Biosequestration of chromium(III) in an aqueous solution using cationic and anionic biosurfactants produced from two different Bacillus sp. – a comparative study

Abstract

Tannery wastewater discharged from chrome tanning section contains Cr(III) in the range of 2100–2300 ppm, and a viable technique for its removal remains a great concern for the leather industry. The ability of a biosurfactant to chelate toxic heavy metal ions and form an insoluble precipitate may be exploited in the treatment of Cr(III) containing wastewater. In the present study, a biosurfactant producing microorganisms, Bacillus subtilis and Bacillus cereus, were isolated from tannery wastewater-contaminated soil using palm oil and coconut oil as the substrates. The biosurfactants produced from palm oil (palm oil BS) and coconut oil (coconut oil BS) were characterized as anionic and cationic biosurfactants, respectively, using the blue agar plate method, agar double diffusion technique and zeta potential measurement. The biosurfactants were characterized for their amino acid composition and elemental (CHNS) composition. The thermal behavior of the biosurfactant was characterized by TGA and DSC. The surface tension of the anionic and cationic biosurfactants were 28.16 ± 0.2 mN m⁻¹ and 23.02 ± 0.2 mN m⁻¹, respectively. The SDS-PAGE and FT-IR analyses confirmed that both the biosurfactants were lipoproteins in nature. The binding ability of the lipoprotein anionic and lipoprotein cationic BS with chromium (Cr(III)) ions in an aqueous solution was then determined. The interaction of Cr(III) with BS was confirmed using FT-IR, SEM-EDX analysis and atomic absorption spectrophotometry (AAS).

---

1. Introduction

The leather manufacturing industry converts putrescible raw skins and hides into non putrescible leather through various chemical and mechanical processes.¹ Basic chromium sulphate is the most common tanning agent used in the leather industry because it enables the faster and cheaper production of highly microbial resistant and durable leather. More than 80% of finished leather goods are tanned using basic chromium sulphate.² The amount of trivalent chromium discharged in the effluent, when processing one ton of wet salted skins/hides, is in the range of 3–7 kg or 5–10 kg Cr₂O_3.³ According to the available reports,⁴ the recommended level for Cr(III) is in the range of 0.5–2.0 ppm regardless of fresh, marine, irrigation and drinking water. In many countries, the treated tannery wastewater is discharged onto the open land or into rivers.⁵

Cr(III) tends to be strongly bound by soil humic acid polymers, and this affinity restricts the availability of Cr(III) to be oxidized and reduces the decomposition of organic matter.⁶ Cr(III) in soils could be leached into surface water or groundwater and be absorbed by plants.⁵ Chromium(III) enters the human food chain through the consumption of the vegetative plants. The symptoms of Cr(III) phytotoxicity include inhibition of seed germination or of early seedling development, reduction of root growth, leaf chlorosis and depressed biomass.⁷

Tannery wastewater from the chrome tanning section contains Cr(III) in the range of 2100–2300 ppm.⁸ The improper treatment of Cr(III) containing effluent, non scientific method of storage of chromium tanned solid waste and the leaching of Cr(III) onto the soil can release Cr(III) to the environment, causing groundwater contamination and adverse biological and ecological effects.⁹ Precipitated Cr(III) hydroxides remain stable in the sediments under aerobic conditions, whereas under acidic and anoxic conditions, Cr(III) hydroxides are soluble and remain as ionic Cr(III) species,¹⁰ which can leach into the groundwater sources.

Cleaner technologies used to reduce Cr(III) in wastewater, such as high exhaustion process and direct or indirect chromium recycling, cannot eliminate Cr(III) completely from tannery wastewater.¹¹ Among the techniques known for the removal of chromium from tannery wastewater, the biosequestration of Cr(III) is one of the proven technologies.

Biosurfactants are a structurally diverse group of surface-active substances produced by microorganisms. They consist of a polar hydrophilic moiety and a non-polar hydrophobic group. The hydrophilic group consists of mono, oligo or polysaccharides, peptides or proteins, and the hydrophobic moiety contains saturated, unsaturated and hydroxylated fatty acids or fatty alcohols.^12,13 The six major types of biosurfactants available are hydroxylated and cross linked fatty acids (mycolic acids), glycolipids, lipopolysaccharides, lipoproteins and phospholipids.¹⁴ Biosurfactants are used in the remediation of pollutants due to their surface active properties.¹⁵

The chemical configuration of biosurfactants allows them to bind with the metal ions. The characteristic feature of biosurfactants to chelate toxic heavy metals and form an insoluble precipitate may be exploited in the treatment of heavy metal containing wastewater.¹⁶ Biosurfactants of anionic nature can capture metal ions through electrostatic interactions or complexation. Cationic biosurfactants can replace the same charged metal ions by competition.¹⁷ To the best of our knowledge, there are no reports on the removal of Cr(III) using cationic biosurfactants.

The focal theme of the present investigation was to compare the Cr(III) binding ability of anionic and cationic biosurfactants (BS) produced from two different substrates such as palm oil and coconut oil.

2. Materials and methods

2.1. Isolation of microorganisms

Biosurfactants producing microorganisms were isolated from tannery wastewater contaminated soil. The soil was acclimatized with palm oil or coconut oil and serially diluted to isolate the microorganisms using nutrient agar by the pour plate method, followed by incubation for 24–48 h at 35 °C.

2.2. Screening and identification of biosurfactant producing microorganisms

The surface active properties of the biosurfactant produced by the isolated bacterial strains were screened by their oil drop collapse activity¹⁷ using four different oils, namely, palm oil, diesel oil, coconut oil, and olive oil. The organisms screened from palm oil and coconut oil acclimatized soil were identified by 16S ribosomal DNA (16S rDNA) sequencing and phylogenetic analyses.

2.3. Optimization studies for the production of biosurfactants

The various parameters that significantly affect the production of biosurfactants using palm oil and coconut oil as substrates, such as time (24–120 h), pH (1–10), temperature (20–50 °C) and concentration of substrate (10–60 g L⁻¹), were optimized using one parameter at a time while the other parameters were kept constant. The most significant range of parameters was further optimized using Response Surface Methodology (RSM) for the production of the biosurfactant.

