based on experimental data, are proposed to predict rate constants k obs,20 and concentrations of sulfide left at equilibrium conditions S -II end , within the range of iron-to-sulfide ratios and pH used in this study.

These values can then be applied in Equation 10 to predict reaction times needed for proper sulfide control in end-of-pipe abatement applications of sewer force mains. Impact Factor 1. CiteScore 3. This paper is Open Access via a Subscribe to Open model. Individuals can help sustain this model by contributing the cost of what would have been author fees. Find out more here. Sign In or Create an Account.

Advanced Search. Sign In. Skip Nav Destination Article Navigation. Close mobile search navigation Article navigation. Volume 78, Issue 5. Previous Article Next Article. Article Navigation. Research Article August 29 Kinetics of sulfide precipitation with ferrous and ferric iron in wastewater Bruno Kiilerich ; Bruno Kiilerich. E-mail: bkiilerich grundfos. This Site. Google Scholar.

Jes Vollertsen Jes Vollertsen. Water Sci Technol 78 5 : — Article history Received:. Cite Icon Cite. Ferric iron is, on the other hand, proposed to have an additional reaction step, as it needs to be reduced to ferrous iron before it precipitates sulfide, as in Equation 1. This reduction can either take place chemically during the oxidation of bisulfide to elemental sulfur S 0 Equation 2 ASCE , or biochemically, where the oxidation power is utilized by a specific group of chemoautotrophic bacteria for respiration Hvitved-Jacobsen et al.

At each experimental run, the concentration of total dissolved sulfides was calculated based on the measured unionized sulfides and the concomitant pH value. A baseline of the total sulfide concentration S -II 0 was determined by root mean square error RMSE fitting of data acquired before ferrous or ferric iron were added. Promptly after the addition of iron, the sulfide concentration declined in a hyperbolic manner.

The initial stoichiometric ratio ratio init for each specific experiment was calculated according to Equation 3. Concentrations are used for the calculations, instead of activities of the species.

This is done as the ionic strength of the buffered water at initial conditions was 0. Wastewater from Frejlev was estimated to be in the range of 0. Precipitation of sulfide using Fe II can be described using rate Equation 4.

For precipitation of sulfides with Fe III , the rate equation Equation 5 is basically an extension of Equation 4 , as it consists of an initial oxidation step of sulfide to elemental sulfur according to Equation 2. For reaction times where off-line analysis is not an option, and the products of precipitation cannot be determined, sulfide precipitation must consequently be described in general terms Equation 6.

Consequently, the different reaction pathways for sulfide precipitation using ferrous and ferric iron, and the dependencies of iron concentrations must all be contained in an overall rate constant for the precipitation reaction. Amorphous FeS K sp 1. It precipitates readily in aqueous solutions and, despite the low solubility constant, it is in equilibrium with measurable levels of sulfide in the presence of the amorphous FeS Davison This corresponds to practical observations where dissolved sulfides could not be removed completely, even with iron-to-sulfide ratios above the stoichiometric requirement ASCE The rate equation can be solved analytically Equation 8.

Using the software R version 3. The nonlinear least square function modified to include the Levenberg-Marquardt type fitting algorithm nlsLM was used. This number outlines the time it takes before the sulfide concentration reaches half of the initial value Equation Table 1 Average values standard deviation of key parameters characterizing the wastewaters sampled for the experiments.

Alkalinity before adjusting pH to approx. View Large. Figure 1. View large Download slide. However, these two models exhibited a reversed trend compared to the models of the remaining groups of media. Scrutinizing the dataset for these series suggests that a single outlier in the datasets explained this effect.

It was hence decided to build the regression models, irrespective of the medium used for sulfide precipitation. As described earlier, an arbitrary constant of 10 was added to the parameter S -II end to perform ln-transformation of the dataset before multiple linear regression.

Upon backward transformation, this constant must consequently be subtracted again as done in Equations 12 and Figure 2. Figure 3. Septicity in sewers: causes, consequences and containment. Search ADS. Dansk Standard. The solubility of iron sulphides in synthetic and natural waters at ambient temperature. Control of sulfide in sewer systems by dosage of iron salts: comparison between theoretical and experimental results, and practical implications.

Chemical dosing for sulfide control in Australia: an industry survey. Dissolved oxygen in gravity sewers — measurement and simulation. Simultaneous online measurement of sulfide and nitrate in sewers for nitrate dosage optimisation.

The formation of iron II sulfides in aqueous solutions. The use of ferrous chloride to control dissolved sulfides in interceptor sewers. Sulfide precipitation in wastewater at short timescales. Chemical sulfide oxidation of wastewater — effects of pH and temperature. Sulfide-iron interactions in domestic wastewater from a gravity sewer.

Effects of iron on chemical sulfide oxidation in wastewater from sewer networks. Effects of pH and iron concentrations on sulfide precipitation in wastewater collection systems. Use of iron salts to control dissolved sulfide in trunk sewers. Iron reduction in activated sludge measured with different extraction techniques.