Biosorption of reactive and basic dyes using fermentation waste <Emphasis Type="Italic">Corynebacterium glutamicum</Emphasis>: the effects of pH and salt concentration and characterization of the binding sites

A low cost biosorbent, Corynebacterium glutamicum, was studied for the sorption of Reactive Red 4 (RR 4) and Methylene Blue (MB). The equilibrium isotherm data were well described by the Langmuir model. pH edge experiments showed that pH of the solution was an important controlling parameter in the sorption process. In the case of RR 4, with increases in the pH from 2 to 10, the uptake decreased from 52 to 1 mg/g; conversely, the uptake of MB increased and the maximum MB uptake was obtained at pH ≥ 9. An increase in the salt concentration strongly influenced the uptake of MB, but had no effect on that of RR 4. In order to identify the binding sites for the dye molecules, the biosorbent was potentiometrically titrated, the results of which showed the presents of four major functional group types on the biomass surface, which were confirmed by FTIR analysis. It was found that positively charged amine groups (Biomass-NH₃ ⁺) were the likely binding sites for anionic RR 4, and negatively charged carboxyl (Biomass-COO⁻) and phosphate groups (Biomass-HPO₄ ⁻) played a role in the electrostatic attraction of cationic MB.     

Biosorption of thorium (IV) by a Pseudomonas biomass

Lyophilized biomass of a Pseudomonas soilisolate adsorbed thorium (IV) (430 mg g^–1 dry wt) optimally at pH 4, with 91% of equilibrium loading being reached in 1 min. Equilibrium metal sorption showing conformity to Langmuir isotherm model suggested a monolayered thorium binding. Thorium binding remained unaffected or slightly affected (< 20% inhibition) in presence of equimolar (430 M) concentration of several interfering ions except Fe³⁺ (40% inhibition). More than 90% of loaded thorium could be recovered using 1 M CaCO₃, though mineral acids and Na₂CO₃ were also effective.     

Immobilization of blue green microalgae on loofa sponge to biosorb cadmium in repeated shake flask batch and continuous flow fixed bed column reactor system

Summary An indigenous strain of blue green microalga, Synechococcus sp., isolated from wastewater, was immobilized onto loofa sponge discs and investigated as a potential biosorbent for the removal of cadmium from aqueous solutions. Immobilization has enhanced the sorption of cadmium and an increase of biosorption (21%) at equilibrium was noted as compared to free biomass. The kinetics of cadmium biosorption was extremely rapid, with (96%) of adsorption within the first 5 min and equilibrium reached at 15 min. Increasing initial pH or initial cadmium concentration resulted in an increase in cadmium uptake. The maximum biosorption capacity of free and loofa immobilized biomass of Synechococcus sp. was found to be 47.73 and 57.76 mg g⁻¹ biomass respectively. The biosorption equilibrium was well described by Langmuir adsorption isotherm model. The biosorbed cadmium was desorbed by washing the immobilized biomass with dilute HCl (0.1 M) and desorbed biomass was reused in five biosorption–desorption cycles without an apparent decrease in its metal biosorption capacity. The metal removing capacity of loofa immobilized biomass was also tested in a continuous flow fixed-bed column bioreactor and was found to be highly effective in removing cadmium from aqueous solution. The results suggested that the loofa sponge-immobilized biomass of Synechococcus sp. could be used as a biosorbent for an efficient removal of heavy metal ions from aqueous solution.     

Characterisation of acidic sites of Pseudomonas biomass capable of binding protons and cadmium and removal of cadmium via biosorption