2.4. Production, extraction and purification of biosurfactants

The production of BS was carried out under optimized conditions and the bacterial cells were removed by centrifugation at 10 000 rpm for 20 min at 4 °C. The cell free supernatant was adjusted using 2 N HCl to obtain the final pH to 2.0 and kept for overnight at 4 °C. The pellet was collected by centrifugation at 9000 rpm for 20 min at 4 °C and the biosurfactant was extracted with acetone. The extracted biosurfactant was purified further using silica gel column chromatography to obtain the purified biosurfactant.

2.5. Characterization of biosurfactant

2.5.1. Determination of ionic character of BS.
Blue agar plate method. The ionic nature of biosurfactant was characterized by a slight modification of the blue agar plate method.¹⁸ Mineral salt agar medium supplemented with glucose as the carbon source (2%) and methylene blue (MB: 0.2 mg mL⁻¹) were used for the preparation of blue agar plate. The anionic and cationic nature of BS was identified by adding cetyl trimethyl ammonium bromide (CTAB: 0.5 mg mL⁻¹) and sodium dodecyl sulphate (SDS: 0.5 mg mL⁻¹) to the respective media. Biosurfactant (50 μL) containing 10 μg was loaded into each well made in blue agar plate and the plates were incubated at 37 °C for 48–72 h. A dark blue halo zone was considered as the positive result for anionic and cationic biosurfactants in their corresponding plates.
Agar double diffusion technique. The ionic character of BS was determined using the agar double diffusion technique.¹⁹ In an agar plate (1% agar of low hardness), two regularly-spaced rows of wells were made. The purified BS of volume 100 μL containing 20 μg was filled in the wells in the lower row, whereas the upper row of the well was filled with a pure compound (SDS or CTAB) of a known ionic charge. The anionic and cationic nature of BS was identified by adding CTAB (0.5 mg mL⁻¹) and SDS (0.5 mg mL⁻¹) in the wells of the upper row. The appearance of precipitation lines between the corresponding wells is considered to be due to the ionic nature of the biosurfactant.

2.5.2. Zeta potential of biosurfactants. The zeta potential of the biosurfactants was determined by mixing with 5 mL of buffer solution and the solution was agitated for 30 min using a magnetic stirrer. The biosurfactants were centrifuged and re-dispersed in 5 mL of pure water and agitated once again for 30 min. Water was used as the dispersant for the particles to measure their zeta potential using a Zetasizer (Nano-ZS from Malvern Instruments, UK).

2.5.3. Emulsification stability index (E24%). The emulsification stability of the biosurfactant was determined in accordance with the method followed by Cooper and Goldenberg.²⁰ Palm oil, olive oil, diesel oil or coconut oil of volume 2 mL was added to the same amount of the cell broth, mixed using a vortex mixer for 2 min and left to stand for 24 h.

2.5.4. Surface tension measurement. The most important property of surfactants is to decrease the interfacial tension of water. BS also reduces the surface tension of water similar to a chemical surfactant. The biosurfactant containing fermentation broth was centrifuged at 10 000 rpm for 20 min and then the surface tension of the supernatant was measured.²¹ Surface tension measurements were recorded by the Wilhelmy Plate method using a Surface Tensiometer (NIMA Technologies Ltd., England).

2.5.5. Biochemical characterization of biosurfactants. The biochemical composition of BS, such as carbohydrate, protein and lipid contents, were estimated using the phenol-sulphuric acid method,²² Lowry's method²³ and phosphoric acid vanillin reagent method,²⁴ respectively.

2.6. Amino acid composition of biosurfactant by HPLC

The amino acid composition of the biosurfactant was determined by HPLC. The biosurfactant was hydrolyzed with 6 N HCl at 100 °C for 20 h and neutralized with 1 M NaOH. The amino acid composition was analyzed using an Agilent 1100 HPLC amino acid analyzer (Agilent Technologies, Middleburg, Netherlands) and a data analysis was performed using the HP chemstation.²⁵

2.7. Molecular weight determination of biosurfactant

The molecular mass of the biosurfactant was determined by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE), in accordance with the method of Laemmli,²⁶ on a 5% stacking gel and 12% resolving gel. The protein marker ranging from 14.3 to 94.7 KDa was used as a standard marker for the determination of molecular weight of the biosurfactant.

2.8. Instrumental characterization of biosurfactant

2.8.1. Elemental composition of biosurfactants. The elemental composition (carbon, hydrogen, nitrogen and sulphur) of the biosurfactants was determined using a CHNS analyser (Euro vector EA 3000 series).

2.8.2. Thermal behaviour of biosurfactants (TGA and DSC). The required quantity (8–10 mg) of BS was loaded onto a platinum TGA pan and gravimetric analysis was carried out under a pure nitrogen atmosphere from 0 °C to 800 °C using a temperature gradient of 10 °C min⁻¹. The thermograms were recorded routinely in duplicate using TGA Universal V4.4A TA instruments.

For DSC analysis, the BS samples (8–10 mg) were loaded in an aluminum DSC pan and gravimetric analysis was performed under a nitrogen atmosphere from 0 °C to 200 °C using a temperature gradient of 10 °C min⁻¹. The scans were routinely recorded in duplicate using a DSC Q200 (V23.10 Build 79).

2.8.3. Circular dichroism (CD) spectroscopy. Circular dichroism was used to study the secondary structures of BS in the range of 150–250 nm with a path length of 1 mm and width of 1 nm using a Peltier temperature controlled system (JASCO J715, Japan).

2.9. Interaction of chromium(III) ions with biosurfactants

The biosequestration of chromium(III) ions from an aqueous solution was confirmed by atomic absorption spectroscopy analysis (AAS). The effect of time (2–12 h), pH (4–6), different concentrations of anionic and cationic biosurfactants (0.5, 1.0, 1.5, 2.0 and 2.5 g) and concentration of Cr(III) (100–500 mg L⁻¹) were studied. The time study was performed using 0.5 g anionic and cationic BS for the removal of Cr(III) from the aqueous solution containing 100 mg L⁻¹ Cr(III). Under the optimized conditions, the concentration of Cr(III) before and after sequestration by biosurfactants were measured using an atomic absorption spectrophotometer (AAS).
C₀ = initial concentration of Cr(III) in mg L⁻¹; C_t = final concentration of Cr(III) in mg L⁻¹.

2.10. UV-visible spectroscopy

The UV-visible scans of cationic BS, anionic BS and BS after the interaction with Cr(III) were recorded in the range of λ_200–800 nm using a UV-visible spectrophotometer (Cary varion; Agilent Technologies, Middleburg, Netherlands).

