Introduction

Metallurgical, chemical and refractories industries discharge wastewater containing chromium (Cr) into the environment (Lyu et al. 2017, Miretzky & Cirelli 2010, Yang & Lee 2006), under its trivalent or hexavalent forms, bringing about severe biological and ecological damages (Kotaś & Stasicka 2000, Lyu et al. 2017). The toxicity of Cr(III) for fishes is reported at concentrations higher than 5.0 mg/L in water (Sari et al. 2008), while serious health concerns for human are associated with Cr(VI), a known carcinogenic for plants and animals (Miretzky & Cirelli 2010, Qiu et al. 2009). Also, reduced microbial growth is observed in the presence of Cr(VI) (Karthik et al. 2017, Wani et al. 2018), which affects soil fertility and decreases crop productivity (Dutton & Fisher 2011, Oves et al. 2013, Takahashi et al. 2012). In aqueous media, Cr(VI) is more soluble and mobile than Cr(III) (Dong et al. 2017). In order to preserve environment and ecosystems quality, Cr(VI) removal from aqueous media is crucial.

Electrochemical processes, ion exchange, membrane filtration, reverse osmosis, flotation and chemical precipitation are different methods employed for the removal of toxic metal ions. For the removal of Cr(VI) in dilute solutions, these methods are either expensive or troublesome. However, adsorption processes are economical, efficient, flexible and reversible; therefore, they are proving significant for chromium removal from wastewater (Kandile & Nasr 2014, Li et al. 2016, Sethy & Sahoo 2019, Sugumaran & Karim 2017).

The adsorption of numerous metal ions from wastes by chelating resins is well studied. An enhanced preconcentration factor at extensive pH ranges is observed when chelating agents are adsorbed onto a solid adsorbent (Jal et al. 2001). On the silica gel (SG) surface, silanol groups (Si–OH) are distributed throughout, providing an opportunity for ligand immobilization by chemical modification of its surface (Dey et al. 2006, Liang & Chen 2005). The stability, immobility, insolubility and high efficiency are key advantages of the covalent grafting of chelating molecules on silica surface (Jal et al. 2001).

Desferrioxamine B (DFO) belongs to the hydroxamate family of siderophores (Powell et al. 1980). It is remarkably avid for trivalent iron, finding its chief role in the treatment of iron overload disorders like thalassaemia, and aluminium poisoning (Aaseth et al. 2015). Many attempts were made to immobilize DFO on multiple supports to develop methods for the treatment of metals overload, biosensing or metal speciation (Alberti et al. 2014, Alberti et al. 2015, Su et al. 2011, Takagai et al. 2007, Wenk et al. 2001). Chelation therapy by another chelator, diethylenetriaminepentaacetic acid (DTPA), has also been extensively studied as a method for removing elements from the body (Parker et al. 1962, Slobodien et al. 1973). DTPA-modified silica gel and chitosan were reported for adsorption of certain metals including Cr(VI) (Bhatt et al. 2015, Repo et al. 2011). DFO immobilized in mesocellular silica foams for chelation of divalent and trivalent metal ions was reported (Schwabe et al. 2018). The stability, immobility, insolubility and high efficiency are key advantages of the covalent grafting of chelating molecules on silica surface (Jal et al. 2001).

In the present work, for the first time, DFO and DTPA were grafted on activated silica gel forming silica gel conjugates (SGC) IDFOSG and IDTPASG, respectively. The optimization of adsorption conditions for chromium species is described, along with their performances in multielemental solutions.