Summary The ability of Pseudomonas aeruginosa to accumulate Cd(II) ions from wastewater industries was experimentally investigated and mathematically modelled. From the potentiometric titration and non-ideal competitive analysis (NICA) model, it was found that the biomass contains three acidic sites. The values of proton binding (pK _i =1.66±3.26×10⁻³, 1.92±1.63×10⁻⁴ and 2.16±3.79×10⁻⁴) and binding constant of cadmium metal ions (pK _M1=1.99±2.45×10⁻³ and pK _M2=1.67±4.08×10⁻³) on the whole surface of biomass showed that protonated functional groups and biosorption of Cd(II) ions could be attributed to a monodentate binding to one acidic site, mainly the carboxylic group. From the isothermal sorption experimental data and Langmuir model, it was also found that the value of Langmuir equilibrium (pK _f) constant is 2.04±2.1×10⁻⁵ suggesting that the carboxyl group is the main active binding site. In addition, results showed that the maximum cadmium capacity (q _max) and affinity of biomass towards cadmium metal ions (b) at pH 5.1 and 20 min were 96.5±0.06 mg/g and 3.40×10⁻³± 2.10×10⁻³, respectively. Finally, interfering metal ions such as Pb(II), Cu(II), Cr(III), Zn(II), Fe(II), Mn(II), Ca(II) and Mg(II) inhibited Cd(II) uptake. Comparing the biosorption of Cd(II) by various Pseudomonas isolates from contaminated environment samples (soil and sewage treatment plant) showed that maximum capacities and equilibrium times were different, indicating that there was a discrepancy in the chemical composition between biomasses of different strains.     

Biosorption of Cd(II) and Pb(II) onto brown seaweed, Lobophora variegata (Lamouroux): kinetic and equilibrium studies

The present work deals with the biosorption performance of raw and chemically modified biomass of the brown seaweed Lobophora variegata for removal of Cd(II) and Pb(II) from aqueous solution. The biosorption capacity was significantly altered by pH of the solution delineating that the higher the pH, the higher the Cd(II) and Pb(II) removal. Kinetic and isotherm experiments were carried out at the optimal pH 5.0. The metal removal rates were conspicuously rapid wherein 90% of the total sorption occurred within 90 min. Biomass treated with CaCl₂ demonstrated the highest potential for the sorption of the metal ions with the maximum uptake capacities i.e. 1.71 and 1.79 mmol g⁻¹ for Cd(II) and Pb(II), respectively. Kinetic data were satisfactorily manifested by a pseudo-second order chemical sorption process. The process mechanism consisting of both surface adsorption and pore diffusion was found to be complex. The sorption data have been analyzed and fitted to sorption isotherm of the Freundlich, Langmuir, and Redlich–Peterson models. The regression coefficient for both Langmuir and Redlich–Peterson isotherms were higher than those secured for Freundlich isotherm implying that the biosorption system is possibly monolayer coverage of the L. variegata surface by the cadmium and lead ions. FT-IR studies revealed that Cd(II) and Pb(II) binding to L. variegata occurred primarily through biomass carboxyl groups accompanied by momentous interactions of the biomass amino and amide groups. In this study, we have observed that L. variegata had maximum biosorption capacity for Cd(II) and Pb(II) reported so far for any marine algae. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Screening of waste biomass from Saccharomyces cerevisiae,Aspergillus oryzae and Bacillus lentus fermentations for removal of Cu,Zn and Cd by biosorption

Three by-products of fermentations containing Bacillus lentus, Aspergillus oryzae or Saccharomyces cerevisiae biomass were tested for the capacity to absorb Cu, Cd and Zn. The composition of the three biomasses was first determined and showed high contents of ashes in both B. lentus and A. oryzae biomass and high amounts of lipids in the bacterial biomass. Metal ion binding experiments were performed by contact of 0.1 g of biomass (protonated for all the metal tests and not protonated only for the Cd test) with 50 ml of solutions containing each of the metals in the concentration range from 10 to 500 mg/ml, at pH 4.5, 3.5 and 2.5. The final metal ion concentrations were determined using a plasma absorption spectrometer, and the metal removal levels for isotherm plots were determined using the Langmuir model. The results showed that B. lentus protonated biomass had the best sorption capacity for Cu and Cd, followed by protonated A. oryzae and S. cerevisiae biomass. The sorption of Zn was low for all tested biomasses, as also was the binding of all metals at acidic pH (2.5 and 3.5). A significant increase in Cd sorption was obtained using non-protonated biomass from B. lentus and A. oryzae.     