2.11. Fluorescence spectroscopy

The fluorescence spectra of cationic BS, anionic BS and BS after the interaction with Cr(III) were recorded using a fluorescence spectrophotometer in the wavelength range of λ_200–800 nm (Cary Eclipse; Agilent Technologies, Middleburg, Netherlands).

2.12. FT-IR spectral analysis

The functional groups of anionic and cationic lipoprotein BS, before and after treating with Cr(III), were characterized using a FT-IR spectrophotometer (Perkin Elmer). The samples were dried and made in the form of pellet with dimensions thickness of 1 mm and a diameter of 13 mm using spectroscopic grade KBr. The spectra were obtained in the spectral range of 400–4000 cm⁻¹.

2.13. SEM analysis

The binding of chromium ions with BS was further confirmed by SEM analysis. The biosurfactant and chromium bound biosurfactant [BS-Cr(III)] samples were coated with a gold foil 120–130 μm in thickness, under an argon atmosphere. SEM images were recorded on a scanning device attached to a JEOL JM-5600 electron microscope at 20 kV accelerating voltage with a 5–6 nm electron beam.

2.14. EDX analysis

The biosurfactant and chromium bound biosurfactant [BS-Cr(III)] samples were coated with a gold foil, 120–130 μm in thickness under an argon atmosphere. The EDX spectra were obtained on a scanning device attached to a JEOL JM-5600 electron microscope at a 15 kV accelerating voltage with a 5–6 nm electron beam.

3. Results and discussion

3.1. Isolation, screening and identification of biosurfactant producing microorganism

The enrichment of bacteria isolated from the tannery wastewater contaminated soil was carried out in 100 mL M9 Minimal media (HiMedia) with palm oil or coconut oil (1 g) as the sole carbon source. The palm/coconut oil concentration in the media was increased from 1 to 5 g with an incremental increase of 1 g. The media was acclimatized for 4–5 weeks and the microbes were then serially diluted (10⁻¹ to 10⁻¹⁰). The biosurfactant producing microorganisms with a non-identical morphology were isolated from palm oil or coconut oil acclimatized soil. About 6 biosurfactant producing microorganisms from palm oil as the substrate and 4 microorganisms from coconut oil as the substrate were isolated. Based on their zone diameter in the oil drop collapse assay (Table 1), microorganisms with high biosurfactant activity (P3 & C2) were used for further studies. The screened microorganisms were identified as Bacillus sp., using 16s rDNA gene sequencing. Phylogenetic analysis shows the evolutionary relationships among various other Bacillus species based upon the similarities and differences in their physical or genetic characteristics. The phylogenetic trees for the identified organisms are presented in Fig. 1 and confirmed that they belong to Bacillus subtilis and Bacillus cereus.

Table 1 Screening of microorganisms by the oil drop collapse assay

Palm oil		Coconut oil
Isolated microorganisms	Zone diameter (cm)	Isolated microorganisms	Zone diameter (cm)
P1	0.9 ± 0.5	C1	1.5 ± 0.4
P2	1.2 ± 0.4	C2	3.4 ± 0.15
P3	2.8 ± 0.3	C3	0.6 ± 0.08
P4	0.6 ± 0.2	C4	1.8 ± 0.21
P5	1.6 ± 0.4
P6	1.3 ± 0.3

---

Fig. 1 Phylogenetic tree for the organisms identified from (a) palm oil substrate, (b) coconut oil substrate.

3.2. Statistical optimization for the production of biosurfactant

The production of biosurfactants from Bacillus sp. was studied in the batch mode. The process parameters, such as time (24–120 h), pH (1–10), temperature (20–50 °C) and concentration of substrate (10–60 g L⁻¹), were optimized using one parameter at a time. The most significant range of parameters was further optimized using RSM (Refer appendix).

RSM clearly states that the biosurfactant production from B. subtilis and B. cereus was significantly (p < 0.05) influenced by pH, temperature and concentration of oil. Nonlinear relationships were significantly (p < 0.05) fitted to the experimental data for describing the changes in biosurfactant yield, when all the experimental variables were simultaneously altered. Among the response variable effects, the interaction effect of pH and palm oil concentration had the most significant (p < 0.05) positive effect on the biosurfactant production from B. subtilis, whereas the interaction effect of temperature and coconut oil concentration had the most significant (p < 0.05) positive effect on the biosurfactant production from B. cereus.

Bacillus subtilis produced BS using 0.32 g per gram of palm oil (12.8 g L⁻¹) as the substrate under the optimized conditions: time, 96 h; pH, 5; temperature, 30 °C; and concentration of palm oil, 40 g L⁻¹. The production of rhamnolipid biosurfactant using palm oil was reported by Thaniyavarn et al. (2.91 g L⁻¹)¹⁶ and by Nawawi (85 g L⁻¹) using palm oil sludge.²⁷ Similarly, Bacillus cereus yielded 0.45 g of BS per gram of coconut oil (22.5g L⁻¹) as the substrate under the optimized conditions: time, 96 h; pH, 7; temperature, 40 °C; and concentration of coconut oil, 50 g L⁻¹. The rhamnolipid biosurfactant produced using coconut oil was reported by Thaniyavarn et al. (2.93 g L⁻¹)¹⁶ and Patil et al. (2.8 g L⁻¹).²⁸ Kannahi and Sherley²⁹ also reported rhamnolipid biosurfactant of 44.2 g L⁻¹ using mixed oil containing 2% coconut oil.

3.2.1. Activation energy. The activation energy required by B. subtilis and B. cereus for BS production using palm oil and coconut oil, respectively, was calculated from the Arrhenius equation
where k is the rate constant, A is the Arrhenius factor and E_a is the activation energy. The activation energy for the BS production from B. subtilis and B. cereus was calculated from a plot of ln k versus 1/T. The activation energy (E_a) for the production of BS from B. subtilis and B. cereus was found to be 80.59 kJ mol⁻¹ and 87.95 kJ mol⁻¹, respectively. The Arrhenius factor was observed to be 6.88 × 10¹¹ and 1.46 × 10¹³ h⁻¹ for the production of BS from B. subtilis and B. cereus, respectively.