Materials and methods

Chemicals

Silica gel, Chelex 100® sodium form, hepes, ferrous ammonium sulphate and acetic acid were from Sigma-Aldrich. Diethylenetriamine pentaacetic acid was from Acros. Aminopropyltriethoxysilane (APTES) and hydrochloric acid were from Fluka. Sodium phosphate dibasic and potassium chloride were from Cromoline (Brazil). Sodium phosphate monobasic was from Reagen (Brazil). Potassium hydrogen phthalate, monopotassium phosphate, sodium hydroxide, nitric acid, acetone, toluene, ethyl alcohol, pyridine, acetic anhydride and sodium chloride were from Synth (Brazil). Calcein and sulphuric acid were from Vetec (Brazil). Glutaraldehyde was from Nuclear (Brazil). Desferrioxamine mesylate salt (Cristália, Brazil) was a gift from Associação Brasileira de Talassemia (ABRASTA). Cr(VI) atomic absorption standard was from SpecSol, and Na(I), Mn(II), Cu(II), Zn(II), Ba(II), Al(III), Fe(III) and Cr(III) standards containing 1000 mg L−1 were from Tritisol. Sodium carbonate was from Fisher Chemicals. The reagents were at least 99% pure and used without further purification. Ultrapure water obtained from a Milli-Q purification system (Millipore) or a OS10LXE reverse osmosis (Gehaka, Brazil) was used throughout the studies. Potassium phthalate buffer was prepared with 0.1 M potassium hydrogen phthalate and pH adjusted to 4. Sodium carbonate buffer was prepared with 0.1 M sodium carbonate, and pH adjusted to 10. Hepes buffered saline (HBS) was hepes 20 mM and NaCl 150 mM, washed with Chelex®100 (1 g/ 100 mL) with pH adjusted to 7 and 7.4.

Instruments

Cr determination in aqueous solutions was performed in a Vario 6 flame atomic absorption spectrometer (Analytik Jena, Germany). Instrument parameters employed include hollow cathode lamp of chromium with 357.9 nm wavelength, 5 mA applied current, 0.8 nm spectral band-pass and deuterium background correction, air flow of 400 L h−1, acetylene flow of 80 L h−1, as recommended by manufacturer. The analytical solutions containing SGC were mixed by using a shaker table (Quimis, Brazil). Bench centrifuges (Q222TM, Quimis and Eppendorf AG 22331) were used to separate the supernatant and SGC solids. All pH measurements were made using a digital pH meter (Gehaka PG1800), and mixing was made by using vortex mixer (Vixer Korea). An analytical balance (AUX220, Shimadzu) was used for weighing purposes.

Surface grafting of DFO and DTPA on SG

The schematics of the grafting are shown in Fig. 1. High-purity silica gel (SG), pore size 60 Å and 230–400 mesh size, was refluxed in 6 M HCl for 3 h. After an extensive washing with ultrapure water, a conventional furnace (EDG10P-S) was used for drying SG at 200 °C for 24 h. The aminopropylation of SG was carried out by adding 15 mL of 3-aminopropyltriethoxy silane (APTES) to 25 g of SG in 150 mL of toluene. Magnetic stirring at 200 rpm was maintained for 16 h. The reaction mixture was filtered to get aminopropyl silica gel (APSG). Several rounds of washing of APSG with toluene and acetone (100 mL each) were conducted before drying in oven at 65 °C for 24 h (Repo et al. 2009).

Fig. 1
figure 1

Scheme for the surface functionalization of DFO and DTPA on SG

Full size image

The surface grafting of DFO on SG was carried out by adding 2 g of APSG to a 2.5% glutaraldehyde solution (final concentration) in 25 mL of 0.1 M sodium phosphate buffer at pH 7. The reaction mixture was kept under continuous stirring for 2 h at room temperature. The product was filtered and washed with 250 mL of water and dried again at 65 °C for 24 h, producing activated APSG. One gram of DFO was mixed with 2 g of activated APSG in 10 mL of water (Ljunggren et al. 1992). The mixture was allowed to react for 4 h, and then filtered and washed with 100 mL of water. Surface grafted immobilized DFO on SG (IDFOSG) was dried at 65 °C for 24 h and kept in an airtight polypropylene vial.