Biosorption of Cr(VI) onto marine Aspergillus niger: experimental studies and pseudo-second order kinetics

The removal of hexavalent chromium from aqueous solution was studied in batch experiments using dead biomass of three different species of marine Aspergillus after alkali treatment. All the cultures exhibited potential to remove Cr(VI), out of which, Aspergillus niger was found to be the most promising one. This culture was further studied employing variation in pH, temperature, metal ion concentration and biomass concentration with a view to understand the effect of these parameters on biosorption of Cr(VI). Higher biosorption percentage was evidenced at lower initial concentration of Cr(VI) ion, while the sorption capacity of the biomass increased with rising concentration of ions. Biomass as low as 0.8 g l⁻¹ could biosorb 95% Cr(VI) ions within 2,880 min from an aqueous solution of 400 mg l⁻¹ Cr(VI) concentration. Optimum pH and temperature for Cr(VI) biosorption were 2.0 and 50°C, respectively. Kinetic studies based on pseudo second order models like Sobkowsk and Czerwinski, Ritchie, Blanchard and Ho and Mckay rate expressions have also been carried out. The nature of the possible cell–metal ion interactions was evaluated by FTIR, SEM and EDAX analysis.     

Enhanced sorption of Cu2+ and Ni2+ by acid-pretreated Chlorella vulgaris from single and binary metal solutions

The influence of HCl pretreatment (0.1 mM) on sorption ofCu²⁺ and Ni²⁺ by Chlorella vulgariswas tested using single and binary metal solutions. The optimal initial pH forsorption was 3.5 for Cu²⁺ and 5.5 for Ni²⁺. Second orderrate kinetics described well sorption by untreated and acid-pretreated cells.The kinetic constant q_e (metal sorption at equilibrium) for sorptionof test metals from single and binary metal solutions was increased afterpretreatment of the biomass with HCl. The Langmuir adsorption isotherm wasdeveloped for describing the various results for metal sorption. In single metalsolution, acid pretreatment enhanced q_max for Cu²⁺ andNi²⁺ sorption by approximately 70% and 65%, respectively.Cu²⁺ and Ni²⁺ mutually interfered with sorption of theother metal in the binary system. The combined presence of Cu²⁺ andNi²⁺ led to their decreased sorption by untreated biomass by 19% and88%, respectively. However, acid-pretreated biomass decreased Cu²⁺and Ni²⁺ sorption by 15 and 22%, respectively, when both the metalswere present in the solution. The results suggest a reduced mutual interferencein sorption of Cu²⁺ and Ni²⁺ from the binary metal systemdue to the acid pretreatment. Acid-pretreated cells sorbed twice the amount ofCu²⁺ and ten times that of Ni²⁺ than the untreated biomassfrom the binary metal system. Acid pretreatment more effectively enhanced thesorption of Ni²⁺ form the binary metal solution. The total metalsorption by untreated and acid-pretreated biomass depended on theCu²⁺ : Ni²⁺ ratio in the binary metal system. Acidpretreatment of C. vulgaris could be an effective andinexpensive strategy for enhancing Cu²⁺ and Ni²⁺ sorptionfrom single and binary metal solutions.     

Sorption of Cu(II) and Cd(II) by extracellular polymeric substances (EPS) from Aspergillus fumigatus

Sorption of Cu(II) and Cd(II) onto the extracellular polymeric substances (EPS) produced by Aspergillus fumigatus was investigated for the initial pH of the solution, EPS concentrations, contact time, NaCl concentration, initial metal ion concentration and the presence of other ions in the solution. The results showed that the adsorption of metal ions was significantly affected by pH, EPS concentrations, initial metal concentration, NaCl concentration and co-ions. The sorption of Cu(II) and Cd(II) increased with increasing pH and initial metal ion concentration but decreased with an increase in the NaCl concentration. The maximum sorption capacities of A. fumigatus EPS calculated from the Langmuir model were 40 mg g⁻¹ EPS and 85.5 mg g⁻¹ EPS for Cu(II) and Cd(II), respectively. The binary metal sorption experiments showed a selective metal binding affinity in the order of Cu(II) > Pb(II) > Cd(II). Both the Freundlich and Langmuir adsorption models described the sorption of Cu(II) and Cd(II) by the EPS of A. fumigatus adequately. Fourier transform infrared spectroscopy (FTIR) analysis revealed that carboxyl, amide and hydroxyl functional groups were mainly correlated with the sorption of Cu(II) and Cd(II). Energy dispersive X-ray (EDX) system analysis revealed that the ion-exchange was an important mechanism involved in the Cu(II) and Cd(II) sorption process taking place on EPS.     

Characterization of natural variation for zinc, iron and manganese accumulation and zinc exposure response in Brassica rapa L.