3.3. Characterization of biosurfactant

3.3.1. Determination of ionic character of BS.
Blue agar plate method. The anionic and cationic BS were characterized by the formation of insoluble ion pair precipitates in an agar plate containing methylene blue exhibited in dark blue color against the light blue background. The biosurfactants produced from palm oil and coconut oil as the substrates exhibited a dark blue halo zone in CTAB (positively charged surfactant) and in SDS (negatively charged surfactant) containing plates, respectively (Fig. 2(I)). These results confirm that the biosurfactant from the palm oil substrate was anionic in nature and the biosurfactant from the coconut oil substrate was cationic in nature.³⁰

Fig. 2 (I) Blue agar plate method (a) CTAB plate (anionic BS) (b) SDS plate (cationic BS); (II) double diffusion on Agar (a) CTAB (b) SDS.

Double diffusion agar method. Agar double diffusion tests are based on the passive diffusion of two compounds bearing similar charges or opposite charges in a weakly concentrated gel. Fig. 2(II)a suggests that the precipitation lines were formed between the BS from palm oil and the cationic compound (CTAB), whereas no precipitation was found with BS from coconut oil. Similarly, the precipitation lines were observed between BS from the coconut oil and the anionic compound (SDS), while no precipitation line was observed between SDS and BS from palm oil (Fig. 2(II)b). Therefore, the results confirm that the biosurfactants produced from B. subtilis and B. cereus were anionic and cationic in nature, respectively.

3.3.2. Zeta potential measurement of biosurfactants

Fig. 3 shows that the zeta potential of the coconut oil BS was +54.7 mV and palm oil BS was −33.01 mV. The zeta potential suggests the polarised nature of the surfactants and the degree of polarization. Furthermore, because of the high positive zeta potential, the cationic BS can be expected to enhance the exchange of Cr(III) ions from the aqueous solution compared to the anionic BS.

Fig. 3 Zeta potential of (a) anionic BS, (b) cationic BS.

3.3.3. Emulsification stability index (E24%). The emulsification stability indices of the anionic and cationic biosurfactants were evaluated by determining the emulsifying activity with different hydrocarbons. Both the biosurfactants exhibited different stabilization properties with the hydrocarbons tested as expressed in terms of the emulsification stability index at 24 h.³¹ BS produced from the coconut oil substrate exhibited better emulsification activity than the palm oil BS, as shown in Table 2.

Table 2 Emulsification stability index of the biosurfactant with different hydrocarbons

Hydrocarbons	Emulsification stability index, E24 (%)
	Palm oil BS	Coconut oil BS
Palm oil	35 ± 1.2	40 ± 1.5
Coconut oil	30 ± 1.0	40 ± 1.0
Olive oil	22 ± 1.8	35 ± 1.5
Diesel oil	35 ± 1.5	38 ± 1.8

---

3.3.4. Surface tension measurement. A good surfactant can reduce the surface tension of water from 75 to 35 mN m⁻¹.³² The cationic and anionic BS were tested for their surface activity. The surface tension of aqueous solutions containing cationic and anionic BS was 28.16 ± 0.2 mN m⁻¹ and 23.02 ± 0.2 mN m⁻¹, respectively. Therefore, the lipoprotein BS reported in this investigation could be classified under efficient and effective surfactants, as per the norms suggested by Kim et al.³³

3.3.5. Biochemical assays of BS. The carbohydrate, lipid and protein contents of cationic BS were 25 mg g⁻¹, 157 mg g⁻¹ and 630 mg g⁻¹, respectively. On the other hand, the anionic BS contained comparatively less amount of carbohydrate, lipid and protein, 30 mg g⁻¹, 270 mg g⁻¹ and 600 mg g⁻¹, respectively. The biochemical assays suggest that BS were lipoproteins in nature. This is in accordance with the study by Nguyen et al.,³⁴ in which the hydrophilic part of cationic BS contains aminoacids or peptides.

3.4. Determination of amino acid composition of BS by HPLC

The amino acid composition showed that the anionic lipoprotein BS contained about 61.5% of polar amino acids and 38.5% of non-polar amino acids. The cationic lipoprotein BS contained 53% polar amino acids and 47% non-polar amino acids (Table 3).

Table 3 Amino acid composition of the biosurfactant from Bacillus sp

Amino acids	μmol g^−1
	Anionic BS	Cationic BS
Aspartic acid	—	10.9
Glutamic acid	23	45
Serine	7.4	5.1
Histidine	—	0.9
Glycine	0.44	8.5
Threonine	4.23	2.6
Arginine	34.5	13.8
Alanine	3.75	6.5
Tyrosine	1.58	15.5
Methionine	0.14	1.6
Valine	1.64	3.1
Phenylalanine	—	2.9
Isoleucine	—	2.6
Leucine	11.3	4.5
Lysine	—	2.2
Cysteine	2.85	—
Glutamine	1.22	—
Asparagine	0.85	—

---

3.5. Molecular weight determination of biosurfactants

The molecular weight of the purified lipoprotein BS was confirmed by SDS-PAGE. The molecular weights of anionic and cationic BS were 18 kDa and 90 KDa, respectively (Fig. 4). The high molecular weight of the biosurfactants confirms that both the biosurfactants were giant molecular lipoproteins. BS from the coconut oil substrate was a high molecular weight lipoprotein compared to the other biosurfactants from other sources reported in the literature.^30,35 The reported research on Bacillus subtilis strain indicated that the production of low molecular weight biosurfactant³² and BS from Bacillus cereus strain was reported to produce plipastatins, a family of lipopeptides.³⁶

Fig. 4 SDS PAGE for the lipoprotein biosurfactant Lane 1: marker, Lane 2: anionic BS-18 kDa, Lane 3: cationic BS-90KDa.

3.6. Instrumental characterization of BS

3.6.1. Elemental analysis of BS. The elemental composition of cationic BS was 41.87% carbon, 8.47% hydrogen and 3.25% nitrogen and that of anionic BS was 20.25% carbon, 4.95% hydrogen and 2.12% nitrogen. The high nitrogen content imparts a certain degree of cationic nature to BS, as reported by Bognolo.³⁷

3.6.2. Thermal behaviour of BS (TGA-DSC). Thermal stability may be regarded as an important property of biosurfactants for their various applications. Both the cationic and anionic BS exhibited high thermal stability. TGA of anionic BS showed an initial weight loss of 1.51% at 39.51 °C due to moisture removal and the maximum weight loss of BS by 41.43% at 461.05 °C due to the decomposition of its constituents. At the end of the scan (800 °C), 55.3% of BS remained as a residue. DTG showed major weight loss of 0.296%/°C at 215.58 °C (Fig. 5a).