DTPA anhydride was prepared by a modified version of the protocol by Tülü et al. (Tülü & Geckeler 1999). Ten grams of DTPA (25 mmol) were suspended in 20 mL of pyridine with subsequent addition of 20 mL of acetic anhydride and refluxed at 65 °C for 24 h. The product was filtered, washed with 25 mL of acetic anhydride and 25 mL of diethyl ether and dried in oven at 65 °C for 24 h. Three grams of APSG and 8 g of DTPA anhydride were added and 100 mL of 1:1 ethanol and acetic acid solution was prepared. A yellowish product formed, indicating the completion of reaction and 4 M NH4OH, 1:1 mixture of ethanol/acetic acid, water and methanol were used for extensive washing after filtration (Hughes & Rosenberg 2007). Surface grafted immobilized DTPA on SG (IDTPASG) was dried at 65 °C for 24 h and kept in an airtight polypropylene vial.

Estimation of DFO and DTPA load on SGC by competition studies with calcein

Upon coordination to iron, the fluorescence of calcein is stoichiometrically quenched and may be regenerated after treatment with other high affinity chelators such as DFO and DTPA (Baccan et al. 2012). Therefore, fluorescence recovery of a known solution of calcein-iron (CAFe), after treatment with IDFOSG or IDTPASG, is a quick method to quantify the amount of immobilized chelator. For this, calcein and ammonium ferrous sulphate in equimolar quantities were used to prepare 5 mM stock solutions of CAFe in HBS/Chelex. All the other CAFe solutions were also prepared in HBS/Chelex. In 1.5 mL vials, weighed amounts of SGC were transferred and treated with 1 mL of 100 μM CAFe. Free chelators DFO and DTPA at 100 μM were used as controls. The suspensions were vortexed for 2 min and centrifuged for 5 min at 4000 rpm at room temperature. Samples of the supernatants (100 μL) were then transferred to clear, flat bottom 96 well microplates, and their fluorescence was registered with a BMG FluoStar Optima instrument (λexc/λem = 485/520 nm).

Sorption optimization

In order to evaluate the effect of SGC on Cr uptake from aqueous solutions, it was conducted optimizations of different parameters: pH, SGC mass, shaking (contact) time and sample volume. The experiments were conducted in 15 mL polypropylene conical centrifuge tubes. Agitation was carried out by a Quimis table shaker. Centrifugation was performed in a Quimis Q222TM at 2790 rcf for 5 min at room temperature. The amount of Cr in the supernatants was quantified by atomic absorption flame spectroscopy. Uptake (α%) is defined as:

$$ \alpha \left(\%\right)=\left(\frac{c_i-{c}_e}{c_i}\right)\times 100 $$
(1)

where ci and ce are the initial and equilibrium concentrations of chromium, respectively (Carvalho et al. 2019).

  • pH optimization: Hydrochloric acid in potassium chloride was used for adjusting the pH to 2. Buffers of pH 4, 7 and 10 were made from potassium hydrogen phthalate, potassium phosphate and sodium carbonate, respectively. Volume of 3 mL of Cr solution (1 mg L−1) in these buffers were mixed with 50 mg of SGC.

  • SGC mass optimization: The effect of different masses of SGC (6.25–200 mg) on the uptake degree of 3 mL of 1 mgL−1 of Cr(VI) at pH 4 was studied.

  • Contact time optimization: The mixtures of 1 mg L−1 Cr(VI) (pH 4) and 100 mg of SGC were shaken for 5–1440 min.

  • Sample volume optimization: Samples (1–10 mL) of 1 mg L−1 of Cr(VI) at pH 4 reacted with 100 mg of SGC for 5 min.

Langmuir isotherms

Adsorption isotherms were determined in the range 0.5–2000 mg L−1 of Cr(VI) solutions treated with 100 mg of SGC in a 5 mL sample volume. Data were fitted with the function LangmuirEXT1 (Origin® 2016; OriginLab Corporation).

Effect of potentially interfering ions

The effect of various interfering ions on Cr(VI) sorption was analyzed in 5 mL of 1 mg L−1 of Cr(VI) solution at pH 4, by adding different amounts of Na(I), Mn(II), Cu(II), Zn(II), Ba(II), Al(III) and Fe(III) according to the ratios listed in Table S1. From the Ordinance 2914/2011 of the Brazilian Ministry of Health, the permissible limits of these ions and Cr(VI) were taken to measure the ratio X/Cr(VI) where X is the interfering ion (Brazil 2011).