Brassica rapa L. is an important vegetable crop in eastern Asia. The objective of this study was to investigate the genetic variation in leaf Zn, Fe and Mn accumulation, Zn toxicity tolerance and Zn efficiency in B. rapa. In total 188 accessions were screened for their Zn-related characteristics in hydroponic culture. In experiment 1, mineral assays on 111 accessions grown under sufficient Zn supply (2 μM ZnSO₄) revealed a variation range of 23.2–155.9 μg g⁻¹ dry weight (d. wt.) for Zn, 60.3–350.1 μg g⁻¹ d. wt. for Fe and 20.9–53.3 μg g⁻¹ d. wt. for the Mn concentration in shoot. The investigation of tolerance to excessive Zn (800 μM ZnSO₄) on 158 accessions, by using visual toxicity symptom parameters (TSPs), identified different levels of tolerance in B. rapa. In experiment 2, a selected sub-set of accessions from experiment 1 was characterized in more detail for their mineral accumulation and tolerance to excessive Zn supply (100 μM and 300 μM ZnSO₄). In this experiment Zn tolerance (ZT) determined by relative root or shoot dry biomass varied about 2-fold. The same six accessions were also examined for Zn efficiency, determined as relative growth under 0 μM ZnSO₄ compared to 2 μM ZnSO₄. Zn efficiency varied 1.8-fold based on shoot dry biomass and 2.6-fold variation based on root dry biomass. Zn accumulation was strongly correlated with Mn and Fe accumulation both under sufficient and deficient Zn supply. In conclusion, there is substantial variation for Zn accumulation, Zn toxicity tolerance and Zn efficiency in Brassica rapa L., which would allow selective breeding for these traits.     

Interactive effects of elevated CO2 and drought stress on leaf water potential and growth in Caragana intermedia

We studied the responses of leaf water potential (Ψ_w), morphology, biomass accumulation and allocation, and canopy productivity index (CPI) to the combined effects of elevated CO₂ and drought stress in Caragana intermedia seedlings. Seedlings were grown at two CO₂ concentrations (350 and 700 μmol mol⁻¹) interacted with three water regimes (60–70%, 45–55%, and 30–40% of field capacity of soil). Elevated CO₂ significantly increased Ψ_w, decreased specific leaf area (SLA) and leaf area ratio (LAR) of drought-stressed seedlings, and increased tree height, basal diameter, shoot biomass, root biomass as well as total biomass under the all the three water regimes. Growth responses to elevated CO₂ were greater in well-watered seedlings than in drought-stressed seedlings. CPI was significantly increased by elevated CO₂, and the increase in CPI became stronger as the level of drought stress increased. There were significant interactions between elevated CO₂ and drought stress on leaf water potential, basal diameter, leaf area, and biomass accumulation. Our results suggest that elevated CO₂ may enhance drought avoidance and improved water relations, thus weakening the effect of drought stress on growth of C. intermedia seedings.     

Assessment of the metal removal capability of two capsulated cyanobacteria, Cyanospira capsulata and Nostoc PCC7936

Two capsulated, exopolysaccharide-producing cyanobacteria, Cyanospira capsulata and Nostoc PCC7936, were tested with regard to their metal removal capability by using copper as model metal. The experiments, carried out with the sole cyanobacterial biomass suspended in distilled water and confined into small dialysis tubings, showed that C. capsulata biomass is characterized by the best efficiency in metal removal, with a q_max (maximum amount of copper removed per biomass unit) of 96 ± 2 mg Cu(II) removed per g of protein in comparison with the value of 79 ± 3 of Nostoc PCC7936 biomass. The experimental data obtained with both cyanobacterial biomass best fit the Langmuir sorption isotherm. The sorption of copper started from the first minutes of contact with the metal and attained the equilibrium state, when no more copper removal was evident, after 5 and 6 hours, for C. capsulata and Nostoc PCC7936, respectively. The best efficiency in Cu(II) removal was obtained at pH 6.1–6.2, while the presence of Mg²⁺ or Ca²⁺ reduced copper removal capability of both species to 60–70% of their q_max. The results showed that the biomass of C. capsulata and Nostoc PCC7936 possesses a high affinity and a high specific uptake for copper, comparable with the best performances shown by other microbial biomass, and suggest the possibility to use the capsulated trichomes of the two cyanobacteria for the bioremoval of heavy metals from polluted water bodies.     