Fig. 5 TGA and DTG thermograms of (a) anionic BS, (b) cationic BS.

TGA of cationic BS showed an initial weight loss of 2.42% at 81.97 °C due to the removal of moisture. The maximum weight loss in BS of 52.25% was due to the decomposition of its constituents over the temperature range from 255.59 °C to 408.74 °C. At the end of the scan, 37.7% of BS remained as a fixed residue at 800 °C. DTG showed major weight loss of 0.4376%/°C at 353.59 °C (Fig. 5b). This suggests that cationic BS was thermally stable compared to anionic BS.

Differential Scanning Calorimetry was used to characterize the phase transition in lipoprotein over the temperature range of 30–300 °C. The transition, which is an exothermic process, was from an amorphous solid to a crystalline solid. The DSC thermogram of anionic BS showed a sharp crystallization temperature at 197 °C (Fig. 6a). The DSC thermogram of cationic BS showed transition temperatures at 107 °C and 197.70 °C (Fig. 6b).

Fig. 6 DSC spectra of (a) anionic BS, (b) cationic BS.

3.6.3. Circular dichorism (CD) of lipoprotein BS. A far UV CD spectra of BS showed the β-sheet form in a phosphate buffer solution (Fig. 7). The negative peak in the region of λ_200–225 nm indicates that the BS molecules were organized by β-sheet formation. The large number of carboxylic groups on the surface due to β-sheet organization may contribute to the special behaviors of BS such as the ease of surface β-sheet micelles formation and the ease of surface adsorption.¹⁴ These surface adsorption properties are important features for binding biosurfactant with metals ions.

Fig. 7 Circular dichroism spectra of the biosurfactant.

3.7. Interaction of trivalent chromium ions with biosurfactant

3.7.1. Effect of time. The biosurfactant for the sequestration of Cr(III) ions from an aqueous solution was confirmed by chromium(III) estimation with AAS. The amount of Cr(III) ions remained at equilibrium after sequestration with the biosurfactant (0.5 g) determined at different contact time, i.e. 2, 4, 6, 8, 10 and 12 h. The maximum percentage removal efficiency of chromium was 50% and 80% by the anionic BS and cationic BS, respectively, at a contact time of 10 h (Fig. 8a). The cationic biosurfactant showed higher sequestering efficiency for Cr(III) compared to that of the anionic biosurfactant.¹²

Fig. 8 Removal percentage of Cr(III) ions (a) time (pH: 7, anionic BS and cationic BS concentration: 0.5 g, Cr(III) concentration: 100 ppm), (b) pH (time: 10 h, anionic BS and cationic BS concentration: 0.5 g, Cr(III) concentration: 100 ppm), (c) different anionic BS and cationic BS concentration (time: 10 h, pH: 5, Cr(III) concentration: 100 ppm), (d) different Cr(III) ion concentration (time: 10 h, pH: 5, anionic BS and cationic BS concentration: 2.5 g).

3.7.2. Effect of pH. The effect of pH on the sequestration of Cr(III) ions by the biosurfactant was studied in the range of pH 4–6. The maximum removal of Cr(III) of 50% was observed at pH 6 with anionic BS and 82% with cationic BS at pH 5 (Fig. 8b). pH plays a very important role in the sequestration of trivalent chromium because of its influence on the protonation of different amino acids and allowing them to acquire charge to facilitate exchange for Cr(III) in a solution. BS was effective in sequestering Cr(III) in the pH range of 4–6.5. The sequestration efficiency of BS was reduced at pH above 6.5 due to the precipitation of Cr³⁺ as Cr(OH)₃.³⁸ The H⁺ concentration in the pH range of 4–6 would facilitate the cationic BS to sequester Cr(III) ions greater than the anionic BS.

3.7.3. Effect of biosurfactant concentration. The effect of biosurfactant concentration on the removal of Cr(III) ions was studied at different concentrations of BS, i.e., 0.5, 1.0, 1.5, 2.0, and 2.5 g L⁻¹. The sequestering efficiency of Cr(III) ions was increased with increasing biosurfactant concentration. The cationic BS at a concentration of 2.5 g L⁻¹ removed 98% of the Cr(III) ions, while anionic BS at the same concentration removed 85% of the chromium ions from the solution containing 100 mg L⁻¹ Cr(III) (Fig. 8c).

3.7.4. Effect of Cr(III) ions concentration. The effect of the concentration of Cr(III) ions on its removal by fixed biosurfactant concentration of 2.5 g was studied. While the percentage removal of Cr(III) using the anionic biosurfactant was 85%, 78%, 77%, 75%, and 75%, the cationic biosurfactant removed 98%, 97.5%, 95%, 93.7%, and 91.6% of the Cr(III) ion at chromium(III) concentrations of 100, 200, 300, 400 and 500 mg L⁻¹, respectively (Fig. 8d). Hence, the biosurfactant concentration of about 2.5 g could sequester 458 mg Cr(III) from an aqueous solution, while the removal of chromium by rhamnolipid biosurfactants were reported by Das et al.¹³ and Juwarkar et al.³⁹ The other potential Cr(III) biosorbents reported were Sargassum seaweed biomass (40 mg g⁻¹),⁴⁰ loofa sponge immobilized biomass of Chlorella sorokiniana (69.26 mg g⁻¹),⁴¹ pretreated bran rice (285.71 mg g⁻¹),⁴² Cassia fistula biomass (107.5 mg g⁻¹)⁴³ and sawdust Acacia Arabica (111.61 mg g⁻¹).⁴³

3.8. Instrumental evidence of the removal of Cr(III) ions using BS

3.8.1. UV-visible spectra of BS. The UV-Visible spectra (Fig. 9a) of cationic and anionic BS showed the characteristic peak at λ₂₉₀ nm. The delocalization of electrons from the non-bonding to π anti-bonding site (π*) caused an absorption at λ₂₉₀ nm. This confirms the presence of aromatic amino acids (phenylalanine) in cationic BS, as shown in Table 3. After interactions of cationic BS with Cr(III), i.e., the blue shift to 285 nm was observed with reduced intensity (Fig. 9b). The reduction in intensity of absorption may be due to the decrease in the delocalization of electrons owing to interaction of cationic BS with Cr(III). The anionic BS after the interaction with Cr(III) showed the characteristic peak at 289 nm.