The calibration curve for Cr(VI) at pH 4 was built from 0.25 to 2.5 mg L−1. The limit of detection (LOD) was estimated by (3 × SD)/a, where SD is the standard deviation of 10 measurements of absorbance of blank solution containing buffered media (pH 4), and a is the calibration curve slope. The limit of quantification (LOQ) was defined as (10 × SD)/a (Currie 1995).

Regeneration studies

The stability and regeneration were assessed by desorption experiments (Bhatt et al. 2015). For this, 400 mg of IDFOSG or IDTPASG were preloaded with 4 mg L−1 Cr(VI) at pH 4 in 20 mL sample volume. After filtering and washing with ultrapure water, 100 mg of the dry Cr-loaded IDFOGC were added to 5 mL of 0.1 M NaOH, H2SO4, HCl and HNO3 separately. After 5 min of table shaking and centrifugation, an aliquot of the supernatant was removed for the quantification of Cr and determination of the % desorption, defined as:

$$ \% desorption=\left(\frac{Cr\ in\ supernatant\ after\ desorption}{Cr\ in\ the\ supernatant\ after\ sorption\ onto\ SGC}\right)\times 100 $$
(2)

(Kumar & Rajesh 2013). The usability of SGC was tested by three successive sorption–desorption cycles.

Results and discussion

Chemical characterization of IDFOSG and IDTPASG

FTIR spectra are illustrated in Fig. 2 according to sequence of reactions for grafting of DFO and DTPA on SG. In SG and in its conjugates, the band around 1634, 1048, 970 and 795 cm−1 are assigned to Si–O–Si bond stretching vibrations which are related with condensed silica network (Bertoluzza et al. 1982, Jitianu et al. 2002, Li et al. 2008). The broad IR absorption band from 3500 to 3055 cm−1 is due to superimposed stretching modes of Si-OH groups and the hydroxyl groups of adsorbed water (Najafi et al. 2012).

Fig. 2
figure 2

FTIR spectra of the intermediates involved in grafting DFO or DTPA on SG

Full size image

The decrease in intensity of peaks between 3400 and 3100 and 970 and 1634 cm−1 and appearance of a less intense N–H bending vibration around 697 cm−1, which can be found in samples of APSG, confirm the incorporation of small amount of amino groups on silica gel surface (Li et al. 2008). The appearance of absorbance peaks around 1000–1200 cm−1 is due to C-N stretching vibrations (Shriner et al. 1948). However, its determination is made difficult by overlapping Si-O-Si stretches at 1000–1130 cm−1 and Si-CH2-R stretches at 1200–1250 cm−1 (Maria Chong & Zhao 2003, White & Tripp 2000). The absorption band at 1635 cm−1 in SG is shifted to 1560 cm−1 in APSG probably due to N-H vibrations.

DFO exhibits the same spectral hallmarks as previously described (Siebner-Freibach et al. 2005, Umemura et al. 2017). The bands at 2875 and 2929 cm−1are due to stretching vibrations of the CH2backbone of DFO. The vibrations typical of the hydroxamic group appear at 1619 cm−1 and are related to stretching vibrations of C=O. For the primary amine, two N–H stretches appear at 3098 and 3307 cm−1. The peak at 1160 cm−1 may be due to C-O or C-C stretching. The bands at 1459, 1160 and 1014 cm−1 are due to N-H, C-C bending vibrations and N-O stretching vibrations of the hydroxamate group, respectively. The appearance of peaks at 2929, 3308, 1623, 1566, 1459 and 1044 cm−1 in IDFOSG which were originally absent in SG and APSG and the retention of peaks due to Si-O-Si stretch vibration modes of the silica gel show the conjugation of DFO to the surface of APSG. The decrease in signal intensity at 3307 cm−1 and the appearance of a relatively intense signal at 1623 cm−1 indicate the formation of the Schiff’s base involving the amino group and the aldehyde group via glutaraldehyde cross-linking between DFO and APSG (Beppu et al. 2007). The absorption of the amide of Schiff’s base is probably overlapping the amide band of DFO. The mixture containing DFO and SG has peaks at 1047, 795, 1640, 2939 and 3371 cm−1 which are mostly related to SG IR spectrum.