Use of Algae for Removing Heavy Metal Ions From Wastewater: Progress and Prospects

ABSTRACT

Many algae have immense capability to sorb metals, and there is considerable potential for using them to treat wastewaters. Metal sorption involves binding on the cell surface and to intracellular ligands. The adsorbed metal is several times greater than intracellular metal. Carboxyl group is most important for metal binding. Concentration of metal and biomass in solution, pH, temperature, cations, anions and metabolic stage of the organism affect metal sorption. Algae can effectively remove metals from multi-metal solutions. Dead cells sorb more metal than live cells. Various pretreatments enhance metal sorption capacity of algae. CaCl₂ pretreatment is the most suitable and economic method for activation of algal biomass. Algal periphyton has great potential for removing metals from wastewaters. An immobilized or granulated biomass-filled column can be used for several sorption/desorption cycles with unaltered or slightly decreased metal removal. Langmuir and Freundlich models, commonly used for fitting sorption data, cannot precisely describe metal sorption since they ignore the effect of pH, biomass concentration, etc. For commercial application of algal technology for metal removal from wastewaters, emphasis should be given to: (i) selection of strains with high metal sorption capacity, (ii) adequate understanding of sorption mechanisms, (iii) development of low-cost methods for cell immobilization, (iv) development of better models for predicting metal sorption, (v) genetic manipulation of algae for increased number of surface groups or over expression of metal binding proteins, and (vi) economic feasibility.     

Microbial Waste Biomass for Removal of Chromium(VI) from Chrome Effluent

ABSTRACT

Microbial waste biomass, a by-product of the fermentation industry, was developed as a biosorbent to remove hexavalent chromium (Cr) from the acidic effluent of a metal processing industry. In batch sorption, 100% Cr(VI) removal was achieved from aqueous solution in 30 min contact at pH 4.0–5.0. The Cr(VI) sorption equilibrium was evaluated using the Langmuir and Freundlich models, indicating the involvement of ion exchange and physicochemical interaction. Fourier transform infrared (FTIR) analysis revealed the presence of amine, hydroxyl, and imine functional groups present on the surface of microbial biomass that are involved in Cr binding. In a continuous sorption system, 95 mg L^?1 of Cr(VI) was adsorbed before the column reached a breakthrough point of 0.1 mg L^?1 Cr(VI) at the column outlet. An overall biosorption capacity of 12.6 mg Cr(VI) g^?1 of dry microbial waste was achieved, including the partially saturated portion of the dynamic sorption zone. Insignificant change in metal removal was observed up to 10 cycles. In pilot-scale studies, 100% removal of Cr(VI) was observed up to 5 weeks, and the method was found to be cost-effective, commercially viable, and environmentally friendly, as it does not generate toxic chrome sludge.     

Removal of Heavy Metals by Waste Tea Leaves from Aqueous Solution

In this paper, tea leaves were shown to be an effective, low‐cost biosorbent. Removal of lead, iron, zinc and nickel from 20 mg/L metal solution by dried biomass of waste tea leaves amounted to 96, 91, 72 and 58 %, respectively, at equilibrium, which followed Langmuir and Freundlich adsorption isotherms. Adsorption of metal was in the order of Pb > Fe > Zn > Ni from 5–100 mg/L of metal solution. From a multi‐metallic mixture, 92.5, 84 and 73.2 % of lead, iron and zinc, respectively, were removed. Fourier transform infrared (FTIR) studies indicated that the carboxyl group was involved in the binding of lead and iron, whereas the amine group was involved in the binding of nickel and zinc. A flow through sorption column packed with dried biomass demonstrated a sorption capacity of 73 mg Pb/g of biomass, indicating its potential in cleaning metal containing wastewater. The metal laden biomass obtained could be disposed off by incineration.     