Fig. 9 UV-visible spectra of (a) anionic BS and cationic BS, (b) anionic and cationic BS with Cr(III) ions.

3.8.2. Fluorescence spectral analysis. The fluorescence spectra (Fig. 10) of cationic and anionic BS showed that BS is a fluorescent inactive compound. After the interaction of cationic BS with Cr(III), the fluorescence spectra revealed the presence of excitation and emission peaks at λ₂₃₂ nm and λ₃₅₇ nm, respectively. In addition, the interaction of anionic BS with Cr(III) revealed the presence of excitation and emission peaks at λ₂₃₂ nm and λ₃₄₉ nm, respectively, which confirms the bonding of Cr(III) ion with BS.

Fig. 10 Fluorescence spectra of (a) cationic BS, (b) anionic BS, (c) cationic BS with Cr(III) and (d) anionic BS with Cr(III) ions.

3.8.3. FT-IR spectral analysis. The FT-IR spectrum of anionic BS (Fig. 11a) revealed the presence of asymmetric N–H stretching of amide at 3451.03 cm⁻¹. C=O stretching vibration of the peptide group was observed at 1651.96 cm⁻¹. The peak at 2924.23 cm⁻¹ resulting from the –C–H stretching of the methylene group reflects the presence of an aliphatic chain. The peak at 1466.10 cm⁻¹ was due to the methylene bending vibration. The absorption peak at 1750.11 cm⁻¹ may be due to the presence of carbonyl stretching in the ester groups. The shouldering at 725 cm⁻¹ may correspond to the C–N stretching vibration. This may be due to the bonding of the protein moiety with the lipid component in lipoprotein BS.²⁴ The peaks at 702 and 2511 cm⁻¹ may be due to the C–S and S–H stretching, respectively, of cysteine present in lipoprotein.

Fig. 11 FTIR spectra of (a) anionic BS, (b) anionic BS with Cr(III) ions, (c) cationic BS and (d) cationic BS with Cr(III) ions.

The FT-IR spectrum of cationic BS (Fig. 11c) showed N–H stretching of the peptide bond at 3297.26 cm⁻¹ and C=O stretching of the peptide group at 1658.41 cm⁻¹. The band at 2952.52 cm⁻¹, resulting from the C–H stretching of methylene group, reflects the presence of an aliphatic chain. The peak at 1465.90 cm⁻¹ was due to methylene bending. The absorption region at 1743.28 cm⁻¹ was due to the carbonyl stretching of ester group. The shouldering at 1107, 1156, 1121, and 1248.28 cm⁻¹ may correspond to the C–N stretching vibrations. The presence of a peptide bond with esters confirms that the biosurfactants were lipoproteins. The characteristic stretching frequency of amides in the regions from 3350 cm⁻¹ to 1500–1650 cm⁻¹ are not normally observed in the FT-IR spectra of rhamnolipid biosurfactants, which differentiates the unique nature of lipoprotein biosurfactant from rhamnolipid biosurfactant.³¹

Fig. 11b shows the FT-IR spectrum of BS-Cr(III) due to the binding of Cr(III) with the NH group and thus the N–H stretching band at 3451.03 cm⁻¹ was shifted to 3395 cm⁻¹. The shift in frequency to a lower value confirms that a stabilized bond formed between Cr(III) and the NH stretching of anionic BS. The characteristic peak of Cr(III) was observed at 620 cm⁻¹. The C–N stretching vibration was shifted from 725 cm⁻¹ to 713 cm⁻¹ after the ionic bonding of Cr(III) with anionic BS.

The FT-IR spectrum in Fig. 11d showed a broadening of the OH stretching and shifting of amide bond from 3297.26 cm⁻¹ to 3139 cm⁻¹ due to the strong binding of Cr(III). The characteristic peak of Cr(III) was observed at 620 cm⁻¹. The shift in the C=O stretching vibration of the peptide group from 1658.41 cm⁻¹ to 1651.41 cm⁻¹ was observed due to the strong binding of cationic biosurfactant with chromium(III) ions. In addition, the band at 1746 cm⁻¹ due to C–O stretching was shortened. This becomes an evidence for the binding of Cr(III) ions with the peptide group of the biosurfactant.

3.8.4. SEM analysis of biosurfactant bonded with chromium ions. The sequestration of Cr(III) ions from an aqueous solution was confirmed by scanning electron microscopy. The SEM images (Fig. 12a and b) showed the morphology of anionic and cationic BS, illustrating the assembly of lipoprotein molecules. The patches in the SEM images (Fig. 12c and d) confirmed the attachment of Cr(III) ions with biosurfactant molecules.

Fig. 12 SEM images of (a) anionic BS, (b) anionic BS with Cr(III) ions (c) cationic BS, and (d) cationic BS with Cr(III) ions.

3.8.5. EDX analysis of the biosurfactant bonded with chromium(III). The EDX analysis further confirmed the removal of Cr(III) ions from the aqueous solution. The EDX spectrum of anionic BS-Cr(III) showed 0.18 weight% Cr(III) (Fig. 13a) and the cationic BS-Cr(III) showed 0.44 weight% Cr(III) at 15 kV accelerating voltage with a 5–6nm electron beam (Fig. 13b). The results confirmed that the removal efficiency of cationic BS was comparatively higher than that of the anionic BS, indicating that cationic BS may be regarded as a more appropriate chelating agent than the anionic BS.

Fig. 13 EDX spectra of (a) anionic BS with Cr(III) ions and (b) cationic BS with Cr(III) ions.