The spectrum of chromium-loaded IDFOSG showed a peak for at 691 cm−1 (Supplementary Fig. S1) attributed to the Cr–O stretch (Bhatt et al. 2015) showing interaction of hydrogen chromate (HCrO4−, the main Cr(VI) species at pH 4 (Michael Sperling 1992, Wu et al. 2009)) with the surface grafted DFO molecule (Fig. S1). On comparing with the metal-loaded sample, it is observed that there is a shift in wave number of dominant peaks due to hydroxamate group binding to Cr(VI) or its adsorption on the surface of IDFOSG. After Cr(VI) loading, a peak at 1460 cm−1 in IDFOSG was displaced to 1554 cm−1, and low intensity narrow peaks at 2929, 2858, 3103 and 3308 cm−1 converged into a broad intense peak around 3366 cm−1 probably due to either to H–O–H stretching vibration bands of adsorbed water molecules, or to the presence of hydrogen chromate (HCrO4−).

ITDPASG was similarly characterized (Fig. 2). Free DTPA exhibits reported carboxylic acid O–H stretch at 3007 cm−1 and bands at 1687 and 1631 cm−1, corresponding to the C = O bending in −COOH. An appearance of an amide C = O stretch at 1624 cm−1 in IDTPASG which is originally absent in APSG shows the formation of amide bond between APSG and DTPA. Beppu et al. (Beppu et al. 2007) reported that this type of linkage showed absorption at 1655 cm−1. The IR spectrum of chromium-loaded IDTPASG showed the same pattern of shifting of wave number of dominant peaks associated with IDFOSG.

Estimation of DFO and DTPA load on SG by competition studies with calcein

The extent of surface functionalization of SG with DFO and DTPA was assessed by the regeneration of calcein fluorescence from ferric calcein, CAFe (Fig. 3). The quenching of calcein fluorescence by iron is stoichiometrical and reversible upon reaction with chelators with higher affinity for iron (e.g., DFO and DTPA); therefore, fluorescence emission from CAFe interacting with SGC can be used to quantify the chelator load of SGC. Initially, as a proof of concept, it was verified that a 100-mg sample of either IDFOSG or IDTPASG (but not untreated SG) was able to regenerate calcein fluorescence (Supplementary Fig. S2), indicating that the binding properties of DFO and DTPA were not affected after conjugation. Then, the dose-dependent relation between regeneration of fluorescence and concentration of conjugated molecules was found to be linear in the range of 1.56 to 50 mg for IDFOSG and IDTPASG (Fig. 3a). The molar amounts of DFO and DTPA per mass unit of SG were calculated using calibration curves relating chelator concentration to fluorescence increase (Supplementary Fig. S3), followed by normalization to the mass of SGC (Fig. 3b). The average DFO and DTPA load in SG was 0.611 ± 0.011 and 3.05 ± 0.43 nmol mg−1, respectively. The DTPA load was greater than DFO per mass unit of SG as the initial reaction conditions were slightly different, with higher chelator/APSG ratio used in the synthesis of IDTPASG.