Effects of leached alginate on metal biosorption

The Cu, Cd and Zn sorption capacity of formaldehyde-crosslinked seaweed biomass (Ascophyllum nodosum : FCAN) was studied using equilibrium methodology. The shape of sorption isotherms and biosorbent mass losses indicated that in the first uptake cycle the biosorbent leached cellular polysaccharides that formed precipitate with metal solutions influencing the sorption study and application. However, since the sorption levels of washed and non-washed biosorbents were comparable, the leached cell wall polysaccharide(s) (alginate) could not be the main component responsible for metal sorption in this case.on leave from Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil     

Modeling the Effect of Glucose Syrup on the Moisture Sorption Isotherm of Figs

Moisture sorption isotherms of figs with and without glucose syrup (at 20% and 40%, w/w) were determined at 5 °C, 25 °C, and 40 °C. A static gravimetric method was used under 0.11–0.84 water activity ranges for the determination of sorption isotherms that were found to be typical type ΙΙΙ for control sample. The inclusion of glucose syrup had significant effects on the sorption isotherms, and the moisture content of samples at each a _w decreased with increasing temperature. The experimental data were fitted well with two-parameter Brunauer–Emmet–Teller, three-parameter Guggenheim–Anderson–de Boer, and four-parameter Peleg models that all had R ² of greater than 0.99. The net isosteric heats of sorption were estimated using the Clausius–Clapeyron equation from the equilibrium data at different temperatures. It was found that the addition of glucose syrup significantly increased the amount of monolayer water and the isosteric heat of sorption. Both water activity and isosteric heat of sorption increased with glucose syrup level and the shape and status of sorption isotherms tend to change toward the typical sigmoid shape of most food systems.     

Adsorption of several actinide (Th,U) and lanthanide (La,Eu, Yb) ions by Mycobacterium smegmatis

Adsorption measurements of several actinide [thorium (Th), uranium (U)] and lanthanide [lanthanum (La), europium (Eu), ytterbium (Yb)] cations by Mycobacterium smegmatis showed that sorption kinetics followed a three-phase pattern. For 5% (w/w) bacterial suspensions at pH 1, maximum cation biosorption per gram dry biomass corresponded to 170 mol Th⁴⁺ and 187 mol UO _inf2 ^sup2+ . Adsorption of all cations studied obeyed the Brunauer-Emmett-Teller isotherm, which assumes multilayer binding at constant energy. Plots for the Scatchard model showed the existence of at least two types of cation complexation site, with strong and weak affinity and negative cooperation. Th⁴⁺ was preferentially adsorbed with respect to the other cations, although all species appeared to compete for the same sites independently of bacterial viability. Adsorption of these cations was accompanied by partial release of magnesium from the cell wall, indicating that exchange reactions occurred at magnesium (Mg)-bonding sites.     

Biosorption of cadmium by a Cd2+-hyperresistant Bacillus cereus strain HQ-1 newly isolated from a lead and zinc mine

A Cd²⁺-hyperresistant bacterial strain HQ-1 was isolated from a lead–zinc mine. The strain was characterized and identified as Bacillus cereus based on morphology, physiological tests and 16S rRNA gene analysis. The minimal inhibitory concentration of Cd²⁺ for the bacterium was 0.012 mol/l. Isotherms for cadmium (Cd) biosorption by cells of B. cereus strain HQ-1 were investigated. The equilibrium data could be fitted by a Langmuir isotherm equation. The possible functional sites that might be influenced by the sorption were determined. The results indicate that this B. cereus strain has excellent potential for biosorption of Cd. Physiological characterization of the isolate also indicates possible application of this strain for bioremediation of sites with Cd contamination.     

News & Notes: Sorption and Desorption of Cobalt by Oscillatoria anguistissima

Oscillatoria anguistissima rapidly adsorbs appreciable amounts of cobalt from the aqueous solutions within 15 min of initial contact with the metal solution. O. anguistissima showed a high sequestration of cobalt at low equilibrium concentrations, and it followed the Freundlich model of adsorption. The adsorption is a strongly pH-dependent and temperature-independent phenomenon. The presence of Mg²⁺ and Ca²⁺ (100–200 ppm) resulted in decline in Co²⁺ adsorption capacity of Oscillatoria biomass. Sulphate and nitrate (0.75–10 mM) drastically reduced the extent of Co²⁺ biosorption. The biosorption of cobalt is an ion-exchange process as the Co²⁺ binding was accompanied by release of a large amounts of Mg²⁺ ions. Na₂CO₃ (1.0 mM) resulted in about 76% desorption of Co²⁺ from the loaded biomass. Received: 30 January 1999 / Accepted: 3 March 1999