3.9. Mechanistic view for removal of Cr(III) ions in an aqueous solution by anionic and cationic BS

The biosurfactant acquired cationic charge (+54.7 mV as evidenced from Fig. 3 relating zeta potential profile) due to the presence of a tertiary amine in histidine, resonance structure of phenylalanine and tyrosine in lipoprotein. The amino acids, such as histidine (pI 7.59), phenylalanine (pI 5.91) and tyrosine (pI 5.66), in cationic BS have isoelectric points much higher than the pH of BS (pH 5). At the optimized pH (pH 5), cationic BS removed Cr(III) by 98%. The positive charges acquired by the cationic BS were due to the overexpression of the abovementioned amino acids (15.63% as evidenced from the composition of amino acids, Table 3), which may explain the enhanced removal of Cr(III) from the aqueous solution. At the optimized pH (pH 5), cationic BS is protonated, as evidenced from FT-IR spectroscopy of histidine. The peak at 3082.6 cm⁻¹ could be attributed to the presence of symmetric N–H stretching and protonated NH₂⁺ group present in histidine.⁴⁴ The in-plane bending of C–H (ring) at 1121 cm⁻¹ and 1156 cm⁻¹ corresponds to the cationic form of phenylalanine.⁴⁵

The protonated amino acids (histidine, phenylalanine and tyrosine) at pH 5, which are lower than their isoelectric points release protons due to the delocalization of electrons caused by resonance (as evidenced from UV-Vis spectroscopy, Fig. 9a). The presence of O⁻ after deprotonation is confirmed from the stretching vibration at 1591.23 cm⁻¹.⁴⁶

The non-bonded electrons at nitrogen of histidine, tyrosine and phenyl alanine stabilize the bonding with Cr(III) through coordinate linkage (as evidenced from the shift in frequency, as recorded by FT-IR spectroscopy).

After the interaction of cationic BS with Cr(III) ions, the peaks such as 3443.35 cm⁻¹ and 3082.6 cm⁻¹ are broadened and also the O⁻ vibration peak disappeared. The in-plane C–H bending vibration corresponding to phenylalanine disappeared. This clearly suggests that Cr(III) binds with nitrogen of histidine, oxygen of tyrosine and nitrogen of phenyl alanine to form a co-ordinate bonding. The shift in the vibrational frequency, corresponding to N–H stretching of the peptide from 3297.26 to 3139 cm⁻¹ and the shift in the C=O stretching of the peptide group from 1658.41 cm⁻¹ to 1651.41 cm⁻¹, were observed after the interaction of Cr(III) with cationic BS. These vibrational shifts in the Cr(III) interacted cationic BS matrix is evidence for stabilization of bonding of Cr(III) with the respective amino acids. The schematic of the removal of Cr(III) by cationic BS is shown in Fig. 14.

Fig. 14 Schematic for the removal of Cr(III) by cationic BS in an aqueous solution.

The anionic charge (−33.01 mV as evidenced from the zeta potential profile, Fig. 3) of BS could be due to the presence of sulphur-containing amino acids, cysteine, in lipoprotein. Cysteine has an isoelectric point (pI 5.02) much lower than the pH of BS. At the optimized pH (pH 6), the anionic BS acquires a negative charge due to the ionization of cysteine alone (2.85% as evidenced from composition of amino acid in anionic BS, Table 3). Under the optimized conditions, anionic BS ionically bonded with Cr(III). Cr(III) has possible bonding only with cysteine of anionic BS, which is poorly expressed in anionic BS and thus accounting for the poor removal (85%) of Cr(III) by the anionic surfactant.

The charged centre of cysteine coordinated with Cr(III) as evidenced from the change in frequency, as recorded by FT-IR spectroscopy. The FT-IR spectrum of anionic BS showed a band at 1560 cm⁻¹ corresponding to the N H bend in cysteine.⁴⁷ In addition, a weak broadened band at 2511 cm⁻¹ may be due to the S–H stretching in cysteine molecule. After the interaction of anionic BS with Cr(III), the shift in the N–H bending vibration to 1552 cm⁻¹ was observed. In addition, the disappearance of weak S–H stretching band clearly indicates that Cr(III) binds with the sulphur containing amino acid through ionic bonding. The vibrational shift in the peak corresponding to N–H stretching of the peptide group from 3451 to 3395 cm⁻¹ was evident. The shift in C–N stretching vibration from 725 cm⁻¹ to 713 cm⁻¹ after ionic bonding of anionic BS with Cr(III) was observed. These vibrational shifts suggest that alteration in the peptide linkage as a consequence of the interaction of Cr(III) with anionic BS.

The anionic BS containing cysteine (Table 3) alone is responsible for ionic bonding with the sulphur group present in BS. Therefore, Cr(III) was removed only by 85% with anionic BS, whereas cationic BS removed Cr(III) by 98% at the same concentration (2.5 g). The mechanism for the removal of Cr(III) by anionic BS is illustrated in Fig. 15.

Fig. 15 Schematic for the removal of Cr(III) by anionic BS in an aqueous solution.

It is evident from the EDX spectra that anionic BS-Cr(III) contained Cr(III) by 0.18% of ions (Fig. 13a) and the cationic BS-Cr(III) contained Cr(III) by 0.44% (Fig. 13b). The fluorescence spectra also confirmed the removal of Cr(III) by cationic and anionic BS (Fig. 10).

4. Conclusions

In the present investigation, the production of anionic and cationic BS using palm oil and coconut oil as substrates, respectively, and their ability to remove Cr(III) ions from aqueous solutions were studied. The agar double diffusion technique and zeta potential measurements of biosurfactants confirmed the charge carried by the surfactants. The composition of the biosurfactants confirmed that they belong to the lipoprotein type. The biosequestration of Cr(III) ions by biosurfactants was confirmed by FT-IR analysis through their shifts in the peptide group of BS. The cationic biosurfactant showed that the maximum removal of Cr(III) ions from an aqueous solution was 98% at a biosurfactant concentration of 2.5 g through coordinate bonding, whereas the anionic biosurfactant removed 85% Cr(III) at the same concentration of biosurfactant through ionic bonding. To the best of our knowledge, this is the first report of interaction of cationic BS with Cr(III) ions in an aqueous solution.

Acknowledgements

P. Saranya is thankful to the Council of Scientific and Industrial Research (CSIR), India. The financial assistance under the STRAIT (CSC0201) programme is also highly acknowledged.