Fig. 3
figure 3

a Fluorescence recovery of CAFe upon iron sequestration by IDFOSG and IDTPASG. b Chelator load of SGC. a.u.: arbitrary fluorescence units

Full size image

Sorption optimization

The influence of pH on Cr(VI) and Cr(III) uptake by SGC is shown in Fig. 4 a. The complexation of several metal ions with DFO has been reported (Alta et al. 2014, Darr et al. 1987, Domingo 2006, Duckworth & Sposito 2005, Gabizon et al. 1988, Randal J. Keller 1991); however, the complexation of Cr(VI) is less frequently discussed (Christoffersen et al. 1987, Ke Jian Liu 1997, Xianglin Shi 1992). At pH < 7, the coordination sites of both IDTPASG and IDFOSG are protonated; therefore, Cr(III) is excluded from binding. This changes in neutral-alkaline pH, because of ligand deprotonation; however, Cr(III) also precipitates as oxides or hydroxides at pH > 7 (Rai et al. 1987). At pH 2–4, Cr(VI) exists mainly (66%) as HCrO4− and also (34%) as Cr2O72−. These proportions shift to mainly CrO42− at pH 10. Since maximum Cr(VI) binding was found at acidic pH when most binding sites are protonated and therefore without charge, it is probable that deprotonation (with increase in the negative charge of the SGC) will not favour the binding of the highly negative CrO42− ion.

Fig. 4
figure 4

Sorption optimization experiments. a Effect of pH on uptake degree of Cr(III) and Cr(VI) by SGC. b Effect of mass on the uptake (α%) of Cr(VI) by IDFOSG, IDTPASG and SG at pH = 4 (n = 3). c Effect of contact time on the uptake (α%) of Cr(VI) (3 mL; 1 mg L−1; pH 4) by 100 mg of SGC (n = 3). d Effect of sample volume on the uptake (α%) of Cr(VI) (1 mg L−1; pH 4) by 100 mg of SGC with shaking time of 5 min (n = 3)

Full size image

Therefore, Cr(VI) and pH = 4 were selected for further studies because at this condition, the uptake degree of hexavalent chromium by IDFOSG was ~ 88% and ~ 75% for IDTPASG.

The effect of different masses of IDFOSG and DTPASG on the uptake degree of Cr(VI) (3 mL; 1 mg L−1; pH 4) was investigated to optimize its value. It was observed that up to 100 mg of SGC, the uptake of Cr(VI) increases, reflecting the increase of interactive sites for Cr(VI) on the surface and then levelling off (Fig. 4b). Less than 10% Cr(VI) uptake was observed for the reference material (SG). Hence, a mass of 100 mg was selected for further studies.

The effect of contact time is an important parameter to optimize adsorbent performances. The interaction of Cr(VI) (3 mL; 1 mg L−1; pH 4) with 100 mg of SGC up to 1440 min of table shaking was determined (Fig. 4c). Two trends were observed. Firstly, no marked change in behaviour was observed in case of IDFOSG. The decrease in Cr(VI) uptake degree was found to be ~ 10% at 1440 min, indicating the stronger interaction of Cr(VI) on surface of IDFOSG and no considerable occurrence of desorption. Secondly, in case of IDTPASG, a steady decrease in Cr(VI) uptake was observed and at the end of the experiment (46% from the original amount), suggesting that this adsorbent is less stable than IDFOSG and some leaching occurred. Considering the results, a contact (shaking) time of 5 min was selected for further experiments.

Optimal sample volume for Cr(VI) uptake was determined by varying the sample volume over a range of 1–10 mL of 1 mg L−1 solutions at the already optimized conditions. From Fig. 4 d, samples of either 3 or 5 mL under 100 mg of SGC have higher uptake degree of Cr(VI), without significant differences. Further increase in sample volume showed a slight trend of decreased Cr(VI) uptake degree, reflecting possibly the need of a longer contact time for bigger samples. The 5 mL-size aliquot was chosen for further experiments.

Effect of potentially interfering ions

The interferences of ions such as Na(I), Mn(II), Cu(II), Zn(II), Ba(II), Al(III) and Fe(III) individually and in a multielemental solution were examined in 5 mL samples containing 1 mg L−1 Cr(VI) (Table 1). Cr(VI) uptake degree was not influenced by Na(I) even at a × 4000 excess over Cr(VI). However, competition was important for divalent ions, and attained a maximum of 13% decrease by trivalent ions and multielemental solution. This is due to two factors. Firstly, the trivalent metal ions (which are harder acids) form much stronger complexes with the hard bases DFO and DTPA than do the divalent metal ions, indicating the significantly more ionic character of the bonds (Kiss & Farkas 1998). In any case, recovery greater than 75% was obtained in every situation, even in multielemental solution. These findings demonstrate that there is a moderate tolerance level towards these ions, and SGC can best serve for Cr removal in an elevated electrolytic environment under optimized conditions.

Table 1 Effect of interfering ions on Cr(VI) (1 mgL−1; pH 4) uptake degree (α%) by IDFOSG and IDTPASG (100 mg)
Full size table

Sorption isotherms

The sorption data for SGC were fitted using the Langmuir monolayer adsorption model (Fig. 5). Cr(VI) uptake degree range was 0.5–2000 mg L−1, and metal concentration in the supernatants was calculated from a calibration curve of Cr(VI) (y = 0.0926 x – 0.0015; R2 = 0.9992). The LOD and LOQ of the method calculated by FAAS were 0.01 mg L−1 and 0.05 mg L−1, respectively. In the Langmuir equation (Eq. 3):

$$ {q}_e=\frac{q_{max}\times {K}_L}{1+\left({K}_L\times {C}_e\right)} $$
(3)
Fig. 5
figure 5

Langmuir adsorption isotherms for hexavalent chromium onto IDFOSG or IDTPASG (100 mg) at pH 4

Full size image

Ce, qe and qmax are equilibrium concentration (mg L−1), adsorption at equilibrium (mg g−1) and maximum adsorption, respectively (Foo & Hameed 2010). KL is the adsorption equilibrium constant which serves as an indication of the relative strength of the adsorption process. The isotherm constants are presented in Table 2. A simple monolayer cover of SGC assisted by the presence of chelators describes satisfactorily the Cr(VI) adsorption behaviour, without significant differences in the affinities and an increase of capacity in the case of IDFOSG.

Table 2 Isotherm constants for IDFOSG and IDTPASG adsorption
Full size table

Our materials compare favourably with other low surface area materials such as agricultural residues (qmax ~ 0.6 mg g−1) and are on par with some nanomaterials (qmax ~ 15 mg g−1). In contrast, synthetic polymers have displayed qmax > 100 mg g−1 for chromium species (Pakade et al. 2019).

Stability and regeneration

To evaluate stability and regeneration of SGC, desorption experiments were performed. It is observed from Fig. 6 that maximum Cr(VI) desorption for IDFOSG or IDTPASG happened with 0.1 M NaOH solution. This is probably due to the changes in charge and concentration of Cr(VI) species as discussed above, resulting in the electrostatic repulsion as well as decreasing levels of HCrO4− for sorption. This can also be due to interference of OH− ions competing with chromate. The net result is the weakening of the interaction of Cr(VI) due to unavailability of active sites for interaction as previously discussed (Bhatt et al. 2015).

Fig. 6
figure 6

Desorption studies of 100 mg IDFOSG and IDTPASG preloaded with 1 mgL−1 Cr(VI) and equilibrated with 0.1 M solutions of NaOH, H2SO4, HNO3 or HCl

Full size image

Three repetitive cycles of sorption-desorption were carried out to measure the capacity of regeneration and stability of both IDFOSG and IDTPASG, and a decrease in percent uptake of Cr(VI) of ~ 17% and 35% was observed for IDFOSG and IDTPASG respectively (Fig. 7). It is obvious from the results that the binding performance of IDFOSG was less affected when compared to IDTPASG after successive cycles; hence, IDFOSG is better suited to remove Cr(VI) from aqueous solutions.

Fig. 7
figure 7

Effect of sorption-desorption repetitive cycles on the uptake degree of Cr(VI) by 100 mg IDFOSG and IDTPASG preloaded with 1 mgL−1 Cr(VI)

Full size image

Conclusion

The surface-grafted siderophore deferrioxamine (DFO) on silica gel had a generally better performance than its DTPA counterpart as a chromium adsorbent. SGC in general and IDFOSG in particular are promising sorption materials for the remediation of chromium pollution.