References

1. J. H. Sharphouse, Leather Technician’s Handbook, Leather Producers Association, Buckland Press Ltd, London, 1989.
2. B. Basaran, M. Ulaş, O. Behzat and A. Band Aslan, Indian J. Chem. Technol., 2008, 15, 511–514.
3. Environment Commission of I.U.L.T.C.S., IUE Recommendations on Cleaner Technologies for Leather Production, London, 1997.
4. Central Pollution Control Board, The Environment (Protection) Rules, India, 1986, p. 418.
5. Z. X. Wang, J. Q. Chen, L. Y. Chai, Z. H. Yang, S. H. Huang and Y. Zheng, J. Hazard. Mater., 2011, 190, 980–985.
6. L. Di Palma, D. Mancini and E. Petrucci, Chem. Eng. Trans., 2012, 28, 145–150.
7. D. C. Sharma, C. Chatterjee and C. P. Sharma, J. Exp. Bot., 1995, 25, 241–251.
8. H. M. Abdulla, E. M. Kamal, A. H. Mohammed, and A. D. EI Bassuony, Proc. fifth Scien. Environ conf. Zagazig uni, 2010, pp. 171–183.
9. J. Kotas and Z. Stasicka, Environ. Pollut., 2000, 107, 263–283.
10. Ecological Analysts, Inc, The sources, chemistry, fate and effects of Chromium in aquatic environments, American Petroleum Institute, 2101 L St., N.W., Washington, DC 20037, 1981, p. 207.
11. M. M. Altaf, F. Masood and A. Malik, Turk. J. Biol., 2008, 32, 1–8.
12. M. P. Plkociniczak, G. A. Płaza, Z. P. Seget and S. S. Cameotra, Int. J. Mol. Sci., 2011, 12, 633–654.
13. P. Das, S. Mukherjee and R. Sen, Bioresour. Technol., 2009, 100, 4887–4890.
14. G. Dehghan Noudeh, M. Housaindokht and B. S. Bazzaz, J. Microbiol. Methods, 2005, 43, 272–276.
15. A. Franzetti, I. Gandolfi, G. Bestetti and I. M. Banat, Trends Biorem. Phytorem., 2010, 145–156.
16. J. Thaniyavarn, A. Chongchin, N. Wanitsuksombut, S. Thaniyavarn, P. Pinphanichakarn, N. Leepipatpiboon, M. Morikawa and S. Kanaya, J. Gen. Appl. Microbiol., 2006, 52, 215–222.
17. V. Walter, C. Syldatk and R. Hausmann, Bioscience and Springer Science, Business Media, 2010.
18. R. D. Rufino, J. M. Luna, G. M. Campos-Takaki, S. R. M. Ferreira and L. A. Sarubbo, Chem. Eng. Trans., 2012, 27, 61–66.
19. T. Meylheuc, C. J. Vanoss and M. M. Bellon-Fontaine, J. Appl. Microbiol., 2001, 91, 832.
20. D. G. Cooper and B. G. Goldenberg, Appl. Environ. Microbiol., 1987, 53, 224–229.
21. T. Varadavenkatesan and V. R. Murty, J. Microbiol. Biotechnol. Res., 2013, 3(4), 63–73.
22. M. Dubois, K. A. Gills, J. K. Hamilton, P. A. Rebers and F. Smith, Anal. Chem., 1956, 28, 350–356.
23. O. H. Lowry, N. J. Rosebrough, A. L. Farr and R. J. Randall, J. Biol. Chem., 1951, 193, 265–275.
24. J. Izard and R. J. Limberger, J. Microbiol. Methods, 2003, 55, 411–418.
25. S. Ramakrishnan, K. N. Sulochana, R. Punitham and K. Arunagiri, Glycoconjugate J., 1996, 13, 519–523.
26. U. K. Laemmli, Nature, 1970, 227, 680–685.
27. W. M. F. W. Nawawi, Optimum Biosurfactant production from sludge palm oil using Klebsiella pneumoniae, International Islamic University Malaysia, Kuala Lumpur, Malaysia, 2011.
28. S. Patil, A. Pendse and K. Aruna, Int. J. Curr. Biotechnol., 2014, 2, 20–30.
29. M. Kannahi and M. Sherley, Int. J. Chem. Pharm. Sci., 2012, 3, 37–42.
30. S. K. Satpute, B. D. Bhawsar, P. K. Dhakephalkar and B. A. Chopade, Indian J. Mar. Sci., 2008, 37, 243–250.
31. K. Ramani, S. Chandan Jain, A. B. Mandal and G. Sekaran, Colloids Surf., B, 2012, 97, 254–263.
32. S. S. Bhadoriya, N. Madoriya, K. Shukla and M. S. Parihar, Biochem. Pharmacol., 2013, 2, 113.
33. S. H. Kim, E. J. Lim, S. O. Lee, J. D. Lee and T. H. Lee, Biotechnol. Appl. Biochem., 2000, 31, 249–253.
34. T. T. Nguyen, N. H. Youssef, M. J. McInerney and D. A. Sabatini, Water Res., 2008, 42, 1735–1743.
35. P. Saranya, S. Swarnalatha and G. Sekaran, RSC Adv., 2014, 4, 34144–34155.
36. C. N. Mulligan and B. F. Gibbs, Proc. Natl. Acad. Sci. U. S. A., 2004, 1, 31–55.
37. G. Bognolo, Colloids Surf., A, 1999, 152, 41–52.
38. E. Z. Ron and E. Rosenberg, Appl. Microbiol. Biotechnol., 1999, 52, 154–162.
39. A. A. Juwarkar, K. V. Dubey, A. Nair and S. Kumar Singh, Indian J. Microbiol., 2008, 32, 142–146.
40. D. Kratochvil, P. Pimentel and B. Volesky, Environ. Sci. Technol., 1998, 32, 2693–2698.
41. A. Nasreen, I. Muhammad, Z. Saeed Iqbal and I. Javed, J. Environ. Sci., 2008, 20, 231–239.
42. K. K. Singh, R. Rastogi and S. H. Hasan, J. Colloid Interface Sci., 2005, 290, 61–68.
43. P. Miretzky and A. Fernandez Cirelli, J. Hazard. Mater., 2010, 180, 1–19.
44. K. Rajarajan, K. Anbarasan, J. Samu Solomon and G. Madhurambal, J. Chem. Pharm. Res., 2012, 4, 4060–4065.
45. S. Olsztynska, M. Komorowska, L. Vrielynck and N. Dupuy, Appl. Spectrosc., 2001, 55, 901–907.
46. A. Barth, Prog. Biophys. Mol. Biol., 2000, 74, 141–173.
47. S. Aryal, B. K. C. Remant, N. Dharmaraj, N. Bhattarai, C. H. Kim and H. Y. Kim, Spectrochim. Acta, Part A, 2006, 63, 160–163.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra07999c

---